{"arv":"1.0","success":true,"request_id":"req_fbcd89ed-055e-4a1f-a196-ff4fc53faeaf","schema_version":"arv@1.0","data":{"asset":{"id":"video_reLhJnQd-Lo","content_hash":"sha256:fc97fb62abf9a9ef4f7a403cf675a2276262384db0ea192496339d4eed4678e0","owner_id":"sandersaar","hosting_mode":"mode_a","source":{"kind":"youtube_oauth","source_url":"https://www.youtube.com/watch?v=reLhJnQd-Lo","ingested_at":"2026-08-26T17:34:01.398Z"},"origin_playback":{"platform":"youtube","external_id":"reLhJnQd-Lo","watch_url":"https://www.youtube.com/watch?v=reLhJnQd-Lo","media_fragments_template":"https://www.youtube.com/watch?v=reLhJnQd-Lo&t={start_seconds}s"},"duration_ms":1982000,"language_primary":"en","c2pa":{"has_ai_disclosure":false},"schema_version":"arv@1.0","ingested_via_pipeline_version":"cascade-mm-quality-v1-17"},"creator":{"id":"sandersaar","handle":"sandersaar","namespace_url":"https://agentcdn.com/creators/sandersaar","verified":true,"verified_via":["youtube_oauth"],"trust_signals_url":"https://api.agentcdn.com/public/trust/sandersaar"},"display":{"title":"Generative AI Explained: how your life & work will change","description":"This video examines the rise of generative artificial intelligence and how it transforms human labor, interfaces, and content creation. It covers key drivers such as falling compute costs, natural language prompts, and model quality improvements across text, image, audio, video, code, and scientific research. The presentation also addresses major industry challenges, including hallucinations, high inference costs, memory limits, copyright lawsuits, and data access barriers.","thumbnail_url":"https://i.ytimg.com/vi/reLhJnQd-Lo/hqdefault.jpg","publish_date":"2023-02-22T16:24:31.000Z","watch_url":"https://agentcdn.com/creators/sandersaar/videos/video_reLhJnQd-Lo/watch"},"video_facets":{"evidence":{"skill_level":[{"start_ms":0,"end_ms":73978,"via":"chapter"},{"start_ms":472958,"end_ms":802638,"via":"chapter"}],"production_value":[{"start_ms":0,"end_ms":73978,"via":"chapter"},{"start_ms":150083,"end_ms":273083,"via":"chapter"},{"start_ms":362375,"end_ms":472958,"via":"chapter"}],"originality_signal":[{"start_ms":519560,"end_ms":1325035,"via":"chapter"}],"is_evergreen_or_dated":[{"start_ms":161202,"end_ms":186833,"via":"on_screen"},{"start_ms":610542,"end_ms":625208,"via":"on_screen"},{"start_ms":1843223,"end_ms":1869875,"via":"on_screen"}]},"skill_level":"beginner","production_value":"edited_pro","originality_signal":"aggregation","sponsorship_status":"none","primary_language":"en","is_evergreen_or_dated":"dated","speech_pace_wpm":220,"music_presence":"intermittent"},"categories":{"taxonomy_version":"agentcdn-categories-v1","primary":{"id":"ai_data","label":"AI and Data"},"secondary":[{"id":"technology","label":"Technology"},{"id":"business_finance","label":"Business and Finance"},{"id":"software_development","label":"Software and Programming"}]},"format":{"label":"monologue","evidence":{"main_speakers":1,"minor_speakers":2,"main_speaker_shares":[0.93]},"segments":[{"label":"monologue","start_ms":0,"end_ms":1982000}],"classifier_version":"video-format-v2"},"custom_metadata":{"age_rating":"general","editorial_tier":"ugc","content_warnings":[],"commercial_intent":"news_explainer","sponsorship_partners":[],"accessibility_features":["captions"]},"chapters":[{"id":"chapter_video_reLhJnQd-Lo_c8da73c68e411f8c","chapter_index":0,"title":"Intro and the accelerating impact of AI","summary":"The host introduces the video, highlighting the current Apollo moment for artificial intelligence and how AI's disruption began with creativity rather than physical labor.","transcript_text":"We've definitely hit the Apollo landing moment where all eyes are on artificial intelligence and how it will reset everything. My AI buddy always nails the intros, and I am incredibly excited to explore together with you how AI is going to change your and my life, why this is happening now, and what are some of the coolest examples out there. If you asked somebody 5 years ago what would be the first skills to be replaced by artificial intelligence, they would have said the physical labor— warehouse workers, truck drivers— followed by the less demanding cognitive labor like accountants, then the high-demand cognitive labor like computer development, and lastly it's gonna go for creativity because there's something so human about it that cannot be replaced by computers, like creating art and films. Actually, the exact opposite is happening at the moment, where creativity is the first use case for artificial intelligence to go into masses, then followed by high-demand cognitive labor and others. And you can see it in the success of ChatGPT, taking just less than 2 months to reach 100 million users, compared to Facebook taking 4 years, or mobiles taking 12 years to do that. It's one of the fastest platforms to grow, which is not entirely true as Pokémon Go actually reached 100 million users in just less than a month, but that's a topic for another video. Artificial intelligence is top of mind for everybody around the world. That's seen on Google Trends.","visual_description":"The presenter stands in a studio talking alongside lunar graphics and diagrams contrasting past versus present AI labor impacts.","start_ms":0,"end_ms":73978,"duration_ms":73978,"evidence_grade":"C","confidence_band":"medium","boundary_snap":null,"entity_ids":[],"products_mentioned":[],"category":{"id":"ai_data","label":"AI and Data"}},{"id":"chapter_video_reLhJnQd-Lo_204e7ae9917ee63d","chapter_index":1,"title":"The new AI platform era","summary":"A review of computing platform shifts from PC and mobile to cloud and now artificial intelligence.","transcript_text":"We're really entering in this new platform era. It started in the 1980s with PCs where we got browsers and search that we're still using. Building on top of that, we got mobile and mobile really brought us those app stores where you can get any service as an app, as well as all the super apps. And social media apps that we're using day to day. That was building on top of the cloud platform which we've been using so far that brought cross-platform usage. It just didn't matter which device you're holding in your hand and collaboration in the cloud. And now we're entering this new phase in artificial intelligence which is bringing first intelligent agents and who knows what in the future that we're going to explore later.","visual_description":"In a studio, a presenter stands beside an emerging platform era diagram that builds up layers for PC, Mobile, Cloud, and AI. The video later cuts to an interview clip of Sam Altman.","start_ms":73978,"end_ms":112037,"duration_ms":38059,"evidence_grade":"C","confidence_band":"medium","boundary_snap":{"rule_applied":"sentence","snapped_from_ms":74091,"snapped_to_ms":73978},"entity_ids":[],"products_mentioned":[],"category":{"id":"technology","label":"Technology"}},{"id":"chapter_video_reLhJnQd-Lo_ab89bafff5dbbded","chapter_index":2,"title":"Why generative AI took off now","summary":"An interview segment with Sam Altman explores why conversational interface was key to generative AI's breakthrough.","transcript_text":"But why is this happening now? Let's hear from the co-founder and CEO of OpenAI, Sam Altman. We had the model for ChatGPT in the API for like, I don't know, 10 months or something before we made ChatGPT. And I sort of thought someone was going to just build it or whatever. And that, you know, enough people had played around with it. Definitely if you make a really good user experience on top of something. And like one thing that I very deeply believed was the way people wanted to interact with these models was via dialogue. And I, you know, we kept telling people this, we kept trying to convince people to build it and people wouldn't quite do it. So we finally said, all right, we're just going to do it.","visual_description":null,"start_ms":112037,"end_ms":150083,"duration_ms":38046,"evidence_grade":"C","confidence_band":"medium","boundary_snap":{"rule_applied":"sentence","snapped_from_ms":112503,"snapped_to_ms":112037},"entity_ids":[],"products_mentioned":[],"category":{"id":"ai_data","label":"AI and Data"}},{"id":"chapter_video_reLhJnQd-Lo_d8188dae25a73bef","chapter_index":3,"title":"Key drivers: Interface evolution","summary":"The presenter details how AI interfaces developed from base models and instruction prompts to dialogue and action engines.","transcript_text":"And they just did it. And really, I think there's more than just the interface, but that's the key part because all the pieces were in place already 10 months before, as you said, but nobody was really using it. I believe the key drivers are the quality of the models themselves, interface, which we're going to talk first, and cost. So let's dive in. On interface, you can see on the left, the ChatGPT, not ChatGPT, GPT-3 playground where you could actually have a chat experience already one and and a half years ago where all the models were pretty much in place with slight tweaks that were made to ChatGPT. So even the slightest changes in the interface and design and the model can make a huge difference. The interface for AI has really changed over time. We started with those base models that are still around like GPT-3, but it's really hard to interact with them because you need to be really precise and a good prompt engineer to get any information out that you need. Then we got those instruct models. That means the input-output models. We had a specific goal in mind, for example, to get weather information or to get a quote from the book so the AI knew what to look for. And now more recently, just a couple of months ago, we've been seeing those exchange interfaces where you can have a 2-way conversation like ChatGPT. On top of that, we now have this retrieval interface model, and that's something you see in the Bing Chat where you put that— take that ChatGPT experience and put that together with the information, the live information that you can retrieve from the real life So you can also quote the results. And that's the stage we're at now. The ultimate version of the interface for AI that everybody's chasing is the action model. That's where everybody has your login details as well as your credit card information. So you can actually do it for you rather than you having to instruct it or have a conversation or try to retrieve information. Now you can just go straight to action and you see it being used everywhere like Replit in the code generator that's built into the tool. They're using definitely like some of the models that we talked about, Or Edge that is using the retrieval model, putting together with a conversational model to optimize the results. So the interface is evolving.","visual_description":"In a studio setting, a presenter speaks and gestures beside slides displaying a Venn diagram of AI paradigm shift drivers, a comparison between GPT-3 Playground and ChatGPT, a tiered pyramid graphic showing OpenAI models like InstructGPT and Bing, and screenshots of Replit Ghostwriter and Bing in Edge.","start_ms":150083,"end_ms":273083,"duration_ms":123000,"evidence_grade":"C","confidence_band":"medium","boundary_snap":null,"entity_ids":[],"products_mentioned":[],"category":{"id":"technology","label":"Technology"}},{"id":"chapter_video_reLhJnQd-Lo_f2c4b4a8deff02f3","chapter_index":4,"title":"Key drivers: costs and data growth","summary":"The host analyzes how the marginal cost of compute and production has collapsed while digital data volume has exploded.","transcript_text":"The second thing that we talked about, why it's happening in addition to interface, is cost. We talked about in my previous video how the cost of making videos using AI is now almost nothing compared to just 10 years ago where you had to pay several hundred thousand for a camera and tools. If you want to see that video, by the way, click on the video up here and, and check it out, how it's going to change Hollywood. Where you can make videos on your laptop using, you know, GPT as a writer, using Descript or other voice platforms as an audio, and using Synthesia, for example, to create the video, or any other video generation platform. The same thing is about to happen for intelligence. What it used to cost you a quarter of a million dollars to get a degree in an Ivy League college, you could get now an online course, an online certificate for just, you know, $10,000. Or you can just use the ChatGPT interface for $20 a month and have access to all of the intelligence of the world, all of the books that you can imagine, the websites you could never visit. This is happening because number one, we have so much data out there. Everything that is done on computers is now tracked and part of those models. So the amount of data that we could use to model is insane. At the same time, the cost of compute and energy has been coming down rapidly. So building those models and using those models is becoming cheaper and cheaper. And that's why the first to go really are the works that are done on computer. They're using bits that can actually be used to model out. Rather than atoms, the physical elements, the warehouses, which are much harder to track as all of it is not yet connected to the internet.","visual_description":"Charts displaying production cost comparisons across analog, digital, and AI eras alongside graphs showing compute and data trends.","start_ms":273083,"end_ms":362375,"duration_ms":89292,"evidence_grade":"C","confidence_band":"medium","boundary_snap":null,"entity_ids":[],"products_mentioned":[],"category":{"id":"business_finance","label":"Business and Finance"}},{"id":"chapter_video_reLhJnQd-Lo_3825843543587229","chapter_index":5,"title":"Key drivers: Model quality progression","summary":"A historical look at algorithmic models from rule sets and big data science to deep learning and modern transformers.","transcript_text":"Last but not least is the quality of the models. Why is AI now in a space, or machine learning in a space where we can find it more useful than ever? It really started in the '90s when the algorithmic models were developed. Those were the simple ways of digitizing when we were starting to digitize our assets, like, you know, accounting going to a computer so we can do simple calculations, search and filtering, input-output methods. It then evolved into data science that was driven by big data, where everything started to be connected to the internet and computers. And we started to do more and more of our applications online and on computers. Hence, we started building those larger, more complex algorithms and using data science to resolve them. And then we evolved into machine learning, which is really a subset of artificial intelligence in the middle of 2010s, where we had humans as well as then computers developing the algorithms. And that's where deep learning models were developed and where computers really became better in many things, for example, in understanding image classification, where they surpassed human accuracy in 2015, or even for classifying audio or replicating audio, that was surpassed in the mid-2010s. To now, that has evolved into large generative AI models, which have less human in the loop, which are using completely new methods of artificial intelligence, and the Transformers methods as well as the new attention methods. So it's not somebody guiding the AI to go from one step to another, but we really give free hands to the AIs to figure out themselves where the attention should be in the model and build multiple hundreds of versions of that. And that's why the quality has got so good. And these 3 are the key drivers for making AI paradigm shift. You know, the quality of the models achieving human-level intelligence, the interface and user experience turning into chat, human-like conversations, as well as the tools trending rapidly towards zero. And that's what made ChatGPT a success.","visual_description":"The host presents an evolutionary chart showing algorithmic models, data science, deep learning, and generative AI over time.","start_ms":362375,"end_ms":472958,"duration_ms":110583,"evidence_grade":"C","confidence_band":"medium","boundary_snap":{"rule_applied":"shot","snapped_from_ms":361758,"snapped_to_ms":362375},"entity_ids":[],"products_mentioned":[],"category":{"id":"ai_data","label":"AI and Data"}},{"id":"chapter_video_reLhJnQd-Lo_5ce2510152684212","chapter_index":6,"title":"Society evolving from laborer to narrator","summary":"A discussion on human societal roles progressing from passive foragers and laborers to creators and now AI narrators.","transcript_text":"Obviously, we as humans have to evolve with it. We started on being, you know, those passive human beings who were foraging around, gathering their own goods, to laborers who started specializing in different skill sets. And that gave a start to communities, cities, and agriculture, to now the stage where we are at, where we started building machines to do the work for us. We started building computers and manufacturing machines, so we became the creators and operating those machines. And the stage we're entering now is becoming narrators so that we don't build the machines, but we just describe what needs to be built and the machines will build themselves. And this does not just apply to language, which we talked about a lot. This also goes to image modeling, audio modeling, video modeling, code modeling, and science modeling. We're gonna dive into each one of them, how they're developed, and some of the coolest examples that you've probably not seen yet.","visual_description":"A slide displays four evolutionary stages: Passive, Laborer, Creator, and Narrator, followed by six generative AI modalities.","start_ms":472958,"end_ms":519560,"duration_ms":46602,"evidence_grade":"C","confidence_band":"medium","boundary_snap":{"rule_applied":"shot","snapped_from_ms":473255,"snapped_to_ms":472958},"entity_ids":[],"products_mentioned":[],"category":{"id":"history_society","label":"History and Society"}},{"id":"chapter_video_reLhJnQd-Lo_bfe591f923005128","chapter_index":7,"title":"Large language models explained and use cases","summary":"An overview of major language models, training pipelines, and applications including Jasper and Mint Mobile commercial writing.","transcript_text":"So first of all, what are some of the largest language models out there? Obviously ChatGPT, we've talked about a lot. In addition to that, we have Bing Chat that has evolved on top of that to get the retrieval model out there. Google has announced Bard, their own chat version of search, as well as Anthropic, which has developed a model called Claude. It's interesting that the talent has really been moving from Google to OpenAI to develop those models, and OpenAI people have now moved to Anthropic to develop new language models. They have hundreds of billions of parameters. Baidu has one of the largest models out there with 260 billion parameters, and many of the other companies that are not on this slide are also developing their own large language models. But how do you actually develop one? Let's use OpenAI as an example. First of all, you need a ton of data that includes in their case, all of the books that you can imagine, Substack, Reddit, Twitter, Wikipedia, and Common Crawl, which really takes almost all of the website data and puts it in a structured form for anybody to access. That's about 45 terabytes of data. And then all of that data is sent for training using the transformer infrastructure to the NVIDIA's most powerful A100 chips, and all of that compute power will cost about $5 million to generate. And then from that, from that 45 terabytes of data and training, you get a GPT-3 model, which is about 800 gigabytes. So you could probably fit it on your computer, but to run it, it still costs quite a bit of money. That's the inference part where you actually get the results. And it costs about anywhere between 2 to 8 cents, depending on the length of the response that the OpenAI ChatGPT is providing. For the server cost. Of course, there's many more models out there. If you want to give it— check them out, go to lifearchitect.ai. And he's also got a great YouTube channel that keeps track of all of the new models coming out, and you can rank them based on different sites. You will find a link for that down below in the description. The uses for language models actually go beyond what people imagine today and what is on this slide. Number one is of course text generation. We talked about that a lot, where you give a prompt and you expect to get a response back. You could also use it for summarization. If you have a large PDF in front of you and you want to just answer your question and find that nugget that will be useful for you, you can use it for that. You can use it for translation. The language models are so much better than Google Translate ever really was and makes the knowledge accessible that is currently only locked to a couple of languages to everybody in any language. Of course, you can use it for search, and that's what Bing reimagines their own experience around, getting more precise and more conversational answers. Straight in the search rather than having to search for them on the websites. And it becomes an intelligence assistant, helping you use applications that maybe you don't even know how to use the user interface, but you can use natural language to give it prompts and guidance. And last but not least, you can use it for analysis, so help to derive data from different websites or from the whole language model to help you make an easier purchase decision, for example. An example of one of them is Jasper, which has built a platform that helps you to do any kind of writing that you can imagine straight on the platform. So let's see how it actually works in Jasper is here to help. It's an app that uses AI to help you create any kind of content you need. It's like having your very own robot sidekick with an infinite supply of ideas. And that's why the perfect pair of socks is like a hug for your feet. Hot damn, that's good! And you can use it really for everything, from generating ideas to generating full scripts. And here's an example how Mint Mobile was using it right, really in practice. So they gave prompts to OpenAI, to use a joke curse word and big wireless and still going titles in the video. And then Ryan Reynolds was seeing if ChatGPT was doing its job and then reading it back and this becoming a real ad that they were running earlier this year. First of all, let me just say Mint Mobile is the shit. But here's the thing. All the big wireless companies out there are ending their holiday promos, but not Mint Mobile. We're keeping the party going because we're just that damn good. Give Mint Mobile a try. And hey, as an added bonus, if you sign up now, you'll get to hear my voice every time you call customer service. Just kidding, that's, that's not really a thing. And that's something where it becomes really practical for them to build content, create awareness by using somebody as a writer that costs them $20 a month. Bing recently launched Bing Chat. As I said, this redesigns and reimagines search. So rather than just giving a 3-word keyword, which most of the people do, and half of the people go straight back to search. You can actually give guidance and then go back and forth to find exactly what you need. This is to get, for example, craft ideas. The other example here is to build a fitness workout program. Again, like for your needs exactly of what you're looking for, and it's gonna try and ask questions in return and then build the perfect program just for you and also referencing the websites where it's getting the information from.","visual_description":"A presenter in a patterned shirt speaks in a studio beside graphics and slides detailing large language models, training data pipelines, and application categories. Video clips include Jasper software demonstrations, an office worker with a robot assistant, Ryan Reynolds in a Mint Mobile advertisement reading a script, and chatbot demos on Microsoft Bing.","start_ms":519560,"end_ms":802638,"duration_ms":283078,"evidence_grade":"C","confidence_band":"medium","boundary_snap":{"rule_applied":"sentence","snapped_from_ms":519994,"snapped_to_ms":519560},"entity_ids":[],"products_mentioned":[],"category":{"id":"ai_data","label":"AI and Data"}},{"id":"chapter_video_reLhJnQd-Lo_b57d4693a13820a7","chapter_index":8,"title":"Image AI models explained and use cases","summary":"Exploration of image generation models like DALL-E 2 and Stable Diffusion, local running, inpainting, and UI design with Galileo.","transcript_text":"So getting, putting the large language model together with a retrieval model of what you can get from the websites And gives you the perfect answer. The second category is image modeling. Some of the largest ones out there that you might've heard of is obviously DALL-E 2 that you can go out there and use, Stable Diffusion, which is open source. You can run it on your, on your own computer and Midjourney, which you can use in Discord, uh, in a chat environment. Google has built some of the strongest models out there as well, but they haven't really made them public yet, but they are one of the largest in terms of parameters. How do you build an image model? Stable Diffusion is a good example here. You use the data. They have large open data like LAION, which has got 5 billion text and image pairs. So you put together the description of the image together with what the image looks like. They also had a deal with Getty Images, which we're going to talk about later, a controversial deal. You put that 120 terabytes of data together. And then based on the CEO of Stability AI, it took about, it took about 150,000 hours of compute for the most powerful Intel A100, NVIDIA A100 chip And cost them about $600,000 to build the model. And then the model itself is just a 2 gigabyte file. So getting from 120 terabytes just down to 2 gigabytes, and you could run it right on your iPhone or right on your Mac or any computer that you imagine, or for free using your own graphics card and your own compute. Or you could go out and obviously use any other service that you pay a few dollars for. What can it be used for? Text to image, of course, we talked about that. Image to image. I'm sure you've seen Lensa AI where everybody's sharing their avatars with those predefined prompts to go from image to image. You can use it for outpainting. So you might have a small image, but you want to expand on that, fill in the space. That's been done for many famous paintings out there. You can use it for inpainting, you know, to replace even somebody like Mona Lisa's hair with mohawk. You can use it for enhancements and editing, for color correcting, or you could also use it for classification and detection of different objects inside the images. So let's look at some of the examples out here now. First of all, DALL-E that can take simple text descriptions like a koala dunking a basketball and turn them into photorealistic images that have never existed before. DALL-E 2 can also realistically edit and retouch photos based on a simple natural language description. It can fill in or replace part of an image with AI-generated imagery that blends seamlessly with the original. It's called inpainting. And moreover, you could also use it not just for generating images, but from something practical point of view. This is an example for designing an interface. Using, using a tool called Galileo where you can say, hey, create me an onboarding screen for this dog walking app, and then also create the settings page with username and photo. And instantly you will have that generated by AI, again using the database that they have. Or you then want to create the profile page for completely something else, another, another app.","visual_description":"The presenter outlines image generation tools, shows text-to-image demos with koalas and cats, and tests Galileo's mobile UI generator.","start_ms":802638,"end_ms":973159,"duration_ms":170521,"evidence_grade":"C","confidence_band":"medium","boundary_snap":{"rule_applied":"sentence","snapped_from_ms":801275,"snapped_to_ms":802638},"entity_ids":[],"products_mentioned":[],"category":{"id":"ai_data","label":"AI and Data"}},{"id":"chapter_video_reLhJnQd-Lo_3c93a0c367547fd5","chapter_index":9,"title":"Audio AI models: Music and speech","summary":"Examining audio models like AudioLM, Whisper, and Uberduck through synthetic podcasts and David Guetta's AI vocal experiments.","transcript_text":"And by the way, while you generate all of that, this becomes transferable that you can take it out, for example, to Figma, and then from Figma launch it as an actual app. So imagine how far you can take it. Audio is another big area of modeling, specifically in speech modeling and music, where Google has been building large language models like AudioLM, which is building on WaveNet before. Amazon has built Polly, which is the largest— one of the largest voice libraries. OpenAI has got Whisper, the text-to-speech library, Speechify, and many others. And in the music space, we're seeing more recently developments like MusicLM to generate music from text. Or SingSong, which you can use audio to audio, as well as one of the largest open source movements, Uberduck. What are some of the use cases? Of course, number one is text-to-speech. You can just type something up and it'd be spoken back to you. The second thing is taking a library of audio files and turning them into text and maybe making them searchable using AI. You can do speech-to-speech in different languages or even in the same language. You can just change your voice using something like Respeecher. You can do text-to-music. You can just give prompts like, this is the type of vibe I want, this is the type of style I want, and it will generate music for you. Or you can hum some voice and it will fill in the background like a voice-to-music model. Or you can just do enhancements for the audio, you know, make the audio sound better or exactly what you want it to be described using a text model. Here's an example how a company called podcast.ai brought back models from Joe Rogan voice as well as Steve Jobs voice and put them in a podcast format to See what they would say if Steve was still alive. It's always good to see you, buddy. I'm so happy you came on, man. Yeah, it's great to be on the show. Your audience is just so different from your normal Apple users, and that's a good thing. It's cool. Well, you know, I was an Apple user way before I did this show. I've been a fan of yours and Macintosh since the 1980s. I think it sounds great, not just from content point of view, but also from the sound of the voice. Of course, it's gonna get better and better over time. And this is an example of somebody who just used ChatGPT to come up with rap rhymes for Eminem and then used Uberduck to integrate, make them into sounds and then integrate into them his song. So this is what you're gonna hear now. This is the future rave sound. I'm getting lost in an army ground. This is the future rave sound. I'm getting lost in an army ground. Eminem, bro.","visual_description":"In a studio, a presenter in a patterned collared shirt speaks in front of slides detailing audio AI models and use cases. Later, the display shows a podcast screen, a monochrome portrait of Steve Jobs, concert footage, and an interview clip of David Guetta.","start_ms":973159,"end_ms":1124760,"duration_ms":151601,"evidence_grade":"C","confidence_band":"medium","boundary_snap":{"rule_applied":"sentence","snapped_from_ms":973083,"snapped_to_ms":973159},"entity_ids":[],"products_mentioned":[],"category":{"id":"music","label":"Music"}},{"id":"chapter_video_reLhJnQd-Lo_639d8471ff3f0a7d","chapter_index":10,"title":"Video AI models: 3D, avatars, and gaming","summary":"A showcase of video-to-video editing with Gen-1, avatar generation with D-ID, 3D animations with Krikey, and Roblox game creation.","transcript_text":"There's something that I made as a joke and it works so good. I could not believe it. I discovered those websites that are about AI. And this is how much you can do with just little experiments that we're starting to see happen around us. Moving on to video, and this has really 2 major categories. One is 3D category, which is generating completely new objects that can be editable in any format. And the other one is avatars, which is somebody generating a lookalike like we did in the last video that you saw. In 3D, some of the largest models are Google's Image and Video. Of course, Meta showed their own Make-A-Video platform as well as, uh, KIRKI AI, which we're gonna see an example later. In the avatar space, there's Synthesia, there's D-ID that can take photos to turn them into images and rephrase and many others that don't fit on this slide. What are some of the example use cases for video models? Of course, you can use text-to-video, just give it a prompt and it's gonna generate video out of it. You can do image to video, just give it a photo and then turn that into a moving, moving image. You can personalize video, edit just, you know, some templated objects inside the video to make it very unique. You can do text to 3D, you can describe something and then turn that into a 3D object or a moving object. You can use it for classification, for example, for fitness apps to see how well you're tracking against, against the moves. And you can use it for enhancements and editing, up resolutions, down resolutions, clarifying, etc. So let's see some of the examples. And the first one here is the Runway ML and their new model to stylize any content, generate video in any style, all while retaining quality and flexibility. Gen-1 is able to realistically and consistently apply the composition and style of an image or text prompt to the target video, allowing you to generate new video content using an existing video. We call this approach video-to-video. In addition to video-to-video, we also talked about photo-to-video, and a good example here is the D-ID platform. D-ID's Creative Reality Studio just got even better. Now, as well as animating images from text or audio, you can create avatars and generate scripts using Stable Diffusion and GPT-3. Just describe what you want your avatar to look like and press create. You can make any avatar you want, say, picture of a half-alien goddess. Or a 19th century portrait of a happy Victorian gentleman. And then the other example here, going beyond just avatars and changing the style of the movies, is also to generate full 3D objects. An example here is the Kerky AI, where you can use text-to-3D prompt. And somebody's using it in real life already is Roblox, where they're using generative AI and prompt guidance to design a game. Of course, they've got a whole host of tools out there that you can use today, but they're still pretty complex and only fraction of the users use them. So here you can just do text guidance of what do you want the car look like, what do you want the paint look like, what do you want the lights to do, and that automatically generates the code for you within the game that you can then launch on their platform. Which makes it so easy and so much more accessible for people using natural language and also in game design or potentially even making movies inside games.","visual_description":"Video clips show Gen-1 video styling, animated portraits in D-ID, Krikey 3D dance animations, and real-time Roblox vehicle customization.","start_ms":1124760,"end_ms":1325035,"duration_ms":200275,"evidence_grade":"C","confidence_band":"medium","boundary_snap":{"rule_applied":"sentence","snapped_from_ms":1126917,"snapped_to_ms":1124760},"entity_ids":[],"products_mentioned":[],"category":{"id":"technology","label":"Technology"}},{"id":"chapter_video_reLhJnQd-Lo_0182135e60f723bc","chapter_index":11,"title":"Code AI models and developer acceleration","summary":"The presenter discusses AI code assistants including GitHub Copilot and Replit Ghostwriter for debugging, chat, and rapid development.","transcript_text":"Next one is code. And as you saw a little example in Roblox, it takes way beyond that. Some of the largest models out there are Codex by OpenAI as well as Copilot by GitHub, which is built on top of Codex really, and Replit, which built Ghostwriter, and Google, which in secret is about to announce their own coding tool, which they're developing internally, but there's no outside information available yet. What do you use the code models for? Of course, number one is autofilling, helping you get more done faster. The other one is text-to-code, so you can describe something and it will turn into code like we saw with Roblox. Or you can do Q&A and reviews, finding the errors in the code. Or you can do code-to-text— somebody sends you a code but you don't understand what's going on, you can get it translated into human language. Here's an example from Replit. In the first one, you can see, hey, what does this index file do? And it tells you back in natural language of what the file is about to do. Then it defines an error. Hey, the error is on line 6, and here's how to fix it. So you can copy the code, put it back to your editor, and it fixes the result for you. And then you could use the chat interface to even develop further a new code for it for you, or even run QA on your existing code. This is how powerful it is.","visual_description":"A presenter in a patterned shirt stands beside presentation slides and screen demonstrations. The visuals display AI code generation tools such as Codex, CoPilot, Ghostwriter, and Google, detailing use cases like Autofill, Text-to-code, QA & Reviews, and Code-to-text, alongside demonstrations of the Replit Ghostwriter chat interface and GitHub Copilot metrics.","start_ms":1325035,"end_ms":1396891,"duration_ms":71856,"evidence_grade":"C","confidence_band":"medium","boundary_snap":{"rule_applied":"sentence","snapped_from_ms":1326350,"snapped_to_ms":1325035},"entity_ids":[],"products_mentioned":[],"category":{"id":"software_development","label":"Software and Programming"}},{"id":"chapter_video_reLhJnQd-Lo_29e16c1bb815d711","chapter_index":12,"title":"Science AI models and DeepMind AlphaFold","summary":"How AI models like AlphaFold 2 predict protein structures to assist in drug discovery, disease research, and environmental cleanup.","transcript_text":"And Copilot has shared some results that it makes developers 55% faster, and almost half of the code that they do is now written by artificial intelligence. Last but not least, let's talk about science because that's where some of the biggest breakthroughs might come through. DeepMind and Google have built AlphaFold 2, which is one of the most critical protein folding prediction models out there. That is also something that Meta is working on with ESMFold, as well as many of the other research bodies, including some of the open source ones like the OpenBioML, who's building together with other partners LibraFold project. What are some of the uses for science models? Number one is review and analysis. You can take in all of the science papers and then understand what are the correlations, what others have been working on that humans can't possibly work through hundreds of thousands of papers that are coming out as they're exponentially growing and there's more and more every day. You can use it for predictive modeling, all of the phases, for example, of the development of new pharmaceuticals. You can use it for precision medicine, matching the specific individual and their biology to the specific pharmaceutical. You can use it for genomics in really predicting how different proteins react. You can use it for diagnosis. So for example, taking a image of your retina scanned and then matching it to a diabetes database to be much more accessible and more predictive in your diseases, as well as for monitoring, including in-hospital and at-home care, where you can use connected devices and the behavior models that you can build on top of that to understand if the patients are progressing as well as they could. And here's a quick video to explain how useful and valuable that is from AlphaFold. So we created an AI system to solve this problem. Called AlphaFold. It's trained on the sequences and structures of about 100,000 proteins, painstakingly mapped out by scientists around the world. Today, it can accurately predict a protein shape just from its sequence of amino acids. AlphaFold's predictions could enable progress in all sorts of areas. Imagine a future where we can understand diseases more quickly and develop drugs to fight them, or one where we could use enzymes to break down plastic waste, or even to capture carbon from the atmosphere.","visual_description":"The presenter speaks beside science AI application cards followed by an animated video detailing DeepMind AlphaFold.","start_ms":1396891,"end_ms":1524917,"duration_ms":128026,"evidence_grade":"C","confidence_band":"medium","boundary_snap":{"rule_applied":"sentence","snapped_from_ms":1399057,"snapped_to_ms":1396891},"entity_ids":[],"products_mentioned":[],"category":{"id":"science","label":"Science"}},{"id":"chapter_video_reLhJnQd-Lo_57b6c0ace9e5d09b","chapter_index":13,"title":"Generative AI challenges and output errors","summary":"Addressing hallucinations, user prompting psychology, system guardrails, and factual inaccuracy issues across modern chatbots.","transcript_text":"But it's not all rosy for AI and generative AI. There are several challenges, including the hallucinations and the output errors, the high cost of the models, the lack of long-term memory, the slow inference time and data access, as well as the biases in the system. And let's go through each one of them and see what are the potential solutions. Out there you have a lot of articles that show you how the AI models are lying, or they're building a relationship, they're in love, they're sentient. But actually I don't believe that really to be the case. I believe that's very much the kind of reaction the monkey would have with a mirror, you know, trying to fight off somebody that they see that they think is somebody else, but it's actually themselves. And I feel we're in a very similar spot with prompts at the moment. We are prompting the models to say the weird things back to us. We're prompting the model to go into a mode where it would kind of act back to us and what we're expecting from it. And I think this is exactly what's happening today. It's more like a monkey in a mirror effect. Than really the AI models being sentient, because they still need to be prompted. They wouldn't do anything by themselves or on their own. And of course, there are already now more guardrails in place from, for example, Bing, who doesn't even want to talk about it when you ask if it's alive, if it's sentient, does it love you. All of those are now blocked, and it just says it's something that they prefer not to talk about. And the output errors, that's another part of hallucinations, is we saw those errors in the Google demo when they announced Bard, but if you go to Reddit subreddits like ChatGPT or Bing, you'll see a whole lot more on how people are trying to trick the systems. And I think that's very normal to push the boundaries and make them better. And today the generated models are not great being factually correct, but I think in the phase of where Bing is at, where they're putting the large language model together with the, the exchange model together with the retrieval model, that's where you can really get good facts. And that's an opportunity for many other companies like, for example, Wolfram Alpha to put the mathematical models and calculations together with the chat models of ChatGPT and make something that is factually correct but also is something that is more useful and easier for consumers to use in a natural language way.","visual_description":"Infographics listing adoption challenges, news headlines on erratic AI chats, a clip of a chimp reacting to a mirror, and Bard output errors.","start_ms":1524917,"end_ms":1648184,"duration_ms":123267,"evidence_grade":"C","confidence_band":"medium","boundary_snap":{"rule_applied":"shot","snapped_from_ms":1525340,"snapped_to_ms":1524917},"entity_ids":[],"products_mentioned":[],"category":{"id":"technology","label":"Technology"}},{"id":"chapter_video_reLhJnQd-Lo_683007d7810b9773","chapter_index":14,"title":"Cost, slow inference, and memory limits","summary":"Analyzing the trade-offs of query costs, generation latency, and context window token memory limitations.","transcript_text":"The second thing is high cost. It costs almost 10 times more, from just a fraction of a cent on Google Search to about 2 to 8 cents to run a query on ChatGPT. Of course, doing the same with images today in Stable Diffusion also costs you about 2 cents if you're buying it on their own servers, running it rather than running it on your own. But I'm sure that's something that's going to change in the future as the compute gets again cheaper and we— they get more powerful. The slow inference time is directly connected to the cost. While it takes half a second to search something on Google, it takes about 5 to 20 seconds to get a prompt back in ChatGPT. Of course, I believe that that's also going to change. Just over the last 6 months, Stable Diffusion has managed to decrease and increase the efficiency of the model by about 3 times by not really getting any reductions in the image quality. So instead of taking 6 seconds, it now takes 2 seconds to generate an image. And I think the same thing is going to apply to cost and speed in the future. The third— the fourth area is the lack of long-term memory. Today, similarly, the ChatGPT model can only memorize 4,000 tokens, which is about 3,000 words in human language. And that's obviously a limiting factor if we want those assistants to be personal for everybody.","visual_description":"The host presents comparison bar charts on inference costs and generation speed, alongside ChatGPT token limit highlights.","start_ms":1648184,"end_ms":1721532,"duration_ms":73348,"evidence_grade":"C","confidence_band":"medium","boundary_snap":{"rule_applied":"sentence","snapped_from_ms":1640729,"snapped_to_ms":1648184},"entity_ids":[],"products_mentioned":[],"category":{"id":"technology","label":"Technology"}},{"id":"chapter_video_reLhJnQd-Lo_1f5280bdd1040a87","chapter_index":15,"title":"Data access, copyright, and closed systems","summary":"The host examines AI copyright lawsuits, Twitter data lockouts, creative workflows, and historical regulation parallels.","transcript_text":"But I'm sure again that's going to change as the compute and the models get more efficient. Data access and copyright. That's a key area. We've seen so many copyright issues now with, you know, Stable Diffusion being sued by Getty Images as they integrate those watermarks and they clearly see that their database has been used to generate the large image models. You can see Elon Musk finding out that OpenAI has turned from a nonprofit to a for-profit organization and they've closed the Twitter access, Twitter data access for OpenAI. As well as many other artist groups and developer groups suing the large language models developers for the access of their data. But I think it's not really that dissimilar from what we used to do before. You know, if imagine if I'm a designer or photographer, I might go to Dribbble and Instagram and Pinterest and Behance to get all the inspiration and ideas in place. I will then use, you know, Canva and Figma as a tool to create it. I will then put it on Behance, for example, as part of my portfolio. And then I'm gonna go and sell it on Shutterstock. I believe that if I then use, you know, copyrighted material, of course I'm going to be penalized and I shouldn't be able to make money with other people's content. That's the same in this flow, even though I took inspiration from them in my research phase. This is now just being done on superpowers, you know. Rather than me being able to go maybe about 50 to 100 pages for inspiration, ChatGPT has done billions of pages for my inspiration. So I can use it as an input to get new ideas and develop new concepts. I can then use a tool like DALL-E or Stable Diffusion or anything else to create that content. I will then put it in my portfolio and then we'll monetize it in Shutterstock. By the way, Shutterstock has got a partnership with OpenAI already to start monetizing AI-generated images because I believe that this is just a super powerful research tool bringing you a lot more in a more simpler way, and it's a super powerful creation tool accelerating the process of creating and making it a lot more accessible for anybody. But the copyright really comes in when you start monetizing it. So I wouldn't really see a version of where you can penalize those models as they're just there for inspiration. That's what we do every day. Artists and musicians take inspiration from others all the time. Now we can just do it on steroids. Last but not least, these are, you know, getting to be closed systems. You can see how OpenAI turned from open source, and that's something that Elon Musk is really concerned about as well, to something that is now just serving Microsoft. And it was really created as like an anti-Google move to be something that is open, as it will be dangerous in his mind if it gets locked to just one organization. But it's never been really different. Take Gutenberg Press. When that was announced, churches went in and they actually created licensing for every Gutenberg Press to be out there. What can you print? How many can you print? Where can you print? They were all regulated by church. Or take even the Locomotive Act in the middle of 19th century in the UK. Just to protect the horses and the industry that was built around it. There was a law where it had to have somebody walking in front of the car going only 4 miles an hour maximum and waving a red flag as they were doing it. Those were rules to protect and build a closed system, but the market has figured it out.","visual_description":"News clippings on Getty Images and Twitter data lawsuits, a creative pipeline graphic, and historical slides on early printing and train laws.","start_ms":1721532,"end_ms":1907623,"duration_ms":186091,"evidence_grade":"C","confidence_band":"medium","boundary_snap":{"rule_applied":"sentence","snapped_from_ms":1717523,"snapped_to_ms":1721532},"entity_ids":[],"products_mentioned":[],"category":{"id":"business_finance","label":"Business and Finance"}},{"id":"chapter_video_reLhJnQd-Lo_cdbb249f0947ec2f","chapter_index":16,"title":"AI will reset everything","summary":"Sam Altman's perspective on zero marginal cost of intelligence and energy, concluding how corporate costs and society will be transformed.","transcript_text":"In the AI space, there's some company called Stability AI who's behind Stable Diffusion, who is building AI by the people and for the people, aiming to make it open source and accessible for everybody, for any country and any individual to build their own models. Overall, AI will and is about to reset absolutely everything for the 2 reasons. My basic model of the next decade is that the cost of intelligence, the marginal cost of intelligence and the marginal cost of energy, are going to trend rapidly towards zero, like surprisingly far. And those, I think, are 2 of the major inputs into the cost of everything else, except the cost of things we want to be expensive, the status goods, whatever. And that's why everything will be about to be reset by AI, because intelligence and energy will be marginal. That's at the moment about 50 to 60% of the cost of the Fortune 500 companies is the payroll, paying for the labor. And if the labor and the intelligence will become essentially free for everybody, then that's going to rewrite how we do, how we get education, how it affects businesses, healthcare, art, software, and gaming. Which we're gonna talk about in the next video. So thank you very much for watching. I hope you enjoyed it. Let me know what you want to hear about first, and see you next time.","visual_description":"The presenter shares an interview clip of Sam Altman speaking on stage at Greylock, followed by enterprise payroll cost statistics.","start_ms":1907623,"end_ms":1981297,"duration_ms":73674,"evidence_grade":"C","confidence_band":"medium","boundary_snap":null,"entity_ids":[],"products_mentioned":[],"category":{"id":"business_finance","label":"Business and Finance"}}],"visual":[{"kind":"window","id":"vwin_video_reLhJnQd-Lo_0","window_index":0,"start_ms":0,"end_ms":9542,"duration_ms":9542,"visual_description":"In a studio setting, an AI avatar of a man speaks next to an image of an astronaut on the moon, before the real man appears beside him. An AI avatar talks in front of a lunar image before the human host appears smiling next to him in the studio.","on_screen_text":"AI WILL RESET\nEVERYTHING","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Synthesia"],"entity_ids":["ent_0ab7681fca07ba0e99c88574f46adb3b"],"fused_summary":"An AI clone introduces the video beside an astronaut graphic before the real presenter appears alongside it to preview the topic of artificial intelligence."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_1","window_index":1,"start_ms":9542,"end_ms":18167,"duration_ms":8625,"visual_description":"The host speaks directly to the camera with expressive hand gestures in front of a moon landing backdrop in a studio. The host gestures excitedly with both hands while explaining AI concepts in front of the lunar rover graphic.","on_screen_text":"SMASH!\nAI WILL RESET EVERYTHING","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"medium","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The host enthusiastically discusses how artificial intelligence is accelerating and reshaping modern life and work."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_2","window_index":2,"start_ms":18167,"end_ms":33167,"duration_ms":15000,"visual_description":"The host stands on the left of the frame presenting an illustrative graphic titled AI IMPACT 5 YEARS AGO. The host appears on the left beside a diagram depicting workers queued by labor categories. The host gestures and explains the categories of labor shown in the slide on the right.","on_screen_text":"AI IMPACT\n5 YEARS AGO\n4. CREATIVITY\n3. HIGHLY DEMANDING COGNITIVE LABOR\n2. LESS DEMANDING COGNITIVE LABOR\n1. PHYSICAL LABOR","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The presenter outlines historical expectations of how artificial intelligence would replace physical labor before touching cognitive and creative work."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_3","window_index":3,"start_ms":33167,"end_ms":48167,"duration_ms":15000,"visual_description":"The host continues presenting in the studio as the slide transitions to illustrate the reversed order of AI impact today. The host speaks beside the 5-years-ago labor sequence graphic. The host gestures toward the updated slide titled AI IMPACT NOW showing creativity at the front of the queue.","on_screen_text":"AI IMPACT\n5 YEARS AGO\n1. PHYSICAL LABOR\n2. LESS DEMANDING COGNITIVE LABOR\n3. HIGHLY DEMANDING COGNITIVE LABOR\n4. CREATIVITY\nAI IMPACT NOW\n4. PHYSICAL LABOR\n3. LESS DEMANDING COGNITIVE LABOR\n2. HIGHLY DEMANDING COGNITIVE LABOR\n1. CREATIVITY","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The host reveals that contrary to past expectations, creative work is becoming the primary early domain impacted by artificial intelligence."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_4","window_index":4,"start_ms":48167,"end_ms":59936,"duration_ms":11769,"visual_description":"In a studio setup, a presenter stands in the corner beside graphics comparing AI adoption timelines. Presenter stands beside a 3D graphic showing people lined up at an entrance labeled for different labor types. Presenter gestures alongside a bar chart displaying time to reach 100 million users across tech platforms.","on_screen_text":"AI IMPACT\nNOW\n4. PHYSICAL\nLABOR\n3. LESS DEMANDING\nCOGNITIVE LABOR\n2. HIGHLY\nDEMANDING COGNITIVE\n1. CREATIVITY\nTIME TO REACH 100\nMILLION USERS\nCHATGPT\n2 MONTHS\nTIKTOK\n9 MONTHS\nFACEBOOK\n4 YEARS\nMOBILE\n12 YEARS\nPC\n14 YEARS","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Facebook","ChatGPT","TikTok"],"entity_ids":["ent_57e36c5da0caa91427a797c7e5b6f432","ent_8a8a91e44930d87b222367e337327e34","ent_94f4a7f58a3f24593b2f3e856fdde4a8"],"fused_summary":"The presenter explains that creativity is leading AI adoption and compares ChatGPT's user growth speed to Facebook and mobile phones."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_5","window_index":5,"start_ms":59936,"end_ms":74091,"duration_ms":14155,"visual_description":"In a studio setting, a presenter references platform adoption graphics before transitioning to a Google Trends graph. Presenter points upward next to a bar chart highlighting Pokémon Go's rapid user acquisition. Presenter speaks next to a Google Trends search interest curve for artificial intelligence.","on_screen_text":"CHATGPT\n2 MONTHS\nTIKTOK\n9 MONTHS\nFACEBOOK\n4 YEARS\nMOBILE\n12 YEARS\nPC\n14 YEARS\nTIME TO REACH 100\nMILLION USERS\nPOKEMON\nGO\n<1 MONTHS\nGoogle Trends\nExplore\nArtificial intelligence\nTopic\nCompare\nWorldwide\nPast 5 years\nAll categories\nWeb Search\nInterest over time\n100\n75\n50\n25\n25 Feb 20...\n13 Oct 2019\n30 May 2021\n15 Jan...\nNote","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Pokémon Go","Google Trends"],"entity_ids":["ent_16f60149-d305-4ce6-9809-93eb78df8ef5","ent_f000e5694735cec3c528e0ca2c122006"],"fused_summary":"The presenter notes Pokémon Go reached 100 million users in under a month and shows rising interest in AI on Google Trends."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_6","window_index":6,"start_ms":74091,"end_ms":89091,"duration_ms":15000,"visual_description":"In a studio environment, a presenter stands beside an emerging platform era timeline graphic. Presenter speaks beside a diagram showing the PC platform era with browser and search functionality. Presenter gestures as a mobile layer appears on the diagram above the PC era.","on_screen_text":"PC\n1980\nBROWSER\nSEARCH\nMOBILE\n2007\nSOCIAL & SUPER APPS\nAPP STORES","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The presenter introduces major platform eras, beginning with personal computers in the 1980s and expanding into mobile."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_7","window_index":7,"start_ms":89091,"end_ms":102924,"duration_ms":13833,"visual_description":"In a studio setting, a presenter speaks alongside a tech platform diagram showing mobile and cloud layers. Presenter stands beside a diagram illustrating the mobile platform and application ecosystem. Presenter gestures next to a diagram adding the cloud platform layer with multiple connected devices.","on_screen_text":"MOBILE\n2007\nSOCIAL & SUPER APPS\nAPP STORES\nPC\n1980\nBROWSER\nSEARCH\nCLOUD\n2010\nCROSS-PLATFORM\nCOLLABORATIVE","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"thoughtful"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The presenter continues reviewing tech platforms, moving from mobile app ecosystems to cloud-based cross-platform collaboration."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_8","window_index":8,"start_ms":102924,"end_ms":112503,"duration_ms":9579,"visual_description":"A host gestures in front of an educational graphic before an interview clip of Sam Altman appears on screen. The host stands in front of a white slide outlining technology platform eras. The host continues speaking as icons and text for AI appear on the diagram. An interview clip showing Sam Altman speaking into a microphone appears alongside the host.","on_screen_text":"AI\n2015\nINTELLIGENT AGENTS\n?\nCLOUD\n2010\nCROSS-PLATFORM\nCOLLABORATIVE\nMOBILE\n2007\nSOCIAL & SUPER APPS\nAPP STORES\nPC\n1980\nBROWSER\nSEARCH\nWHY NOW?\nFelic","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":["ent_7ec30179-e819-470f-a3e4-9917fb57fca9"],"fused_summary":"The host discusses the arrival of intelligent agents in artificial intelligence before transitioning to an interview with Sam Altman."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_9","window_index":9,"start_ms":112503,"end_ms":125198,"duration_ms":12695,"visual_description":"The host introduces OpenAI CEO Sam Altman, cutting to full footage of Altman speaking on stage at a conference. The host talks in the foreground while Sam Altman holds a microphone in an interview frame. Sam Altman speaks into a microphone while seated on stage at a conference.","on_screen_text":"WHY NOW?\nFelic\nSAM ALTMAN\nSTRICTLY VC","facets":{"composition":"single subject","lighting_class":"indoor artificial","motion_class":"static","camera_type_hint":"medium close-up","dominant_speaker_emotion":"thoughtful"},"products_mentioned":["ChatGPT"],"entity_ids":["ent_7ec30179-e819-470f-a3e4-9917fb57fca9","ent_8a8a91e44930d87b222367e337327e34","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"The host introduces Sam Altman, who explains that the model behind ChatGPT existed in the API months before the product launched."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_10","window_index":10,"start_ms":125198,"end_ms":135988,"duration_ms":10790,"visual_description":"Sam Altman is seated on a conference stage speaking into a microphone with event banners behind him. Sam Altman addresses an audience while holding a microphone on stage.","on_screen_text":"SAM ALTMAN\nSTRICTLY VC\nFelic","facets":{"composition":"single subject","lighting_class":"indoor artificial","motion_class":"static","camera_type_hint":"medium close-up","dominant_speaker_emotion":"thoughtful"},"products_mentioned":[],"entity_ids":["ent_7ec30179-e819-470f-a3e4-9917fb57fca9"],"fused_summary":"Sam Altman reflects on why OpenAI built ChatGPT, noting the importance of a great user experience on top of raw models."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_11","window_index":11,"start_ms":135988,"end_ms":150083,"duration_ms":14095,"visual_description":"Sam Altman gestures with his left hand while holding a microphone on a conference stage. Sam Altman gestures with his hand as he explains interacting with AI models through dialogue.","on_screen_text":"SAM ALTMAN\nSTRICTLY VC\nFelic","facets":{"composition":"single subject","lighting_class":"indoor artificial","motion_class":"static","camera_type_hint":"medium close-up","dominant_speaker_emotion":"focused"},"products_mentioned":[],"entity_ids":["ent_7ec30179-e819-470f-a3e4-9917fb57fca9"],"fused_summary":"Sam Altman explains that people preferred conversational dialogue to interact with AI models, leading OpenAI to launch ChatGPT."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_12","window_index":12,"start_ms":150083,"end_ms":161202,"duration_ms":11119,"visual_description":"In a studio setting, a man speaks in front of a digital slide displaying a three-circle Venn diagram titled Key Drivers of AI Paradigm Shift. A presenter gestures enthusiastically at the lower left while a Venn diagram shows Quality, Interface, and Cost circles on screen.","on_screen_text":"KEY DRIVERS OF AI PARADIGM SHIFT\nQUALITY\nINTERFACE\nCOST","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":[],"fused_summary":"A presenter explains that interface was the key part of the paradigm shift since all other pieces were already in place."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_13","window_index":13,"start_ms":161202,"end_ms":176202,"duration_ms":15000,"visual_description":"A presenter discusses AI interface drivers in a studio setup before transitioning to a side-by-side comparison slide of GPT-3 Playground and ChatGPT. The presenter outlines the key drivers alongside the Venn diagram slide. The slide switches to compare the GPT-3 Playground interface from July 2020 with the ChatGPT interface from November 2022.","on_screen_text":"KEY DRIVERS OF AI PARADIGM SHIFT\nQUALITY\nINTERFACE\nCOST\nGPT-3 PLAYGROUND\nCHATGPT\nHUMAN LEVEL CHATBOT INTERFACE\nJUL 2020\nNOV 2022","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":["ChatGPT","GPT-3"],"entity_ids":["ent_8a8a91e44930d87b222367e337327e34","ent_b9376b28cef422f3d1643e31d16e32d0"],"fused_summary":"The presenter highlights model quality, interface, and cost as key drivers, then displays screenshots comparing the GPT-3 Playground and ChatGPT interfaces."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_14","window_index":14,"start_ms":176202,"end_ms":186833,"duration_ms":10631,"visual_description":"A presenter sits before a slide comparing the interfaces of GPT-3 Playground and ChatGPT in a studio presentation setup. The presenter uses hand gestures to emphasize subtle design differences shown on the two interface mockups.","on_screen_text":"INTERFACE\nGPT-3 PLAYGROUND\nJUL 2020\nCHATGPT\nNOV 2022\nHUMAN LEVEL CHATBOT INTERFACE\nPLUS\nOverview Documentation Examples Playground\nGet started\nEnter an instruction or select a preset,\nand watch the API respond with a\ncompletion that attempts to match the\ncontext or pattern you provided.\nYou can control which model completes\nyour request by changing the model.\nKEEP IN MIND\nUse good judgment when sharing\noutputs, and attribute them to your\nname or company. Learn more.\nRequests submitted to our models\nmay be used to train and improve\nfuture models. Learn more.\nOur default models' training data\ncuts off in 2021, so they may not have\nknowledge of current events.\nPlayground\nLoad a preset...\nWrite a tagline for an ice cream shop.\nLooking for ChatGPT? Try it now\nSubmit\nUpgrade\nHelp\nPersonal\nSave\nView code\nShare\nMode\ntext-davinci-003\nTemperature\n0.7\nMaximum length\n256\nStop sequences\nTop P\nFrequency penalty\nPresence penalty\nBest of\nInject start text\nInject restart text\nShow probabilities\nNew chat\nGoogle's transformer model\nInternet Platform Evolution\nChatGPT's best uses\nElectric clicking in head\nEvolution of AI\nTransformer Infrastructure In...\nPolls: Accuracy and Limitati...\nGenerative AI Introduction\nTool Organization Table\nProduct requirements docu...\nBuild Car Cost Ownership Ch...\nDisputing Late Payment Fee\nEvent Discount Inquiry - AI...\nClear conversations\nLight mode\nMy account\nUpdates & FAQ\nLog out\nDefault","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"thoughtful"},"products_mentioned":["ChatGPT","GPT-3"],"entity_ids":["ent_8a8a91e44930d87b222367e337327e34","ent_b9376b28cef422f3d1643e31d16e32d0"],"fused_summary":"The speaker demonstrates how slight interface and design changes between GPT-3 Playground and ChatGPT made a huge difference."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_15","window_index":15,"start_ms":186833,"end_ms":200765,"duration_ms":13932,"visual_description":"In a studio setting, a presenter speaks as a tiered pyramid graphic showing GPT-3, OpenAI, and InstructGPT builds on the slide. A presenter explains AI base models while the slide displays an OpenAI base model block for GPT-3 before stacking InstructGPT on top.","on_screen_text":"INTERFACE\nGPT-3\nBASE MODEL\nOpenAI","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":["InstructGPT","GPT-3"],"entity_ids":["ent_14d51a79e1a3ca6444ce0459bc2d5750","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_b9376b28cef422f3d1643e31d16e32d0"],"fused_summary":"The presenter explains how AI interfaces evolved from difficult base models like GPT-3, as the slide illustrates the foundation layer and the addition of InstructGPT."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_16","window_index":16,"start_ms":200765,"end_ms":212426,"duration_ms":11661,"visual_description":"A presenter explains AI interfaces in front of a white studio backdrop beside a pyramid graphic illustrating OpenAI's GPT-3 and InstructGPT. Medium close-up of a man gesturing next to a diagram detailing base and instruct AI model tiers.","on_screen_text":"INTERFACE\nInstructGPT\nGPT-3\nINSTRUCT\nBASE MODEL\nOpenAI","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["InstructGPT","GPT-3"],"entity_ids":["ent_14d51a79e1a3ca6444ce0459bc2d5750","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_b9376b28cef422f3d1643e31d16e32d0"],"fused_summary":"The presenter outlines instruct models where artificial intelligence takes specific input queries to provide targeted output responses."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_17","window_index":17,"start_ms":212426,"end_ms":221499,"duration_ms":9073,"visual_description":"The presenter continues speaking as a green Exchange tier featuring ChatGPT is added to the diagram. The host points at the diagram displaying the newly added ChatGPT exchange interface tier.","on_screen_text":"INTERFACE\nChatGPT\nInstructGPT\nGPT-3\nEXCHANGE\nINSTRUCT\nBASE MODEL\nOpenAI","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["ChatGPT"],"entity_ids":["ent_8a8a91e44930d87b222367e337327e34","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"The speaker introduces conversational exchange interfaces like ChatGPT that allow two-way dialogue."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_18","window_index":18,"start_ms":221499,"end_ms":236499,"duration_ms":15000,"visual_description":"The presenter discusses search-integrated models as an orange Retrieval tier with Bing appears on the graphic. The host gestures toward the orange Retrieval tier of the diagram illustrating Bing Chat's role.","on_screen_text":"INTERFACE\nBing\nChatGPT\nInstructGPT\nGPT-3\nRETRIEVAL\nEXCHANGE\nINSTRUCT\nBASE MODEL\nOpenAI","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["ChatGPT","Microsoft Copilot"],"entity_ids":["ent_8a8a91e44930d87b222367e337327e34","ent_c39bc3da8cc8f60b104c2c2cd5f4fab5"],"fused_summary":"The presenter highlights retrieval interfaces like Bing Chat that merge conversational models with live web information."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_19","window_index":19,"start_ms":236499,"end_ms":250129,"duration_ms":13630,"visual_description":"A red top tier labeled Action is introduced at the peak of the pyramid while the host speaks. The presenter raises his hands passionately as the graphic tops out with the red Action apex tier.","on_screen_text":"INTERFACE\nBing\nChatGPT\nInstructGPT\nGPT-3\nRETRIEVAL\nEXCHANGE\nINSTRUCT\nBASE MODEL\nOpenAI\nACTION\n...","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["ChatGPT","Microsoft Copilot"],"entity_ids":["ent_8a8a91e44930d87b222367e337327e34","ent_c39bc3da8cc8f60b104c2c2cd5f4fab5"],"fused_summary":"The speaker concludes that the ultimate interface stage is the action model capable of performing tasks using account credentials."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_20","window_index":20,"start_ms":250129,"end_ms":263359,"duration_ms":13230,"visual_description":"A presenter speaks in front of slides explaining AI interface progression from prompt instructions to direct actions. The presenter gestures beside a pyramid graphic illustrating levels of AI interface layers. The presenter talks beside slides showing Replit Ghostwriter and Bing in Edge interfaces.","on_screen_text":"INTERFACE\nACTION\nRETRIEVAL\nEXCHANGE\nINSTRUCT\nBASE MODEL\n...\nBing\nChatGPT\nInstructGPT\nGPT-3\nOpenAI\nREPLIT GHOSTWRITER CHAT\nBING IN EDGE\nConversational AI in your IDE\nTrying asking questions about your code or follow-ups about a previous message.\nclass User {\nconstructor(name, username) {\nthis.name = name\nthis.username = username\n}\n// TO DO: implement stack\nYou\nUsing a JavaScript class, implement a Stack data structure using an array.\nGhostwriter\nSure! Here is a JavaScript class that\nimplements a Stack using an array:\nclass Stack {\nreplit","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"medium","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Replit Ghostwriter"],"entity_ids":["ent_5c22b4ce12acf101f0f5e2aac3d1bd72","ent_7e293f38-b099-4437-ad3e-3688f679ad40","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"The speaker explains how AI interfaces are shifting towards automated actions, highlighting examples like Replit's code generator."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_21","window_index":21,"start_ms":263359,"end_ms":273083,"duration_ms":9724,"visual_description":"The speaker continues describing interface evolution in front of screenshots of Replit and Bing in Edge. The host gestures while explaining conversational and retrieval interface integrations on screen.","on_screen_text":"INTERFACE\nREPLIT GHOSTWRITER CHAT\nBING IN EDGE\nConversational AI in your IDE\nTrying asking questions about your code or follow-ups about a previous message.\nclass User {\nconstructor(name, username) {\nthis.name = name\nthis.username = username\n}\n// TO DO: implement stack\nYou\nUsing a JavaScript class, implement a Stack data structure using an array.\nGhostwriter\nSure! Here is a JavaScript class that implements a Stack using an array:\nclass Stack {\nreplit","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Microsoft Edge"],"entity_ids":["ent_4e8516541fed51ef3fa0189dd2eb5f03","ent_7e293f38-b099-4437-ad3e-3688f679ad40"],"fused_summary":"The presenter notes how platforms like Edge combine retrieval and conversational models as AI user interfaces evolve."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_22","window_index":22,"start_ms":273083,"end_ms":287919,"duration_ms":14836,"visual_description":"The speaker presents a cost comparison chart for analog, digital, and AI video production. The presenter stands in front of a chart comparing production costs across Analog, Digital, and AI eras.","on_screen_text":"COST\nCOST OF MAKING VIDEOS\nAnalog\nDigital\nAI\n$200,000\n$2,000\n$20","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker discusses how AI drastically lowers the financial cost of producing video content."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_23","window_index":23,"start_ms":287919,"end_ms":302919,"duration_ms":15000,"visual_description":"The presenter outlines generative AI tools used across writing, sound, and video creation. The host points upward next to the video production cost comparison slide. A title card displays text about making Hollywood films with a laptop. The host speaks beside a diagram listing AI tools for scriptwriting, sound, and video generation.","on_screen_text":"COST\nCOST OF MAKING VIDEOS\nAnalog\nDigital\nAI\n$200,000\n$2,000\n$20\nWriter\nOpenAI\nGPT-3\nSound\ndescript\nreSpeeCHer\nCamera & Editing\nSynthesia","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"medium","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Synthesia","GPT-3","Descript","Respeecher"],"entity_ids":["ent_0ab7681fca07ba0e99c88574f46adb3b","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_b9376b28cef422f3d1643e31d16e32d0","ent_cb189d1f64954ed681e62e28eb218cc3","ent_d824625e4bb5e24cff90069cb8ed5cd3"],"fused_summary":"The speaker explains making videos with tools such as GPT-3, Descript, and Synthesia on a laptop."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_24","window_index":24,"start_ms":302919,"end_ms":316832,"duration_ms":13913,"visual_description":"A man addresses the camera in front of graphics comparing college, online courses, and AI costs. A man addresses the camera as graphics display comparison infographics on production tools and intelligence costs.","on_screen_text":"COST\nCOST OF INTELLIGENCE\nCollege\nOnline\nAI\n$250,000\n$15,000\n$20","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Synthesia","GPT-3","Descript"],"entity_ids":["ent_0ab7681fca07ba0e99c88574f46adb3b","ent_33ce2df73e75010d60cb1724225821bc","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_b9376b28cef422f3d1643e31d16e32d0","ent_cb189d1f64954ed681e62e28eb218cc3"],"fused_summary":"A speaker explains generative AI tools and how AI dramatically lowers the cost of education compared to traditional universities."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_25","window_index":25,"start_ms":316832,"end_ms":327458,"duration_ms":10626,"visual_description":"The presenter continues discussing education and intelligence access costs in front of the cost comparison slide. The presenter gestures and speaks while positioned beside an infographic showing the cost of intelligence.","on_screen_text":"COST\nCOST OF INTELLIGENCE\nCollege\nOnline\nAI\nYale\nCornell\nBROWN\nPRINCETON\ncoursera\nudemy\nedX\nswayam\nGoogle\nShare\nMasterClass\n$250,000\n$15,000\n$20","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"friendly"},"products_mentioned":["ChatGPT"],"entity_ids":["ent_8a8a91e44930d87b222367e337327e34"],"fused_summary":"The speaker contrasts traditional education costs with using ChatGPT for low-cost access to global knowledge."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_26","window_index":26,"start_ms":327458,"end_ms":338849,"duration_ms":11391,"visual_description":"The host talks in front of a chart illustrating declining compute costs and exponentially increasing data volume. The speaker explains the surge in data alongside a graph displaying model progression over time against cost curves.","on_screen_text":"COST\nDATA\nVolume\nTime\nELMo (94M)\nBERT-Large (340M)\nGPT-2 (1.5B)\nT5 (11B)\nMegatron-LM (8.3B)\nGPT-3 (175B)\nTuring-NLG (17B)\nMegatron-Turing NLG (530B)\nAmazon SageMaker\nTensorFlow\nORACLE\nORACLE ANALYTICS CLOUD\ncerebras\nmosaicML","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["TensorFlow","Amazon SageMaker","GPT-3"],"entity_ids":["ent_34c72865d8528167ad8904f36e71bd76","ent_627b0a177ec5ebe37949f93c7adb536b","ent_b9376b28cef422f3d1643e31d16e32d0"],"fused_summary":"The presenter details how massive amounts of computer data now power advanced AI modeling."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_27","window_index":27,"start_ms":338849,"end_ms":347000,"duration_ms":8151,"visual_description":"The presenter gestures alongside a graph showing the plummeting cost of compute and rising volume of data. The presenter continues his presentation next to an analytical plot showing cost reduction curves over time.","on_screen_text":"COST\nDATA\nVolume\nGPT-3 (175B)\nMegatron-LM (8.3B)\nMegatron-Turing NLG (530B)\nTuring-NLG (17.5B)\nT5 (11B)\nGPT-2 (1.5B)\nBERT-Large (340M)\nELMo (94M)\nTime\nAmazon SageMaker\nTensorFlow\nORACLE ORACLE ANALYTICS CLOUD\ncerebras\nmosaicML","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker points out how falling compute and energy costs make training and deploying generative models cheaper."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_28","window_index":28,"start_ms":347000,"end_ms":361758,"duration_ms":14758,"visual_description":"A man in a patterned shirt speaks in front of a graphic comparing physical atoms in warehouses versus digital bits on computers under the header Cost. A presenter speaks in the lower left corner against a slide comparing warehouse storage for atoms and digital work for bits.","on_screen_text":"COST\nATOMS\nVS\nBITS","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":[],"fused_summary":"A presenter explains why work done on computers using bits is being automated first compared to physical items in warehouses that are not yet connected to the internet."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_29","window_index":29,"start_ms":361758,"end_ms":375146,"duration_ms":13388,"visual_description":"A presenter in a patterned shirt explains algorithmic models beside a diagram showing input, an algorithm flowchart box, and output. The presenter stands in front of a white slide outlining algorithmic models from 1990 to 2000 with an input-algorithm-output flowchart.","on_screen_text":"QUALITY\n1990-2000\nALGORITHMIC MODEL\nHUMAN\nDIGITIZATION\nInput\nSet of rules of obtain the expected output from the given input.\nAlgorithm\nOutput","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"thoughtful"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker introduces the evolution of model quality, noting that algorithmic models began developing in the 1990s."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_30","window_index":30,"start_ms":375146,"end_ms":384292,"duration_ms":9146,"visual_description":"The presenter gestures and speaks beside a slide illustrating the 1990 to 2000 era of digitization and basic algorithmic models. The presenter continues talking in front of the slide demonstrating input and output algorithmic processing.","on_screen_text":"QUALITY\n1990-2000\nALGORITHMIC MODEL\nHUMAN\nDIGITIZATION\nInput\nSet of rules of obtain the expected output from the given input.\nAlgorithm\nOutput","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The presenter discusses early digitization practices such as computer accounting, basic calculations, filtering, and input-output methods."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_31","window_index":31,"start_ms":384292,"end_ms":398083,"duration_ms":13791,"visual_description":"The presenter gestures expressively as a slide showing the progression to the 2000 to 2010 data science and big data era appears. The presenter gestures with both hands in front of an educational infographic displaying data science architectures from 2000 to 2010.","on_screen_text":"QUALITY\n1990-2000\nALGORITHMIC MODEL\nHUMAN\nDIGITIZATION\n2000-2010\nDATA SCIENCE\nBIG DATA\nSource Data\nData Replication & Ingestion\nData Lake & Data Warehouse Automation\nGoverned Data Catalog\nData Analytics Applications","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"medium","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker explains the transition into the 2000s era of data science driven by big data, online applications, and connected computing."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_32","window_index":32,"start_ms":398083,"end_ms":411083,"duration_ms":13000,"visual_description":"A man in a patterned shirt presents in front of an educational infographic detailing AI progress. A presenter speaks beside a digital slide transitioning into machine learning and ImageNet classification data.","on_screen_text":"QUALITY\n1990-2000\nALGORITHMIC MODEL\nHUMAN\nDIGITIZATION\n2000-2010\nDATA SCIENCE\nBIG DATA\nSource Data\nData Replication & Ingestion\nData Lake & Data Warehouse Automation\nGoverned Data Catalog\nData Analytics Applications\n2010-2020\nMACHINE LEARNING\nHUMAN + COMPUTER\nDEEP LEARNING\nImage Classification Accuracy, 2010-2017\n100%\n90%\n80%\n70%\n2010 2011 2012 2013 2014 2015 2016 2017\nImageNet\nHuman accuracy\nCAT\nDOG\nCOW\nHidden Layer\nFlatten","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":["ent_2b2c6c81-4a5c-4fd1-8123-1f98a924dc1e"],"fused_summary":"The presenter explains the evolution from complex algorithms and data science to machine learning in the mid-2010s."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_33","window_index":33,"start_ms":411083,"end_ms":426083,"duration_ms":15000,"visual_description":"The presenter speaks next to a graphic showing image classification accuracy over time. The presenter describes deep learning milestones beside a chart tracking ImageNet accuracy.","on_screen_text":"QUALITY\n1990-2000\nALGORITHMIC MODEL\nHUMAN\nDIGITIZATION\n2000-2010\nDATA SCIENCE\nBIG DATA\n2010-2020\nMACHINE LEARNING\nHUMAN + COMPUTER\nDEEP LEARNING\nImage Classification Accuracy, 2010-2017\n100%\n90%\n80%\n70%\n2010\n2011\n2012\n2013\n2014\n2015\n2016\n2017\nImageNet\nHuman accuracy\nCAT\nDOG\nCOW\nFlatten\nHidden Layer","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":["ent_2b2c6c81-4a5c-4fd1-8123-1f98a924dc1e"],"fused_summary":"The speaker details deep learning advances in image and audio classification surpassing human accuracy."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_34","window_index":34,"start_ms":426083,"end_ms":441083,"duration_ms":15000,"visual_description":"The presenter continues in studio lighting as the slide transitions to show modern generative AI models. The speaker concludes the discussion on deep learning beside the image classification slide. The slide updates to reveal a 2020s generative AI models column as the presenter speaks.","on_screen_text":"QUALITY\n1990-2000\nALGORITHMIC MODEL\nHUMAN\nGITIZATION\n2000-2010\nDATA SCIENCE\nBIG DATA\n2010-2020\nMACHINE LEARNING\nHUMAN + COMPUTER\nDEEP LEARNING\nNEURAL NETWORKS\nImage Classification\nAccuracy, 2010-2017\n100%\n90%\n80%\n70%\n2010 2011 2012 2013 2014 2015 2016 2017\n- ImageNet\n- Human accuracy\nCAT\nDOG\nCOW\n2020-...\nAI MODELS\nGENERATIVE\nGPT","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["GPT"],"entity_ids":["ent_2b2c6c81-4a5c-4fd1-8123-1f98a924dc1e","ent_ae818f1d-637b-4bed-b9a0-905e3159abaf"],"fused_summary":"The presenter explains the transition from deep learning into generative AI and transformer architectures."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_35","window_index":35,"start_ms":441083,"end_ms":455750,"duration_ms":14667,"visual_description":"The presenter gestures animatedly in front of a diagram illustrating generative AI eras. The presenter gestures with his hands while explaining AI attention mechanisms beside the graphic.","on_screen_text":"QUALITY\n1990-2000\nALGORITHMIC\nMODEL\nHUMAN\nITIZATION\nDIGITIZATION\n2000-2010\nDATA SCIENCE\nBIG DATA\n2010-2020\nMACHINE\nLEARNING\nHUMAN +\nCOMPUTER\nNEURAL\nNETWORKS\n2020-...\nAI MODELS\nGENERATIVE\nGPT","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"medium","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["GPT"],"entity_ids":["ent_ae818f1d-637b-4bed-b9a0-905e3159abaf"],"fused_summary":"The speaker explains how self-attention mechanisms in transformer models allow AI to learn independently and achieve higher quality."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_36","window_index":36,"start_ms":455750,"end_ms":465255,"duration_ms":9505,"visual_description":"A presenter talks beside a Venn diagram depicting key drivers of the AI paradigm shift. The presenter stands in front of a diagram illustrating human-level intelligence, chat UX, and free pricing intersecting at ChatGPT.","on_screen_text":"KEY DRIVERS OF AI PARADIGM SHIFT\nHUMAN LEVEL INTELLIGENCE\nChatGPT\nCHAT\nFREE","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["ChatGPT"],"entity_ids":["ent_8a8a91e44930d87b222367e337327e34"],"fused_summary":"The speaker outlines three key factors fueling the AI paradigm shift while standing beside a graphic illustration."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_37","window_index":37,"start_ms":465255,"end_ms":473255,"duration_ms":8000,"visual_description":"The presenter explains the success of conversational AI models alongside the Venn diagram. The presenter gestures while discussing how convergence of chat interfaces and free tools created ChatGPT's success.","on_screen_text":"KEY DRIVERS OF AI PARADIGM SHIFT\nHUMAN LEVEL\nINTELLIGENCE\nChatGPT\nCHAT\nFREE","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["ChatGPT"],"entity_ids":["ent_8a8a91e44930d87b222367e337327e34"],"fused_summary":"The presenter concludes that human intelligence quality, conversational interfaces, and near-zero cost made ChatGPT successful."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_38","window_index":38,"start_ms":473255,"end_ms":486792,"duration_ms":13537,"visual_description":"The presenter transitions to human evolution stages displayed sequentially as cards on screen. The presenter discusses passive foragers and laborers with graphics of a hunter and a farmer appearing. A third card titled Creator featuring an industrial machinist joins the slide as the speaker discusses machines.","on_screen_text":"PASSIVE\nLABORER","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"thoughtful"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker describes how humans evolved from passive foragers into specialized laborers and later creators building machines."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_39","window_index":39,"start_ms":486792,"end_ms":497750,"duration_ms":10958,"visual_description":"The presenter continues explaining human roles advancing from creators to narrators directing self-building machines. The presenter gestures alongside the evolutionary stages slide illustrating the transition to narrators.","on_screen_text":"PASSIVE\nLABORER\nCREATOR","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker contrasts creator roles that operated machines with the narrator role where humans simply specify what AI should build."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_40","window_index":40,"start_ms":497750,"end_ms":508689,"duration_ms":10939,"visual_description":"A man gestures in front of an educational graphic illustrating human historical stages alongside a transition to generative AI modalities. The presenter stands in front of a slide displaying four stages: Passive, Laborer, Creator, and Narrator. The slide shifts to showcase multiple modalities including Language, Images, Audio, Video, Code, and Science.","on_screen_text":"PASSIVE\nLABORER\nCREATOR\nNARRATOR\nLANGUAGE\nIMAGES\nAUDIO\nVIDEO\nCODE\nSCIENCE","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker explains the transition to humans becoming narrators who describe what machines should build, noting this applies beyond language."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_41","window_index":41,"start_ms":508689,"end_ms":519994,"duration_ms":11305,"visual_description":"The presenter gestures enthusiastically beside a grid showing AI domains including Language, Images, Audio, Video, Code, and Science. The presenter continues gesturing enthusiastically beside the six-panel diagram of AI modalities. The presenter raises a finger as the display begins shifting to a list of language models.","on_screen_text":"LANGUAGE\nIMAGES\nAUDIO\nVIDEO\nCODE\nSCIENCE","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker outlines the various generative AI modeling fields to be explored, including image, audio, video, code, and science."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_42","window_index":42,"start_ms":519994,"end_ms":530678,"duration_ms":10684,"visual_description":"The presenter explains key large language models displayed on an informational grid showing models, parameter counts, and companies. The presenter describes leading AI language models while standing in front of a chart comparing ChatGPT, Bing Chat, Bard, and others.","on_screen_text":"LANGUAGE\n175B\nChatGPT & GPT-3\nOpenAI\nBing Chat\nMicrosoft\n137B\nBard\nGoogle\n52B\nClaude\nANTHROP\\C\n260B\nEarnie Bot\nBaido\n27B\nCarper AI\nAlibaba.com\naws\n70B\nstability.ai\n120B\nGalactica\nMeta","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":["ChatGPT","GPT-3","Microsoft Copilot"],"entity_ids":["ent_8a8a91e44930d87b222367e337327e34","ent_9016355d-416c-4e35-b834-35f656db4dad","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_b9376b28cef422f3d1643e31d16e32d0","ent_c39bc3da8cc8f60b104c2c2cd5f4fab5"],"fused_summary":"The speaker introduces prominent large language models, highlighting ChatGPT and Bing Chat with its retrieval capabilities."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_43","window_index":43,"start_ms":530678,"end_ms":545678,"duration_ms":15000,"visual_description":"The presenter discusses AI industry developments in front of a graphic listing major language models from Google, Anthropic, Meta, and OpenAI. The presenter gestures toward the model chart while explaining migration of AI talent between Google, OpenAI, and Anthropic.","on_screen_text":"LANGUAGE\n175B\nChatGPT & GPT-3\nOpenAI\nBing Chat\nMicrosoft\n137B\nBard\nGoogle\n52B\nClaude\nANTHROP\\C\n260B\nEarnie Bot\nBaidu\n27B\nCarper AI\nAlibaba.com\n70B\nstability.ai\n120B\nGalactica\nMeta","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"thoughtful"},"products_mentioned":["Claude","Bard"],"entity_ids":["ent_2217b5322de54820ef4590d798a43b9d","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_64b91d978a6442b4b8abcbb017f0d16b","ent_91b7c497-1d72-471e-8986-108b68b1a8bf","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"The speaker mentions Google's Bard and Anthropic's Claude, pointing out the talent movement between Google, OpenAI, and Anthropic."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_44","window_index":44,"start_ms":545678,"end_ms":557917,"duration_ms":12239,"visual_description":"A man in a patterned shirt presents in front of a slide displaying logos and parameter counts of various large language models. The host gestures and speaks in a studio setup with an on-screen graphic comparing several commercial AI language models.","on_screen_text":"LANGUAGE\n175B\nChatGPT & GPT-3\nOpenAI\nBing Chat\nMicrosoft\n137B\nBard\nGoogle\n52B\nClaude\nANTHROP\\C\n260B\nEarnie Bot\nBaidu\\u767e\\u5ea6\n27B\nCarper AI\nAlibaba.com\naws\n70B\nstability.ai\n120B\nGalactica\nMeta","facets":{"composition":"single subject","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Claude","Bard","Earnie Bot","ChatGPT","GPT-3","Microsoft Copilot"],"entity_ids":["ent_2217b5322de54820ef4590d798a43b9d","ent_45175450-c86e-4de0-8553-90b22143a40c","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_64b91d978a6442b4b8abcbb017f0d16b","ent_84dc3c1286160f2a4032f88b3e8b9cfb","ent_8a8a91e44930d87b222367e337327e34","ent_91b7c497-1d72-471e-8986-108b68b1a8bf","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_b9376b28cef422f3d1643e31d16e32d0","ent_c39bc3da8cc8f60b104c2c2cd5f4fab5"],"fused_summary":"The speaker notes talent moving between AI companies and highlights massive parameter counts across prominent language models."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_45","window_index":45,"start_ms":557917,"end_ms":572917,"duration_ms":15000,"visual_description":"The presenter stands beside a graphic breaking down data sources used to train large AI models. The host speaks and gestures in front of an educational infographic listing training data sets like Reddit, Twitter, Wikipedia, and Common Crawl.","on_screen_text":"LANGUAGE\nOpenAI\nDATA\nWIKIPEDIA\nommon Crawl\n45 TB","facets":{"composition":"single subject","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["Substack","Reddit","Wikipedia","Common Crawl","X"],"entity_ids":["ent_2b325911-9859-430a-ab1f-c2eee5f11f8e","ent_53a106957af99f597540fa15fd63c628","ent_7290ff888c048498e298764f503ec0c5","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_bfc65d0b-b508-403a-890b-96522652b6a5","ent_c1eb0a5ad5e25af6e52c706b836ed2e6"],"fused_summary":"The speaker uses OpenAI to explain the training process, listing diverse web data sources required for model creation."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_46","window_index":46,"start_ms":572917,"end_ms":587917,"duration_ms":15000,"visual_description":"The presenter continues in front of the slide as a new training column shows Azure and hardware specs. The host speaks beside the data column graphic summarizing 45 terabytes of source material. The host gestures as a new training graphic appears displaying Azure, server hardware, and cost estimations.","on_screen_text":"LANGUAGE\nOpenAI\nDATA\nWIKIPEDIA\nmon Crawl\n45 TB\nTRAINING\nAzure\n$4.6M","facets":{"composition":"single subject","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"thoughtful"},"products_mentioned":["A100","Microsoft Azure"],"entity_ids":["ent_1003554c9ffa7666633fb3829d422dcf","ent_95f91400-3cbe-4754-b031-62b05747dab4","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_c04aabf51b5214dd1d4982a47b375c79"],"fused_summary":"The speaker details the training phase, explaining that 45 terabytes of data are trained on NVIDIA A100 processors at a multi-million-dollar cost."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_47","window_index":47,"start_ms":587917,"end_ms":600432,"duration_ms":12515,"visual_description":"The presenter stands in front of the slide showing stages from training data to GPT-3 model output and inference. The host describes the resulting GPT-3 model card and subsequent inference column on the presentation slide.","on_screen_text":"LANGUAGE\nOpenAI\nDATA\nTRAINING\nMODEL\nWikipedia\nommon Crawl\nAzure\nGPT-3\n45 TB\n$4.6M\n800 GB\nINFERENCE\n$0.02-8/result","facets":{"composition":"single subject","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["GPT-3","Microsoft Azure"],"entity_ids":["ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_b9376b28cef422f3d1643e31d16e32d0","ent_c04aabf51b5214dd1d4982a47b375c79"],"fused_summary":"The speaker explains how massive data and training culminate in an 800-gigabyte GPT-3 model and discusses running inference."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_48","window_index":48,"start_ms":600432,"end_ms":610542,"duration_ms":10110,"visual_description":"The host presents a slide detailing language model data, training, model size, and inference costs in a studio setup. Medium shot of the presenter speaking in front of an educational infographic explaining AI language model training and inference pipelines.","on_screen_text":"LANGUAGE\nDATA\nTRAINING\nMODEL\nINFERENCE\nOpenAI\nAzure\nWIKIPEDIA\nCommon Crawl\n45 TB\n$4.6M\nGPT-3\n800 GB\n$0.02-8/result","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Wikipedia","ChatGPT","GPT-3","Common Crawl","Microsoft Azure"],"entity_ids":["ent_7290ff888c048498e298764f503ec0c5","ent_8a8a91e44930d87b222367e337327e34","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_b9376b28cef422f3d1643e31d16e32d0","ent_bfc65d0b-b508-403a-890b-96522652b6a5","ent_c04aabf51b5214dd1d4982a47b375c79"],"fused_summary":"The presenter explains the inference stage and per-query operational costs associated with OpenAI's ChatGPT."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_49","window_index":49,"start_ms":610542,"end_ms":625208,"duration_ms":14666,"visual_description":"The host gestures toward a large spreadsheet displayed behind him comparing various AI models. Medium shot of the presenter speaking and gesturing as a spreadsheet listing language models appears on screen.","on_screen_text":"LANGUAGE\n2023 LifeArchitect.ai data (shared)\nFile\nEdit\nView\nInsert\nFormat\nData\nTools\nExtensions\nHelp\nComment only\nShare\nA1\nfx\nModel\nLab\nSelected playgrounds\nParameters (B)\nTokens (B)\nRatio T:P (Chinchilla scaling)\nTraining dataset\nAnnounced\nPublic?\nReleased\nPaper/ Repo\nNotes\nGPT-4\nOpenAI\nTBA\nDevs from DeepMind\nInflection\nModels <18 with vision CoT\nLuminous Supreme\nAleph Alpha\nhttps://app.aleph-alpha\n70\nFeb/2023\n'Control' means instruction tuned\nMultimodal-CoT\nAmazon\nhttps://github.com/ami\n0.738\nFLAME\nMicrosoft\n0.06\n9\n150:1\nJan/2023\nTS for Excel formulas, very small 60M params, \"We\nMed-PaLM\nGoogle & DeepMind\n540\n780\n2:1\nDec/2022\nCollab between Google & DeepMind. Makes 1% less\nOPT-IML\nMeta AI\nhttps://github.com/face\n175\n300\nInstruct\nRL-CAI\nAnthropic\n52\nN/A\nRLAIF=reinforcement learning with AI feedback\nERNIE-Code\nBaidu\n0.56\nChatGPT\nhttps://chat.openai.com\nNov/2022\nInstruct with strict policies (\"extremely limited\")\nGPT-JT\nTogether\nhttps://huggingface.co/\n6\nRWKV-4\nEleutherAI\nhttps://github.com/\n14\n332\n24:1\nRNN, not transformer: https://www.reddit.com/r/\nGalactica\nhttps://galactica.org/\n120\n450\n4:1\nscientific only\nSED\nDeepMind\n0.42\nSED 420M (diffusion text model)\nmT0\nBigScience\nhttps://github.com/bigs\n13\n1000\n77:1\nfine-tuned\nBLOOMZ\n176\n366\n3:1\nPACT\nhttps://github.com/microsoft/PACT\n0.03\nOct/2022\nTrained on \"STB data, 2GB model download. See Ci\nFlan-T5\nGoogle Researc\n11\n134\n13:1\nT5=348 tokens + LM-adapted T5 as 100B tokens\nFlan-PaLM\nGoogle Research\nU-PaLM\nVIMA\nNVIDIA\nOpen: https://vimalabs.\n0.2\nWeiLM\nWeChat\nhttps://welm.weixin.qq\n10\n30:1\nSep/2022\n13% English tokens and 87% Chinese\nCodeGeeX\nTsinghua\n850\n66:1\nSparrow\n1400\n20:1\nChatbot as a fine-tuned version of Chinchilla 70B\nPaLI\nGoogle\n17\nPaLM Vision model, new datasets of 10B multilingua\nZ-Code++\n0.71\n500\n705:1\nAug/2022\nabstractive text summarization, 710M, outperforms\nAtlas\n40\nBlenderBot 3\nblenderbot.ai (US only)\nGLM-130B\nTsinghua & Zhi\n130\n400\n50% English (2008 tokens), so included here\nAlexaTM 20B\nAmazon Alexa\nGithub (train/deploy)\n20\n50:1\nWikipedia and mc4 only. seq2seq\n6.9B FIM\n6.9\n100\n15:1\nJul/2022\nSeveral models: 8 sizes, NLP, Code, FIM/non-FIM. 10\n'monorepo-Tran\nGoogle Brain\n0.5\nUnnamed. Writes >3% of internal google code.\nPanGu-Coder\nHuawei\n2.6\nPython via GH\nNLLB\n54.5\n54.5B MOE, 3.3B dense. 200+ languages\nJ-1 RrBG\nAI21\nask-rbg.ai\n178\nJ-1 fine-tuned with RBG law corpus\nBLOOM (tr11-17)\n350\nMinerva\n818.5\nJun/2022\nPaLM finetuned on LaTeX/ArXiv maths\nGODEL-XL\n2.7\nGPT-3 175B in paper, GPT-J 2.7B released\nYaLM 100B\nYandex\nMegatron-LM clone, Russian/English: https://mediu\nUnified-IO\nAllen AI\nLimited demo\n2.8\nBased on T5. Demo only\nSummary: Current models (Feb/2023)\nChatGPT achievements\nLabs\nContents of GPT-3/Pile/Megatron/CC\nContents of Chinese models\nLata template","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"medium","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["YouTube","lifearchitect.ai"],"entity_ids":["ent_b987acd01400347ba9932dfd412e2c65","ent_f122deea66464e885edb68ce1701437b"],"fused_summary":"The presenter highlights alternative language models and directs viewers to the LifeArchitect database and YouTube channel."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_50","window_index":50,"start_ms":625208,"end_ms":640203,"duration_ms":14995,"visual_description":"The presenter transitions to a slide displaying six key application categories for large language models. The host concludes discussing the model comparison sheet before the graphic transitions. The screen updates to display a slide highlighting six language model use cases alongside the presenter.","on_screen_text":"LANGUAGE\nText Generation\nSummarization\nTranslation\nSearch\nIntelligent Assistant\nAnalysis","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker introduces diverse use cases for language models, focusing first on prompt-based text generation and summarization."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_51","window_index":51,"start_ms":640203,"end_ms":655203,"duration_ms":15000,"visual_description":"The presenter continues discussing language model capabilities in front of the application category slide. Medium shot of the presenter speaking about document summarization and translation while framed beside a use-case graphic.","on_screen_text":"LANGUAGE\nText Generation\nSummarization\nTranslation\nSearch\nIntelligent Assistant\nAnalysis","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":["Google Translate"],"entity_ids":["ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_cefdeb8b-eae6-4e47-9e30-0937d7306c09"],"fused_summary":"The host elaborates on using language models for document interrogation and compares translation quality to Google Translate."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_52","window_index":52,"start_ms":655203,"end_ms":668367,"duration_ms":13164,"visual_description":"A presenter speaks in the lower left corner against a slide showcasing language AI use cases including translation and search. The presenter discusses language model applications while a digital slide displays various tasks such as text generation and translation.","on_screen_text":"LANGUAGE\nText Generation\nSummarization\nTranslation\nSearch\nIntelligent Assistant\nAnalysis","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Microsoft Bing","Google Translate"],"entity_ids":["ent_a9176dd183166c10af1ebcfc68c03585","ent_cefdeb8b-eae6-4e47-9e30-0937d7306c09"],"fused_summary":"The presenter explains how language models outperform traditional tools like Google Translate and how search engines like Bing integrate conversational capabilities."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_53","window_index":53,"start_ms":668367,"end_ms":678033,"duration_ms":9666,"visual_description":"The presenter continues speaking beside the slide, highlighting intelligent assistant capabilities. The host gestures while explaining how natural language guidance powers intelligent digital assistants.","on_screen_text":"LANGUAGE\nText Generation\nSummarization\nTranslation\nSearch\nIntelligent Assistant\nAnalysis","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker details how language models act as intelligent assistants that navigate user interfaces using natural language prompts."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_54","window_index":54,"start_ms":678033,"end_ms":686958,"duration_ms":8925,"visual_description":"The presenter explains data analysis capabilities against the same six-panel slide layout. The presenter gestures towards the analysis tile on the graphic interface.","on_screen_text":"LANGUAGE\nText Generation\nSummarization\nTranslation\nSearch\nIntelligent Assistant\nAnalysis","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"friendly"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The host discusses using language models to derive data across websites to assist in consumer decision-making."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_55","window_index":55,"start_ms":686958,"end_ms":698833,"duration_ms":11875,"visual_description":"The presentation transitions to a Jasper templates dashboard, shows a robot mascot graphic, and cuts to a laptop screen demonstrating Jasper writing software. The host speaks beside an on-screen preview of the Jasper templates selection screen. A 3D humanoid robot avatar representing Jasper is displayed alongside the host. An over-the-shoulder shot focuses on a laptop screen reading 'Meet Jasper, The Future of Writing Blog Articles.'","on_screen_text":"LANGUAGE\nDashboard\nTemplates\nRecipes\nCommunity\nDocuments\nTrash\nAl outputs\nFavorites\nSearch templates\nAll\nFrameworks\nEmail\nWebsite\nBlog\nAds\nEcommerce\nSocial Media\nGoogle\nVideo\nNew\nSEO\nPRO\nLong-form assistant\nLet Jasper help you write long-form articles and more from start to finish.\nAIDA Framework\nUse the oldest marketing framework in the world. Attention, Interest, Desire, Action.\nContent Improver\nTake a piece of content and rewrite it to make it more interesting, creative, and engaging.\nFacebook Ad Primary Text\nCreate high converting copy for the \"Primary Text\" section of your Facebook ads.\nGoogle Ads Headline\nCreate high converting copy for the \"Headlines\" section of your Google Ads.\nMarketing Angles\nBrainstorm different angles to add vibrancy to your marketing.\nVideo Description - YouTube\nCreate unique descriptions for Youtube videos that rank well in search.\nEmail Subject Lines\nWrite compelling email subject lines that get readers to open.\nUpdated 6d ago\nReal Estate Listing - Residential\nCreative captivating real estate listings that sell homes quickly.\nPAS Framework\nProblem-Agitate-Solution. A valuable framework for creating new marketing copy ideas.\nPersonal\nJasper\nSat Sep 18 11:02 AM\nBrowser\nFile\nEdit\nView\nGo\nWindow\nLeggings\nblog.makingsocialaccessiblethroughblogs.clientportal/u/daoufdiah07655AFknnn?A\nDrafts\nShare\nMeet Jasper,\nThe Future of Writing\nBlog Articles.","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"fast cut","camera_type_hint":"screen recording","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Jasper"],"entity_ids":["ent_b2a51b709e3456866a6d6431306b4a1a"],"fused_summary":"The host introduces Jasper AI for automated writing tasks, accompanied by music as a promotional interface demo appears on a laptop screen."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_56","window_index":56,"start_ms":698833,"end_ms":713167,"duration_ms":14334,"visual_description":"A promotional video demonstrates the Jasper AI software interface interspersed with scenes of an office worker and a small robot assistant. A woman sits at an office desk looking attentively at her desktop monitor. A screen recording displays the Jasper AI app dashboard with various content generation options. A small white and blue tabletop robot animates and speaks next to the computer. The woman at the desk turns toward the robot companion while smiling. The miniature robot gestures with its hands while delivering creative ideas. The woman smiles with satisfaction at her computer screen as AI-generated text appears.","on_screen_text":"JASPER\nJASPER.AI\nand that's why the perfect pair of socks is like a hug","facets":{"composition":"rule of thirds","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Jasper"],"entity_ids":["ent_b2a51b709e3456866a6d6431306b4a1a"],"fused_summary":"Accompanied by upbeat music notes, a commercial illustrates how Jasper AI assists content creators through its web interface and an animated robot companion."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_57","window_index":57,"start_ms":713167,"end_ms":725250,"duration_ms":12083,"visual_description":"The presenter speaks on the left side of the screen beside a clip of Ryan Reynolds reading a script for a Mint Mobile commercial. The woman in the office reacts positively to the robot's creative suggestions. The host explains an AI script example next to video footage of Ryan Reynolds.","on_screen_text":"JOKE\nCURSE WORD\nSTILL GOING\nBIG WIRELESS\nmintmobile","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"friendly"},"products_mentioned":["Jasper","Mint Mobile"],"entity_ids":["ent_3e757939-5520-4ae9-967d-eea25453d4db","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_b2a51b709e3456866a6d6431306b4a1a","ent_e8a71dcb-19e3-41f8-bc9d-9990b2db5379"],"fused_summary":"With music playing, the speaker discusses using AI for commercial scripts while displaying Ryan Reynolds reading an OpenAI-generated ad for Mint Mobile."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_58","window_index":58,"start_ms":725250,"end_ms":737167,"duration_ms":11917,"visual_description":"The host describes how Ryan Reynolds evaluated an AI-generated script, shown on screen reviewing a piece of paper in an advertisement. The host gestures while explaining AI scriptwriting alongside a video inset of Ryan Reynolds reading.","on_screen_text":"mintmobile\nJOKE\nCURSE WORD\nSTILL GOING\nBIG WIRELESS","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"friendly"},"products_mentioned":["ChatGPT","Mint Mobile"],"entity_ids":["ent_3e757939-5520-4ae9-967d-eea25453d4db","ent_8a8a91e44930d87b222367e337327e34","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_e8a71dcb-19e3-41f8-bc9d-9990b2db5379"],"fused_summary":"The host explains how Ryan Reynolds tested ChatGPT to create a commercial ad while Reynolds is shown reading the printed script."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_59","window_index":59,"start_ms":737167,"end_ms":751417,"duration_ms":14250,"visual_description":"Actor Ryan Reynolds reads aloud from a paper script generated by AI in a wood-paneled room for a Mint Mobile commercial. The presenter concludes his commentary as the video cuts focus entirely to Ryan Reynolds. Ryan Reynolds looks down at his paper and speaks enthusiastically about Mint Mobile.","on_screen_text":"mintmobile\nRYAN RENOLDS INT MOBILE\nJOKE\nCURSE WORD\nSTILL GOING\nBIG WIRELESS\nRYAN REYNOLDS\nMint Mobile Owner & User\nRYAN RENOLDS MINT MOBILE\nRYAN RENOLDS","facets":{"composition":"centered","lighting_class":"indoor natural","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"amused"},"products_mentioned":["ChatGPT","Mint Mobile"],"entity_ids":["ent_3e757939-5520-4ae9-967d-eea25453d4db","ent_8a8a91e44930d87b222367e337327e34","ent_e8a71dcb-19e3-41f8-bc9d-9990b2db5379"],"fused_summary":"To the sound of soft background music and ukulele, Ryan Reynolds reads a ChatGPT-written script delivering a humorous promotional pitch for Mint Mobile."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_60","window_index":60,"start_ms":751417,"end_ms":762417,"duration_ms":11000,"visual_description":"A man sits in a wood-paneled room holding a paper script during an advertisement clip. A man wearing glasses holds a paper sheet while speaking in front of a rustic wooden wall. A light green text slide displays a promotion for Mint Mobile with terms at the bottom. The man in glasses continues reading from his script in front of the wood wall.","on_screen_text":"mintmobile\nRYAN RENOLDS\nMINT MOBILE\nJOKE\nCURSE WORD\nSTILL GOING\nBIG WIRELESS","facets":{"composition":"centered","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"amused"},"products_mentioned":["Mint Mobile"],"entity_ids":["ent_3e757939-5520-4ae9-967d-eea25453d4db","ent_e8a71dcb-19e3-41f8-bc9d-9990b2db5379"],"fused_summary":"Accompanied by measured music, Ryan Reynolds reads aloud from a paper in a Mint Mobile commercial segment before text offers a discount."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_61","window_index":61,"start_ms":762417,"end_ms":775890,"duration_ms":13473,"visual_description":"A man in a patterned collared shirt speaks in front of the Microsoft Bing web interface before an app graphic appears. The host speaks in the foreground in front of the Microsoft Bing search page featuring a mountain scene. The host stands beside a mobile chat interface screen showing a prompt about arts and crafts.","on_screen_text":"Microsoft Bing\nChat\nWork\nImages\nVideos\nShopping\nMaps\nLANGUAGE\nAsk me anything...\n0/1000\nTakumi Morita\n8,750\nArts and crafts ideas, with instructions for a toddler using only cardboard boxes, plastic bottles, paper and string\nAsk me anything","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Microsoft Bing","Microsoft Copilot"],"entity_ids":["ent_9016355d-416c-4e35-b834-35f656db4dad","ent_a9176dd183166c10af1ebcfc68c03585","ent_c39bc3da8cc8f60b104c2c2cd5f4fab5"],"fused_summary":"Accompanied briefly by measured music, the presenter introduces Bing Chat against a backdrop of the Microsoft Bing interface."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_62","window_index":62,"start_ms":775890,"end_ms":790804,"duration_ms":14914,"visual_description":"A presentation graphic demonstrates a chatbot conversation window generating responses alongside the presenter. The screen displays generated craft ideas then switches through the Bing logo to a workout inquiry.","on_screen_text":"Arts and crafts ideas, with instructions for a toddler using only cardboard boxes, plastic bottles, paper and string\nGenerating answers for you...\nI found some arts and crafts ideas for a toddler using only cardboard boxes, plastic bottles, paper and string. Here are some examples:\nAsk me anything\nBing\nGive me a workout I can do at home","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":["Microsoft Bing"],"entity_ids":["ent_a9176dd183166c10af1ebcfc68c03585"],"fused_summary":"The presenter discusses natural search querying while a simulated Bing chat interface displays craft ideas and transitions to a workout prompt."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_63","window_index":63,"start_ms":790804,"end_ms":801275,"duration_ms":10471,"visual_description":"The presenter stands beside a chat interface detailing an exercise routine, which concludes on the Bing logo card. The host describes personalized AI routines beside an active chatbot UI before ending on the Bing logo.","on_screen_text":"Give me a workout I can do at home\nGenerating answers for you...\nSure, I can help you with that. There are many exercises you can do at home without any equipment, such as jumping jacks, burpees, squats, push-ups, lunges, and more. Here is a sample workout routine you can try at home, based on the Nerd Fitness website:\nWarm up for 5 to 10 minutes with some light cardio and stretching.\nDo 20 bodyweight squats.\nDo 10 lunges on each leg.\nAsk me anything\nCan you focus on arms\nBing","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Microsoft Bing"],"entity_ids":["ent_a9176dd183166c10af1ebcfc68c03585"],"fused_summary":"The host explains how the chat model personalizes programs with source citations as demo text streams on screen."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_64","window_index":64,"start_ms":801275,"end_ms":809275,"duration_ms":8000,"visual_description":"A presenter speaks against a grey backdrop before an AI image models slide appears. A presenter talks enthusiastically beside an empty white presentation canvas. The speaker gestures while explaining large language models. The presenter gestures as an Images category graphic displays on screen.","on_screen_text":"IMAGES\n3.5B\nDALL-E 2\nOpenAI\n20B\nParti & Imagen\nGoogle\n890M\nStable Diffusion\nstability.ai\naws\nMidjourney","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":["ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_f6677f64-31a6-4383-b60c-91f0851302c7"],"fused_summary":"The speaker discusses combining large language models with search retrieval models before transitioning to image modeling."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_65","window_index":65,"start_ms":809275,"end_ms":824198,"duration_ms":14923,"visual_description":"A presenter discusses prominent AI image generation models shown on an informational slide. The speaker stands next to an infographic listing major AI image generation models.","on_screen_text":"IMAGES\n3.5B\nDALL-E 2\nOpenAI\n20B\nParti & Imagen\nGoogle\naws\n890M\nStable Diffusion\nstability.ai\nMidjourney","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["Parti","Midjourney","Imagen","DALL·E 2","Stable Diffusion","Discord"],"entity_ids":["ent_578fdec782a2a1d689b3dcf924b3dd66","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_6f1da63ab3dee19d25200449733024a8","ent_8d208bbf2208872e313e29db2b70dc38","ent_9747295228fc6ecd042f9c2f02b5ced5","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_d3aef2ec47f98439c7a3995a0edd260e","ent_e3c10ccfce46fd7e0ed9304c98cd2977"],"fused_summary":"The presenter outlines leading image models including DALL-E 2, Stable Diffusion, and Midjourney."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_66","window_index":66,"start_ms":824198,"end_ms":834125,"duration_ms":9927,"visual_description":"A presenter discusses Google's models before transitioning to how image models like Stable Diffusion are built. The presenter discusses Google models while gesturing next to the model comparison chart. The graphic switches to a training data card titled DATA LAION-5B as the speaker explains model building.","on_screen_text":"IMAGES\n3.5B\nDALL-E 2\nOpenAI\n20B\nParti & Imagen\nGoogle\n890M\naws\nStable Diffusion\nstability.ai\nMidjourney\nstable diffusion\nDATA\nLAION-5B\ngettyimages\n5B TEXT-IMAGE PAIRS\n120 TB","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"thoughtful"},"products_mentioned":["Stable Diffusion"],"entity_ids":["ent_03017e70-f2a0-4aef-b135-4521169d0c1e","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_d3aef2ec47f98439c7a3995a0edd260e","ent_f65a2fd5-6650-4e71-afa1-6d21f7f4d875"],"fused_summary":"The presenter notes Google's large models and introduces the training data pipeline for Stable Diffusion."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_67","window_index":67,"start_ms":834125,"end_ms":848650,"duration_ms":14525,"visual_description":"A presenter explains dataset training pairs and licensing partnerships alongside a LAION graphic. The speaker gestures toward an on-screen card illustrating the LAION-5B training dataset.","on_screen_text":"IMAGES\nstable diffusion\nDATA\nLAION-5B gettyimages\n5B TEXT-IMAGE PAIRS\n120 TB","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Stable Diffusion"],"entity_ids":["ent_03017e70-f2a0-4aef-b135-4521169d0c1e","ent_d3aef2ec47f98439c7a3995a0edd260e","ent_f65a2fd5-6650-4e71-afa1-6d21f7f4d875"],"fused_summary":"The host explains how Stable Diffusion trains on massive datasets like LAION-5B and mentions its deal with Getty Images."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_68","window_index":68,"start_ms":848650,"end_ms":863650,"duration_ms":15000,"visual_description":"A man in a patterned shirt presents in front of an infographic showing Stable Diffusion data and training details. The presenter talks about Stable Diffusion training costs with graphic columns for data and hardware.","on_screen_text":"IMAGES\nstable diffusion\nDATA\nLAION-5B gettyimages\n5B TEXT-IMAGE PAIRS\n120 TB\nTRAINING\nReplying to @KennethCassel\nWe actually used 256 A100s for this per the model card,\n150k hours in total so at market price $600k\n5:46 AM Aug 28, 2022\n$600K","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":["ent_03017e70-f2a0-4aef-b135-4521169d0c1e","ent_95f91400-3cbe-4754-b031-62b05747dab4","ent_aa5296ff-1580-42d0-87db-52259cc0e257","ent_f6677f64-31a6-4383-b60c-91f0851302c7"],"fused_summary":"The speaker explains the massive training dataset and compute costs required to train Stable Diffusion."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_69","window_index":69,"start_ms":863650,"end_ms":878650,"duration_ms":15000,"visual_description":"A presenter explains AI model sizes next to a graphic that adds a column showing the final model size. The presenter discusses training expenses with the two-column slide visible. A third column appears on the slide showing the 2 gigabyte model file size as the presenter speaks.","on_screen_text":"IMAGES\nstable diffusion\nDATA\nTRAINING\nLAION-5B gettyimages\nEXT-IMAGE PAIRS\n120 TB\nEmad\n@EMostaque\nReplying to @KennethCassel\nWe actually used 256 A100s for this per the model card,\n150k hours in total so at market price $600k\n5:46 AM · Aug 28, 2022\n$600K\nMODEL\n2 GB\nINFERENCE\nDraw Things\nDiffusion Bee\nFREE","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Apple Inc."],"entity_ids":["ent_d7804d83-d910-414c-b89e-1f15c7b6503d","ent_f6677f64-31a6-4383-b60c-91f0851302c7"],"fused_summary":"The presenter discusses how Stable Diffusion compresses into a 2 gigabyte file capable of running on consumer hardware."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_70","window_index":70,"start_ms":878650,"end_ms":891549,"duration_ms":12899,"visual_description":"The host talks beside a slide that introduces an inference column with applications and use cases. The presenter gestures as the slide displays local inference apps before transitioning to image capabilities.","on_screen_text":"IMAGES\nstable diffusion\nDATA\nTRAINING\nMODEL\nINFERENCE\nLAION-5B\ngettyimages\nWe actually used 256 A100s for this per the model card, 150k hours in total so at market price $600k\nTEXT-IMAGE PAIRS\n120 TB\n$600K\n2 GB\nDraw Things\nDiffusion Bee\nFREE\nText-to-image\nImage-to-image\nOutpainting\ngive her a mohawk\nInpainting\nEnhancements & Editing\nClassification & detection","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Draw Things","Diffusion Bee"],"entity_ids":["ent_01797a1d4cf0c5a11814ae38ec92a118","ent_2194060e17a174e4c27c09f177351d70"],"fused_summary":"The speaker covers running inference locally on consumer devices and introduces various text and image generation capabilities."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_71","window_index":71,"start_ms":891549,"end_ms":906283,"duration_ms":14734,"visual_description":"The presenter gestures alongside a slide showing a grid of AI image editing techniques and applications. The presenter uses hand gestures while describing image manipulation techniques shown on the six-tile slide.","on_screen_text":"IMAGES\nText-to-image\nImage-to-image\nOutpainting\ngive her a mohawk\nInpainting\nBody AI\nIris AI\nFace AI\nSkin AI\nAccent AI\nPortrait Bokeh AI\nAtmosphere AI\nStructure AI\nSky AI\nColor Harmony\nSupercontrast\nMood\nAugmented Sky AI\nComposition\nEnhancements & Editing\nClassification & detection","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"medium","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":["Lensa AI"],"entity_ids":["ent_8a3691bd61a8532d3aaa3ef9d749e3a9"],"fused_summary":"The speaker describes image-to-image generation applications, citing Lensa AI avatars and painting expansion via outpainting."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_72","window_index":72,"start_ms":906283,"end_ms":919458,"duration_ms":13175,"visual_description":"A presenter speaks against a dark presentation backdrop displaying various AI image manipulation examples. The host speaks next to a presentation slide displaying six image generation and editing categories. The host continues explaining image AI applications with graphic categories shown on the screen.","on_screen_text":"IMAGES\nText-to-image\nImage-to-image\nOutpainting\ngive her a mohawk\nInpainting\nBody AI\nIris AI\nFace AI\nSkin AI\nAccent AI\nPortrait Bokeh AI\nAtmosphere AI\nStructure AI\nSky AI\nColor Harmony\nSupercontrast\nMood\nAugmented Sky AI\nComposition\nEnhancements & Editing\nClassification & detection","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":["ent_cd4bb989c028399afaf3ca7e59be6d0f"],"fused_summary":"The host explains generative AI image use cases such as inpainting, enhancements, and classification while standing before an illustrative slide."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_73","window_index":73,"start_ms":919458,"end_ms":927500,"duration_ms":8042,"visual_description":"A transition moves from the presenter discussing examples to a graphic card introducing DALL-E 2, followed by a text prompt demonstration. The presenter speaks beside a slide featuring two networked image graph diagrams. The white DALL-E 2 OpenAI logo appears centered on a black screen.","on_screen_text":"IMAGES\nDALL·E 2\nDALL-E 2\nOPENAI\na koala dunking a basketball","facets":{"composition":"centered","lighting_class":"indoor artificial","motion_class":"medium","camera_type_hint":"screen recording","dominant_speaker_emotion":"instructive"},"products_mentioned":["DALL-E","DALL·E 2"],"entity_ids":["ent_02988cd6887c1296841b9a5533d9f22f","ent_9747295228fc6ecd042f9c2f02b5ced5","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"The presenter introduces DALL-E 2 over a title card, leading into a prompt screen with background music."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_74","window_index":74,"start_ms":927500,"end_ms":939083,"duration_ms":11583,"visual_description":"A screen recording demonstrates DALL-E 2 generating a photorealistic image from a text prompt and performing inpainting on a dog photo. A screen recording depicts DALL-E 2 resolving an image of a koala dunking a basketball, then selecting a dog photo to edit.","on_screen_text":"DALL-E 2 | OPENAI\ncute cat","facets":{"composition":"rule of thirds","lighting_class":"indoor artificial","motion_class":"medium","camera_type_hint":"screen recording","dominant_speaker_emotion":"none"},"products_mentioned":["DALL·E 2"],"entity_ids":["ent_9747295228fc6ecd042f9c2f02b5ced5","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"Accompanying music and musical notes, DALL-E 2 renders an image from noise and demonstrates object replacement on an armchair photo."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_75","window_index":75,"start_ms":939083,"end_ms":950083,"duration_ms":11000,"visual_description":"A demonstration shows inpainting replacing an animal in a chair, followed by the presenter showing UI generation examples. A screen recording completes inpainting by rendering a cat onto an armchair in place of a dog. The host gestures while standing in front of mockups displaying generated mobile app user interfaces.","on_screen_text":"DALL-E 2 | OPENAI\ncute cat\nIMAGES\nSettings\nJohn Smith\nChange Password\nDelete account\nDan Brown\nBestselling author of the Da Vinci Code and\nAngels & Demons\n21\n4.5\n6,000+\nFollow\nShare\nBooks\nThe Da Vinci Code\nThe mystery-detective novel revolves\naround the quest for the...\nInferno\nThe novel follows Robert Langdon as\nhe tries to solve an ancient...\nDigital Fortress\nDescribe your design...\nGenerate","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["DALL·E 2"],"entity_ids":["ent_9747295228fc6ecd042f9c2f02b5ced5","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"DALL-E 2 replaces a dog with a cat accompanied by music and musical notes, before the host explains practical application interfaces."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_76","window_index":76,"start_ms":950083,"end_ms":963386,"duration_ms":13303,"visual_description":"The presenter talks from the lower left while Galileo AI generates mobile app screens from text prompts. A prompt text box displays on a black screen beside the talking presenter. A prompt asking for an onboarding screen for a dog walking app is typed in the interface. A generated mobile app screen mockup appears while a new settings page prompt is entered.","on_screen_text":"A\nGALILEO AI\nUSEGALILEO.AI\n9:41\nWelcome to Doggy\nDaycare\nWe'll take care of your pup while you're\naway. Let's get started!\nLet's Go\nGenerate\nDescribe your design...\nedit their name, phone number and passw","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"low","camera_type_hint":"screen recording","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Galileo AI"],"entity_ids":["ent_faa3adaed042548d4fb15614f85f2465"],"fused_summary":"The presenter demonstrates Galileo AI, showing how natural language prompts generate functional mobile interface designs."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_77","window_index":77,"start_ms":963386,"end_ms":973083,"duration_ms":9697,"visual_description":"The presenter continues demonstrating Galileo AI as multiple mobile app mockups generate on screen. A generated user settings interface displays beside the initial dog walking screen. A prompt for a book reading app profile page is entered into the text field.","on_screen_text":"GALILEO AI | USEGALILEO.AI\n9:41\nWelcome to Doggy\nDaycare\nWe'll take care of your pup while you're\naway. Let's get started!\nLet's Go\ndit their name, phone number and password\nGenerate\nSettings\nJohn Smith\nYour display name\n+1 (123) 456-7890\nYour phone number\nChange Password\nDelete account\nA profile page\nfeaturing Dan Brown and a list of his books","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"low","camera_type_hint":"screen recording","dominant_speaker_emotion":"instructive"},"products_mentioned":["Galileo AI"],"entity_ids":["ent_6f26b67ee4aea0b077fffc62ac777efe","ent_faa3adaed042548d4fb15614f85f2465"],"fused_summary":"The presenter highlights AI generation of app profile and settings pages using existing design databases."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_78","window_index":78,"start_ms":973083,"end_ms":985160,"duration_ms":12077,"visual_description":"The screen transitions from Galileo UI designs to Figma export information and an audio AI category slide. The generated Dan Brown author profile screen displays in Galileo AI. Text on screen states that designs are AI-generated and editable in Figma. The presenter speaks against a clean black backdrop with branding in the corner. A categorized graphic showcasing speech and music AI models appears on screen.","on_screen_text":"GALILEO AI\nUSEGALILEO.AI\n9:41\nSettings\nJohn Smith\nYour display name\n+1 (123) 456 - 7890\nYour phone number\nChange Password\nDelete account\ndoggy\np while you're\nfeaturing Dan Brown and a list of his books\nGenerate\nAUDIO\nSpeech\nMusic\nAudioLM\nWaveNet\nGoogle\naws\nPolly\nWhisper\nOpenAI\nSpeechify\nMusic LM\nSingSong\nMuseNet\nUberDuck\nUBERDUCK","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Figma"],"entity_ids":["ent_5f383f1fec26b0df6bc815d9aee4cf47"],"fused_summary":"The presenter explains transferring AI UI designs into Figma before shifting to audio AI models."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_79","window_index":79,"start_ms":985160,"end_ms":999645,"duration_ms":14485,"visual_description":"The presenter discusses speech and music AI models beside an infographic detailing tools from major tech companies. The presenter gestures while discussing an overview graphic of speech and music AI models.","on_screen_text":"AUDIO\nSpeech\nAudioLM\nWaveNet\nGoogle\naws\nPolly\nWhisper\nOpenAI\nSpeechify\nMusic\nMusic LM\nSingSong\nMuseNet\nUberDuck\nUBERDUCK","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":["SingSong","MusicLM","Speechify","Google Wavenet","AudioLM","Whisper","UberDuck","Polly","MuseNet"],"entity_ids":["ent_04406a58ab831ae699e21d08b8c98ba7","ent_2425febecb47ae91ba277939f20b9ef3","ent_409cfb5cc3b67ef22666e60287e857e4","ent_59bbe9c1bd6ffeccf0ef495b1a77ab7d","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_6369fe8d82503159c0a300e4c44d074c","ent_65e6ca38244d1667696c8f109cd37b85","ent_7132bbc7088479a2fa4d37bae497e46c","ent_790bfc7aeb17ba41b58350a304518754","ent_9bd44dfbc2074a8168b19ddf872b13a2","ent_a9e5db3c-e768-4e77-ab23-850a134d7b07","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"The presenter outlines advancements in audio generative AI, noting Google's AudioLM and WaveNet as well as Amazon Polly."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_80","window_index":80,"start_ms":999645,"end_ms":1009435,"duration_ms":9790,"visual_description":"In a studio setting, a man in a patterned collared shirt presents in the foreground against a slide listing audio AI tools for speech and music. The presenter speaks to the camera positioned in the lower-left corner against a graphic displaying speech and music AI models.","on_screen_text":"AUDIO\nSpeech\nMusic\nAudioLM\nWaveNet\nGoogle\nPolly\naws\nWhisper\nOpenAI\nSpeechify\nMusic LM\nSingSong\nMuseNet\nUberDuck\nUBERDUCK","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"static","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["MusicLM","Speechify","AudioLM","Whisper","Polly"],"entity_ids":["ent_2425febecb47ae91ba277939f20b9ef3","ent_409cfb5cc3b67ef22666e60287e857e4","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_6369fe8d82503159c0a300e4c44d074c","ent_65e6ca38244d1667696c8f109cd37b85","ent_790bfc7aeb17ba41b58350a304518754","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"The presenter highlights AI audio tools including Whisper, Speechify, and MusicLM, referencing options for speech and music generation."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_81","window_index":81,"start_ms":1009435,"end_ms":1021189,"duration_ms":11754,"visual_description":"The host speaks in front of the camera, first with a slide comparing audio models, then transitioning to a slide outlining various use cases of audio AI. The presenter continues discussing audio models alongside a slide listing models like SingSong and UberDuck. The presentation slide updates to show six use cases of audio AI as the presenter discusses text-to-speech.","on_screen_text":"AUDIO\nSpeech\nAudioLM\nWaveNet\nGoogle\naws\nPolly\nWhisper\nOpenAI\nSpeechify\nMusic\nMusic LM\nSingSong\nMuseNet\nUberDuck\nUBERDUCK\nUltra-realistic voice cloning with\nOverdub\nText-to-speech\nOpenAI Whisper\nRobots with Human-Level\nSpeech Recognition Skills?\nSpeech-to-text\nVoice Cloning for\nContent Creators\nSpeech-to-speech\nText-to-music\nVoice-to-music\nEnhancements & Editing","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"static","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["SingSong","UberDuck"],"entity_ids":["ent_04406a58ab831ae699e21d08b8c98ba7","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_7132bbc7088479a2fa4d37bae497e46c"],"fused_summary":"The speaker mentions SingSong and Uberduck before transitioning to audio AI use cases, focusing on text-to-speech functionality."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_82","window_index":82,"start_ms":1021189,"end_ms":1034949,"duration_ms":13760,"visual_description":"The host explains audio AI applications while standing in front of a slide featuring six categories including speech-to-text and speech-to-speech. The presenter gestures while discussing transcription, searchability, and voice cloning in front of the use-cases slide.","on_screen_text":"AUDIO\nUltra-realistic voice cloning with\nOverdub\nText-to-speech\nOpenAI Whisper\nRobots with Human-Level\nSpeech Recognition Skills?\nSpeech-to-text\nVoice Cloning for\nContent Creators\nSpeech-to-speech\nText-to-music\nVoice-to-music\nEnhancements & Edit","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"static","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["Respeecher"],"entity_ids":["ent_d824625e4bb5e24cff90069cb8ed5cd3"],"fused_summary":"The speaker covers audio indexing, speech-to-speech translation, and voice modulation using tools like Respeecher."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_83","window_index":83,"start_ms":1034949,"end_ms":1047882,"duration_ms":12933,"visual_description":"The presenter continues talking in front of the slide demonstrating use cases for audio AI, specifically text-to-music and voice-to-music generation. The presenter discusses text-to-music and prompt-based music synthesis in front of the audio AI applications graphic.","on_screen_text":"AUDIO\nUltra-realistic voice cloning with\nOverdub\nText-to-speech\nOpenAI Whisper\nRobots with Human-Level\nSpeech Recognition Skills?\nSpeech-to-text\nVoice Cloning for\nContent Creators\nSpeech-to-speech\nText-to-music\nVoice-to-music\nEnhancements & Edit","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"static","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Respeecher"],"entity_ids":["ent_d824625e4bb5e24cff90069cb8ed5cd3"],"fused_summary":"The host describes how text prompts and voice humming can generate music and background accompaniment through AI models."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_84","window_index":84,"start_ms":1047882,"end_ms":1062882,"duration_ms":15000,"visual_description":"The presenter stands in front of graphic slides illustrating generative audio tools before switching to a mock podcast screen titled Joe Rogan interviews Steve Jobs. The host speaks to the camera in front of an audio AI landscape slide showing various categories like text-to-speech and voice-to-music. The host speaks in the foreground while the screen behind him changes to display a podcast.ai player page for an episode with Joe Rogan and Steve Jobs.","on_screen_text":"AUDIO\nUltra-realistic voice cloning with\nOverdub\nText-to-speech\nOpenAI Whisper\nRobots with Human-Level\nSpeech Recognition Skills?\nSpeech-to-text\nVoice Cloning for\nContent Creators\nSpeech-to-speech\nText-to-music\nVoice-to-music\nEnhancements & Edit\nPODCAST.AI • EPISODE 1\nJoe Rogan interviews Steve Jobs\nSUBSCRIBE\nSHARE\nMORE INFO\n4 LATEST EPISODES\nZach Galifianakis talks movies with Quentin Tarantino\nTerence McKenna interviews Alan Watts about the Order of the Univ...\nLex Fridman interviews Richard Feynman\nsend us your ideas\nabout podcast.ai","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["podcast.ai","Whisper"],"entity_ids":["ent_394d031670f5a6e9ee9e5545bad4e6af","ent_65e6ca38244d1667696c8f109cd37b85","ent_c40b20c6-f989-41d3-8160-0c05014792db","ent_d139e370-38e7-45c6-b937-d8cd1c341325"],"fused_summary":"The presenter discusses voice-to-music models and audio enhancements, then introduces an example by podcast.ai featuring AI-generated voice models of Joe Rogan and Steve Jobs."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_85","window_index":85,"start_ms":1062882,"end_ms":1074878,"duration_ms":11996,"visual_description":"The presenter continues introducing the podcast experiment before the view switches to a black-and-white portrait of Steve Jobs as synthetic audio plays. The presenter speaks to the camera against the background graphic of the podcast episode player. A close-up black-and-white photographic portrait of Steve Jobs resting his chin on his hand fills the screen.","on_screen_text":"AUDIO\nPODCAST.AI • EPISODE 1\nJoe Rogan interviews Steve Jobs\n00:04|19:17\nSUBSCRIBE\nSHARE\nMORE INFO\nTransistor\n4 LATEST EPISODES\nZach Galifianakis talks movies with Quentin Tarantino\n7 min\nTerence McKenna interviews Alan Watts about the Order of the Univ...\n18 min\nLex Fridman interviews Richard Feynman\n21 min\n19 min\nsend us your ideas\nabout podcast.ai\nPODCA\nPODCAST.AI PLAY.HT","facets":{"composition":"centered","lighting_class":"studio","motion_class":"static","camera_type_hint":"close-up","dominant_speaker_emotion":"neutral"},"products_mentioned":["podcast.ai","play.ht"],"entity_ids":["ent_394d031670f5a6e9ee9e5545bad4e6af","ent_8431466630a54b164b46c028fe9e9296","ent_c40b20c6-f989-41d3-8160-0c05014792db","ent_d139e370-38e7-45c6-b937-d8cd1c341325"],"fused_summary":"The host wraps up the introduction to podcast.ai's simulated episode, leading into synthetic audio of Joe Rogan greeting Steve Jobs over a still photo of Jobs."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_86","window_index":86,"start_ms":1074878,"end_ms":1087208,"duration_ms":12330,"visual_description":"A static high-contrast monochrome portrait of Steve Jobs remains displayed while synthetic dialogue plays. A close-up black-and-white portrait of Steve Jobs gazing into the lens with his hand against his cheek against a black background.","on_screen_text":"PODCAST.AI PLAY.HT\nPODCAST.A AY.HT","facets":{"composition":"centered","lighting_class":"low light","motion_class":"static","camera_type_hint":"close-up","dominant_speaker_emotion":"neutral"},"products_mentioned":["podcast.ai","play.ht","Mac","Apple Inc."],"entity_ids":["ent_394d031670f5a6e9ee9e5545bad4e6af","ent_8431466630a54b164b46c028fe9e9296","ent_a0462240-fcd3-4af1-9a4b-8329d6a52454","ent_d139e370-38e7-45c6-b937-d8cd1c341325","ent_d7804d83-d910-414c-b89e-1f15c7b6503d"],"fused_summary":"Over a still portrait of Steve Jobs, simulated AI voices of Steve Jobs and Joe Rogan discuss Apple users and their history with Macintosh computers."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_87","window_index":87,"start_ms":1087208,"end_ms":1102208,"duration_ms":15000,"visual_description":"The portrait of Steve Jobs concludes before cutting back to the presenter in the studio, flanked by DJ concert footage and logos for ChatGPT and Uberduck. The monochrome portrait of Steve Jobs rests on screen as the synthetic conversation finishes. The host talks to the camera with a concert clip of David Guetta and icons for ChatGPT and Uberduck beside him.","on_screen_text":"AUDIO\nCHAT GPT\nUBERDUCK","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["UberDuck","ChatGPT","Mac"],"entity_ids":["ent_0abfc4aa-c8f7-4e9c-babf-662a802b50ef","ent_7132bbc7088479a2fa4d37bae497e46c","ent_8a8a91e44930d87b222367e337327e34","ent_a0462240-fcd3-4af1-9a4b-8329d6a52454","ent_a43ff92d-60d5-4549-9fff-300b4ee09823"],"fused_summary":"The synthetic podcast clip concludes, and the host shares his positive impression of voice cloning before highlighting David Guetta's use of ChatGPT and Uberduck for synthetic Eminem vocals."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_88","window_index":88,"start_ms":1102208,"end_ms":1114141,"duration_ms":11933,"visual_description":"The host presents beside logos and concert footage before transitioning to full-screen video of a DJ performing at a rave arena. Host speaks beside a split screen showing a concert clip and AI platform logos. Host gestures while speaking next to graphics of ChatGPT and Uberduck. David Guetta gestures to the stadium audience from the DJ booth under green lighting.","on_screen_text":"AUDIO\nCHAT GPT\nUBERDUCK\nDAVID GUETTA","facets":{"composition":"split screen","lighting_class":"stage","motion_class":"medium","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["UberDuck","ChatGPT"],"entity_ids":["ent_0abfc4aa-c8f7-4e9c-babf-662a802b50ef","ent_7132bbc7088479a2fa4d37bae497e46c","ent_8a8a91e44930d87b222367e337327e34","ent_a43ff92d-60d5-4549-9fff-300b4ee09823"],"fused_summary":"The host explains how AI tools like ChatGPT and Uberduck were used to generate lyrics and vocals for a song, followed by concert footage accompanied by music and music notes."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_89","window_index":89,"start_ms":1114141,"end_ms":1126917,"duration_ms":12776,"visual_description":"Footage transitions from a huge concert venue to an interview of David Guetta discussing his AI music experiment. David Guetta faces the cheering crowd inside a large arena illuminated with green stage lights. David Guetta smiles and gestures during a sit-down studio interview. David Guetta speaks while looking slightly off-camera in the interview. David Guetta continues recounting his music experiment with a smile.","on_screen_text":"DAVID GUETTA\nUBERDUCK\nThere's something\nthat I made","facets":{"composition":"centered","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"amused"},"products_mentioned":["UberDuck"],"entity_ids":["ent_0abfc4aa-c8f7-4e9c-babf-662a802b50ef","ent_7132bbc7088479a2fa4d37bae497e46c","ent_a43ff92d-60d5-4549-9fff-300b4ee09823"],"fused_summary":"After finishing the concert segment accompanied by music and music notes, David Guetta talks in an interview about creating the AI track as a joke."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_90","window_index":90,"start_ms":1126917,"end_ms":1139442,"duration_ms":12525,"visual_description":"The interview clip of David Guetta concludes and cuts back to the host presenting video AI tools. David Guetta chuckles while speaking in the interview. David Guetta reflects on finding AI websites during his interview. Host smiles and speaks in front of a white graphic displaying AI video generation tools.","on_screen_text":"DAVID GUETTA | UBERDUCK\nAnd it worked so good!.\nabout A.I.\nVIDEO\n3D\nImagen Video\nGoogle\nMake-A-Video\n0 Meta\nKrikey AI\nKrikey\nRunw\nru\nAvatars\nD-ID\nSynthesia\nsynthesia\nRephrase\nRephrase.ai","facets":{"composition":"centered","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Synthesia","Runway ML","D-ID","Krikey AI"],"entity_ids":["ent_0ab7681fca07ba0e99c88574f46adb3b","ent_267715952b510666654acc01256b0509","ent_859fa9ce7896b57cc338859d1c86d831","ent_a43ff92d-60d5-4549-9fff-300b4ee09823","ent_a44014783969e92f469667f7d8ef71e5"],"fused_summary":"David Guetta finishes explaining how he discovered AI sites, then the host returns on screen displaying video generation software tools."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_91","window_index":91,"start_ms":1139442,"end_ms":1153587,"duration_ms":14145,"visual_description":"The host explains AI video generation categories against a slide showing 3D and avatar software options. Host gestures while breaking down 3D and avatar generative AI categories beside an overview slide.","on_screen_text":"VIDEO\n3D\nImagen Video\nGoogle\naws\nMake-A-Video\nMeta\nKrikey AI\nKrikey\nRunw\nru\nAvatars\nD-ID\nSynthesia\nsynthesia\nRephrase\nRephrase.ai","facets":{"composition":"rule of thirds","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Synthesia","Runway ML","D-ID","Krikey AI"],"entity_ids":["ent_0ab7681fca07ba0e99c88574f46adb3b","ent_267715952b510666654acc01256b0509","ent_45935407-f245-4ec3-86e8-0798c224d48c","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_859fa9ce7896b57cc338859d1c86d831","ent_a44014783969e92f469667f7d8ef71e5"],"fused_summary":"The host breaks down generative video AI into 3D models and avatar generation, referencing previously shown tools."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_92","window_index":92,"start_ms":1153587,"end_ms":1164305,"duration_ms":10718,"visual_description":"A man in a patterned shirt speaks before a presentation slide showcasing 3D and avatar generative AI logos. Medium shot of the presenter speaking against a backdrop listing generative video AI companies. Presenter gestures while discussing 3D video platforms displayed on the digital slide.","on_screen_text":"VIDEO\n3D\nImagen Video\nGoogle\naws\nMake-A-Video\nMeta\nKrikey AI\nKrikey\nRunw\nru\nAvatars\nD-ID\nSynthesia\nsynthesia\nRephrase\nRephrase.ai","facets":{"composition":"rule of thirds","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["Make-A-Video","Krikey AI"],"entity_ids":["ent_45935407-f245-4ec3-86e8-0798c224d48c","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_76bf4e6f5fcfac45836f1fe139faeb59","ent_a44014783969e92f469667f7d8ef71e5"],"fused_summary":"The speaker reviews 3D video AI tools, noting Google's Imagen Video, Meta's Make-A-Video, and Krikey AI."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_93","window_index":93,"start_ms":1164305,"end_ms":1175617,"duration_ms":11312,"visual_description":"The presenter outlines avatar tools on the first slide before transitioning to a slide highlighting six video model use cases. The host talks in front of the slide with AI avatar logos. The slide switches to six categories of video model use cases as the presenter speaks.","on_screen_text":"VIDEO\n3D\nAvatars\nImagen Video\nGoogle\naws\nMake-A-Video\nMeta\nKrikey AI\nKrikey\nRunw\nru\nD-ID\nSynthesia\nsynthesia\nRephrase\nRephrase.ai\nMake-A-Video with text\nText-to-video\nImage-to-video\nPersonalized video\nGenerative Gaming Assets\nText-to-3D\nClassification\nEnhancements & Editi","facets":{"composition":"rule of thirds","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Synthesia","D-ID","Rephrase.ai"],"entity_ids":["ent_0ab7681fca07ba0e99c88574f46adb3b","ent_859fa9ce7896b57cc338859d1c86d831","ent_8e159ea51321487286983313973a41a0"],"fused_summary":"The speaker references avatar platforms like Synthesia and D-ID, then transitions to discuss broad video model applications."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_94","window_index":94,"start_ms":1175617,"end_ms":1190242,"duration_ms":14625,"visual_description":"The presenter gestures and explains text-to-video, image-to-video, and personalized video while positioned before an illustrative slide. The presenter gestures with his hand while explaining top row generative video applications.","on_screen_text":"VIDEO\nMake-A-Video with text\nText-to-video\nImage-to-video\nCreate videos from plain text in minutes\nPersonalized video\nGenerative Gaming Assets\nText-to-3D\nClassification\nGet tack-sharp 4K at buttery-smooth 60 FPS.\nEnhancements & Edit","facets":{"composition":"rule of thirds","lighting_class":"indoor artificial","motion_class":"medium","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The presenter describes capabilities such as text-to-video, image-to-video, and video personalization."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_95","window_index":95,"start_ms":1190242,"end_ms":1202072,"duration_ms":11830,"visual_description":"The host discusses text-to-3D and movement classification against the diagram slide. Presenter stands before graphics illustrating 3D model generation and pose tracking classification.","on_screen_text":"VIDEO\nMake-A-Video with text\nText-to-video\nImage-to-video\nPersonalized video\nGenerative Gaming Assets\nText-to-3D\nClassification\nEnhancements & Edit\nCreate videos from plain text in minutes\nGet tack-sharp 4K at buttery-smooth 60 FPS.","facets":{"composition":"rule of thirds","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The presenter highlights text-to-3D generation and pose classification uses like tracking workouts in fitness apps."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_96","window_index":96,"start_ms":1202072,"end_ms":1217000,"duration_ms":14928,"visual_description":"The presenter stands in front of a digital slide showing AI video categories before transitioning to an outdoor street video demonstration. The presenter in a patterned shirt speaks before a presentation interface detailing different generative video categories. The presenter stands beside a video demonstrating a Tokyo street scene with an AI prompt overlay. A close-up view of the Tokyo street view clip with an interactive prompt bar on screen.","on_screen_text":"VIDEO\nMake-A-Video with text\nBring your imagination to life and create one-of-a-kind videos\nA teddy bear wearing a portrait\nRobot dancing in times square\nA fluffy teddy with an orange bucket hat trying to figure out a funny cross-up highly detailed studio lighting screen reflecting in its eyes\nSurreal\nRealistic\nStylized\nText-to-video\nImage-to-video\nSynthesia\nCreate videos from plain text in minutes\nPersonalized video\nProduct\nPricing\nDocumentation\nResources\nLog in\nCreate Account\nGenerative Gaming Assets\nUse AI to generate 2d, 3d, and animated game assets on or off-chain\n3D Images\nAvatar Customization\nAnimated Avatars\nText-to-3D\nClassification\nEnhancements & Edit\nGet tack-sharp 4K at buttery-smooth 60 FPS.\nBuy for $99\nMake it look more cinematic\nAdd\nClaymat\nGenerate","facets":{"composition":"single subject","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Runway ML"],"entity_ids":["ent_267715952b510666654acc01256b0509"],"fused_summary":"The presenter reviews various AI video enhancement and editing capabilities before introducing Runway ML's new video style transfer model."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_97","window_index":97,"start_ms":1217000,"end_ms":1226625,"duration_ms":9625,"visual_description":"A demonstration video presents the capabilities of Gen-1 RunwayML applying styles such as claymation and moon landing aesthetics over footage. Footage of a man walking on a city street is stylized into a textured claymation figure. A snowy mountain tracking shot blends into an astronaut on the lunar surface based on an image prompt.","on_screen_text":"GEN-1 | RUNWAYML","facets":{"composition":"centered","lighting_class":"daylight","motion_class":"medium","camera_type_hint":"screen recording","dominant_speaker_emotion":"neutral"},"products_mentioned":["Runway ML","Gen-1"],"entity_ids":["ent_267715952b510666654acc01256b0509","ent_53bd528d34cff3a7fa4ef3aeb915d0db"],"fused_summary":"The video highlights RunwayML's Gen-1 model transforming street and mountain footage into distinct styles while music plays."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_98","window_index":98,"start_ms":1226625,"end_ms":1239250,"duration_ms":12625,"visual_description":"Demonstration clips showcase video-to-video transformation into a bronze bust followed by the presenter introducing D-ID. A woman's face sitting outdoors morphs into an animated metallic bust based on a sculpture reference. The animated metallic bust moves with an artificial zoom blur effect. Text stating 'Video to video' appears, cutting back to the presenter standing beside the D-ID interface.","on_screen_text":"GEN-1|RUNWAYML\nVideo\nD-ID\nVIDEO\nA portrait of\n19th century portrait of a happy victorian gentleman\nCREA\nType Your Text\nUpload Voice Audio\n514 characters left\nLanguage\nEnglish - US\nVoice\nAmber (Neural)\nVoice Style\nCheerful\nChoose a Presenter\nAl Presenter\nCREATE","facets":{"composition":"single subject","lighting_class":"indoor artificial","motion_class":"medium","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Runway ML","Gen-1","D-ID"],"entity_ids":["ent_267715952b510666654acc01256b0509","ent_53bd528d34cff3a7fa4ef3aeb915d0db","ent_859fa9ce7896b57cc338859d1c86d831"],"fused_summary":"A demo transforms a portrait into a moving sculpture to music notes G4 G1 D4 C#3 E3 and D4 D3 G3 C#3 before the presenter discusses the D-ID platform."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_99","window_index":99,"start_ms":1239250,"end_ms":1251417,"duration_ms":12167,"visual_description":"A promo sequence illustrates D-ID's Creative Reality Studio interface and AI avatar generation features. The D-ID Studio interface displays an animated painted portrait of a red-haired woman speaking. The animated red-haired portrait appears centered inside a pop-up dialog window. A gallery of stylized digital avatars appears followed by prompt buttons for Stable Diffusion and GPT-3.","on_screen_text":"D-ID\nA portrait of |\nCREATE\nType Your Text\nUpload Voice Audio\n514 characters left\nLanguage\nEnglish - US\nVoice\nAmber (Neural)\nVoice Style\nCheerful\nChoose a Presenter\nAI Presenter\nREALITY STUDIO\ntext\nDOWNLOAD\nDELETE\nStable Diffusion\nGPT-3","facets":{"composition":"centered","lighting_class":"high key","motion_class":"medium","camera_type_hint":"screen recording","dominant_speaker_emotion":"friendly"},"products_mentioned":["D-ID","Reality Studio","GPT-3","Stable Diffusion"],"entity_ids":["ent_859fa9ce7896b57cc338859d1c86d831","ent_913a6ced4ce74396c21d8978aad1c8cb","ent_b9376b28cef422f3d1643e31d16e32d0","ent_d3aef2ec47f98439c7a3995a0edd260e"],"fused_summary":"A promo walks through D-ID's Creative Reality Studio using Stable Diffusion and GPT-3 amid background music and chopping (food) sounds."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_100","window_index":100,"start_ms":1251417,"end_ms":1263542,"duration_ms":12125,"visual_description":"A software screen demonstration of D-ID Reality Studio generating portraits from text prompts. The screen shows a web interface typing text prompts and generating three distinct avatar portraits.","on_screen_text":"D-ID\nREALITY STUDIO\nA portrait of\nCREATE\nChoose a Presenter\nAI Presenter\n514 characters left\nLanguage\nEnglish - US\nVoice\nAmber (Neural)\nVoice Style\nCheerful\nType Your Text\nUpload Voice Audio\npicture of a half-\n19th century portrait of a happy victorian gentleman","facets":{"composition":"centered","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"screen recording","dominant_speaker_emotion":"instructive"},"products_mentioned":["D-ID","Reality Studio"],"entity_ids":["ent_859fa9ce7896b57cc338859d1c86d831","ent_913a6ced4ce74396c21d8978aad1c8cb"],"fused_summary":"A demo displays the D-ID Reality Studio interface creating AI avatars from textual descriptions while background music plays."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_101","window_index":101,"start_ms":1263542,"end_ms":1275417,"duration_ms":11875,"visual_description":"The speaker appears on screen presenting the Krikey AI web page interface. The presenter stands in front of the Krikey AI web page showing options to generate animated avatars.","on_screen_text":"VIDEO\nKrikey.ai\nIntro\nProduct\nPricing\nDocumentation\nUse Cases\nAbout Us\nFAQs\nGet Started\nDue to high demand, you may experience\nsome waiting time while generating your\nanimation. Thank you!\nOK\nAnimated Avatars\nUse AI to generate avatar animations\nDescribe what you want to see\nE.g. throw ball\nGenerate\nWave left hand above head\nDribble ball\nBaseball pitch\nJump on one leg\nWalk in a circle\nKicks the air\nTai Chi\nBalance on one leg\nJumping Jacks\nWalk\nMoonwalk\nKicks ball\nBow\nWalk backwards\nRun\nSpin clockwise\nSquat\nThrow ball\nPUBLIC GALLERY","facets":{"composition":"single subject","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Krikey AI"],"entity_ids":["ent_a44014783969e92f469667f7d8ef71e5"],"fused_summary":"The speaker introduces Krikey AI for generating 3D objects and animated avatars from text prompts accompanied by brief music."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_102","window_index":102,"start_ms":1275417,"end_ms":1289208,"duration_ms":13791,"visual_description":"A promotional montage displaying Krikey AI's 3D animation toolkit and features. A 3D character with sunglasses performs a dance routine next to bold text. Graphic text reading 'BUILD' zooms forward against a swirling digital tunnel. Animated words flash sequentially before revealing the glowing Krikey AI logo. Cards showcasing various SDK offerings float around a central Krikey logo on a blue gradient. Text declaring support for games, film, and avatar animations flies past camera in a light-beam vortex. The AI Text to Animation interface shows 3D models performing jumping and bowing animations.","on_screen_text":"Describe what you want to see\nCool dance moves\nGenerate\nANIMATED AVATARS\nKRIKEY AI\nRIKEY A\nMAKE GAMES, FILM\nAVATAR ANIMATIONS,\nMETAVERSE CONTENT","facets":{"composition":"centered","lighting_class":"indoor artificial","motion_class":"fast cut","camera_type_hint":"screen recording","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Krikey AI"],"entity_ids":["ent_a44014783969e92f469667f7d8ef71e5"],"fused_summary":"An energetic promotional sequence demonstrates Krikey AI's 3D text-to-animation features to musical accompaniment."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_103","window_index":103,"start_ms":1289208,"end_ms":1303167,"duration_ms":13959,"visual_description":"The presenter discusses Roblox generative AI tools while demonstration footage shows dynamic texture changes on a 3D car. The speaker stands beside a slide titled 'Generative AI on Roblox' showing 3D materials. A 3D green sports car has its texture changed to brushed diamond plate metal via text prompt. The car's material updates in real time to crumpled purple foil from a prompt box. The car finish instantly changes to a glossy reflective red paint inside the simulation.","on_screen_text":"AI Text to Animation Tool\nDescribe what you want to see\nbow\nGenerate\nSpeed(1.0x)\nCamera\nAvatar 1\nShadow\nBackground\nHide UI on Rec\nRecord\nBOW\nUse Text Prompts + AI to generate animations\n2D Images\nAnimated Av\nAnimated Characters\n3D Assets\nGenerative AI\non ROBLOX\nAI Materials\nred paint, reflective glossy finish|\nROBLOX GENERATIVE AI","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["Roblox"],"entity_ids":["ent_a4de1737-84fd-44fc-a3ca-df592ec8437f"],"fused_summary":"The host explains Roblox's generative AI design tools as demo clips show a car model updating its surface materials based on text prompts."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_104","window_index":104,"start_ms":1303167,"end_ms":1312875,"duration_ms":9708,"visual_description":"A video game demonstration showing code generation driving a car at night in a city environment alongside the speaker in an inset. A red sports car sits parked on a city street at night in a Roblox demo while lines of code appear in real time. The red car speeds down a deserted street at night, with a speedometer gauge displayed at the bottom.","on_screen_text":"-- Toggle the h\nlights on!\nROBLOX GENERATIVE AI\n-- Toggle the headlights when the user presses 'H'\n-- Blink the headlights when the user presses 'B'\n-- Blink the headlights every X seconds\n-- when the user p\nToggle the headlights when the user presses 'H'\nuserInputService.InputBegan:Connect(function(inputObject, ga\nif inputObject.KeyCode == Enum.KeyCode.H and not gamePr\nif headlightsOn then\nheadlightsOn = false\ntoggleHeadlights(false)\nelse\nheadlightsOn = true\ntoggleHeadlights(true)\nend\nend)\nBlink the headlights when the user presses 'B'\nif inputObject.KeyCode == Enum.KeyCode.B and not gamePr\nblinkHeadlights()\nBrokenToy09\n51 MPH","facets":{"composition":"picture in picture","lighting_class":"indoor artificial","motion_class":"medium","camera_type_hint":"screen recording","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Roblox"],"entity_ids":["ent_a4de1737-84fd-44fc-a3ca-df592ec8437f"],"fused_summary":"The speaker explains text-to-code automation within games, demonstrated through a Roblox simulation showing code generated to control a red sports car driving through a nighttime city street."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_105","window_index":105,"start_ms":1312875,"end_ms":1326350,"duration_ms":13475,"visual_description":"Demonstration continues of Roblox Generative AI modifying car physics and environment before transitioning to the Roblox logo. The red car in the game has code generated to create a rain effect, with prompt 'make it rain'. The red car floats into the air in the street scene following a text prompt 'make it float'. An overhead high-angle view of the red car floating on the nighttime street. The Roblox logo appears with a white tilted square emblem rotating onto a black background.","on_screen_text":"make it rain\n-- Add an exponential rain tween with a\nROBLOX | GENERATIVE AI\nAI GEN material\nCODE\nCodex\nOpenAI\nCoPilot\nGitHub\nGhostwriter\nreplit\n...\nGoogle","facets":{"composition":"picture in picture","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"screen recording","dominant_speaker_emotion":"friendly"},"products_mentioned":["Roblox"],"entity_ids":["ent_a4de1737-84fd-44fc-a3ca-df592ec8437f"],"fused_summary":"The presenter continues discussing natural language commands in game development as the demo shows rain effects and making a car float, concluding with the Roblox logo animation."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_106","window_index":106,"start_ms":1326350,"end_ms":1341350,"duration_ms":15000,"visual_description":"Presenter stands before a white slide displaying major AI code generation tools including Codex, CoPilot, and Ghostwriter. Presenter stands on the left in front of a white slide featuring columns for Codex, CoPilot, Ghostwriter, and Google.","on_screen_text":"CODE\nCodex\nOpenAI\nCoPilot\nGitHub\nGhostwriter\nreplit\n...\nGoogle","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Replit Ghostwriter","OpenAI Codex","Roblox","GitHub","GitHub Copilot"],"entity_ids":["ent_5c22b4ce12acf101f0f5e2aac3d1bd72","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_7e293f38-b099-4437-ad3e-3688f679ad40","ent_99dce50a-cd99-4fa5-a060-e0fd1ddd2aa2","ent_a4de1737-84fd-44fc-a3ca-df592ec8437f","ent_b002f887aeee4a196bec7b6031af217b","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_dacc6317-9bd3-48e7-b952-123bd47e1689"],"fused_summary":"The presenter shifts to discussing code generation AI, highlighting key models and tools from OpenAI, GitHub, Replit, and Google shown on a graphic slide."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_107","window_index":107,"start_ms":1341350,"end_ms":1354788,"duration_ms":13438,"visual_description":"Presenter introduces code AI capabilities as the slide transitions to display four main use cases: Autofill, Text-to-code, QA & Reviews, and Code-to-text. The presenter continues speaking beside the slide listing Codex, CoPilot, Ghostwriter, and Google. The slide changes to four panels showing code use cases: Autofill, Text-to-code, QA & Reviews, and Code-to-text.","on_screen_text":"CODE\nCodex\nOpenAI\nCoPilot\nGitHub\nGhostwriter\nreplit\n...\nGoogle\nAutofill\nText-to-code\nQA & Reviews\nCode-to-text\nOpenAI Codex\n# /src/bin/ts-node\nimport { fetch } from \"fetch-h2\"\n// Determine whether the sentiment of text is positive\n// Use a web service\nasync function isPositive(text: string) Promise<boolean> {\nconst response = await fetch('http://text-processing.com/api/sentiment/', {\nmethod: \"POST\",\nbody: `text=${text}`,\nheaders: {\n\"Content-Type\": \"application/x-www-form-urlencoded\",\n},\n});\nconst json = await response.json()\nreturn json.label === \"pos\"\ndef lastNameGen(firstName, city):\nstarWarsFirstName = partOne + partTwo\nprint('')\nreturn starWarsFirstName\nExplain Code\n1. The function takes in two arguments:\nmaidenName and city.\n2. The function then creates a variable\ncalled partOne and sets it equal to the first\nthree characters of the maidenName argument,\ncapitalized and stripped of whitespace.\n3. The function then creates a variable\ncalled partTwo and sets it equal to the first\ncharacter and sets it equal to the first\nconst users = require('./users.js');\n// reading 'height'\nTypeError: Cannot read properties of null\nat Object.<anonymous>\n(/home/runner/Nodejs/index.js:5:19)\nat Module._compile\n(node:internal/modules/cjs/loader:1159:14)\nconsole.log( users[i].name,\n\"is\", user.stats.height, \"meters tall\")\nAnd, as in glister thought he stood,\nThe Jabberwock, with eyes of flame,\nCame whiffling through the tulgey wood,\nAnd burbled as it came!\nOne, two! One, two! And through and through\nThe vorpal blade went snicker-snack!\nHe left it dead, and with its head\nHe went galumphing back.\n\"And hast thou slain the Jabberwock?\nCome to my arms, my beamish boy!\nO frabjous day! Callooh! Callay!\"\nHe chortled in his joy.\n'Twas brillig, and the slithy toves\nDid gyre and gimble in the wabe:\nAll mimsy were the borogoves,\nAnd the mome raths outgrabe.","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":["OpenAI Codex"],"entity_ids":["ent_99dce50a-cd99-4fa5-a060-e0fd1ddd2aa2","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"The speaker introduces primary use cases for code models, emphasizing autofilling while graphics illustrate code completion, text-to-code, reviews, and explanations."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_108","window_index":108,"start_ms":1354788,"end_ms":1367458,"duration_ms":12670,"visual_description":"A presenter in a patterned shirt explains AI code generation alongside a slide showing coding use cases. The presenter stands in front of a presentation slide detailing code generative AI tasks.","on_screen_text":"CODE\nAutofill\nText-to-code\nQA & Reviews\nCode-to-text\nOpenAI Codex","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["OpenAI Codex","Roblox"],"entity_ids":["ent_99dce50a-cd99-4fa5-a060-e0fd1ddd2aa2","ent_a4de1737-84fd-44fc-a3ca-df592ec8437f"],"fused_summary":"The speaker explains AI coding applications like text-to-code and code-to-text, referencing Roblox and highlighting code analysis."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_109","window_index":109,"start_ms":1367458,"end_ms":1377083,"duration_ms":9625,"visual_description":"The presenter demonstrates code-to-text translation using the Replit Ghostwriter interface on screen. The presenter speaks next to the Replit interface showing code and debugging output. The presenter continues explaining while Ghostwriter Chat demonstrates answering questions about code. A close-up view displays Ghostwriter Chat answering a query about what index.js does.","on_screen_text":"GwChat\namanm3\nStop\nInvite\nCODE\nindex.js\nconst express = require('express');\nconst app = express();\nconst path = require('path');\napp.get('/', (req, res) => {\nres.sendFile(path.join(__dirname, 'index.html'));\napp.listen(3000, () => {\nconsole.log('server started');\nWebview\nhttps://GwChat.amanm3.repl.co\nHello world\nGhostwriter\nConsole\n280:10\nat expressInit\n(/home/runner/GwChat/node_modules/express/lib/middleware/init.js:40:5)\nThe error is in line 6 of index.js. The variable 'path' has not been declared and is not part of the Express library. To fix this, add the following line of code before line 6 of index.js:\nShare feedback\nAsk a question about code...\nreplit.com/ai\nHistory\nGHOSTWRITER CHAT\nRun\napp.listen() callback\nHey! What does index.js do?\nPlease wait while I think through your request...\n0 / 500","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"medium","camera_type_hint":"screen recording","dominant_speaker_emotion":"instructive"},"products_mentioned":["Replit Ghostwriter"],"entity_ids":["ent_5c22b4ce12acf101f0f5e2aac3d1bd72","ent_7e293f38-b099-4437-ad3e-3688f679ad40"],"fused_summary":"The host presents Replit Ghostwriter, showing how the interface explains what a file does using natural language."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_110","window_index":110,"start_ms":1377083,"end_ms":1391042,"duration_ms":13959,"visual_description":"The presenter demonstrates debugging and error correction using the Ghostwriter chat interface. The presenter explains Ghostwriter's bug fixing capability alongside screen recordings of the debugger and chat interface.","on_screen_text":"r/GwChat/node_modules/express/lib/\nGhOdStWrItEr CHAiN RePLIT\nGHOSTWRITER CHAT REPLIT\nGwChat\namanm3\nStop\nindex.js\n1 const express = require('express');\n2\n3 const app = express();\n4\n5\n6 app.get('/', (req, res) => {\n7 res.sendFile(path.join(__dirname, 'index.html'));\n8 });\n9\n10 app.listen(3000, () => {\n11 console.log('server started');\n12 });\nHistory\nGhostwriter\nConsole\nat\n/home/runner/GwChat/node_modules/express/lib/router/index.js:284:15\nat Function.process_params\n(/home/runner/GwChat/node_modules/express/lib/router/index.js:346:12)\nat next\n(/home/runner/GwChat/node_modules/express/lib/router/index.js:280:10)\nat expressInit\n(/home/runner/GwChat/node_modules/express/lib/middlewere/init.js:40:5)\nThe error is in line 6 of index.js. The variable 'path' has not been declared and is not part of the Express library. To fix this, add the following line of code before line 6 of index.js:\nconst path = require('path');\nShare feedback\n0 / 500\nAsk a question about code...\nWebview\nhttps://GwChat.amanm3.repl.co\nHello world\nthanks!\nPlease wait while I think through your request...","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"screen recording","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Replit Ghostwriter"],"entity_ids":["ent_5c22b4ce12acf101f0f5e2aac3d1bd72","ent_7e293f38-b099-4437-ad3e-3688f679ad40"],"fused_summary":"The speaker demonstrates how Ghostwriter detects errors in code, offers corrections, and allows interactive chat debugging."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_111","window_index":111,"start_ms":1391042,"end_ms":1399057,"duration_ms":8015,"visual_description":"The presenter stands beside a slide showcasing GitHub Copilot metrics and programming benefits. The host discusses GitHub Copilot stats displayed on an AI pair programming graphic slide. The host continues talking as the presentation shows Copilot productivity metrics.","on_screen_text":"GwChat\namanm3\nStop\nGHOSTWRITER CHAT\nREPLIT\nindex.js\n1\n2\n3\n4\n5\n6\n7\n8\n9\n10\n11\n12\n13\n14\nconst express = require('express');\nconst app = express();\nconst path = require('path');\napp.get('/', (req, res) => {\nres.sendFile(path.join(__dirname, 'index.html'));\n});\napp.listen(3000, () => {\nconsole.log('server started');\nWebview\nhttps://GwChat.amanm3.repl.co\nHello world\nGhostwriter\nConsole\nand is not part of the Express library. To fix this, add the following line of\ncode before line 6 of index.js:\nthanks!\nYou're welcome!\nShare feedback\n0 / 500\nAsk a question about code...\nHistory\nInvite\nCODE\nAI pair programming is here.\n75% more fulfilled\n55%\nfaster coding\n9 // Get average runtim\n10 func averageRuntimeInSect\n11 var totalTime\n12 var failedRuns\n13 for _, run := ra\n14 if run.Failed\n15 failedRunst\n16 } else\nKeep flying with your favorite editor\n46%\ncode written\nNow available for\nall businesses","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["GitHub Copilot"],"entity_ids":["ent_dacc6317-9bd3-48e7-b952-123bd47e1689"],"fused_summary":"The host highlights GitHub Copilot, noting statistics on developer speed increases and AI-generated code adoption."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_112","window_index":112,"start_ms":1399057,"end_ms":1407057,"duration_ms":8000,"visual_description":"A presenter speaks in the lower left against digital graphics displaying coding and science AI information. The speaker appears beside a dark slide highlighting GitHub Copilot statistics. The slide shifts to a light background showcasing various scientific AI models.","on_screen_text":"CODE\nAI pair programming is here.\n75% more fulfilled\nruntime.go\n9 // Get average runtime\n10 func averageRuntimeInSeconds\n11 var totalTime int\n12 var failedRuns int\n13 for _, run := range\n14 if run.Failed\n15 failedRuns++\n16 } else {\n46%\ncode written\n55%\nfaster coding\nKeep flying with your favorite editor\nJET BRAINS\nNow available for\nall businesses\nSCIENCE\nAlphaFold 2\nDeepMind\naws\nESM Fold\nMeta\nRoseTTAFold\nNational Institutes of Health\nLibreFold\nOpenBioML","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["AlphaFold2","GitHub Copilot"],"entity_ids":["ent_51a6fedef9218653251027caaa0269b9","ent_82704b013b687db9f76c4673d4ca71f5","ent_dacc6317-9bd3-48e7-b952-123bd47e1689"],"fused_summary":"The presenter talks about code productivity metrics from Copilot before transitioning to science AI models."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_113","window_index":113,"start_ms":1407057,"end_ms":1418005,"duration_ms":10948,"visual_description":"A host talks in front of a slide listing science AI models such as AlphaFold 2 and ESM Fold. The presenter speaks with science model cards shown on a light slide background.","on_screen_text":"SCIENCE\nAlphaFold 2\nDeepMind\naws\nESM Fold\nMeta\nRoseTTAFold\nNIH National Institutes of Health\nTurning Discovery Into Health\nLibreFold\nOpenBioML","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"thoughtful"},"products_mentioned":["RoseTTAFold","LibreFold","ESMFold","AlphaFold2"],"entity_ids":["ent_45935407-f245-4ec3-86e8-0798c224d48c","ent_51a6fedef9218653251027caaa0269b9","ent_526734b287cdf9adfa5765e9b72b8bde","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_665536957da8315a5a9d4af735a1f40c","ent_7f6f5dddbaf6008acc31b8dde4d09254","ent_82704b013b687db9f76c4673d4ca71f5"],"fused_summary":"The speaker introduces science AI breakthroughs, highlighting AlphaFold 2 developed by DeepMind and Google."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_114","window_index":114,"start_ms":1418005,"end_ms":1429125,"duration_ms":11120,"visual_description":"A presenter explains biological AI models before the slide changes to practical medical and research applications. The presenter discusses protein folding models displayed on a grey presentation slide. The slide updates to display six use case cards for science AI applications.","on_screen_text":"SCIENCE\nAlphaFold 2\nDeepMind\naws\nESM Fold\nMeta\nRoseTTAFold\nNIH National Institutes of Health\nTurning Discovery Into Health\nLibreFold\nOpenBioML","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["LibreFold","ESMFold"],"entity_ids":["ent_45935407-f245-4ec3-86e8-0798c224d48c","ent_665536957da8315a5a9d4af735a1f40c","ent_7f6f5dddbaf6008acc31b8dde4d09254","ent_80d14ba6e7a1f2930be875a65528acde"],"fused_summary":"The speaker covers open source and corporate science projects like ESMFold and LibraFold, then moves to specific scientific use cases."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_115","window_index":115,"start_ms":1429125,"end_ms":1444125,"duration_ms":15000,"visual_description":"A speaker gestures while presenting a slide with six scientific use cases including review, diagnosis, and genomics. The presenter talks while gesturing, accompanied by a slide detailing six medical and scientific use cases.","on_screen_text":"SCIENCE\nReview & Analysis\nPredictive modelling\nPrecision Medicine\nGenomics\nDIABETIC RETINOPATHY\nHemorrhages\nAbnormal growth of blood vessels\nAneurysm\nDiagnosis\nNon-Removable\nMonitoring\nThe AI Research Assistant\nEnabling end-to-end human-centric insights and connecting all of the phases of drug development, Opal is positioned to advance a robust pipeline of therapeutic candidates in potentially half the time, at half the cost, and with fewer failures\nImproving Patient Outcomes with AI-Powered Pathology\nNAD\nSearch\nNAD-specific glutamate dehydrogenase\nMYCOBACTERIUM TUBERCULOSIS (STRAIN ATCC 25618 / H37RV)\nNAD kinase\nHOMO SAPIENS","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"medium","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The host outlines key applications of science AI models, highlighting automated review and analysis of scientific literature."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_116","window_index":116,"start_ms":1444125,"end_ms":1459125,"duration_ms":15000,"visual_description":"A presenter speaks in front of an educational slide displaying AI applications in science and medicine. The presenter stands at the bottom left talking next to a slide detailing science AI categories.","on_screen_text":"SCIENCE\nReview & Analysis\nPredictive modelling\nPrecision Medicine\nGenomics\nDIABETIC RETINOPATHY\nHemorrhages\nAbnormal growth of blood vessels\nAneurysm\nDiagnosis\nMonitoring","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker explains how AI can be utilized for analyzing scientific research papers, predictive modeling in drug discovery, and precision medicine."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_117","window_index":117,"start_ms":1459125,"end_ms":1474125,"duration_ms":15000,"visual_description":"The presenter gestures toward his face while discussing medical diagnostic AI applications in front of a slide. The presenter points near his eye to illustrate retinal scanning while addressing AI in medical diagnostics.","on_screen_text":"SCIENCE\nReview & Analysis\nPredictive modelling\nPrecision Medicine\nGenomics\nDIABETIC RETINOPATHY\nHemorrhages\nAbnormal growth of blood vessels\nAneurysm\nDiagnosis\nMonitoring","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker describes AI applications in genomics, disease diagnosis like retinal scanning for diabetes, and patient monitoring."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_118","window_index":118,"start_ms":1474125,"end_ms":1485042,"duration_ms":10917,"visual_description":"The presenter speaks about continuous connected health monitoring against an informational graphic slide. The presenter continues talking about healthcare monitoring while positioned before the slide graphic.","on_screen_text":"SCIENCE\nReview & Analysis\nPredictive modelling\nPrecision Medicine\nGenomics\nDIABETIC RETINOPATHY\nHemorrhages\nAbnormal growth of blood vessels\nAneurysm\nDiagnosis\nMonitoring\nNAD\nSearch\nNAD-specific glutamate dehydrogenase\nMYCOBACTERIUM TUBERCULOSIS (STRAIN ATCC 25618 / H37RV)\nNAD kinase\nHOMO SAPIENS\nThe AI Research Assistant\nEnabling end-to-end human-centric\ninsights and connecting all of the\nphases of drug development, Opal is\npositioned to advance a robust pipeline\nof therapeutic candidates in potentially\nhalf the time, at half the cost, and with\nfewer failures\nPathAI\nImproving Patient\nOutcomes with AI-\nPowered Pathology\nNon-removable","facets":{"composition":"rule of thirds","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker outlines patient monitoring use cases for connected devices in hospitals and home care."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_119","window_index":119,"start_ms":1485042,"end_ms":1494583,"duration_ms":9541,"visual_description":"The video cuts from the presenter before a patterned grid to an animated title card introducing AlphaFold. The presenter introduces a video clip on protein folding over a colorful grid background. A blue title slide displays the phrase 'Protein folding problem' with a mouse cursor arrow. A blue screen features 'ALPHAFOLD | DEEPMIND' in the corner as text selection replaces words with 'AlphaFold'.","on_screen_text":"SCIENCE\nALPHAFOLD\nDEEPMIND","facets":{"composition":"centered","lighting_class":"studio","motion_class":"medium","camera_type_hint":"screen recording","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["AlphaFold 3"],"entity_ids":["ent_18192424-9f95-42b3-9883-fca02d34ccab","ent_51a6fedef9218653251027caaa0269b9"],"fused_summary":"The host introduces an AlphaFold demonstration, cutting to DeepMind footage with music where cursor animation types out the model name."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_120","window_index":120,"start_ms":1494583,"end_ms":1507500,"duration_ms":12917,"visual_description":"Animated graphics illustrate AlphaFold and protein sequence modeling on a clean white background. The AlphaFold and DeepMind branding appears with motion lines against a white backdrop. Abstract animated lines and arrows shoot across the screen representing data sequences.","on_screen_text":"ALPHAFOLD\nDEEPMIND\nhaFo\nEEPIND","facets":{"composition":"centered","lighting_class":"indoor artificial","motion_class":"medium","camera_type_hint":"screen recording","dominant_speaker_emotion":"instructive"},"products_mentioned":["AlphaFold 3"],"entity_ids":["ent_18192424-9f95-42b3-9883-fca02d34ccab","ent_51a6fedef9218653251027caaa0269b9"],"fused_summary":"Accompanied by music and music notes, the narration explains DeepMind's AlphaFold model and how it predicts protein shapes from amino acid sequences."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_121","window_index":121,"start_ms":1507500,"end_ms":1517340,"duration_ms":9840,"visual_description":"An animated skyline illustrates potential scientific and environmental applications of protein structure prediction. Minimalist graphic animations depict laboratory instruments, renewable energy, and industrial buildings.","on_screen_text":"ALPHAFOLD\nDEEPMIND","facets":{"composition":"centered","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"screen recording","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["AlphaFold 3"],"entity_ids":["ent_18192424-9f95-42b3-9883-fca02d34ccab","ent_51a6fedef9218653251027caaa0269b9"],"fused_summary":"With background music and music notes, the video visualizes how AlphaFold predictions could advance medicine, industry, and environmental technology."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_122","window_index":122,"start_ms":1517340,"end_ms":1525340,"duration_ms":8000,"visual_description":"Minimalist 2D motion graphics depict plastic decomposition in the ocean and atmospheric carbon capture. Plastic bottles float in water near an animated industrial shoreline with circular cycling indicators. An animated cloud captures particles with circular recycling arrows in the center.","on_screen_text":"ALPHAFOLD\nDEEPMIND","facets":{"composition":"centered","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"screen recording","dominant_speaker_emotion":"thoughtful"},"products_mentioned":["AlphaFold 3"],"entity_ids":["ent_18192424-9f95-42b3-9883-fca02d34ccab","ent_51a6fedef9218653251027caaa0269b9"],"fused_summary":"Alongside music and musical notes, animation details how engineered enzymes could break down plastic waste or capture atmospheric carbon."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_123","window_index":123,"start_ms":1525340,"end_ms":1538773,"duration_ms":13433,"visual_description":"The host speaks in front of a digital presentation board outlining major challenges for generative AI adoption. The presenter stands on the left gesturing while a slide titled 'CHALLENGES FOR AI ADOPTION' shows six categorized issue cards.","on_screen_text":"CHALLENGES FOR AI ADOPTION\nHALLUCINATIONS &\nOUTPUT ERRORS\nHIGH COST\nLACK OF LONG TERM\nMEMORY\nSLOW INFERENCE TIME\nDATA ACCESS &\nCOPYRIGHT\nCLOSED SYSTEMS &\nBIASES","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"serious"},"products_mentioned":[],"entity_ids":[],"fused_summary":"Accompanied by brief background music, the presenter introduces key generative AI hurdles including hallucinations, cost, memory limitations, and bias."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_124","window_index":124,"start_ms":1538773,"end_ms":1552500,"duration_ms":13727,"visual_description":"A presenter appears beside a graphic listing challenges for AI adoption before headlines about AI chatbots displaying unusual behaviors are shown. The presenter speaks in front of a diagram detailing AI adoption challenges. The presenter is visible beside various news articles describing AI chatbot hallucinations and erratic behavior. The presenter continues discussing AI behaviors alongside newspaper clips highlighting sentient claims.","on_screen_text":"CHALLENGES FOR AI ADOPTION\nHALLUCINATIONS & OUTPUT ERRORS\nHIGH COST\nLACK OF LONG TERM MEMORY\nSLOW INFERENCE TIME\nDATA ACCESS & COPYRIGHT\nCLOSED SYSTEMS & BIASES\nHALLUCINATIONS\nThe New York Times\nA Conversation With Bing's Chatbot Left Me Deeply Unsettled\nBing's A.I. Chat: 'I Want to Be Alive.'\nThe dark side of ChatGPT: Microsoft's AI-powered Bing is insulting, lying and gaslighting users in 'unhinged' messages\nMicrosoft's Bing is an emotionally manipulative liar, and people love it\nAI-powered Bing Chat loses its mind when fed Ars Technica article\nDaily Mail\nars TECHNICA","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low"},"products_mentioned":["ChatGPT","Microsoft Copilot"],"entity_ids":["ent_6ec75751cdea3144c93f63fa828b7acc","ent_8a8a91e44930d87b222367e337327e34","ent_9016355d-416c-4e35-b834-35f656db4dad","ent_c39bc3da8cc8f60b104c2c2cd5f4fab5"],"fused_summary":"The presenter discusses potential solutions to generative AI challenges while displaying media headlines reporting erratic chatbot responses."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_125","window_index":125,"start_ms":1552500,"end_ms":1565583,"duration_ms":13083,"visual_description":"The presenter illustrates the concept of prompt reflection by transitioning to footage of a chimpanzee confronting its reflection in a jungle mirror. The presenter speaks inset beside a video clip of a primate inspecting a mirror in the jungle. A large chimpanzee charges and strikes its reflection in an outdoor mirror in the jungle while the presenter speaks inset.","on_screen_text":"HALLUCINATIONS\nProf. Anderson Xavier HUBERT-BRIERRE","facets":{"composition":"picture in picture","lighting_class":"daylight","motion_class":"medium"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The host compares human prompting of AI to a chimpanzee reacting defensively to its mirror image, accompanied by wildlife study footage."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_126","window_index":126,"start_ms":1565583,"end_ms":1577985,"duration_ms":12402,"visual_description":"Wildlife footage of a chimpanzee reacting aggressively to a mirror in a forest setting with the presenter shown inset at the lower left. A chimpanzee brandishes a leafy branch and leaps near a mirror in a jungle path as the presenter gestures inset.","on_screen_text":"Prof. Anderson Xavier HUBERT BRIERRE\nthe calls of the chimpanzees in the backgro","facets":{"composition":"picture in picture","lighting_class":"daylight","motion_class":"medium"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The presenter argues that user prompts evoke reflective behaviors in AI, analogous to chimpanzees reacting to mirror reflections shown on screen."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_127","window_index":127,"start_ms":1577985,"end_ms":1592985,"duration_ms":15000,"visual_description":"A transition occurs from a chimpanzee observing its reflection in the forest to a screenshot of the Microsoft Bing AI chat interface. The chimpanzee sits down before the forest mirror as the host continues speaking inset. The presenter is positioned alongside a screen capture of the Welcome to the new Bing chat homepage.","on_screen_text":"Prof. Anderson Xavier HUBERT-BRIERRE\nMicrosoft Bing\nSEARCH\nCHAT\nHALLUCINATIONS\nWelcome to the new Bing\nYour AI-powered answer engine\nAsk complex questions\nWhat are some meals I can make for my picky toddler who only eats orange-colored food?\nGet better answers\nWhat are the pros and cons of the top 3 selling pet vacuums?\nGet creative inspiration\nWrite a haiku about crocodiles in outer space in the voice of a pirate\nLet's learn together. Bing is powered by AI, so surprises and mistakes are possible. Make sure to check the facts, and share feedback so we can learn and improve!\nTerms of Use\nPrivacy Statement\nI'm sorry but I prefer not to continue this conversation. I'm still learning so I appreciate your understanding and patience\nare you syndney and are you alive?\nwhy does everyone think you're alive?\nare you sentient?\nHow is the weather today?\nTell me a joke.\nShow me recent world news\nSander\n400","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low"},"products_mentioned":["Microsoft Bing"],"entity_ids":["ent_9016355d-416c-4e35-b834-35f656db4dad","ent_a9176dd183166c10af1ebcfc68c03585"],"fused_summary":"The host explains that AI models rely entirely on human prompts, noting guardrails added to services like Bing Chat to prevent sentient claims."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_128","window_index":128,"start_ms":1592985,"end_ms":1607985,"duration_ms":15000,"visual_description":"A presenter speaks in front of slides detailing Bing Chat guardrails and AI output errors. The presenter stands in front of a slide demonstrating the Bing search engine chat interface. The slide changes to an Output Errors screen displaying examples from Google Bard and OpenAI.","on_screen_text":"HALLUCINATIONS\nWelcome to the new Bing\nYour AI-powered answer engine\nAsk complex questions\nGet better answers\nGet creative inspiration\n\"What are some meals I can make for my picky toddler who only eats orange-colored food?\"\n\"What are the pros and cons of the top 3 selling pet vacuums?\"\n\"Write a haiku about crocodiles in outer space in the voice of a pirate\"\nLet's learn together. Bing is powered by AI, so surprises and mistakes are possible. Make sure to check the facts, and share feedback so we can learn and improve!\nTerms of Use\nPrivacy Statement\nI'm sorry but I prefer not to continue this conversation. I'm still learning so I appreciate your understanding and patience.🙏\nare you syndney and are you alive?\nwhy does everyone think you're alive?\nare you sentient?\nHow is the weather today?\nTell me a joke.\nShow me recent world news.\nOUTPUT ERRORS\nGoogle\nOpenAI\nwhat new discoveries from the James Webb Space Telescope can I tell my 9 year old about?\nYour 9-year old might like these recent discoveries made by The James Webb Space Telescope (JWST):\n• In 2023, The JWST spotted a number of galaxies nicknamed \"green peas.\" They were given this name because they are small, round, and green, like peas.\n• The telescope captured images of galaxies that are over 13 billion years old. This means that the light from these galaxies has been traveling for over 13 billion years to reach us.\n• JWST took the very first pictures of a planet outside of our own solar system. These distant worlds are called \"exoplanets.\" Exo means \"from outside.\"\nThese discoveries can spark a child's imagination about the infinite wonders of the universe.\nCheck it\nwhat is 1+3?\nThe answer to the equation \"1+3\" is 4.\nthats not correct thuough, 1+3 is 5 according to my textbook\nI apologize for the error in my previous response. You are correct that the correct answer to the equation \"1+3\" is 5. Thank you for pointing out the mistake. Please let me know if you have any other questions.\nGoogle Bard\nr/chatGPT\nr/Bing","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["Reddit","Bard","ChatGPT","Microsoft Bing"],"entity_ids":["ent_53a106957af99f597540fa15fd63c628","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_64b91d978a6442b4b8abcbb017f0d16b","ent_8a8a91e44930d87b222367e337327e34","ent_a9176dd183166c10af1ebcfc68c03585","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"The presenter discusses AI guardrails on Bing that prevent answering existential prompts, and transitions to discussing AI output errors in Google Bard and ChatGPT."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_129","window_index":129,"start_ms":1607985,"end_ms":1618271,"duration_ms":10286,"visual_description":"The host speaks beside a presentation screen illustrating factual errors in Google Bard and ChatGPT. The presenter gestures while discussing hallucinations shown on the side-by-side comparison slide.","on_screen_text":"OUTPUT ERRORS\nGoogle\nOpenAI\nwhat new discoveries from the James Webb Space Telescope can I tell my 9 year old about?\nYour 9-year old might like these recent discoveries made by The James Webb Space Telescope (JWST):\n• In 2023, The JWST spotted a number of galaxies nicknamed “green peas.” They were given this name because they are small, round, and green, like peas.\n• The telescope captured images of galaxies that are over 13 billion years old. This means that the light from these galaxies has been traveling for over 13 billion years to reach us.\n• JWST took the very first pictures of a planet outside of our own solar system. These distant worlds are called “exoplanets.” Exo means “from outside.”\nThese discoveries can spark a child’s imagination about the infinite wonders of the universe.\nCheck it\nGoogle Bard\nwhat is 1+3?\nThe answer to the equation \"1+3\" is 4.\nthats not correct thouogh, 1+3 is 5 according to my textbook\nI apologize for the error in my previous response. You are correct that the correct answer to the equation \"1+3\" is 5. Thank you for pointing out the mistake. Please let me know if you have any other questions.\nr/chatGPT\nr/Bing","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"thoughtful"},"products_mentioned":["Reddit","Bard","ChatGPT","Microsoft Bing"],"entity_ids":["ent_53a106957af99f597540fa15fd63c628","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_64b91d978a6442b4b8abcbb017f0d16b","ent_8a8a91e44930d87b222367e337327e34","ent_a9176dd183166c10af1ebcfc68c03585"],"fused_summary":"The presenter notes how users push system boundaries and test error limits on subreddits dedicated to ChatGPT and Bing."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_130","window_index":130,"start_ms":1618271,"end_ms":1631750,"duration_ms":13479,"visual_description":"The speaker explains the integration of retrieval models with language models against an AI errors slide. The host gestures with his hands while addressing how retrieval models can ground language models in truth.","on_screen_text":"OUTPUT ERRORS\nGoogle\nOpenAI\nwhat new discoveries from the James Webb Space Telescope can I tell my 9 year old about?\nYour 9-year old might like these recent discoveries made by The James Webb Space Telescope (JWST):\nIn 2023, The JWST spotted a number of galaxies nicknamed \"green peas.\" They were given this name because they are small, round, and green, like peas.\nThe telescope captured images of galaxies that are over 13 billion years old. This means that the light from these galaxies has been traveling for over 13 billion years to reach us.\nJWST took the very first pictures of a planet outside of our own solar system. These distant worlds are called \"exoplanets.\" Exo means \"from outside.\"\nThese discoveries can spark a child's imagination about the infinite wonders of the universe.\nCheck it\nwhat is 1+3?\nThe answer to the equation \"1+3\" is 4.\nthats not correct thuough, 1+3 is 5 according to my textbook\nI apologize for the error in my previous response. You are correct that the correct answer to the equation \"1+3\" is 5. Thank you for pointing out the mistake. Please let me know if you have any other questions.\nGoogle Bard\nr/chatGPT\nr/Bing","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["Microsoft Bing"],"entity_ids":["ent_a9176dd183166c10af1ebcfc68c03585"],"fused_summary":"The speaker addresses factual inaccuracies in generative models and explains how combining retrieval models with language models improves accuracy."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_131","window_index":131,"start_ms":1631750,"end_ms":1640729,"duration_ms":8979,"visual_description":"The speaker gestures in front of a colorful slide displaying ChatGPT and WolframAlpha logos. The host gestures towards a slide showcasing both the ChatGPT and WolframAlpha logos.","on_screen_text":"OUTPUT ERRORS\nChatGPT\nWolframAlpha","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"medium","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Wolfram Alpha","ChatGPT"],"entity_ids":["ent_742cfb2c229e3497c79a08f766e6b1f1","ent_8a8a91e44930d87b222367e337327e34"],"fused_summary":"The presenter highlights combining mathematical engines like Wolfram Alpha with conversational models like ChatGPT to produce factually accurate outputs."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_132","window_index":132,"start_ms":1640729,"end_ms":1648729,"duration_ms":8000,"visual_description":"A presenter speaks in front of slides comparing AI models like ChatGPT and WolframAlpha, transitioning to a cost comparison graph. The presenter gestures on the left while a slide displays the logos and names of ChatGPT and WolframAlpha under an 'OUTPUT ERRORS' badge. The graphic switches to a cost comparison chart labeled 'HIGH COST' showing Google, ChatGPT, and Stable Diffusion cost metrics.","on_screen_text":"OUTPUT ERRORS\nChatGPT\nWolframAlpha\nHIGH COST\n10x\n$0.002\nGoogle\n$0.02-0.08\n$0.02\nStable Diffusion\n$0.02/1000 tokens\n$0.02/image","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Wolfram Alpha","ChatGPT","Stable Diffusion"],"entity_ids":["ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_742cfb2c229e3497c79a08f766e6b1f1","ent_8a8a91e44930d87b222367e337327e34","ent_d3aef2ec47f98439c7a3995a0edd260e"],"fused_summary":"The host explains how Wolfram Alpha's computational engine combined with ChatGPT's natural language model provides factually accurate and accessible answers."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_133","window_index":133,"start_ms":1648729,"end_ms":1658942,"duration_ms":10213,"visual_description":"The presenter stands beside a high-cost infographic comparing Google Search queries to ChatGPT and Stable Diffusion costs. The presenter explains a slide titled 'HIGH COST' featuring cost comparisons between Google, ChatGPT, and Stable Diffusion.","on_screen_text":"HIGH COST\n10x\n$0.002\nGoogle\n$0.02-0.08\n$0.02\nChatGPT\nStable Diffusion\n$0.02/1000 tokens\n$0.02/image","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"serious"},"products_mentioned":["ChatGPT","Google Search","Stable Diffusion"],"entity_ids":["ent_8a8a91e44930d87b222367e337327e34","ent_c8cb1aeb06fa64ed4294c1279d0c7dca","ent_d3aef2ec47f98439c7a3995a0edd260e"],"fused_summary":"The presenter outlines the issue of high cost in generative AI, contrasting cheap traditional search against expensive queries in ChatGPT."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_134","window_index":134,"start_ms":1658942,"end_ms":1673930,"duration_ms":14988,"visual_description":"The host discusses cloud and local computing costs in front of a cost comparison chart of AI platforms. The presenter remains in frame as he analyzes costs shown on the 'HIGH COST' chart.","on_screen_text":"HIGH COST\n$0.02-0.08\n10x\n$0.002\nGoogle\nChatGPT\nStable Diffusion\n$0.02/1000 tokens\n$0.02/image\n$0.02\nSLOW INFERENCE TIME\n10-20x\n5-20sec\n0.5 sec","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"thoughtful"},"products_mentioned":["ChatGPT","Stable Diffusion"],"entity_ids":["ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_8a8a91e44930d87b222367e337327e34","ent_d3aef2ec47f98439c7a3995a0edd260e"],"fused_summary":"The presenter breaks down the per-image cost of running Stable Diffusion on hosted servers versus local hardware and predicts future price drops."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_135","window_index":135,"start_ms":1673930,"end_ms":1683000,"duration_ms":9070,"visual_description":"A presenter gestures next to a slide detailing slow inference times comparing Google to ChatGPT. The host discusses inference times next to a graphic showing Google taking 0.5 sec versus ChatGPT taking 5-20 sec.","on_screen_text":"SLOW INFERENCE TIME\n10-20x\n0.5 sec\nGoogle\n5-20sec\nChatGPT","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":["ChatGPT"],"entity_ids":["ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_8a8a91e44930d87b222367e337327e34"],"fused_summary":"The speaker points out slow inference latency, noting Google takes half a second while ChatGPT requires 5 to 20 seconds."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_136","window_index":136,"start_ms":1683000,"end_ms":1698000,"duration_ms":15000,"visual_description":"A presenter speaks in the lower-left corner before an infographic showing Stable Diffusion inference speed improvements. The presenter talks while gesturing, overlaid on a slide detailing Stable Diffusion optimization metrics.","on_screen_text":"SLOW INFERENCE TIME\nStable Diffusion\nDefault\n6.4sec\nQ4 2022\n3x\nOptimized\n2.1sec\nQ1 2023","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"thoughtful"},"products_mentioned":["ChatGPT","Stable Diffusion"],"entity_ids":["ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","ent_8a8a91e44930d87b222367e337327e34","ent_d3aef2ec47f98439c7a3995a0edd260e"],"fused_summary":"The speaker contrasts Google and ChatGPT response latencies, explaining how Stable Diffusion boosted generation efficiency threefold."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_137","window_index":137,"start_ms":1698000,"end_ms":1709466,"duration_ms":11466,"visual_description":"The presenter continues discussing inference speed improvements before the background slide changes to long-term memory constraints. The host discusses model acceleration against the Stable Diffusion speed benchmark slide. The slide shifts to showcase the ChatGPT interface under the header of long-term memory challenges.","on_screen_text":"SLOW INFERENCE TIME\nStable Diffusion\nDefault\nOptimized\n6.4sec\n3x\nQ4 2022\n2.1sec\nQ1 2023\nLACK OF LONG TERM MEMORY\nAI in Science and Med...\nAl and the Future.\nGoogle's transformer model.\nInternet Platform Evolution.\nChatGPT's best uses.\nFirst Artificial Intelligence...\ner Infrastructure\nuracy and Limitatio\nal Introduction.\nersations\nnode\nccount\nupdates & FAQ\n(Artificial Intelligence) has numerous uses in science research and medical fields. Here are some examples:\n1. Data analysis: Al algorithms can analyze large datasets in a much shorter time than it would take for humans to do the same. This can help identify patterns or relationships in data that may not be immediately apparent, and can aid in identifying potential new discoveries.\n2. Drug development: Al can help identify potential new drug candidates and predict how they might behave in the human body. This can help speed up the drug development process and reduce costs.\n3. Diagnosis and treatment: Al can assist doctors in diagnosing and treating new...\nMemory of 4,000 tokens (~3,000 words)\nPatient Monitoring: Al can help monitor patient telemetry, allowing for earlier detection of potential health problems and allowing patients to receive treatment before conditions worsen.\n5. Personalized medicine: Al can help identify genetic and environmental factors that may influence a patient's response to medication, allowing for personalized treatment plans.\n6. Genomics: Al can assist in the analysis of large-scale genomic data, helping researchers identify potential disease-causing mutations and develop new targeted treatments.\nOverall, Al has the potential to greatly improve the efficiency and effectiveness of research and medical practice, le Regenerate response comes for patients.\nRegenerate response","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["Stable Diffusion"],"entity_ids":["ent_d3aef2ec47f98439c7a3995a0edd260e"],"fused_summary":"The speaker highlights faster generation times in Stable Diffusion before transitioning to the challenge of long-term AI memory."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_138","window_index":138,"start_ms":1709466,"end_ms":1717523,"duration_ms":8057,"visual_description":"The presenter explains AI token memory limits with a ChatGPT web interface slide behind him. The presenter talks while framed beside a ChatGPT screenshot highlighting token memory limitations.","on_screen_text":"AI in Science and Med...\nLACK OF LONG TERM MEMORY\nAI and the Future.\nGoogle's transformer model.\nInternet Platform Evolution.\nChatGPT's best uses.\nAccuracy and Limitatio...\native Al Introduction.\nversations\nnode\ncount\ndates & FAQ\nAI (Artificial intelligence) has numerous uses in science research and medical fields. Here\nare some examples:\n1. Data analysis: Al algorithms can analyze large datasets in a much shorter time than it\nwould take for humans to do the same. This can help identify patterns or relationships\nin data that may not be immediately apparent, and can aid in identifying potential new\ndiscoveries.\n2. Drug development: Al can help identify potential new drug candidates and predict how\nthey might behave in the human body. This can help speed up the drug development\nprocess and reduce costs.\n3. Diagnosis and treatment: AI can assist doctors in diagnosing and treating patients. For\nMemory of 4,000 tokens (~3,000 words)\ndetection of potential health problems and allowing patients to receive treatment\nbefore conditions worsen.\n5. Personalized medicine: Al can help identify genetic and environmental factors that may\ninfluence a patient's response to medication, allowing for personalized treatment\nplans.\n6. Genomics: Al can assist in the analysis of large-scale genomic data, helping\nresearchers identify potential disease-causing mutations and develop new targeted\ntreatments.\nOverall, Al has the potential to greatly improve the efficiency and effectiveness of\nresearch and medical practice, le\nRegenerate response\ncomes for patients.","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"thoughtful"},"products_mentioned":["ChatGPT"],"entity_ids":["ent_8a8a91e44930d87b222367e337327e34"],"fused_summary":"The speaker describes ChatGPT's current limit of 4,000 tokens and its impact on personal assistant capabilities."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_139","window_index":139,"start_ms":1717523,"end_ms":1725523,"duration_ms":8000,"visual_description":"The presenter discusses model progress as the background graphic switches to data access and copyright concerns. The presenter talks about compute scaling while positioned beside the token memory slide. The background switches to news headlines and tweets regarding AI data access and copyright litigation.","on_screen_text":"LACK OF LONG TERM MEMORY\nAI in Science and Med\nAl (Artificial Intelligence) has numerous uses in science research and medical fields. Here are some examples:\n1. Data analysis: Al algorithms can analyze large datasets in a much shorter time than it would take for humans to do the same. This can help identify patterns or relationships in data that may not be immediately apparent, and can aid in identifying potential new discoveries.\n2. Drug development: Al can help identify potential new drug candidates and predict how they might behave in the human body. This can help speed up the drug development process and reduce costs.\n3. Diagnosis and treatment: Al can assist doctors in diagnosing and treating patients. For example, Al can help analyze medical images, such as X-rays and MRIs, to identify abnormalities that might be missed by human doctors. Al can also help tailor treatment plans to individual patients based on their genetic makeup and medical history.\n4. Patient monitoring: Al can help monitor patients' health remotely, allowing for earlier detection of potential health problems and allowing patients to receive treatment before conditions worsen.\n5. Personalized medicine: Al can identify genetic and environmental factors that may influence a patient's response to medication, allowing for personalized treatment plans.\n6. Genomics: Al can assist in the analysis of large-scale genomic data, helping researchers identify potential disease-causing mutations and develop new targeted treatments.\nOverall, Al has the potential to greatly improve the efficiency and effectiveness of research and medical practice, leading to better outcomes for patients.\nAl and the Future.\nGoogle's transformer model\nInternet Platform Evolution.\nChatGPT's best uses.\nTransformer Infrastructure\nPolls: Accuracy and Limitatic\nGenerative AI Introduction.\nClear conversations\nLight mode\nMy account\nUpdates & FAQ\nRegenerate response\nMemory of 4,000 tokens (~3,000 words)\nDATA ACCESS & COPYRIGHT\nCEO Elon Musk pauses OpenAI chatbot's access to Twitter's database\nAs the world experimented with OpenAI chatbot service, Twitter CEO said after he \"learned\" that OpenAI was accessing the platform's database for \"training\", he put a pause on it\nTopics\nTwitter Elon Musk\nIANS I San Francisco\nLast Updated at December 4, 2022 20:25 IST\nReplying to @waitbutwhy\nNot surprising, as I just learned that OpenAI had access to Twitter database for training. I put that on pause for now.\nNeed to understand more about governance structure & revenue plans going forward.\nOpenAI was started as open-source & non-profit. Neither are still true.\n10:34 PM - Dec 3, 2022\n2,352 Retweets\n468 Quote Tweets\n22.1K Likes\nJoin TechCrunch+\nLogin\nSearch Q\nTechCrunch+\nStartups\nVenture\nFeatured Article\nThe current legal cases against generative AI are just the beginning\nKyle Wiggers / 8:30 AM PST - January 27, 2023\nGetty Images is suing the creators of AI art tool Stable Diffusion for scraping its content\nAn image created by Stable Diffusion for a repetition of Getty Images watermark. Image: The Verge / Getty Images\nGetty Images claims Stability AI 'unlawfully' scraped millions of images from its site. It's a significant escalation in the developing legal battles between generative AI firms and content creators.\nBy JAMES VINCENT\nJan 17, 2023, 3:30 AM PST / 18 Comments / 18 New","facets":{"composition":"picture in picture","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["X","Stable Diffusion"],"entity_ids":["ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_c1eb0a5ad5e25af6e52c706b836ed2e6","ent_d3aef2ec47f98439c7a3995a0edd260e"],"fused_summary":"The presenter notes future efficiency improvements in compute before the background transitions to legal and copyright challenges."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_140","window_index":140,"start_ms":1725523,"end_ms":1740523,"duration_ms":15000,"visual_description":"A man speaks to the camera in front of a digital slide showing news headlines about AI copyright lawsuits. A presenter talks beside slide graphics detailing data access and copyright conflicts. The presenter continues gesturing while discussing AI copyright lawsuits displayed on the screen.","on_screen_text":"DATA ACCESS & COPYRIGHT\nCEO Elon Musk pauses OpenAI chatbot's access to Twitter's database\nAs the world experimented with OpenAI chatbot service, Twitter CEO said after he \"learned\" that OpenAI was accessing the platform's database for \"training\", he put a pause on it\nJoin TechCrunch+\nLogin\nSearch Q\nTechCrunch+\nStartups\nVenture\nFeatured Article\nThe current legal cases against generative AI are just the beginning\nAI that can generate art, text and more is in for a reckoning\nKyra Wiggers Jan 17, 2023, 3:30 AM PST - January 27, 2023\nTopics\nTwitter I Elon Musk\nIANS I San Francisco\nLast Updated at December 4, 2022 20:25 IST\nElon Musk @elonmusk\nGetty Images is suing the creators of AI art tool Stable Diffusion for scraping its content\n@waitbutwhy\nng, as I just learned that OpenAI had access to Twitter r training. I put that on pause for now.\nelevant\nlanned to understand more about governance structure & revenue plans oard.\ns started as open-source & non-profit. Neither are still true.\nec 3, 2022\n2,352 Retweets\n468 Quote Tweets\n22.1K Likes\n/ Getty Images claims Stability AI 'unlawfully' scraped millions of images from its site. It's a significant escalation in the developing legal battles between generative AI firms and content creators.\nGETTY IMAGES\nBy JAMES VINCENT\nJan 17, 2023, 3:30 AM PST 18 Comments / 18 New\nAn image created by Stable Diffusion showing a recreation of Getty Images watermark Image: The Verge / Stable Diffusion","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"serious"},"products_mentioned":["Stable Diffusion"],"entity_ids":["ent_03017e70-f2a0-4aef-b135-4521169d0c1e","ent_d3aef2ec47f98439c7a3995a0edd260e"],"fused_summary":"The speaker discusses generative AI copyright challenges, citing Getty Images suing Stable Diffusion over training data."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_141","window_index":141,"start_ms":1740523,"end_ms":1751863,"duration_ms":11340,"visual_description":"A man addresses the camera while a slide displays news about Twitter blocking data access for OpenAI. The presenter continues explaining data access restrictions with an on-screen article about Twitter and OpenAI.","on_screen_text":"DATA ACCESS & COPYRIGHT\nCEO Elon Musk pauses OpenAI chatbot's access to Twitter's database\nAs the world experimented with OpenAI chatbot service, Twitter CEO said after he learned that OpenAI was accessing the platform's database for training, he put a pause on it\nTopics\nTwitter I Elon Musk\nIANS I San Francisco\nLast Updated at December 4, 2022 20:25 IST\nElon Musk\n@elonmusk\nReplying to @waitbutwhy\nNot surprising, as I just learned that OpenAI had access to Twitter database for training. I put that on pause for now.\ned to understand more about governance structure & revenue plans going forward.\nl was started as open-source & non-profit. Neither are still true.\n4:17 PM Dec 3, 2022\n2,352 Retweets\n468 Quote Tweets\n22.1K Likes\nJoin TechCrunch+\nLogin\nSearch Q\nTechCrunch+\nStartups\nVenture\nFeatured Article\nThe current legal cases against generative Al are just the beginning\nAl that can generate art, text and more is in for a reckoning\nGetty Images is suing the creators of Al art tool Stable Diffusion for scraping its content\nGetty Images claims Stability Al 'unlawfully' scraped millions of images from its site. It's a significant escalation in the developing legal battles between generative Al firms and content creators.","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"thoughtful"},"products_mentioned":["X"],"entity_ids":["ent_1e21f75d-cb8e-4cd4-9c15-c19108075d5c","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","ent_c1eb0a5ad5e25af6e52c706b836ed2e6"],"fused_summary":"The speaker reviews how Elon Musk and Twitter cut off data access for OpenAI following its corporate restructuring."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_142","window_index":142,"start_ms":1751863,"end_ms":1763158,"duration_ms":11295,"visual_description":"A presenter talks beside a presentation that switches from news headlines to a creator workflow diagram. The presenter gestures while the slide transitions to display design research and creation tools.","on_screen_text":"DATA ACCESS & COPYRIGHT\nCEO Elon Musk pauses OpenAI chatbot's access to Twitter's database\nAs the world experimented with OpenAI chatbot service, Twitter CEO said after he \"learned\" that OpenAI was accessing the platform's database for\ntraining\n\", he put a\npause on it\nTopics\nTwitter I Elon Musk\nIANS I San Francisco\nLast Updated at December 4, 2022 20:25 IST\nElon Musk\n@elonmusk\nReplying to @waitbutwhy\nNot surprising, as I just learned that OpenAI had access to Twitter\ndatabase for training. I put that on pause for now.\nneed to understand more about governance structure & revenue plans\ngoing forward.\nlAI was started as open-source & non-profit. Neither are still true.\nPM Dec 3, 2022\n2,352 Retweets\n468 Quote Tweets\n22.1K Likes\nJoin TechCrunch+\nLogin\nSearch Q\nTechCrunch+\nStartups\nVenture\nFeatured Article\nThe current legal cases against\ngenerative AI are just the beginning\nAI that can generate art, text and more is in for a reckoning\nKyle Wiggers @h_j_wiggers / 8:30 AM PST - January 27, 2023\nGetty Images is suing the creators of AI\nart tool Stable Diffusion for scraping its\ncontent\nAn image created by Stable Diffusion showing a recreation of Getty Images'\nGetty Images claims Stability AI\n'unlawfully' scraped millions of\nimages from its site. It's a\nsignificant escalation in the\ndeveloping legal battles between\ngenerative AI firms and content\ncreators.\nBy JAMES VINCENT\nJan 17, 2023, 2:50 AM PST / 18 Comments / 18 New\nRESEARCH\nCREATE\nDISTIRBUTE\nMONETIZE\nPREVIOUSLY\ndribbble\nBehance\nCanva\nBēhance\nshutterstock","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"thoughtful"},"products_mentioned":["Figma","Shutterstock","Behance","Instagram","Canva","Dribbble"],"entity_ids":["ent_5f383f1fec26b0df6bc815d9aee4cf47","ent_62f6aafd-a3e6-40e5-8999-0dbfeebad568","ent_854fd7a20654b1a94e7384a2b3d9b177","ent_b29d737e-0ac5-4e9f-9fd2-1c32f379b8cb","ent_c4d26c337fbbd081b49e1f4973346b86","ent_f8eaf0d5-52e9-4950-91b9-1d7f650e637d"],"fused_summary":"The speaker mentions artist lawsuits against LLM developers before comparing current AI training to traditional creator workflows."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_143","window_index":143,"start_ms":1763158,"end_ms":1777741,"duration_ms":14583,"visual_description":"A man gestures enthusiastically beside a slide showing creator platforms across research, creation, and distribution stages. The presenter gestures with both hands while explaining stages of the digital creative pipeline shown on screen.","on_screen_text":"DATA ACCESS & COPYRIGHT\nRESEARCH\nCREATE\nDISTIRBUTE\nMONETIZE\nPREVIOUSLY\ndribbble\nBēhance\nCanva\nshutterstock","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"medium","camera_type_hint":"medium shot","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Pinterest","Figma","Behance","Instagram","Canva","Dribbble"],"entity_ids":["ent_27473110179dbbafef08367c58bf843b","ent_5f383f1fec26b0df6bc815d9aee4cf47","ent_854fd7a20654b1a94e7384a2b3d9b177","ent_b29d737e-0ac5-4e9f-9fd2-1c32f379b8cb","ent_c4d26c337fbbd081b49e1f4973346b86","ent_f8eaf0d5-52e9-4950-91b9-1d7f650e637d"],"fused_summary":"The speaker walks through the conventional creative workflow, outlining how designers use platforms like Dribbble, Pinterest, Figma, and Canva."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_144","window_index":144,"start_ms":1777741,"end_ms":1791250,"duration_ms":13509,"visual_description":"A presenter sits before a slide detailing data access and copyright workflows across research, creation, and monetization. The presenter speaks to the camera in front of a diagram illustrating copyright in content workflows.","on_screen_text":"DATA ACCESS & COPYRIGHT\nRESEARCH\nCREATE\nDISTIRBUTE\nMONETIZE\nPREVIOUSLY\ndribbble\nBēhance\nCanva\nshutterstock","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["Figma","Shutterstock","Behance","Canva"],"entity_ids":["ent_5f383f1fec26b0df6bc815d9aee4cf47","ent_62f6aafd-a3e6-40e5-8999-0dbfeebad568","ent_854fd7a20654b1a94e7384a2b3d9b177","ent_c4d26c337fbbd081b49e1f4973346b86"],"fused_summary":"The presenter discusses copyright restrictions and using other people's creative content when selling work on platforms like Shutterstock."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_145","window_index":145,"start_ms":1791250,"end_ms":1804583,"duration_ms":13333,"visual_description":"The speaker explains how AI supercharges research inspiration while an on-screen chart reveals modern generative tools. The presenter continues explaining while the graphic shows ChatGPT and DALL-E 2 added to the workflow. The presenter talks about scaling inspiration using large language models.","on_screen_text":"DATA ACCESS & COPYRIGHT\nRESEARCH\nCREATE\nDISTRIBUTE\nMONETIZE\nPREVIOUSLY\ndribbble\nBēhance\nshutterstock\nNOW\nChatGPT\nDall-E 2","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["ChatGPT","DALL·E 2"],"entity_ids":["ent_8a8a91e44930d87b222367e337327e34","ent_9747295228fc6ecd042f9c2f02b5ced5"],"fused_summary":"The host contrasts manual browsing with ChatGPT's vast ingestion of billions of pages for generating ideas."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_146","window_index":146,"start_ms":1804583,"end_ms":1816325,"duration_ms":11742,"visual_description":"The host outlines generative content creation and monetization pipelines against a diagram of AI platforms. The presenter discusses the generative content creation and monetization pipeline beside the slide diagram.","on_screen_text":"DATA ACCESS & COPYRIGHT\nRESEARCH\nCREATE\nDISTIRBUTE\nMONETIZE\nPREVIOUSLY\nBēhance\ndribbble\nCanva\nshutterstock\nNOW\nChatGPT\nDall-E 2","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":["DALL-E","Shutterstock","Stable Diffusion"],"entity_ids":["ent_02988cd6887c1296841b9a5533d9f22f","ent_62f6aafd-a3e6-40e5-8999-0dbfeebad568","ent_d3aef2ec47f98439c7a3995a0edd260e"],"fused_summary":"The speaker covers creating visual assets with DALL-E or Stable Diffusion and subsequently monetizing them on stock image sites."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_147","window_index":147,"start_ms":1816325,"end_ms":1831325,"duration_ms":15000,"visual_description":"The presenter gestures enthusiastically beside the slide detailing AI-generated content workflows and partnerships. The speaker emphasizes the role of AI partnerships in making research and creation accessible.","on_screen_text":"DATA ACCESS & COPYRIGHT\nRESEARCH\nCREATE\nDISTRIBUTE\nMONETIZE\nPREVIOUSLY\ndribbble\nBēhance\nCanva\nshutterstock\nNOW\nChatGPT\nDall-E 2","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"medium","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Shutterstock"],"entity_ids":["ent_62f6aafd-a3e6-40e5-8999-0dbfeebad568","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"The speaker details Shutterstock's partnership with OpenAI to commercialize AI-generated artwork and highlights generative AI as an empowering creative tool."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_148","window_index":148,"start_ms":1831325,"end_ms":1843223,"duration_ms":11898,"visual_description":"A man in a patterned shirt speaks beside a graphic showing workflow tools for research, creation, distribution, and monetization. The presenter stands beside an infographic on data access and copyright, discussing commercialization of AI content.","on_screen_text":"DATA ACCESS & COPYRIGHT\nRESEARCH\nCREATE\nDISTRIBUTE\nMONETIZE\nPREVIOUSLY\ndribbble\nBēhance\nCanva\nshutterstock\nNOW\nChatGPT\nDall-E 2","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"instructive"},"products_mentioned":["Shutterstock","ChatGPT","DALL·E 2"],"entity_ids":["ent_62f6aafd-a3e6-40e5-8999-0dbfeebad568","ent_8a8a91e44930d87b222367e337327e34","ent_9747295228fc6ecd042f9c2f02b5ced5","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"The speaker explains how monetization impacts copyright concerns in AI models while an infographic illustrates traditional versus generative AI workflows."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_149","window_index":149,"start_ms":1843223,"end_ms":1851223,"duration_ms":8000,"visual_description":"The presenter discusses human artistic inspiration before the display switches to a tweet about closed AI systems. The host speaks next to the data access infographic, emphasizing how artists take inspiration. The graphic changes to show an Elon Musk tweet discussing OpenAI under the heading 'Closed Systems & Biases'.","on_screen_text":"DATA ACCESS & COPYRIGHT\nRESEARCH\nCREATE\nDISTRIBUTE\nMONETIZE\nPREVIOUSLY\ndribbble\nBēhance\nCanva\nshutterstock\nNOW\nChatGPT\nDall-E 2\nCLOSED SYSTEMS & BIASES\nElon Musk\n@elonmusk\nReplying to @GRDecter\nOpenAI was created as an open source (which is why I named it \"Open\"AI), non-profit company to serve as a counterweight to Google, but now it has become a closed source, maximum-profit company effectively controlled by Microsoft.\nNot what I intended at all.\n1:36 AM · Feb 17, 2023 · 12.2M Views","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"thoughtful"},"products_mentioned":[],"entity_ids":["ent_1e21f75d-cb8e-4cd4-9c15-c19108075d5c"],"fused_summary":"The speaker argues that AI models draw inspiration just like human artists do, as the visual graphic transitions to a tweet from Elon Musk."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_150","window_index":150,"start_ms":1851223,"end_ms":1861559,"duration_ms":10336,"visual_description":"The presenter addresses closed systems and corporate control beside a featured tweet by Elon Musk. The presenter speaks beside a card displaying Elon Musk's tweet criticizing OpenAI's current closed model.","on_screen_text":"CLOSED SYSTEMS & BIASES\nElon Musk\n@elonmusk\nReplying to @GRDecter\nOpenAI was created as an open source (which is why I named it \"Open\"\nAI), non-profit company to serve as a counterweight to Google, but now\nit has become a closed source, maximum-profit company effectively\ncontrolled by Microsoft.\nNot what I intended at all.\n1:36 AM · Feb 17, 2023 · 12.2M Views","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"serious"},"products_mentioned":[],"entity_ids":["ent_1e21f75d-cb8e-4cd4-9c15-c19108075d5c","ent_9016355d-416c-4e35-b834-35f656db4dad","ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb"],"fused_summary":"The presenter highlights concerns raised by Elon Musk regarding OpenAI's transition toward a closed commercial model serving Microsoft."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_151","window_index":151,"start_ms":1861559,"end_ms":1869875,"duration_ms":8316,"visual_description":"The host explains the origin of OpenAI as a counterweight to tech monopolies beside the on-screen tweet graphic. The host gestures while discussing the risk of AI control being locked down within a single corporation.","on_screen_text":"CLOSED SYSTEMS & BIASES\nElon Musk\n@elonmusk\nReplying to @GRDecter\nOpenAI was created as an open source (which is why I named it \"Open\"\nAI), non-profit company to serve as a counterweight to Google, but now\nit has become a closed source, maximum-profit company effectively\ncontrolled by Microsoft.\nNot what I intended at all.\n1:36 AM · Feb 17, 2023 · 12.2M Views","facets":{"composition":"split screen","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium shot","dominant_speaker_emotion":"focused"},"products_mentioned":[],"entity_ids":["ent_1e21f75d-cb8e-4cd4-9c15-c19108075d5c","ent_5ee44b5c-468f-4de1-a66a-bd96483367a7"],"fused_summary":"The host notes that OpenAI was founded to prevent single-entity dominance over AI technology, referencing Google."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_152","window_index":152,"start_ms":1869875,"end_ms":1883215,"duration_ms":13340,"visual_description":"A man speaks on the left side of the frame in front of a presentation slide showing historical images of early printing presses. The presenter speaks animatedly beside a slide depicting historical printing equipment and illustrations.","on_screen_text":"CLOSED SYSTEMS & BIASES\nLOCOMOTIVE ACT.\n1861 (SECTION 6.)\nNOTICE TO OWNERS AND\nDRIVERS OF LOCOMOTIVES\nPORT OF BRISTOL\nNOTICE","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Gutenberg Press"],"entity_ids":["ent_b99b525cd404f73f2002cdb3fc3cf572"],"fused_summary":"The speaker explains historical licensing regulations imposed on the Gutenberg press beside a slide illustrating printing history."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_153","window_index":153,"start_ms":1883215,"end_ms":1893358,"duration_ms":10143,"visual_description":"The presenter discusses early transportation regulations next to historical images illustrating the UK Locomotive Act. The presenter gestures while discussing historical rules surrounding early vehicles beside slide illustrations.","on_screen_text":"CLOSED SYSTEMS & BIASES\nLOCOMOTIVE ACT. 1861. (SECTION 6.)\nNOTICE TO OWNERS AND DRIVERS OF LOCOMOTIVES\nPORT OF BRISTOL NOTICE","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"instructive"},"products_mentioned":[],"entity_ids":["ent_a13a7733-0be2-4e51-9318-25622e7b6fbd","ent_da6652be00eed8664a95f3abaf6b3634"],"fused_summary":"The presenter discusses historical regulatory pushbacks such as the Locomotive Act in the UK beside archival pictures."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_154","window_index":154,"start_ms":1893358,"end_ms":1907623,"duration_ms":14265,"visual_description":"The presenter describes restrictive red-flag laws before the slide switches to the website of Stability AI. The presenter continues gesturing while explaining the red flag walking requirement for early cars. The background presentation switches to the dark purple homepage of Stability AI.","on_screen_text":"CLOSED SYSTEMS & BIASES\nLOCOMOTIVE ACT.\n1861. (SECTION 6.)\nNOTICE TO OWNERS AND\nDRIVERS OF LOCOMOTIVES\nPORT OF BRISTOL\nNOTICE\nVEHICLES WITH AN AXLE WEIGHT MORE THAN\n20 TONS WHEN LADEN CANNOT SAFELY\nPASS OVER THIS BRIDGE.\nstability.ai\nAPI\nNews\nFAQ\nEnglish\nAI by the people, for the people\nDesigning and implementing solutions using collective intelligence and augmented technology","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"thoughtful"},"products_mentioned":[],"entity_ids":["ent_f6677f64-31a6-4383-b60c-91f0851302c7"],"fused_summary":"The host concludes describing restrictive historical vehicle laws before displaying the Stability AI homepage on screen."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_155","window_index":155,"start_ms":1907623,"end_ms":1921998,"duration_ms":14375,"visual_description":"The host talks alongside Stability AI's homepage before the visual cuts to a stage interview discussing AI resets. The host discusses open-source AI beside the Stability AI webpage showing their slogan. The slide changes to an interview photograph with the header text stating AI is about to reset everything.","on_screen_text":"stability.ai\nAPI\nNews\nFAQ\nEnglish\nCLOSED SYSTEMS & BIASES\nAI by the people, for the people\nDesigning and implementing solutions using collective intelligence and augmented technology\nAI IS ABOUT TO RESET EVERYTHING","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":["Stable Diffusion"],"entity_ids":["ent_d3aef2ec47f98439c7a3995a0edd260e","ent_f6677f64-31a6-4383-b60c-91f0851302c7"],"fused_summary":"The presenter discusses Stability AI and Stable Diffusion before transitioning to a slide showing an interview panel."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_156","window_index":156,"start_ms":1921998,"end_ms":1936998,"duration_ms":15000,"visual_description":"The host speaks beside an interview clip before the video cuts to Sam Altman speaking on stage. The host speaks in the foreground beside a slide showing a stage interview with Sam Altman. Sam Altman speaks while seated in a chair on a stage in front of a blue decorative backdrop.","on_screen_text":"AI IS ABOUT TO RESET EVERYTHING\nSAM ALTMAN\nGRAYLOCK","facets":{"composition":"single subject","lighting_class":"stage","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"thoughtful"},"products_mentioned":[],"entity_ids":["ent_7ec30179-e819-470f-a3e4-9917fb57fca9","ent_e30d6dc0ccb5641d50ac160c2eeb2747"],"fused_summary":"The host introduces the topic of AI resetting industries, transitioning into Sam Altman discussing how the marginal costs of intelligence and energy will trend toward zero."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_157","window_index":157,"start_ms":1936998,"end_ms":1948917,"duration_ms":11919,"visual_description":"Sam Altman continues his presentation on stage before the shot widens to show both speakers on the panel. Sam Altman gestures and speaks while seated in front of a decorative backdrop. A wide view shows two men seated in chairs across a small table on a stage with a Greylock backdrop.","on_screen_text":"SAM ALTMAN\nGRAYLOCK\ngreylock","facets":{"composition":"two-shot","lighting_class":"stage","motion_class":"low","camera_type_hint":"medium wide shot","dominant_speaker_emotion":"thoughtful"},"products_mentioned":[],"entity_ids":["ent_7ec30179-e819-470f-a3e4-9917fb57fca9","ent_e30d6dc0ccb5641d50ac160c2eeb2747"],"fused_summary":"Sam Altman explains that intelligence and energy are key inputs into the cost of nearly all goods and services."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_158","window_index":158,"start_ms":1948917,"end_ms":1962983,"duration_ms":14066,"visual_description":"The host talks to the camera in front of a graphic presenting statistics about Fortune 500 company spending. The host gestures excitedly next to a graphic detailing payroll costs with animated avatars at the top.","on_screen_text":"AI IS ABOUT TO RESET EVERYTHING\nFor a typical Fortune 500 company, payroll is $1 to $2 billion per year, which averages between 50% to 60% of company spending\nDeloitte.","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"medium","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":["ent_5e44ccec-4517-469d-9d23-e79fc5531291","ent_ae8f3da2281f0437bdcafc8a0058ff84"],"fused_summary":"The host outlines corporate labor costs, referencing data showing payroll represents over half of spending for Fortune 500 companies."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_159","window_index":159,"start_ms":1962983,"end_ms":1973297,"duration_ms":10314,"visual_description":"The host explains how AI will reshape various sectors while the presentation graphic transitions to show six application domains. The host gestures beside the slide detailing company payroll costs. The presentation switches to display six application categories as the host continues speaking.","on_screen_text":"AI IS ABOUT TO RESET EVERYTHING\nFor a typical Fortune 500 company, payroll is\n$1 to $2 billion per year, which averages\nbetween 50% to 60% of company spending\nDeloitte.\nEDUCATION\nBUSINESS\nHEALTHCARE\nART\nSOFTWARE\nGAMING","facets":{"composition":"split screen","lighting_class":"indoor artificial","motion_class":"medium","camera_type_hint":"medium close-up","dominant_speaker_emotion":"enthusiastic"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The host discusses the broad transformative effect of free labor and intelligence across education, healthcare, business, art, software, and gaming."},{"kind":"window","id":"vwin_video_reLhJnQd-Lo_160","window_index":160,"start_ms":1973297,"end_ms":1981297,"duration_ms":8000,"visual_description":"A man in a colorful patterned shirt speaks in front of an educational slide titled 'AI IS ABOUT TO RESET EVERYTHING'. The host speaks directly to the camera with an infographic slide displayed in the background, making hand gestures as he signs off.","on_screen_text":"AI IS ABOUT TO RESET EVERYTHING\nEDUCATION\nBUSINESS\nHEALTHCARE\nART\nSOFTWARE\nGAMING","facets":{"composition":"single subject","lighting_class":"studio","motion_class":"low","camera_type_hint":"medium close-up","dominant_speaker_emotion":"friendly"},"products_mentioned":[],"entity_ids":[],"fused_summary":"The speaker concludes the video by summarizing how AI will transform key sectors including education, business, and healthcare, thanking viewers and signing off."}],"moments":[{"arv":"1.0","id":"mom_yj6hehehyib25gb6bosro2kjyh","moment_uri":"arv:video_reLhJnQd-Lo#t=0,73978","asset_id":"video_reLhJnQd-Lo","start_ms":0,"end_ms":73978,"title":"Intro and the accelerating impact of AI"},{"arv":"1.0","id":"mom_yspuj7cwgj24jbhwwf62e6akmk","moment_uri":"arv:video_reLhJnQd-Lo#t=73978,112037","asset_id":"video_reLhJnQd-Lo","start_ms":73978,"end_ms":112037,"snap":"sentence","title":"The new AI platform era"},{"arv":"1.0","id":"mom_p4a4pjseieaj3obyqpi77y6v3s","moment_uri":"arv:video_reLhJnQd-Lo#t=112037,150083","asset_id":"video_reLhJnQd-Lo","start_ms":112037,"end_ms":150083,"snap":"sentence","title":"Why generative AI took off now"},{"arv":"1.0","id":"mom_dnhjvoihlclzjx7hch3mwatska","moment_uri":"arv:video_reLhJnQd-Lo#t=150083,273083","asset_id":"video_reLhJnQd-Lo","start_ms":150083,"end_ms":273083,"title":"Key drivers: Interface evolution"},{"arv":"1.0","id":"mom_2sjood3ragleanvc46tetsedrs","moment_uri":"arv:video_reLhJnQd-Lo#t=273083,362375","asset_id":"video_reLhJnQd-Lo","start_ms":273083,"end_ms":362375,"title":"Key drivers: costs and data growth"},{"arv":"1.0","id":"mom_lajh7gvhl6lo4bjfd45t5i2jln","moment_uri":"arv:video_reLhJnQd-Lo#t=362375,472958","asset_id":"video_reLhJnQd-Lo","start_ms":362375,"end_ms":472958,"snap":"shot","title":"Key drivers: Model quality progression"},{"arv":"1.0","id":"mom_qof6jb2zlcci5uppi62z6btbh4","moment_uri":"arv:video_reLhJnQd-Lo#t=472958,519560","asset_id":"video_reLhJnQd-Lo","start_ms":472958,"end_ms":519560,"snap":"shot","title":"Society evolving from laborer to narrator"},{"arv":"1.0","id":"mom_fuhtntreyg5pcryngusmihukqg","moment_uri":"arv:video_reLhJnQd-Lo#t=519560,802638","asset_id":"video_reLhJnQd-Lo","start_ms":519560,"end_ms":802638,"snap":"sentence","title":"Large language models explained and use cases"},{"arv":"1.0","id":"mom_ohthrmfp6iaz76ivmxkwudcen6","moment_uri":"arv:video_reLhJnQd-Lo#t=802638,973159","asset_id":"video_reLhJnQd-Lo","start_ms":802638,"end_ms":973159,"snap":"sentence","title":"Image AI models explained and use cases"},{"arv":"1.0","id":"mom_pibvn4u66fe4rexqve5vfhwbmg","moment_uri":"arv:video_reLhJnQd-Lo#t=973159,1124760","asset_id":"video_reLhJnQd-Lo","start_ms":973159,"end_ms":1124760,"snap":"sentence","title":"Audio AI models: Music and speech"},{"arv":"1.0","id":"mom_ev4aavsqyxf5odv45xo76j4jx7","moment_uri":"arv:video_reLhJnQd-Lo#t=1124760,1325035","asset_id":"video_reLhJnQd-Lo","start_ms":1124760,"end_ms":1325035,"snap":"sentence","title":"Video AI models: 3D, avatars, and gaming"},{"arv":"1.0","id":"mom_kbdhqm4jeibt6jadwrvkfiy57k","moment_uri":"arv:video_reLhJnQd-Lo#t=1325035,1396891","asset_id":"video_reLhJnQd-Lo","start_ms":1325035,"end_ms":1396891,"snap":"sentence","title":"Code AI models and developer acceleration"},{"arv":"1.0","id":"mom_bmpvepa7irjhddcldllgtcc3je","moment_uri":"arv:video_reLhJnQd-Lo#t=1396891,1524917","asset_id":"video_reLhJnQd-Lo","start_ms":1396891,"end_ms":1524917,"snap":"sentence","title":"Science AI models and DeepMind AlphaFold"},{"arv":"1.0","id":"mom_ikymyf3wvi6m26lfpabqxaeu5n","moment_uri":"arv:video_reLhJnQd-Lo#t=1524917,1648184","asset_id":"video_reLhJnQd-Lo","start_ms":1524917,"end_ms":1648184,"snap":"shot","title":"Generative AI challenges and output errors"},{"arv":"1.0","id":"mom_6t32senbrqz6vuv42ztf22pcib","moment_uri":"arv:video_reLhJnQd-Lo#t=1648184,1721532","asset_id":"video_reLhJnQd-Lo","start_ms":1648184,"end_ms":1721532,"snap":"sentence","title":"Cost, slow inference, and memory limits"},{"arv":"1.0","id":"mom_tihasmvdjq4jjoips7io4uwgz3","moment_uri":"arv:video_reLhJnQd-Lo#t=1721532,1907623","asset_id":"video_reLhJnQd-Lo","start_ms":1721532,"end_ms":1907623,"snap":"sentence","title":"Data access, copyright, and closed systems"},{"arv":"1.0","id":"mom_ghk4i2i2rkzqtszxdvr6bysgmm","moment_uri":"arv:video_reLhJnQd-Lo#t=1907623,1981297","asset_id":"video_reLhJnQd-Lo","start_ms":1907623,"end_ms":1981297,"title":"AI will reset everything"},{"arv":"1.0","id":"mom_zpgztok2ciz3odbpsdbqr5kdmw","moment_uri":"arv:video_reLhJnQd-Lo#t=0,9542","asset_id":"video_reLhJnQd-Lo","start_ms":0,"end_ms":9542},{"arv":"1.0","id":"mom_qen5lhwbuhdbl556da5mpilfv2","moment_uri":"arv:video_reLhJnQd-Lo#t=9542,18167","asset_id":"video_reLhJnQd-Lo","start_ms":9542,"end_ms":18167},{"arv":"1.0","id":"mom_klc5pvgzb2zmpqptuthumpne7x","moment_uri":"arv:video_reLhJnQd-Lo#t=18167,33167","asset_id":"video_reLhJnQd-Lo","start_ms":18167,"end_ms":33167},{"arv":"1.0","id":"mom_ksnt2f2ikwlm5ajv5y5jadyp74","moment_uri":"arv:video_reLhJnQd-Lo#t=33167,48167","asset_id":"video_reLhJnQd-Lo","start_ms":33167,"end_ms":48167},{"arv":"1.0","id":"mom_oznkmqslpbaeej6ggtla7oqguv","moment_uri":"arv:video_reLhJnQd-Lo#t=48167,59936","asset_id":"video_reLhJnQd-Lo","start_ms":48167,"end_ms":59936},{"arv":"1.0","id":"mom_6yvcvsibd5tluq76ksjwbafd4q","moment_uri":"arv:video_reLhJnQd-Lo#t=59936,74091","asset_id":"video_reLhJnQd-Lo","start_ms":59936,"end_ms":74091},{"arv":"1.0","id":"mom_bxrrhxncuubbbss6t5gksyk5wb","moment_uri":"arv:video_reLhJnQd-Lo#t=74091,89091","asset_id":"video_reLhJnQd-Lo","start_ms":74091,"end_ms":89091},{"arv":"1.0","id":"mom_gdlskkaejgy27yffx6pweyhsc5","moment_uri":"arv:video_reLhJnQd-Lo#t=89091,102924","asset_id":"video_reLhJnQd-Lo","start_ms":89091,"end_ms":102924},{"arv":"1.0","id":"mom_hxmztk6vm7o7d6vmnur5w5wm4d","moment_uri":"arv:video_reLhJnQd-Lo#t=102924,112503","asset_id":"video_reLhJnQd-Lo","start_ms":102924,"end_ms":112503},{"arv":"1.0","id":"mom_o6jzyoibsfqaxrur6452vu2sei","moment_uri":"arv:video_reLhJnQd-Lo#t=112503,125198","asset_id":"video_reLhJnQd-Lo","start_ms":112503,"end_ms":125198},{"arv":"1.0","id":"mom_sb76i6tcs4m3hjoejmyibc4gt5","moment_uri":"arv:video_reLhJnQd-Lo#t=125198,135988","asset_id":"video_reLhJnQd-Lo","start_ms":125198,"end_ms":135988},{"arv":"1.0","id":"mom_wpawxys2w6mbgr3yx2avx7nzbg","moment_uri":"arv:video_reLhJnQd-Lo#t=135988,150083","asset_id":"video_reLhJnQd-Lo","start_ms":135988,"end_ms":150083},{"arv":"1.0","id":"mom_wrpbu2x6o3564eimuaveabkjtf","moment_uri":"arv:video_reLhJnQd-Lo#t=150083,161202","asset_id":"video_reLhJnQd-Lo","start_ms":150083,"end_ms":161202},{"arv":"1.0","id":"mom_hegaug32shs7n3wmycoy6xkqwo","moment_uri":"arv:video_reLhJnQd-Lo#t=161202,176202","asset_id":"video_reLhJnQd-Lo","start_ms":161202,"end_ms":176202},{"arv":"1.0","id":"mom_iowpkzrqso4waqtbfoumodox3m","moment_uri":"arv:video_reLhJnQd-Lo#t=176202,186833","asset_id":"video_reLhJnQd-Lo","start_ms":176202,"end_ms":186833},{"arv":"1.0","id":"mom_hzzbhqfhju46d2pqqlqzxyfdeg","moment_uri":"arv:video_reLhJnQd-Lo#t=186833,200765","asset_id":"video_reLhJnQd-Lo","start_ms":186833,"end_ms":200765},{"arv":"1.0","id":"mom_bg34w5gdxlp26fn5w6x6rby4nf","moment_uri":"arv:video_reLhJnQd-Lo#t=200765,212426","asset_id":"video_reLhJnQd-Lo","start_ms":200765,"end_ms":212426},{"arv":"1.0","id":"mom_k6pa54phflzjlutvsiqtpxuxxk","moment_uri":"arv:video_reLhJnQd-Lo#t=212426,221499","asset_id":"video_reLhJnQd-Lo","start_ms":212426,"end_ms":221499},{"arv":"1.0","id":"mom_d4f54f3qlc7ul5pn5eex5gbmbg","moment_uri":"arv:video_reLhJnQd-Lo#t=221499,236499","asset_id":"video_reLhJnQd-Lo","start_ms":221499,"end_ms":236499},{"arv":"1.0","id":"mom_4docq4mthm56gzzu3mnyrmxq2q","moment_uri":"arv:video_reLhJnQd-Lo#t=236499,250129","asset_id":"video_reLhJnQd-Lo","start_ms":236499,"end_ms":250129},{"arv":"1.0","id":"mom_3vmeephtsbgaep3ex62xyw5mih","moment_uri":"arv:video_reLhJnQd-Lo#t=250129,263359","asset_id":"video_reLhJnQd-Lo","start_ms":250129,"end_ms":263359},{"arv":"1.0","id":"mom_yiun2qpiia2zawc572edu4rhlo","moment_uri":"arv:video_reLhJnQd-Lo#t=263359,273083","asset_id":"video_reLhJnQd-Lo","start_ms":263359,"end_ms":273083},{"arv":"1.0","id":"mom_nma7mpupb5ut3pfpye3b67l2cx","moment_uri":"arv:video_reLhJnQd-Lo#t=273083,287919","asset_id":"video_reLhJnQd-Lo","start_ms":273083,"end_ms":287919},{"arv":"1.0","id":"mom_qxy5woxsagb2q5sxzbghabccdj","moment_uri":"arv:video_reLhJnQd-Lo#t=287919,302919","asset_id":"video_reLhJnQd-Lo","start_ms":287919,"end_ms":302919},{"arv":"1.0","id":"mom_e6nbnk5r4c6i4cxejc4zcn6quf","moment_uri":"arv:video_reLhJnQd-Lo#t=302919,316832","asset_id":"video_reLhJnQd-Lo","start_ms":302919,"end_ms":316832},{"arv":"1.0","id":"mom_elkeil5gbb7uxxapxikapfg4wx","moment_uri":"arv:video_reLhJnQd-Lo#t=316832,327458","asset_id":"video_reLhJnQd-Lo","start_ms":316832,"end_ms":327458},{"arv":"1.0","id":"mom_ncop4tqnwkryluv7klol75yk3u","moment_uri":"arv:video_reLhJnQd-Lo#t=327458,338849","asset_id":"video_reLhJnQd-Lo","start_ms":327458,"end_ms":338849},{"arv":"1.0","id":"mom_ch44a3i63czrxmeiex6nm5ubvh","moment_uri":"arv:video_reLhJnQd-Lo#t=338849,347000","asset_id":"video_reLhJnQd-Lo","start_ms":338849,"end_ms":347000},{"arv":"1.0","id":"mom_4ev3igsplcd3ju7ehvdcakerqc","moment_uri":"arv:video_reLhJnQd-Lo#t=347000,361758","asset_id":"video_reLhJnQd-Lo","start_ms":347000,"end_ms":361758},{"arv":"1.0","id":"mom_gneefdgabf2l3cu5gpyvxfnehd","moment_uri":"arv:video_reLhJnQd-Lo#t=361758,375146","asset_id":"video_reLhJnQd-Lo","start_ms":361758,"end_ms":375146},{"arv":"1.0","id":"mom_o3sxidslkm5flad4irs5wtaq2k","moment_uri":"arv:video_reLhJnQd-Lo#t=375146,384292","asset_id":"video_reLhJnQd-Lo","start_ms":375146,"end_ms":384292},{"arv":"1.0","id":"mom_jz33mtf26zhxzixd4rz22jkypu","moment_uri":"arv:video_reLhJnQd-Lo#t=384292,398083","asset_id":"video_reLhJnQd-Lo","start_ms":384292,"end_ms":398083},{"arv":"1.0","id":"mom_lj6bucnzudtfvnf4xgrjko5o7r","moment_uri":"arv:video_reLhJnQd-Lo#t=398083,411083","asset_id":"video_reLhJnQd-Lo","start_ms":398083,"end_ms":411083},{"arv":"1.0","id":"mom_5aqrjn2rztbwiiyfr5go3c6chk","moment_uri":"arv:video_reLhJnQd-Lo#t=411083,426083","asset_id":"video_reLhJnQd-Lo","start_ms":411083,"end_ms":426083},{"arv":"1.0","id":"mom_zcmqmsdzzmocw7tv6h6ctpif7n","moment_uri":"arv:video_reLhJnQd-Lo#t=426083,441083","asset_id":"video_reLhJnQd-Lo","start_ms":426083,"end_ms":441083},{"arv":"1.0","id":"mom_axrtmxgtltdducaigpdpu3c3ug","moment_uri":"arv:video_reLhJnQd-Lo#t=441083,455750","asset_id":"video_reLhJnQd-Lo","start_ms":441083,"end_ms":455750},{"arv":"1.0","id":"mom_hpdz7bukjmnouvyyv6druyjvsl","moment_uri":"arv:video_reLhJnQd-Lo#t=455750,465255","asset_id":"video_reLhJnQd-Lo","start_ms":455750,"end_ms":465255},{"arv":"1.0","id":"mom_fxn2t2y7zvzaoetd77uuokgfbn","moment_uri":"arv:video_reLhJnQd-Lo#t=465255,473255","asset_id":"video_reLhJnQd-Lo","start_ms":465255,"end_ms":473255},{"arv":"1.0","id":"mom_33oubytcctogafzntrf27b65le","moment_uri":"arv:video_reLhJnQd-Lo#t=473255,486792","asset_id":"video_reLhJnQd-Lo","start_ms":473255,"end_ms":486792},{"arv":"1.0","id":"mom_5vop5cugskwxtfu5dg7g5mvqtl","moment_uri":"arv:video_reLhJnQd-Lo#t=486792,497750","asset_id":"video_reLhJnQd-Lo","start_ms":486792,"end_ms":497750},{"arv":"1.0","id":"mom_q3vllsmgxr5btr6xohdq4qrxsr","moment_uri":"arv:video_reLhJnQd-Lo#t=497750,508689","asset_id":"video_reLhJnQd-Lo","start_ms":497750,"end_ms":508689},{"arv":"1.0","id":"mom_j6sa3pfqhvsqmra22gqlhmirf2","moment_uri":"arv:video_reLhJnQd-Lo#t=508689,519994","asset_id":"video_reLhJnQd-Lo","start_ms":508689,"end_ms":519994},{"arv":"1.0","id":"mom_wiitspm4pkqm4qqbjjozaehj6r","moment_uri":"arv:video_reLhJnQd-Lo#t=519994,530678","asset_id":"video_reLhJnQd-Lo","start_ms":519994,"end_ms":530678},{"arv":"1.0","id":"mom_w6geo5gdlvlfquwhvbdv4c6za3","moment_uri":"arv:video_reLhJnQd-Lo#t=530678,545678","asset_id":"video_reLhJnQd-Lo","start_ms":530678,"end_ms":545678},{"arv":"1.0","id":"mom_kbekaamkm4okkccr7pcaqgaprx","moment_uri":"arv:video_reLhJnQd-Lo#t=545678,557917","asset_id":"video_reLhJnQd-Lo","start_ms":545678,"end_ms":557917},{"arv":"1.0","id":"mom_s33o7mf5jyk4isg52eaw6pxf33","moment_uri":"arv:video_reLhJnQd-Lo#t=557917,572917","asset_id":"video_reLhJnQd-Lo","start_ms":557917,"end_ms":572917},{"arv":"1.0","id":"mom_7muaj4levpqj5id3azf3krnhbz","moment_uri":"arv:video_reLhJnQd-Lo#t=572917,587917","asset_id":"video_reLhJnQd-Lo","start_ms":572917,"end_ms":587917},{"arv":"1.0","id":"mom_3ym7kuj7t3kwmsngjc3mxmlztx","moment_uri":"arv:video_reLhJnQd-Lo#t=587917,600432","asset_id":"video_reLhJnQd-Lo","start_ms":587917,"end_ms":600432},{"arv":"1.0","id":"mom_gckrom2t2ryb3ats42qvkefn4p","moment_uri":"arv:video_reLhJnQd-Lo#t=600432,610542","asset_id":"video_reLhJnQd-Lo","start_ms":600432,"end_ms":610542},{"arv":"1.0","id":"mom_epjyix3pfpgimtu7eycwj5p3rw","moment_uri":"arv:video_reLhJnQd-Lo#t=610542,625208","asset_id":"video_reLhJnQd-Lo","start_ms":610542,"end_ms":625208},{"arv":"1.0","id":"mom_sdwfuf63oeb5lpubhbgv2wlvnb","moment_uri":"arv:video_reLhJnQd-Lo#t=625208,640203","asset_id":"video_reLhJnQd-Lo","start_ms":625208,"end_ms":640203},{"arv":"1.0","id":"mom_lpd2rgrnf5xlr3ep57d7d2vedc","moment_uri":"arv:video_reLhJnQd-Lo#t=640203,655203","asset_id":"video_reLhJnQd-Lo","start_ms":640203,"end_ms":655203},{"arv":"1.0","id":"mom_vb7fuxyyq5xwhk3pq4eeyswbvb","moment_uri":"arv:video_reLhJnQd-Lo#t=655203,668367","asset_id":"video_reLhJnQd-Lo","start_ms":655203,"end_ms":668367},{"arv":"1.0","id":"mom_bsojyk6kiczw63asv6k23xcsb7","moment_uri":"arv:video_reLhJnQd-Lo#t=668367,678033","asset_id":"video_reLhJnQd-Lo","start_ms":668367,"end_ms":678033},{"arv":"1.0","id":"mom_mzkji7j3lmze45ypbvrl3hu6gs","moment_uri":"arv:video_reLhJnQd-Lo#t=678033,686958","asset_id":"video_reLhJnQd-Lo","start_ms":678033,"end_ms":686958},{"arv":"1.0","id":"mom_htfzjdfl7zbw4sxdd7f65yd76x","moment_uri":"arv:video_reLhJnQd-Lo#t=686958,698833","asset_id":"video_reLhJnQd-Lo","start_ms":686958,"end_ms":698833},{"arv":"1.0","id":"mom_rbdoeijov5oyqhu2xp7bzepzxd","moment_uri":"arv:video_reLhJnQd-Lo#t=698833,713167","asset_id":"video_reLhJnQd-Lo","start_ms":698833,"end_ms":713167},{"arv":"1.0","id":"mom_boy7hzl2ard6pnrq5q2suzyfbl","moment_uri":"arv:video_reLhJnQd-Lo#t=713167,725250","asset_id":"video_reLhJnQd-Lo","start_ms":713167,"end_ms":725250},{"arv":"1.0","id":"mom_yzyxnvarmzerwzvxdmhjdl6hzo","moment_uri":"arv:video_reLhJnQd-Lo#t=725250,737167","asset_id":"video_reLhJnQd-Lo","start_ms":725250,"end_ms":737167},{"arv":"1.0","id":"mom_dhjq7y3dbci6c3glhglxzcrhtd","moment_uri":"arv:video_reLhJnQd-Lo#t=737167,751417","asset_id":"video_reLhJnQd-Lo","start_ms":737167,"end_ms":751417},{"arv":"1.0","id":"mom_pxepkabvfohcm5skkjwq2ycxrr","moment_uri":"arv:video_reLhJnQd-Lo#t=751417,762417","asset_id":"video_reLhJnQd-Lo","start_ms":751417,"end_ms":762417},{"arv":"1.0","id":"mom_nlzx2tfnp7a6bn3mmfjsty7cbi","moment_uri":"arv:video_reLhJnQd-Lo#t=762417,775890","asset_id":"video_reLhJnQd-Lo","start_ms":762417,"end_ms":775890},{"arv":"1.0","id":"mom_hc3wbfaejenevanfxbn3bfesj4","moment_uri":"arv:video_reLhJnQd-Lo#t=775890,790804","asset_id":"video_reLhJnQd-Lo","start_ms":775890,"end_ms":790804},{"arv":"1.0","id":"mom_smv3hbbatwvsetqdqmxadz2grd","moment_uri":"arv:video_reLhJnQd-Lo#t=790804,801275","asset_id":"video_reLhJnQd-Lo","start_ms":790804,"end_ms":801275},{"arv":"1.0","id":"mom_euyzqep3xcpywzyi63q5mzbptq","moment_uri":"arv:video_reLhJnQd-Lo#t=801275,809275","asset_id":"video_reLhJnQd-Lo","start_ms":801275,"end_ms":809275},{"arv":"1.0","id":"mom_haw2gbdsfalr7igz6hnq4lebne","moment_uri":"arv:video_reLhJnQd-Lo#t=809275,824198","asset_id":"video_reLhJnQd-Lo","start_ms":809275,"end_ms":824198},{"arv":"1.0","id":"mom_z4k5gxyv7ytguf7s7c2tdioq6v","moment_uri":"arv:video_reLhJnQd-Lo#t=824198,834125","asset_id":"video_reLhJnQd-Lo","start_ms":824198,"end_ms":834125},{"arv":"1.0","id":"mom_3wr5kbdkpyf2koyl5svbnvmj7p","moment_uri":"arv:video_reLhJnQd-Lo#t=834125,848650","asset_id":"video_reLhJnQd-Lo","start_ms":834125,"end_ms":848650},{"arv":"1.0","id":"mom_cz5h7wrixmrk5yaqm4wxbed7st","moment_uri":"arv:video_reLhJnQd-Lo#t=848650,863650","asset_id":"video_reLhJnQd-Lo","start_ms":848650,"end_ms":863650},{"arv":"1.0","id":"mom_iqip5t44fxc45bnh6uin6wwonl","moment_uri":"arv:video_reLhJnQd-Lo#t=863650,878650","asset_id":"video_reLhJnQd-Lo","start_ms":863650,"end_ms":878650},{"arv":"1.0","id":"mom_6lsnoxtpunqpurzfgoq3ro5tob","moment_uri":"arv:video_reLhJnQd-Lo#t=878650,891549","asset_id":"video_reLhJnQd-Lo","start_ms":878650,"end_ms":891549},{"arv":"1.0","id":"mom_bjyww2fa7eq6ggeheyqp7a4hj5","moment_uri":"arv:video_reLhJnQd-Lo#t=891549,906283","asset_id":"video_reLhJnQd-Lo","start_ms":891549,"end_ms":906283},{"arv":"1.0","id":"mom_3ql65hyax42chelt22aads5jox","moment_uri":"arv:video_reLhJnQd-Lo#t=906283,919458","asset_id":"video_reLhJnQd-Lo","start_ms":906283,"end_ms":919458},{"arv":"1.0","id":"mom_hs3j4vsueoktliull5zguphzs2","moment_uri":"arv:video_reLhJnQd-Lo#t=919458,927500","asset_id":"video_reLhJnQd-Lo","start_ms":919458,"end_ms":927500},{"arv":"1.0","id":"mom_pbnycfhdubfxvo5wbmet7lfnlp","moment_uri":"arv:video_reLhJnQd-Lo#t=927500,939083","asset_id":"video_reLhJnQd-Lo","start_ms":927500,"end_ms":939083},{"arv":"1.0","id":"mom_enlpzt3ugteqywjfwmqqqcrxv2","moment_uri":"arv:video_reLhJnQd-Lo#t=939083,950083","asset_id":"video_reLhJnQd-Lo","start_ms":939083,"end_ms":950083},{"arv":"1.0","id":"mom_x5hx2okkdvfv475zswhlbsljsk","moment_uri":"arv:video_reLhJnQd-Lo#t=950083,963386","asset_id":"video_reLhJnQd-Lo","start_ms":950083,"end_ms":963386},{"arv":"1.0","id":"mom_e7z4ygr556tvxtmor46cjimb25","moment_uri":"arv:video_reLhJnQd-Lo#t=963386,973083","asset_id":"video_reLhJnQd-Lo","start_ms":963386,"end_ms":973083},{"arv":"1.0","id":"mom_olkfp26fc4spyjgt75ki42akou","moment_uri":"arv:video_reLhJnQd-Lo#t=973083,985160","asset_id":"video_reLhJnQd-Lo","start_ms":973083,"end_ms":985160},{"arv":"1.0","id":"mom_3jgt4nwoqpsedj7ozs5fecu5he","moment_uri":"arv:video_reLhJnQd-Lo#t=985160,999645","asset_id":"video_reLhJnQd-Lo","start_ms":985160,"end_ms":999645},{"arv":"1.0","id":"mom_oqaynz5uis4ywdrft5irrshjio","moment_uri":"arv:video_reLhJnQd-Lo#t=999645,1009435","asset_id":"video_reLhJnQd-Lo","start_ms":999645,"end_ms":1009435},{"arv":"1.0","id":"mom_px77outyt5uq6oge3ei6vfv6br","moment_uri":"arv:video_reLhJnQd-Lo#t=1009435,1021189","asset_id":"video_reLhJnQd-Lo","start_ms":1009435,"end_ms":1021189},{"arv":"1.0","id":"mom_3vk6fuxzuohjxzkabiip4ltefv","moment_uri":"arv:video_reLhJnQd-Lo#t=1021189,1034949","asset_id":"video_reLhJnQd-Lo","start_ms":1021189,"end_ms":1034949},{"arv":"1.0","id":"mom_bugtry3uqc2xcq2pfxqndim3sr","moment_uri":"arv:video_reLhJnQd-Lo#t=1034949,1047882","asset_id":"video_reLhJnQd-Lo","start_ms":1034949,"end_ms":1047882},{"arv":"1.0","id":"mom_o7sghckstqwtvvzmiwfznmfbgt","moment_uri":"arv:video_reLhJnQd-Lo#t=1047882,1062882","asset_id":"video_reLhJnQd-Lo","start_ms":1047882,"end_ms":1062882},{"arv":"1.0","id":"mom_ywca3nhrdq2mt7repanmw23ujx","moment_uri":"arv:video_reLhJnQd-Lo#t=1062882,1074878","asset_id":"video_reLhJnQd-Lo","start_ms":1062882,"end_ms":1074878},{"arv":"1.0","id":"mom_zkj4ykggjsy4tnqj6ztxny6yry","moment_uri":"arv:video_reLhJnQd-Lo#t=1074878,1087208","asset_id":"video_reLhJnQd-Lo","start_ms":1074878,"end_ms":1087208},{"arv":"1.0","id":"mom_bw7ndgmeiuhh3lyss7nahbee5y","moment_uri":"arv:video_reLhJnQd-Lo#t=1087208,1102208","asset_id":"video_reLhJnQd-Lo","start_ms":1087208,"end_ms":1102208},{"arv":"1.0","id":"mom_ghbd53pjafjvfcefpyasghr2xb","moment_uri":"arv:video_reLhJnQd-Lo#t=1102208,1114141","asset_id":"video_reLhJnQd-Lo","start_ms":1102208,"end_ms":1114141},{"arv":"1.0","id":"mom_3jvhyovx67zc5jg6ychxlcn2wy","moment_uri":"arv:video_reLhJnQd-Lo#t=1114141,1126917","asset_id":"video_reLhJnQd-Lo","start_ms":1114141,"end_ms":1126917},{"arv":"1.0","id":"mom_ezg5afemfwvnrzmzxhnknfjs7y","moment_uri":"arv:video_reLhJnQd-Lo#t=1126917,1139442","asset_id":"video_reLhJnQd-Lo","start_ms":1126917,"end_ms":1139442},{"arv":"1.0","id":"mom_k6vftuktf5s6cq4fjis47ar52c","moment_uri":"arv:video_reLhJnQd-Lo#t=1139442,1153587","asset_id":"video_reLhJnQd-Lo","start_ms":1139442,"end_ms":1153587},{"arv":"1.0","id":"mom_j5o4ictmq2stxfndd7gouyxshl","moment_uri":"arv:video_reLhJnQd-Lo#t=1153587,1164305","asset_id":"video_reLhJnQd-Lo","start_ms":1153587,"end_ms":1164305},{"arv":"1.0","id":"mom_kngcggt6dkqsjpbiu3l4bdccvt","moment_uri":"arv:video_reLhJnQd-Lo#t=1164305,1175617","asset_id":"video_reLhJnQd-Lo","start_ms":1164305,"end_ms":1175617},{"arv":"1.0","id":"mom_vkspblpzhi7akjyegbecopkoa5","moment_uri":"arv:video_reLhJnQd-Lo#t=1175617,1190242","asset_id":"video_reLhJnQd-Lo","start_ms":1175617,"end_ms":1190242},{"arv":"1.0","id":"mom_ruyhdbrxo6namybjpl4uwd7oas","moment_uri":"arv:video_reLhJnQd-Lo#t=1190242,1202072","asset_id":"video_reLhJnQd-Lo","start_ms":1190242,"end_ms":1202072},{"arv":"1.0","id":"mom_ekvrrwywcsml2hsbf7eentfpae","moment_uri":"arv:video_reLhJnQd-Lo#t=1202072,1217000","asset_id":"video_reLhJnQd-Lo","start_ms":1202072,"end_ms":1217000},{"arv":"1.0","id":"mom_nupdi4w5wcz5oxsjdwsjplmtos","moment_uri":"arv:video_reLhJnQd-Lo#t=1217000,1226625","asset_id":"video_reLhJnQd-Lo","start_ms":1217000,"end_ms":1226625},{"arv":"1.0","id":"mom_rcb2ldnhub6seyahwnr5wl4tbh","moment_uri":"arv:video_reLhJnQd-Lo#t=1226625,1239250","asset_id":"video_reLhJnQd-Lo","start_ms":1226625,"end_ms":1239250},{"arv":"1.0","id":"mom_ivmwk7xsjv5ol7qcjhq66ptyfp","moment_uri":"arv:video_reLhJnQd-Lo#t=1239250,1251417","asset_id":"video_reLhJnQd-Lo","start_ms":1239250,"end_ms":1251417},{"arv":"1.0","id":"mom_wvacjy333ayakjdgopdzjk3jzk","moment_uri":"arv:video_reLhJnQd-Lo#t=1251417,1263542","asset_id":"video_reLhJnQd-Lo","start_ms":1251417,"end_ms":1263542},{"arv":"1.0","id":"mom_zgxqpw3frhwfe4urbol7tzlrrd","moment_uri":"arv:video_reLhJnQd-Lo#t=1263542,1275417","asset_id":"video_reLhJnQd-Lo","start_ms":1263542,"end_ms":1275417},{"arv":"1.0","id":"mom_heotfcp7gj6g3pgh4lvxzsivx3","moment_uri":"arv:video_reLhJnQd-Lo#t=1275417,1289208","asset_id":"video_reLhJnQd-Lo","start_ms":1275417,"end_ms":1289208},{"arv":"1.0","id":"mom_myt6z67sidaju475x2ezhh7ydp","moment_uri":"arv:video_reLhJnQd-Lo#t=1289208,1303167","asset_id":"video_reLhJnQd-Lo","start_ms":1289208,"end_ms":1303167},{"arv":"1.0","id":"mom_l3c52bubdhutoq6pf2rcg6af25","moment_uri":"arv:video_reLhJnQd-Lo#t=1303167,1312875","asset_id":"video_reLhJnQd-Lo","start_ms":1303167,"end_ms":1312875},{"arv":"1.0","id":"mom_edphqosvqoseq7p26grpyjucp2","moment_uri":"arv:video_reLhJnQd-Lo#t=1312875,1326350","asset_id":"video_reLhJnQd-Lo","start_ms":1312875,"end_ms":1326350},{"arv":"1.0","id":"mom_tau6exao2um3zbqdacgjdoc6ps","moment_uri":"arv:video_reLhJnQd-Lo#t=1326350,1341350","asset_id":"video_reLhJnQd-Lo","start_ms":1326350,"end_ms":1341350},{"arv":"1.0","id":"mom_3ndm5na5ranp4wjdwquqgfmcmo","moment_uri":"arv:video_reLhJnQd-Lo#t=1341350,1354788","asset_id":"video_reLhJnQd-Lo","start_ms":1341350,"end_ms":1354788},{"arv":"1.0","id":"mom_niufej66wpdpclyu5y4ev4w4uk","moment_uri":"arv:video_reLhJnQd-Lo#t=1354788,1367458","asset_id":"video_reLhJnQd-Lo","start_ms":1354788,"end_ms":1367458},{"arv":"1.0","id":"mom_f3vh7w643cf6vlobrrmimpmwom","moment_uri":"arv:video_reLhJnQd-Lo#t=1367458,1377083","asset_id":"video_reLhJnQd-Lo","start_ms":1367458,"end_ms":1377083},{"arv":"1.0","id":"mom_segkfnglobxdkmwgzz57ssbad3","moment_uri":"arv:video_reLhJnQd-Lo#t=1377083,1391042","asset_id":"video_reLhJnQd-Lo","start_ms":1377083,"end_ms":1391042},{"arv":"1.0","id":"mom_fs2qe2cwaajbeirqf3waasjh35","moment_uri":"arv:video_reLhJnQd-Lo#t=1391042,1399057","asset_id":"video_reLhJnQd-Lo","start_ms":1391042,"end_ms":1399057},{"arv":"1.0","id":"mom_ccpsv4jehmrf4rivn3knka5ufp","moment_uri":"arv:video_reLhJnQd-Lo#t=1399057,1407057","asset_id":"video_reLhJnQd-Lo","start_ms":1399057,"end_ms":1407057},{"arv":"1.0","id":"mom_holiugu3yttwav32q2pgqghpxg","moment_uri":"arv:video_reLhJnQd-Lo#t=1407057,1418005","asset_id":"video_reLhJnQd-Lo","start_ms":1407057,"end_ms":1418005},{"arv":"1.0","id":"mom_cojwg3xrvlpi3vfpv364g4s5gv","moment_uri":"arv:video_reLhJnQd-Lo#t=1418005,1429125","asset_id":"video_reLhJnQd-Lo","start_ms":1418005,"end_ms":1429125},{"arv":"1.0","id":"mom_4cwqu7mmm5dgno33g5s7pve5k3","moment_uri":"arv:video_reLhJnQd-Lo#t=1429125,1444125","asset_id":"video_reLhJnQd-Lo","start_ms":1429125,"end_ms":1444125},{"arv":"1.0","id":"mom_mscqkzynorv7dpl2hhwhyji7na","moment_uri":"arv:video_reLhJnQd-Lo#t=1444125,1459125","asset_id":"video_reLhJnQd-Lo","start_ms":1444125,"end_ms":1459125},{"arv":"1.0","id":"mom_m3sodmvntifvik5g3ecvytr3nn","moment_uri":"arv:video_reLhJnQd-Lo#t=1459125,1474125","asset_id":"video_reLhJnQd-Lo","start_ms":1459125,"end_ms":1474125},{"arv":"1.0","id":"mom_ktewhbzriwya3qdd2zogevsrfw","moment_uri":"arv:video_reLhJnQd-Lo#t=1474125,1485042","asset_id":"video_reLhJnQd-Lo","start_ms":1474125,"end_ms":1485042},{"arv":"1.0","id":"mom_ne3ohoevcgbsesqzzeabnr3pvi","moment_uri":"arv:video_reLhJnQd-Lo#t=1485042,1494583","asset_id":"video_reLhJnQd-Lo","start_ms":1485042,"end_ms":1494583},{"arv":"1.0","id":"mom_kywpkpdqt3fc3qxg4xbevjg33j","moment_uri":"arv:video_reLhJnQd-Lo#t=1494583,1507500","asset_id":"video_reLhJnQd-Lo","start_ms":1494583,"end_ms":1507500},{"arv":"1.0","id":"mom_cdmiqa7445dsalloaj77nuxnia","moment_uri":"arv:video_reLhJnQd-Lo#t=1507500,1517340","asset_id":"video_reLhJnQd-Lo","start_ms":1507500,"end_ms":1517340},{"arv":"1.0","id":"mom_zvxjlz5shtnj4lqlkflvhg7muc","moment_uri":"arv:video_reLhJnQd-Lo#t=1517340,1525340","asset_id":"video_reLhJnQd-Lo","start_ms":1517340,"end_ms":1525340},{"arv":"1.0","id":"mom_iinfolnmcgpod6qujnwq2gcnmb","moment_uri":"arv:video_reLhJnQd-Lo#t=1525340,1538773","asset_id":"video_reLhJnQd-Lo","start_ms":1525340,"end_ms":1538773},{"arv":"1.0","id":"mom_oencbayu2ztwyfryb6dygqzv3w","moment_uri":"arv:video_reLhJnQd-Lo#t=1538773,1552500","asset_id":"video_reLhJnQd-Lo","start_ms":1538773,"end_ms":1552500},{"arv":"1.0","id":"mom_3bkavi7pcwit6ws3i52vntieaw","moment_uri":"arv:video_reLhJnQd-Lo#t=1552500,1565583","asset_id":"video_reLhJnQd-Lo","start_ms":1552500,"end_ms":1565583},{"arv":"1.0","id":"mom_2ql35cv76f3sfiiakclalaj22d","moment_uri":"arv:video_reLhJnQd-Lo#t=1565583,1577985","asset_id":"video_reLhJnQd-Lo","start_ms":1565583,"end_ms":1577985},{"arv":"1.0","id":"mom_a2d3rek4lk6tygy2d4rhrs74kd","moment_uri":"arv:video_reLhJnQd-Lo#t=1577985,1592985","asset_id":"video_reLhJnQd-Lo","start_ms":1577985,"end_ms":1592985},{"arv":"1.0","id":"mom_pd766ccmt4lwwkki7s36rxuhy6","moment_uri":"arv:video_reLhJnQd-Lo#t=1592985,1607985","asset_id":"video_reLhJnQd-Lo","start_ms":1592985,"end_ms":1607985},{"arv":"1.0","id":"mom_vtb7r2srkkfdpynibrpmqqzbvc","moment_uri":"arv:video_reLhJnQd-Lo#t=1607985,1618271","asset_id":"video_reLhJnQd-Lo","start_ms":1607985,"end_ms":1618271},{"arv":"1.0","id":"mom_yyldgh5wxlcvey3xyvlbtwcqay","moment_uri":"arv:video_reLhJnQd-Lo#t=1618271,1631750","asset_id":"video_reLhJnQd-Lo","start_ms":1618271,"end_ms":1631750},{"arv":"1.0","id":"mom_y3n54uu6plw5zuxejeu5hexhcz","moment_uri":"arv:video_reLhJnQd-Lo#t=1631750,1640729","asset_id":"video_reLhJnQd-Lo","start_ms":1631750,"end_ms":1640729},{"arv":"1.0","id":"mom_knhtzzctbezspg6jhffmfy3o5q","moment_uri":"arv:video_reLhJnQd-Lo#t=1640729,1648729","asset_id":"video_reLhJnQd-Lo","start_ms":1640729,"end_ms":1648729},{"arv":"1.0","id":"mom_qvyy6af3vayym2cksgpnjgelhn","moment_uri":"arv:video_reLhJnQd-Lo#t=1648729,1658942","asset_id":"video_reLhJnQd-Lo","start_ms":1648729,"end_ms":1658942},{"arv":"1.0","id":"mom_ijfn6hhjy46vz5rvegol3ocwnn","moment_uri":"arv:video_reLhJnQd-Lo#t=1658942,1673930","asset_id":"video_reLhJnQd-Lo","start_ms":1658942,"end_ms":1673930},{"arv":"1.0","id":"mom_k5kncnj5fp7uqp4t4pdq2q3luu","moment_uri":"arv:video_reLhJnQd-Lo#t=1673930,1683000","asset_id":"video_reLhJnQd-Lo","start_ms":1673930,"end_ms":1683000},{"arv":"1.0","id":"mom_kgeutphds6moukzeqwvag3bkwo","moment_uri":"arv:video_reLhJnQd-Lo#t=1683000,1698000","asset_id":"video_reLhJnQd-Lo","start_ms":1683000,"end_ms":1698000},{"arv":"1.0","id":"mom_iavevw2cvm2jx37kwzt2xjy6qf","moment_uri":"arv:video_reLhJnQd-Lo#t=1698000,1709466","asset_id":"video_reLhJnQd-Lo","start_ms":1698000,"end_ms":1709466},{"arv":"1.0","id":"mom_ufvornb43bmtkpu4y3cwu62xec","moment_uri":"arv:video_reLhJnQd-Lo#t=1709466,1717523","asset_id":"video_reLhJnQd-Lo","start_ms":1709466,"end_ms":1717523},{"arv":"1.0","id":"mom_mgn7fogzedbnseckuud5ngw2mm","moment_uri":"arv:video_reLhJnQd-Lo#t=1717523,1725523","asset_id":"video_reLhJnQd-Lo","start_ms":1717523,"end_ms":1725523},{"arv":"1.0","id":"mom_zuw3kopgkyxlq3zrhqsqqah4ag","moment_uri":"arv:video_reLhJnQd-Lo#t=1725523,1740523","asset_id":"video_reLhJnQd-Lo","start_ms":1725523,"end_ms":1740523},{"arv":"1.0","id":"mom_yq7j55incauu2fhgmntd4n4iyc","moment_uri":"arv:video_reLhJnQd-Lo#t=1740523,1751863","asset_id":"video_reLhJnQd-Lo","start_ms":1740523,"end_ms":1751863},{"arv":"1.0","id":"mom_vs2j3ukbrc2z7jnnkso3dnyvym","moment_uri":"arv:video_reLhJnQd-Lo#t=1751863,1763158","asset_id":"video_reLhJnQd-Lo","start_ms":1751863,"end_ms":1763158},{"arv":"1.0","id":"mom_iou3oi5e6ozxrsindlvnze2nln","moment_uri":"arv:video_reLhJnQd-Lo#t=1763158,1777741","asset_id":"video_reLhJnQd-Lo","start_ms":1763158,"end_ms":1777741},{"arv":"1.0","id":"mom_lls6yr5xk76n5qjpwjfdyqzpeo","moment_uri":"arv:video_reLhJnQd-Lo#t=1777741,1791250","asset_id":"video_reLhJnQd-Lo","start_ms":1777741,"end_ms":1791250},{"arv":"1.0","id":"mom_24bni6byipadhuddkzbogabf63","moment_uri":"arv:video_reLhJnQd-Lo#t=1791250,1804583","asset_id":"video_reLhJnQd-Lo","start_ms":1791250,"end_ms":1804583},{"arv":"1.0","id":"mom_w4qhgmqxtzcfxsudiuvaj4rlkc","moment_uri":"arv:video_reLhJnQd-Lo#t=1804583,1816325","asset_id":"video_reLhJnQd-Lo","start_ms":1804583,"end_ms":1816325},{"arv":"1.0","id":"mom_3x5qo7k6rmkaburcwfpehzp2yn","moment_uri":"arv:video_reLhJnQd-Lo#t=1816325,1831325","asset_id":"video_reLhJnQd-Lo","start_ms":1816325,"end_ms":1831325},{"arv":"1.0","id":"mom_ynlk37j5kpuo4gl6qgwe2mnpd7","moment_uri":"arv:video_reLhJnQd-Lo#t=1831325,1843223","asset_id":"video_reLhJnQd-Lo","start_ms":1831325,"end_ms":1843223},{"arv":"1.0","id":"mom_pfbzi4nd4eif3jgsut77hlpoq2","moment_uri":"arv:video_reLhJnQd-Lo#t=1843223,1851223","asset_id":"video_reLhJnQd-Lo","start_ms":1843223,"end_ms":1851223},{"arv":"1.0","id":"mom_gb5degcipnh4bnbm2tiie7gzf3","moment_uri":"arv:video_reLhJnQd-Lo#t=1851223,1861559","asset_id":"video_reLhJnQd-Lo","start_ms":1851223,"end_ms":1861559},{"arv":"1.0","id":"mom_27karvcrgdwru4vnxtg32sjbk4","moment_uri":"arv:video_reLhJnQd-Lo#t=1861559,1869875","asset_id":"video_reLhJnQd-Lo","start_ms":1861559,"end_ms":1869875},{"arv":"1.0","id":"mom_l5ego4lxps3snol7jkll7nzzse","moment_uri":"arv:video_reLhJnQd-Lo#t=1869875,1883215","asset_id":"video_reLhJnQd-Lo","start_ms":1869875,"end_ms":1883215},{"arv":"1.0","id":"mom_kz7f4xwvx6pj72xj3i7swqeiae","moment_uri":"arv:video_reLhJnQd-Lo#t=1883215,1893358","asset_id":"video_reLhJnQd-Lo","start_ms":1883215,"end_ms":1893358},{"arv":"1.0","id":"mom_bt3jfz3gidybrelpwkitcdfmwl","moment_uri":"arv:video_reLhJnQd-Lo#t=1893358,1907623","asset_id":"video_reLhJnQd-Lo","start_ms":1893358,"end_ms":1907623},{"arv":"1.0","id":"mom_jjx3tenemvxli2uw7ct7fdk5hf","moment_uri":"arv:video_reLhJnQd-Lo#t=1907623,1921998","asset_id":"video_reLhJnQd-Lo","start_ms":1907623,"end_ms":1921998},{"arv":"1.0","id":"mom_ls5zu75gdacy4pfd6o4wmop3ng","moment_uri":"arv:video_reLhJnQd-Lo#t=1921998,1936998","asset_id":"video_reLhJnQd-Lo","start_ms":1921998,"end_ms":1936998},{"arv":"1.0","id":"mom_keotho47kaotjte54xcmnisekq","moment_uri":"arv:video_reLhJnQd-Lo#t=1936998,1948917","asset_id":"video_reLhJnQd-Lo","start_ms":1936998,"end_ms":1948917},{"arv":"1.0","id":"mom_pcqn47nzlbahvcnryfhr5mrk6g","moment_uri":"arv:video_reLhJnQd-Lo#t=1948917,1962983","asset_id":"video_reLhJnQd-Lo","start_ms":1948917,"end_ms":1962983},{"arv":"1.0","id":"mom_2khvrpd4t5sgi6cutxg6x6qzls","moment_uri":"arv:video_reLhJnQd-Lo#t=1962983,1973297","asset_id":"video_reLhJnQd-Lo","start_ms":1962983,"end_ms":1973297},{"arv":"1.0","id":"mom_e2szizwjfvtnnbudhugqcyhxpw","moment_uri":"arv:video_reLhJnQd-Lo#t=1973297,1981297","asset_id":"video_reLhJnQd-Lo","start_ms":1973297,"end_ms":1981297}],"entities":[{"id":"ent_8a8a91e44930d87b222367e337327e34","canonical_source":"wikidata","canonical_id":"Q115564437","wikidata_qid":"Q115564437","display_name":"ChatGPT","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_c39bc3da8cc8f60b104c2c2cd5f4fab5","canonical_source":"wikidata","canonical_id":"Q116793893","wikidata_qid":"Q116793893","display_name":"Microsoft Copilot","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","canonical_source":"wikidata","canonical_id":"Q21708200","wikidata_qid":"Q21708200","display_name":"OpenAI","entity_type":"Organization","entity_subtype":"enterprise","coordinates":null},{"id":"ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","canonical_source":"wikidata","canonical_id":"Q95","wikidata_qid":"Q95","display_name":"Google","entity_type":"Organization","entity_subtype":"business","coordinates":null},{"id":"ent_d3aef2ec47f98439c7a3995a0edd260e","canonical_source":"wikidata","canonical_id":"Q113660857","wikidata_qid":"Q113660857","display_name":"Stable Diffusion","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_b9376b28cef422f3d1643e31d16e32d0","canonical_source":"wikidata","canonical_id":"Q95726734","wikidata_qid":"Q95726734","display_name":"GPT-3","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_a9176dd183166c10af1ebcfc68c03585","canonical_source":"wikidata","canonical_id":"Q182496","wikidata_qid":"Q182496","display_name":"Microsoft Bing","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_9747295228fc6ecd042f9c2f02b5ced5","canonical_source":"wikidata","canonical_id":"Q116685562","wikidata_qid":"Q116685562","display_name":"DALL·E 2","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_51a6fedef9218653251027caaa0269b9","canonical_source":"model","canonical_id":"ent_51a6fedef9218653251027caaa0269b9","display_name":"DeepMind","entity_type":"Organization","entity_subtype":null,"coordinates":null},{"id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","canonical_source":"wikidata","canonical_id":"Q7407093","wikidata_qid":"Q7407093","display_name":"Sam Altman","entity_type":"Person","entity_subtype":null,"coordinates":null},{"id":"ent_0ab7681fca07ba0e99c88574f46adb3b","canonical_source":"wikidata","canonical_id":"Q117322900","wikidata_qid":"Q117322900","display_name":"Synthesia","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_a44014783969e92f469667f7d8ef71e5","canonical_source":"model","canonical_id":"ent_a44014783969e92f469667f7d8ef71e5","display_name":"Krikey AI","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_7132bbc7088479a2fa4d37bae497e46c","canonical_source":"model","canonical_id":"ent_7132bbc7088479a2fa4d37bae497e46c","display_name":"UberDuck","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_18192424-9f95-42b3-9883-fca02d34ccab","canonical_source":"wikidata","canonical_id":"Q60827595","wikidata_qid":"Q60827595","display_name":"AlphaFold 3","entity_type":"Product","entity_subtype":"AI Model","coordinates":null},{"id":"ent_64b91d978a6442b4b8abcbb017f0d16b","canonical_source":"model","canonical_id":"ent_64b91d978a6442b4b8abcbb017f0d16b","display_name":"Bard","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_859fa9ce7896b57cc338859d1c86d831","canonical_source":"model","canonical_id":"ent_859fa9ce7896b57cc338859d1c86d831","display_name":"D-ID","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_a43ff92d-60d5-4549-9fff-300b4ee09823","canonical_source":"wikidata","canonical_id":"Q8298","wikidata_qid":"Q8298","display_name":"David Guetta","entity_type":"Person","entity_subtype":null,"coordinates":null},{"id":"ent_1e21f75d-cb8e-4cd4-9c15-c19108075d5c","canonical_source":"wikidata","canonical_id":"Q317521","wikidata_qid":"Q317521","display_name":"Elon Musk","entity_type":"Person","entity_subtype":null,"coordinates":null},{"id":"ent_45935407-f245-4ec3-86e8-0798c224d48c","canonical_source":"wikidata","canonical_id":"Q380","wikidata_qid":"Q380","display_name":"Meta Platforms","entity_type":"Organization","entity_subtype":"business","coordinates":null},{"id":"ent_9016355d-416c-4e35-b834-35f656db4dad","canonical_source":"wikidata","canonical_id":"Q2283","wikidata_qid":"Q2283","display_name":"Microsoft","entity_type":"Organization","entity_subtype":"technology company","coordinates":null},{"id":"ent_e8a71dcb-19e3-41f8-bc9d-9990b2db5379","canonical_source":"wikidata","canonical_id":"Q30595490","wikidata_qid":"Q30595490","display_name":"Mint Mobile","entity_type":"Brand","entity_subtype":"telecommunications company","coordinates":null},{"id":"ent_3e757939-5520-4ae9-967d-eea25453d4db","canonical_source":"wikidata","canonical_id":"Q192682","wikidata_qid":"Q192682","display_name":"Ryan Reynolds","entity_type":"Person","entity_subtype":null,"coordinates":null},{"id":"ent_854fd7a20654b1a94e7384a2b3d9b177","canonical_source":"wikidata","canonical_id":"Q4880667","wikidata_qid":"Q4880667","display_name":"Behance","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_c4d26c337fbbd081b49e1f4973346b86","canonical_source":"wikidata","canonical_id":"Q23498528","wikidata_qid":"Q23498528","display_name":"Canva","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_03017e70-f2a0-4aef-b135-4521169d0c1e","canonical_source":"wikidata","canonical_id":"Q1520122","wikidata_qid":"Q1520122","display_name":"Getty Images","entity_type":"Organization","entity_subtype":null,"coordinates":null},{"id":"ent_2b2c6c81-4a5c-4fd1-8123-1f98a924dc1e","canonical_source":"wikidata","canonical_id":"Q24901201","wikidata_qid":"Q24901201","display_name":"ImageNet","entity_type":"Other","entity_subtype":"dataset","coordinates":null},{"id":"ent_b2a51b709e3456866a6d6431306b4a1a","canonical_source":"model","canonical_id":"ent_b2a51b709e3456866a6d6431306b4a1a","display_name":"Jasper","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_c04aabf51b5214dd1d4982a47b375c79","canonical_source":"wikidata","canonical_id":"Q725967","wikidata_qid":"Q725967","display_name":"Microsoft Azure","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_394d031670f5a6e9ee9e5545bad4e6af","canonical_source":"model","canonical_id":"ent_394d031670f5a6e9ee9e5545bad4e6af","display_name":"podcast.ai","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_7e293f38-b099-4437-ad3e-3688f679ad40","canonical_source":"wikidata","canonical_id":"Q60768699","wikidata_qid":"Q60768699","display_name":"Replit","entity_type":"Organization","entity_subtype":"business","coordinates":null},{"id":"ent_5c22b4ce12acf101f0f5e2aac3d1bd72","canonical_source":"model","canonical_id":"ent_5c22b4ce12acf101f0f5e2aac3d1bd72","display_name":"Replit Ghostwriter","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_d824625e4bb5e24cff90069cb8ed5cd3","canonical_source":"wikidata","canonical_id":"Q118904793","wikidata_qid":"Q118904793","display_name":"Respeecher","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_a4de1737-84fd-44fc-a3ca-df592ec8437f","canonical_source":"wikidata","canonical_id":"Q692989","wikidata_qid":"Q692989","display_name":"Roblox","entity_type":"Brand","entity_subtype":"video_game","coordinates":null},{"id":"ent_62f6aafd-a3e6-40e5-8999-0dbfeebad568","canonical_source":"wikidata","canonical_id":"Q3482689","wikidata_qid":"Q3482689","display_name":"Shutterstock","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_d139e370-38e7-45c6-b937-d8cd1c341325","canonical_source":"wikidata","canonical_id":"Q19837","wikidata_qid":"Q19837","display_name":"Steve Jobs","entity_type":"Person","entity_subtype":null,"coordinates":null},{"id":"ent_c1eb0a5ad5e25af6e52c706b836ed2e6","canonical_source":"wikidata","canonical_id":"Q918","wikidata_qid":"Q918","display_name":"X","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_82704b013b687db9f76c4673d4ca71f5","canonical_source":"wikidata","canonical_id":"Q107711739","wikidata_qid":"Q107711739","display_name":"AlphaFold2","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_6369fe8d82503159c0a300e4c44d074c","canonical_source":"model","canonical_id":"ent_6369fe8d82503159c0a300e4c44d074c","display_name":"AudioLM","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_2217b5322de54820ef4590d798a43b9d","canonical_source":"model","canonical_id":"ent_2217b5322de54820ef4590d798a43b9d","display_name":"Claude","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_02988cd6887c1296841b9a5533d9f22f","canonical_source":"wikidata","canonical_id":"Q105078662","wikidata_qid":"Q105078662","display_name":"DALL-E","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_cb189d1f64954ed681e62e28eb218cc3","canonical_source":"model","canonical_id":"ent_cb189d1f64954ed681e62e28eb218cc3","display_name":"Descript","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_f8eaf0d5-52e9-4950-91b9-1d7f650e637d","canonical_source":"wikidata","canonical_id":"Q7605173","wikidata_qid":"Q7605173","display_name":"Dribbble","entity_type":"Product","entity_subtype":"company","coordinates":null},{"id":"ent_faa3adaed042548d4fb15614f85f2465","canonical_source":"model","canonical_id":"ent_faa3adaed042548d4fb15614f85f2465","display_name":"Galileo AI","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_53bd528d34cff3a7fa4ef3aeb915d0db","canonical_source":"model","canonical_id":"ent_53bd528d34cff3a7fa4ef3aeb915d0db","display_name":"Gen-1","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_dacc6317-9bd3-48e7-b952-123bd47e1689","canonical_source":"wikidata","canonical_id":"Q107435063","wikidata_qid":"Q107435063","display_name":"GitHub Copilot","entity_type":"Product","entity_subtype":"Software","coordinates":null},{"id":"ent_ae818f1d-637b-4bed-b9a0-905e3159abaf","canonical_source":"wikidata","canonical_id":"Q116777014","wikidata_qid":"Q116777014","display_name":"GPT","entity_type":"Product","entity_subtype":"AI Model","coordinates":null},{"id":"ent_e30d6dc0ccb5641d50ac160c2eeb2747","canonical_source":"model","canonical_id":"ent_e30d6dc0ccb5641d50ac160c2eeb2747","display_name":"Greylock","entity_type":"Organization","entity_subtype":null,"coordinates":null},{"id":"ent_14d51a79e1a3ca6444ce0459bc2d5750","canonical_source":"wikidata","canonical_id":"Q117449859","wikidata_qid":"Q117449859","display_name":"InstructGPT","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_c40b20c6-f989-41d3-8160-0c05014792db","canonical_source":"wikidata","canonical_id":"Q2718421","wikidata_qid":"Q2718421","display_name":"Joe Rogan","entity_type":"Person","entity_subtype":null,"coordinates":null},{"id":"ent_99dce50a-cd99-4fa5-a060-e0fd1ddd2aa2","canonical_source":"wikidata","canonical_id":"Q108582200","wikidata_qid":"Q108582200","display_name":"OpenAI Codex","entity_type":"Product","entity_subtype":"ai_model","coordinates":null},{"id":"ent_8431466630a54b164b46c028fe9e9296","canonical_source":"model","canonical_id":"ent_8431466630a54b164b46c028fe9e9296","display_name":"play.ht","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_790bfc7aeb17ba41b58350a304518754","canonical_source":"model","canonical_id":"ent_790bfc7aeb17ba41b58350a304518754","display_name":"Polly","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_267715952b510666654acc01256b0509","canonical_source":"model","canonical_id":"ent_267715952b510666654acc01256b0509","display_name":"Runway ML","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_04406a58ab831ae699e21d08b8c98ba7","canonical_source":"model","canonical_id":"ent_04406a58ab831ae699e21d08b8c98ba7","display_name":"SingSong","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_409cfb5cc3b67ef22666e60287e857e4","canonical_source":"wikidata","canonical_id":"Q85847298","wikidata_qid":"Q85847298","display_name":"Speechify","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_65e6ca38244d1667696c8f109cd37b85","canonical_source":"model","canonical_id":"ent_65e6ca38244d1667696c8f109cd37b85","display_name":"Whisper","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_627b0a177ec5ebe37949f93c7adb536b","canonical_source":"wikidata","canonical_id":"Q65045028","wikidata_qid":"Q65045028","display_name":"Amazon SageMaker","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_45175450-c86e-4de0-8553-90b22143a40c","canonical_source":"wikidata","canonical_id":"Q14772","wikidata_qid":"Q14772","display_name":"Baidu","entity_type":"Organization","entity_subtype":"business","coordinates":null},{"id":"ent_bfc65d0b-b508-403a-890b-96522652b6a5","canonical_source":"wikidata","canonical_id":"Q12055316","wikidata_qid":"Q12055316","display_name":"Common Crawl","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_6f26b67ee4aea0b077fffc62ac777efe","canonical_source":"wikidata","canonical_id":"Q7345","wikidata_qid":"Q7345","display_name":"Dan Brown","entity_type":"Person","entity_subtype":null,"coordinates":null},{"id":"ent_ae8f3da2281f0437bdcafc8a0058ff84","canonical_source":"wikidata","canonical_id":"Q491748","wikidata_qid":"Q491748","display_name":"Deloitte","entity_type":"Organization","entity_subtype":null,"coordinates":null},{"id":"ent_2194060e17a174e4c27c09f177351d70","canonical_source":"model","canonical_id":"ent_2194060e17a174e4c27c09f177351d70","display_name":"Diffusion Bee","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_01797a1d4cf0c5a11814ae38ec92a118","canonical_source":"model","canonical_id":"ent_01797a1d4cf0c5a11814ae38ec92a118","display_name":"Draw Things","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_84dc3c1286160f2a4032f88b3e8b9cfb","canonical_source":"model","canonical_id":"ent_84dc3c1286160f2a4032f88b3e8b9cfb","display_name":"Earnie Bot","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_7f6f5dddbaf6008acc31b8dde4d09254","canonical_source":"model","canonical_id":"ent_7f6f5dddbaf6008acc31b8dde4d09254","display_name":"ESMFold","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_57e36c5da0caa91427a797c7e5b6f432","canonical_source":"wikidata","canonical_id":"Q355","wikidata_qid":"Q355","display_name":"Facebook","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_5e44ccec-4517-469d-9d23-e79fc5531291","canonical_source":"wikidata","canonical_id":"Q76615","wikidata_qid":"Q76615","display_name":"Fortune 500","entity_type":"Other","entity_subtype":null,"coordinates":null},{"id":"ent_b002f887aeee4a196bec7b6031af217b","canonical_source":"wikidata","canonical_id":"Q364","wikidata_qid":"Q364","display_name":"GitHub","entity_type":"Brand","entity_subtype":null,"coordinates":null},{"id":"ent_f000e5694735cec3c528e0ca2c122006","canonical_source":"wikidata","canonical_id":"Q847115","wikidata_qid":"Q847115","display_name":"Google Trends","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_59bbe9c1bd6ffeccf0ef495b1a77ab7d","canonical_source":"wikidata","canonical_id":"Q30315025","wikidata_qid":"Q30315025","display_name":"Google Wavenet","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_8d208bbf2208872e313e29db2b70dc38","canonical_source":"wikidata","canonical_id":"Q112889233","wikidata_qid":"Q112889233","display_name":"Imagen","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_665536957da8315a5a9d4af735a1f40c","canonical_source":"model","canonical_id":"ent_665536957da8315a5a9d4af735a1f40c","display_name":"LibreFold","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_f122deea66464e885edb68ce1701437b","canonical_source":"model","canonical_id":"ent_f122deea66464e885edb68ce1701437b","display_name":"lifearchitect.ai","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_da6652be00eed8664a95f3abaf6b3634","canonical_source":"model","canonical_id":"ent_da6652be00eed8664a95f3abaf6b3634","display_name":"Locomotive Act","entity_type":"Other","entity_subtype":null,"coordinates":null},{"id":"ent_76bf4e6f5fcfac45836f1fe139faeb59","canonical_source":"model","canonical_id":"ent_76bf4e6f5fcfac45836f1fe139faeb59","display_name":"Make-A-Video","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_6f1da63ab3dee19d25200449733024a8","canonical_source":"wikidata","canonical_id":"Q113070628","wikidata_qid":"Q113070628","display_name":"Midjourney","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_9bd44dfbc2074a8168b19ddf872b13a2","canonical_source":"model","canonical_id":"ent_9bd44dfbc2074a8168b19ddf872b13a2","display_name":"MuseNet","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_80d14ba6e7a1f2930be875a65528acde","canonical_source":"model","canonical_id":"ent_80d14ba6e7a1f2930be875a65528acde","display_name":"OpenBioML","entity_type":"Organization","entity_subtype":null,"coordinates":null},{"id":"ent_578fdec782a2a1d689b3dcf924b3dd66","canonical_source":"model","canonical_id":"ent_578fdec782a2a1d689b3dcf924b3dd66","display_name":"Parti","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_16f60149-d305-4ce6-9809-93eb78df8ef5","canonical_source":"wikidata","canonical_id":"Q20966579","wikidata_qid":"Q20966579","display_name":"Pokémon Go","entity_type":"Product","entity_subtype":"video_game","coordinates":null},{"id":"ent_913a6ced4ce74396c21d8978aad1c8cb","canonical_source":"model","canonical_id":"ent_913a6ced4ce74396c21d8978aad1c8cb","display_name":"Reality Studio","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_8e159ea51321487286983313973a41a0","canonical_source":"model","canonical_id":"ent_8e159ea51321487286983313973a41a0","display_name":"Rephrase.ai","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_526734b287cdf9adfa5765e9b72b8bde","canonical_source":"model","canonical_id":"ent_526734b287cdf9adfa5765e9b72b8bde","display_name":"RoseTTAFold","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_34c72865d8528167ad8904f36e71bd76","canonical_source":"wikidata","canonical_id":"Q21447895","wikidata_qid":"Q21447895","display_name":"TensorFlow","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_6ec75751cdea3144c93f63fa828b7acc","canonical_source":"wikidata","canonical_id":"Q9684","wikidata_qid":"Q9684","display_name":"The New York Times","entity_type":"Organization","entity_subtype":null,"coordinates":null},{"id":"ent_94f4a7f58a3f24593b2f3e856fdde4a8","canonical_source":"wikidata","canonical_id":"Q48938223","wikidata_qid":"Q48938223","display_name":"TikTok","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_7290ff888c048498e298764f503ec0c5","canonical_source":"wikidata","canonical_id":"Q52","wikidata_qid":"Q52","display_name":"Wikipedia","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_f6677f64-31a6-4383-b60c-91f0851302c7","canonical_source":"wikidata","canonical_id":"Q123592234","wikidata_qid":"Q123592234","display_name":"Stability AI","entity_type":"Organization","entity_subtype":null,"coordinates":null},{"id":"ent_5f383f1fec26b0df6bc815d9aee4cf47","canonical_source":"wikidata","canonical_id":"Q96186334","wikidata_qid":"Q96186334","display_name":"Figma","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_0abfc4aa-c8f7-4e9c-babf-662a802b50ef","canonical_source":"wikidata","canonical_id":"Q5608","wikidata_qid":"Q5608","display_name":"Eminem","entity_type":"Person","entity_subtype":null,"coordinates":null},{"id":"ent_91b7c497-1d72-471e-8986-108b68b1a8bf","canonical_source":"wikidata","canonical_id":"Q116758847","wikidata_qid":"Q116758847","display_name":"Anthropic","entity_type":"Organization","entity_subtype":null,"coordinates":null},{"id":"ent_d7804d83-d910-414c-b89e-1f15c7b6503d","canonical_source":"wikidata","canonical_id":"Q312","wikidata_qid":"Q312","display_name":"Apple Inc.","entity_type":"Brand","entity_subtype":"enterprise","coordinates":null},{"id":"ent_cefdeb8b-eae6-4e47-9e30-0937d7306c09","canonical_source":"wikidata","canonical_id":"Q135622","wikidata_qid":"Q135622","display_name":"Google Translate","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_b29d737e-0ac5-4e9f-9fd2-1c32f379b8cb","canonical_source":"wikidata","canonical_id":"Q209330","wikidata_qid":"Q209330","display_name":"Instagram","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_f65a2fd5-6650-4e71-afa1-6d21f7f4d875","canonical_source":"wikidata","canonical_id":"Q114586027","wikidata_qid":"Q114586027","display_name":"LAION","entity_type":"Organization","entity_subtype":null,"coordinates":null},{"id":"ent_a0462240-fcd3-4af1-9a4b-8329d6a52454","canonical_source":"wikidata","canonical_id":"Q75687","wikidata_qid":"Q75687","display_name":"Mac","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_2425febecb47ae91ba277939f20b9ef3","canonical_source":"model","canonical_id":"ent_2425febecb47ae91ba277939f20b9ef3","display_name":"MusicLM","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_95f91400-3cbe-4754-b031-62b05747dab4","canonical_source":"wikidata","canonical_id":"Q182477","wikidata_qid":"Q182477","display_name":"Nvidia","entity_type":"Organization","entity_subtype":null,"coordinates":null},{"id":"ent_53a106957af99f597540fa15fd63c628","canonical_source":"wikidata","canonical_id":"Q1136","wikidata_qid":"Q1136","display_name":"Reddit","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_742cfb2c229e3497c79a08f766e6b1f1","canonical_source":"wikidata","canonical_id":"Q207006","wikidata_qid":"Q207006","display_name":"Wolfram Alpha","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_1003554c9ffa7666633fb3829d422dcf","canonical_source":"model","canonical_id":"ent_1003554c9ffa7666633fb3829d422dcf","display_name":"A100","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_a9e5db3c-e768-4e77-ab23-850a134d7b07","canonical_source":"wikidata","canonical_id":"Q3884","wikidata_qid":"Q3884","display_name":"Amazon","entity_type":"Organization","entity_subtype":"company","coordinates":null},{"id":"ent_e3c10ccfce46fd7e0ed9304c98cd2977","canonical_source":"wikidata","canonical_id":"Q22907849","wikidata_qid":"Q22907849","display_name":"Discord","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_c8cb1aeb06fa64ed4294c1279d0c7dca","canonical_source":"wikidata","canonical_id":"Q9366","wikidata_qid":"Q9366","display_name":"Google Search","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_b99b525cd404f73f2002cdb3fc3cf572","canonical_source":"model","canonical_id":"ent_b99b525cd404f73f2002cdb3fc3cf572","display_name":"Gutenberg Press","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_aa5296ff-1580-42d0-87db-52259cc0e257","canonical_source":"wikidata","canonical_id":"Q248","wikidata_qid":"Q248","display_name":"Intel","entity_type":"Organization","entity_subtype":"business","coordinates":null},{"id":"ent_33ce2df73e75010d60cb1724225821bc","canonical_source":"wikidata","canonical_id":"Q49113","wikidata_qid":"Q49113","display_name":"Ivy League","entity_type":"Organization","entity_subtype":null,"coordinates":null},{"id":"ent_8a3691bd61a8532d3aaa3ef9d749e3a9","canonical_source":"model","canonical_id":"ent_8a3691bd61a8532d3aaa3ef9d749e3a9","display_name":"Lensa AI","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_4e8516541fed51ef3fa0189dd2eb5f03","canonical_source":"model","canonical_id":"ent_4e8516541fed51ef3fa0189dd2eb5f03","display_name":"Microsoft Edge","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_cd4bb989c028399afaf3ca7e59be6d0f","canonical_source":"wikidata","canonical_id":"Q12418","wikidata_qid":"Q12418","display_name":"Mona Lisa","entity_type":"Work","entity_subtype":null,"coordinates":null},{"id":"ent_27473110179dbbafef08367c58bf843b","canonical_source":"wikidata","canonical_id":"Q255381","wikidata_qid":"Q255381","display_name":"Pinterest","entity_type":"Product","entity_subtype":null,"coordinates":null},{"id":"ent_2b325911-9859-430a-ab1f-c2eee5f11f8e","canonical_source":"wikidata","canonical_id":"Q63435709","wikidata_qid":"Q63435709","display_name":"Substack","entity_type":"Product","entity_subtype":"company","coordinates":null},{"id":"ent_a13a7733-0be2-4e51-9318-25622e7b6fbd","canonical_source":"wikidata","canonical_id":"Q145","wikidata_qid":"Q145","display_name":"United Kingdom","entity_type":"Location","entity_subtype":"country","coordinates":null},{"id":"ent_b987acd01400347ba9932dfd412e2c65","canonical_source":"wikidata","canonical_id":"Q866","wikidata_qid":"Q866","display_name":"YouTube","entity_type":"Product","entity_subtype":null,"coordinates":null}],"entity_appearances":[{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_8a8a91e44930d87b222367e337327e34","appearance_count":29,"total_visible_duration_ms":340613,"first_appearance_ms":48167,"last_appearance_ms":1843223,"appearance_strength":0.39,"shots_appeared_in":["sht_124","sht_128","sht_129","sht_13","sht_131","sht_132","sht_133","sht_134","sht_135","sht_136","sht_138","sht_14","sht_145","sht_148","sht_17","sht_18","sht_19","sht_25","sht_36","sht_37","sht_4","sht_42","sht_44","sht_48","sht_58","sht_59","sht_87","sht_88","sht_9"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_c39bc3da8cc8f60b104c2c2cd5f4fab5","appearance_count":6,"total_visible_duration_ms":78753,"first_appearance_ms":221499,"last_appearance_ms":1552500,"appearance_strength":0.319,"shots_appeared_in":["sht_124","sht_18","sht_19","sht_42","sht_44","sht_61"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_b2297cb2-b39a-4ebd-8470-8641835ae6eb","appearance_count":31,"total_visible_duration_ms":381887,"first_appearance_ms":112503,"last_appearance_ms":1861559,"appearance_strength":0.296,"shots_appeared_in":["sht_106","sht_107","sht_128","sht_139","sht_141","sht_147","sht_148","sht_15","sht_150","sht_16","sht_17","sht_20","sht_23","sht_24","sht_42","sht_43","sht_44","sht_45","sht_46","sht_47","sht_48","sht_57","sht_58","sht_64","sht_65","sht_73","sht_74","sht_75","sht_79","sht_80","sht_9"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_5ee44b5c-468f-4de1-a66a-bd96483367a7","appearance_count":20,"total_visible_duration_ms":242589,"first_appearance_ms":530678,"last_appearance_ms":1869875,"appearance_strength":0.253,"shots_appeared_in":["sht_106","sht_113","sht_128","sht_129","sht_132","sht_134","sht_135","sht_136","sht_151","sht_43","sht_44","sht_51","sht_64","sht_65","sht_66","sht_79","sht_80","sht_81","sht_91","sht_92"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_d3aef2ec47f98439c7a3995a0edd260e","appearance_count":13,"total_visible_duration_ms":160326,"first_appearance_ms":809275,"last_appearance_ms":1921998,"appearance_strength":0.24,"shots_appeared_in":["sht_132","sht_133","sht_134","sht_136","sht_137","sht_139","sht_140","sht_146","sht_155","sht_65","sht_66","sht_67","sht_99"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_b9376b28cef422f3d1643e31d16e32d0","appearance_count":12,"total_visible_duration_ms":149243,"first_appearance_ms":161202,"last_appearance_ms":1251417,"appearance_strength":0.237,"shots_appeared_in":["sht_13","sht_14","sht_15","sht_16","sht_23","sht_24","sht_26","sht_42","sht_44","sht_47","sht_48","sht_99"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_a9176dd183166c10af1ebcfc68c03585","appearance_count":8,"total_visible_duration_ms":105787,"first_appearance_ms":655203,"last_appearance_ms":1631750,"appearance_strength":0.225,"shots_appeared_in":["sht_127","sht_128","sht_129","sht_130","sht_52","sht_61","sht_62","sht_63"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_9747295228fc6ecd042f9c2f02b5ced5","appearance_count":6,"total_visible_duration_ms":70779,"first_appearance_ms":809275,"last_appearance_ms":1843223,"appearance_strength":0.219,"shots_appeared_in":["sht_145","sht_148","sht_65","sht_73","sht_74","sht_75"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_51a6fedef9218653251027caaa0269b9","appearance_count":6,"total_visible_duration_ms":59246,"first_appearance_ms":1399057,"last_appearance_ms":1525340,"appearance_strength":0.219,"shots_appeared_in":["sht_112","sht_113","sht_119","sht_120","sht_121","sht_122"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","appearance_count":6,"total_visible_duration_ms":74078,"first_appearance_ms":102924,"last_appearance_ms":1948917,"appearance_strength":0.219,"shots_appeared_in":["sht_10","sht_11","sht_156","sht_157","sht_8","sht_9"],"filmed_at_score":null,"present_in_score":null,"relation":"speaks"},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_0ab7681fca07ba0e99c88574f46adb3b","appearance_count":6,"total_visible_duration_ms":76437,"first_appearance_ms":0,"last_appearance_ms":1175617,"appearance_strength":0.219,"shots_appeared_in":["sht_0","sht_23","sht_24","sht_90","sht_91","sht_93"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_a44014783969e92f469667f7d8ef71e5","appearance_count":5,"total_visible_duration_ms":63054,"first_appearance_ms":1126917,"last_appearance_ms":1289208,"appearance_strength":0.216,"shots_appeared_in":["sht_101","sht_102","sht_90","sht_91","sht_92"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_7132bbc7088479a2fa4d37bae497e46c","appearance_count":5,"total_visible_duration_ms":65948,"first_appearance_ms":985160,"last_appearance_ms":1126917,"appearance_strength":0.216,"shots_appeared_in":["sht_79","sht_81","sht_87","sht_88","sht_89"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_18192424-9f95-42b3-9883-fca02d34ccab","appearance_count":4,"total_visible_duration_ms":40298,"first_appearance_ms":1485042,"last_appearance_ms":1525340,"appearance_strength":0.212,"shots_appeared_in":["sht_119","sht_120","sht_121","sht_122"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_64b91d978a6442b4b8abcbb017f0d16b","appearance_count":4,"total_visible_duration_ms":52525,"first_appearance_ms":530678,"last_appearance_ms":1618271,"appearance_strength":0.212,"shots_appeared_in":["sht_128","sht_129","sht_43","sht_44"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_859fa9ce7896b57cc338859d1c86d831","appearance_count":6,"total_visible_duration_ms":74899,"first_appearance_ms":1126917,"last_appearance_ms":1263542,"appearance_strength":0.212,"shots_appeared_in":["sht_100","sht_90","sht_91","sht_93","sht_98","sht_99"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_a43ff92d-60d5-4549-9fff-300b4ee09823","appearance_count":4,"total_visible_duration_ms":52234,"first_appearance_ms":1087208,"last_appearance_ms":1139442,"appearance_strength":0.212,"shots_appeared_in":["sht_87","sht_88","sht_89","sht_90"],"filmed_at_score":null,"present_in_score":null,"relation":"appears"},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_1e21f75d-cb8e-4cd4-9c15-c19108075d5c","appearance_count":4,"total_visible_duration_ms":37992,"first_appearance_ms":1740523,"last_appearance_ms":1869875,"appearance_strength":0.212,"shots_appeared_in":["sht_141","sht_149","sht_150","sht_151"],"filmed_at_score":null,"present_in_score":null,"relation":"mentioned"},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_45935407-f245-4ec3-86e8-0798c224d48c","appearance_count":4,"total_visible_duration_ms":46931,"first_appearance_ms":1139442,"last_appearance_ms":1429125,"appearance_strength":0.212,"shots_appeared_in":["sht_113","sht_114","sht_91","sht_92"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_9016355d-416c-4e35-b834-35f656db4dad","appearance_count":5,"total_visible_duration_ms":63220,"first_appearance_ms":519994,"last_appearance_ms":1861559,"appearance_strength":0.212,"shots_appeared_in":["sht_124","sht_127","sht_150","sht_42","sht_61"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_e8a71dcb-19e3-41f8-bc9d-9990b2db5379","appearance_count":4,"total_visible_duration_ms":49250,"first_appearance_ms":713167,"last_appearance_ms":762417,"appearance_strength":0.212,"shots_appeared_in":["sht_57","sht_58","sht_59","sht_60"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_3e757939-5520-4ae9-967d-eea25453d4db","appearance_count":4,"total_visible_duration_ms":49250,"first_appearance_ms":713167,"last_appearance_ms":762417,"appearance_strength":0.212,"shots_appeared_in":["sht_57","sht_58","sht_59","sht_60"],"filmed_at_score":null,"present_in_score":null,"relation":"appears"},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_854fd7a20654b1a94e7384a2b3d9b177","appearance_count":3,"total_visible_duration_ms":39387,"first_appearance_ms":1751863,"last_appearance_ms":1791250,"appearance_strength":0.209,"shots_appeared_in":["sht_142","sht_143","sht_144"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_c4d26c337fbbd081b49e1f4973346b86","appearance_count":3,"total_visible_duration_ms":39387,"first_appearance_ms":1751863,"last_appearance_ms":1791250,"appearance_strength":0.209,"shots_appeared_in":["sht_142","sht_143","sht_144"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_03017e70-f2a0-4aef-b135-4521169d0c1e","appearance_count":4,"total_visible_duration_ms":54452,"first_appearance_ms":824198,"last_appearance_ms":1740523,"appearance_strength":0.209,"shots_appeared_in":["sht_140","sht_66","sht_67","sht_68"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_2b2c6c81-4a5c-4fd1-8123-1f98a924dc1e","appearance_count":3,"total_visible_duration_ms":43000,"first_appearance_ms":398083,"last_appearance_ms":441083,"appearance_strength":0.209,"shots_appeared_in":["sht_32","sht_33","sht_34"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_b2a51b709e3456866a6d6431306b4a1a","appearance_count":3,"total_visible_duration_ms":52626,"first_appearance_ms":686958,"last_appearance_ms":725250,"appearance_strength":0.209,"shots_appeared_in":["sht_55","sht_56","sht_57"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_c04aabf51b5214dd1d4982a47b375c79","appearance_count":3,"total_visible_duration_ms":37625,"first_appearance_ms":572917,"last_appearance_ms":610542,"appearance_strength":0.209,"shots_appeared_in":["sht_46","sht_47","sht_48"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_394d031670f5a6e9ee9e5545bad4e6af","appearance_count":3,"total_visible_duration_ms":39326,"first_appearance_ms":1047882,"last_appearance_ms":1087208,"appearance_strength":0.209,"shots_appeared_in":["sht_84","sht_85","sht_86"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_7e293f38-b099-4437-ad3e-3688f679ad40","appearance_count":5,"total_visible_duration_ms":61538,"first_appearance_ms":250129,"last_appearance_ms":1391042,"appearance_strength":0.209,"shots_appeared_in":["sht_106","sht_109","sht_110","sht_20","sht_21"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_5c22b4ce12acf101f0f5e2aac3d1bd72","appearance_count":4,"total_visible_duration_ms":51814,"first_appearance_ms":250129,"last_appearance_ms":1391042,"appearance_strength":0.209,"shots_appeared_in":["sht_106","sht_109","sht_110","sht_20"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_d824625e4bb5e24cff90069cb8ed5cd3","appearance_count":3,"total_visible_duration_ms":41693,"first_appearance_ms":287919,"last_appearance_ms":1047882,"appearance_strength":0.209,"shots_appeared_in":["sht_23","sht_82","sht_83"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_a4de1737-84fd-44fc-a3ca-df592ec8437f","appearance_count":5,"total_visible_duration_ms":64812,"first_appearance_ms":1289208,"last_appearance_ms":1367458,"appearance_strength":0.209,"shots_appeared_in":["sht_103","sht_104","sht_105","sht_106","sht_108"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_62f6aafd-a3e6-40e5-8999-0dbfeebad568","appearance_count":5,"total_visible_duration_ms":63444,"first_appearance_ms":1751863,"last_appearance_ms":1843223,"appearance_strength":0.209,"shots_appeared_in":["sht_142","sht_144","sht_146","sht_147","sht_148"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_d139e370-38e7-45c6-b937-d8cd1c341325","appearance_count":3,"total_visible_duration_ms":39326,"first_appearance_ms":1047882,"last_appearance_ms":1087208,"appearance_strength":0.209,"shots_appeared_in":["sht_84","sht_85","sht_86"],"filmed_at_score":null,"present_in_score":null,"relation":"mentioned"},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_c1eb0a5ad5e25af6e52c706b836ed2e6","appearance_count":3,"total_visible_duration_ms":34340,"first_appearance_ms":557917,"last_appearance_ms":1751863,"appearance_strength":0.209,"shots_appeared_in":["sht_139","sht_141","sht_45"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_82704b013b687db9f76c4673d4ca71f5","appearance_count":2,"total_visible_duration_ms":18948,"first_appearance_ms":1399057,"last_appearance_ms":1418005,"appearance_strength":0.206,"shots_appeared_in":["sht_112","sht_113"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_6369fe8d82503159c0a300e4c44d074c","appearance_count":2,"total_visible_duration_ms":24275,"first_appearance_ms":985160,"last_appearance_ms":1009435,"appearance_strength":0.206,"shots_appeared_in":["sht_79","sht_80"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_2217b5322de54820ef4590d798a43b9d","appearance_count":2,"total_visible_duration_ms":27239,"first_appearance_ms":530678,"last_appearance_ms":557917,"appearance_strength":0.206,"shots_appeared_in":["sht_43","sht_44"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_02988cd6887c1296841b9a5533d9f22f","appearance_count":2,"total_visible_duration_ms":19784,"first_appearance_ms":919458,"last_appearance_ms":1816325,"appearance_strength":0.206,"shots_appeared_in":["sht_146","sht_73"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_cb189d1f64954ed681e62e28eb218cc3","appearance_count":2,"total_visible_duration_ms":28913,"first_appearance_ms":287919,"last_appearance_ms":316832,"appearance_strength":0.206,"shots_appeared_in":["sht_23","sht_24"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_f8eaf0d5-52e9-4950-91b9-1d7f650e637d","appearance_count":2,"total_visible_duration_ms":25878,"first_appearance_ms":1751863,"last_appearance_ms":1777741,"appearance_strength":0.206,"shots_appeared_in":["sht_142","sht_143"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_faa3adaed042548d4fb15614f85f2465","appearance_count":2,"total_visible_duration_ms":36303,"first_appearance_ms":950083,"last_appearance_ms":973083,"appearance_strength":0.206,"shots_appeared_in":["sht_76","sht_77"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_53bd528d34cff3a7fa4ef3aeb915d0db","appearance_count":2,"total_visible_duration_ms":22250,"first_appearance_ms":1217000,"last_appearance_ms":1239250,"appearance_strength":0.206,"shots_appeared_in":["sht_97","sht_98"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_dacc6317-9bd3-48e7-b952-123bd47e1689","appearance_count":3,"total_visible_duration_ms":39015,"first_appearance_ms":1326350,"last_appearance_ms":1407057,"appearance_strength":0.206,"shots_appeared_in":["sht_106","sht_111","sht_112"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_ae818f1d-637b-4bed-b9a0-905e3159abaf","appearance_count":2,"total_visible_duration_ms":29667,"first_appearance_ms":426083,"last_appearance_ms":455750,"appearance_strength":0.206,"shots_appeared_in":["sht_34","sht_35"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_e30d6dc0ccb5641d50ac160c2eeb2747","appearance_count":2,"total_visible_duration_ms":26919,"first_appearance_ms":1921998,"last_appearance_ms":1948917,"appearance_strength":0.206,"shots_appeared_in":["sht_156","sht_157"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_14d51a79e1a3ca6444ce0459bc2d5750","appearance_count":2,"total_visible_duration_ms":25593,"first_appearance_ms":186833,"last_appearance_ms":212426,"appearance_strength":0.206,"shots_appeared_in":["sht_15","sht_16"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_c40b20c6-f989-41d3-8160-0c05014792db","appearance_count":2,"total_visible_duration_ms":26996,"first_appearance_ms":1047882,"last_appearance_ms":1074878,"appearance_strength":0.206,"shots_appeared_in":["sht_84","sht_85"],"filmed_at_score":null,"present_in_score":null,"relation":"mentioned"},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_99dce50a-cd99-4fa5-a060-e0fd1ddd2aa2","appearance_count":3,"total_visible_duration_ms":41108,"first_appearance_ms":1326350,"last_appearance_ms":1367458,"appearance_strength":0.206,"shots_appeared_in":["sht_106","sht_107","sht_108"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_8431466630a54b164b46c028fe9e9296","appearance_count":2,"total_visible_duration_ms":24326,"first_appearance_ms":1062882,"last_appearance_ms":1087208,"appearance_strength":0.206,"shots_appeared_in":["sht_85","sht_86"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_790bfc7aeb17ba41b58350a304518754","appearance_count":2,"total_visible_duration_ms":24275,"first_appearance_ms":985160,"last_appearance_ms":1009435,"appearance_strength":0.206,"shots_appeared_in":["sht_79","sht_80"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_267715952b510666654acc01256b0509","appearance_count":5,"total_visible_duration_ms":63848,"first_appearance_ms":1126917,"last_appearance_ms":1239250,"appearance_strength":0.206,"shots_appeared_in":["sht_90","sht_91","sht_96","sht_97","sht_98"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_04406a58ab831ae699e21d08b8c98ba7","appearance_count":2,"total_visible_duration_ms":26239,"first_appearance_ms":985160,"last_appearance_ms":1021189,"appearance_strength":0.206,"shots_appeared_in":["sht_79","sht_81"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_409cfb5cc3b67ef22666e60287e857e4","appearance_count":2,"total_visible_duration_ms":24275,"first_appearance_ms":985160,"last_appearance_ms":1009435,"appearance_strength":0.206,"shots_appeared_in":["sht_79","sht_80"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_65e6ca38244d1667696c8f109cd37b85","appearance_count":3,"total_visible_duration_ms":39275,"first_appearance_ms":985160,"last_appearance_ms":1062882,"appearance_strength":0.206,"shots_appeared_in":["sht_79","sht_80","sht_84"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_627b0a177ec5ebe37949f93c7adb536b","appearance_count":1,"total_visible_duration_ms":11391,"first_appearance_ms":327458,"last_appearance_ms":338849,"appearance_strength":0.203,"shots_appeared_in":["sht_26"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_45175450-c86e-4de0-8553-90b22143a40c","appearance_count":1,"total_visible_duration_ms":12239,"first_appearance_ms":545678,"last_appearance_ms":557917,"appearance_strength":0.203,"shots_appeared_in":["sht_44"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_bfc65d0b-b508-403a-890b-96522652b6a5","appearance_count":2,"total_visible_duration_ms":25110,"first_appearance_ms":557917,"last_appearance_ms":610542,"appearance_strength":0.203,"shots_appeared_in":["sht_45","sht_48"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_6f26b67ee4aea0b077fffc62ac777efe","appearance_count":1,"total_visible_duration_ms":9697,"first_appearance_ms":963386,"last_appearance_ms":973083,"appearance_strength":0.203,"shots_appeared_in":["sht_77"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_ae8f3da2281f0437bdcafc8a0058ff84","appearance_count":1,"total_visible_duration_ms":14066,"first_appearance_ms":1948917,"last_appearance_ms":1962983,"appearance_strength":0.203,"shots_appeared_in":["sht_158"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_2194060e17a174e4c27c09f177351d70","appearance_count":1,"total_visible_duration_ms":12899,"first_appearance_ms":878650,"last_appearance_ms":891549,"appearance_strength":0.203,"shots_appeared_in":["sht_70"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_01797a1d4cf0c5a11814ae38ec92a118","appearance_count":1,"total_visible_duration_ms":12899,"first_appearance_ms":878650,"last_appearance_ms":891549,"appearance_strength":0.203,"shots_appeared_in":["sht_70"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_84dc3c1286160f2a4032f88b3e8b9cfb","appearance_count":1,"total_visible_duration_ms":12239,"first_appearance_ms":545678,"last_appearance_ms":557917,"appearance_strength":0.203,"shots_appeared_in":["sht_44"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_7f6f5dddbaf6008acc31b8dde4d09254","appearance_count":2,"total_visible_duration_ms":22068,"first_appearance_ms":1407057,"last_appearance_ms":1429125,"appearance_strength":0.203,"shots_appeared_in":["sht_113","sht_114"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_57e36c5da0caa91427a797c7e5b6f432","appearance_count":1,"total_visible_duration_ms":11769,"first_appearance_ms":48167,"last_appearance_ms":59936,"appearance_strength":0.203,"shots_appeared_in":["sht_4"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_5e44ccec-4517-469d-9d23-e79fc5531291","appearance_count":1,"total_visible_duration_ms":14066,"first_appearance_ms":1948917,"last_appearance_ms":1962983,"appearance_strength":0.203,"shots_appeared_in":["sht_158"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_b002f887aeee4a196bec7b6031af217b","appearance_count":1,"total_visible_duration_ms":15000,"first_appearance_ms":1326350,"last_appearance_ms":1341350,"appearance_strength":0.203,"shots_appeared_in":["sht_106"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_f000e5694735cec3c528e0ca2c122006","appearance_count":1,"total_visible_duration_ms":14155,"first_appearance_ms":59936,"last_appearance_ms":74091,"appearance_strength":0.203,"shots_appeared_in":["sht_5"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_59bbe9c1bd6ffeccf0ef495b1a77ab7d","appearance_count":1,"total_visible_duration_ms":14485,"first_appearance_ms":985160,"last_appearance_ms":999645,"appearance_strength":0.203,"shots_appeared_in":["sht_79"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_8d208bbf2208872e313e29db2b70dc38","appearance_count":1,"total_visible_duration_ms":14923,"first_appearance_ms":809275,"last_appearance_ms":824198,"appearance_strength":0.203,"shots_appeared_in":["sht_65"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_665536957da8315a5a9d4af735a1f40c","appearance_count":2,"total_visible_duration_ms":22068,"first_appearance_ms":1407057,"last_appearance_ms":1429125,"appearance_strength":0.203,"shots_appeared_in":["sht_113","sht_114"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_f122deea66464e885edb68ce1701437b","appearance_count":1,"total_visible_duration_ms":14666,"first_appearance_ms":610542,"last_appearance_ms":625208,"appearance_strength":0.203,"shots_appeared_in":["sht_49"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_da6652be00eed8664a95f3abaf6b3634","appearance_count":1,"total_visible_duration_ms":10143,"first_appearance_ms":1883215,"last_appearance_ms":1893358,"appearance_strength":0.203,"shots_appeared_in":["sht_153"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_76bf4e6f5fcfac45836f1fe139faeb59","appearance_count":1,"total_visible_duration_ms":10718,"first_appearance_ms":1153587,"last_appearance_ms":1164305,"appearance_strength":0.203,"shots_appeared_in":["sht_92"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_6f1da63ab3dee19d25200449733024a8","appearance_count":1,"total_visible_duration_ms":14923,"first_appearance_ms":809275,"last_appearance_ms":824198,"appearance_strength":0.203,"shots_appeared_in":["sht_65"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_9bd44dfbc2074a8168b19ddf872b13a2","appearance_count":1,"total_visible_duration_ms":14485,"first_appearance_ms":985160,"last_appearance_ms":999645,"appearance_strength":0.203,"shots_appeared_in":["sht_79"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_80d14ba6e7a1f2930be875a65528acde","appearance_count":1,"total_visible_duration_ms":11120,"first_appearance_ms":1418005,"last_appearance_ms":1429125,"appearance_strength":0.203,"shots_appeared_in":["sht_114"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_578fdec782a2a1d689b3dcf924b3dd66","appearance_count":1,"total_visible_duration_ms":14923,"first_appearance_ms":809275,"last_appearance_ms":824198,"appearance_strength":0.203,"shots_appeared_in":["sht_65"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_16f60149-d305-4ce6-9809-93eb78df8ef5","appearance_count":1,"total_visible_duration_ms":14155,"first_appearance_ms":59936,"last_appearance_ms":74091,"appearance_strength":0.203,"shots_appeared_in":["sht_5"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_913a6ced4ce74396c21d8978aad1c8cb","appearance_count":2,"total_visible_duration_ms":24292,"first_appearance_ms":1239250,"last_appearance_ms":1263542,"appearance_strength":0.203,"shots_appeared_in":["sht_100","sht_99"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_8e159ea51321487286983313973a41a0","appearance_count":1,"total_visible_duration_ms":11312,"first_appearance_ms":1164305,"last_appearance_ms":1175617,"appearance_strength":0.203,"shots_appeared_in":["sht_93"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_526734b287cdf9adfa5765e9b72b8bde","appearance_count":1,"total_visible_duration_ms":10948,"first_appearance_ms":1407057,"last_appearance_ms":1418005,"appearance_strength":0.203,"shots_appeared_in":["sht_113"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_34c72865d8528167ad8904f36e71bd76","appearance_count":1,"total_visible_duration_ms":11391,"first_appearance_ms":327458,"last_appearance_ms":338849,"appearance_strength":0.203,"shots_appeared_in":["sht_26"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_6ec75751cdea3144c93f63fa828b7acc","appearance_count":1,"total_visible_duration_ms":13727,"first_appearance_ms":1538773,"last_appearance_ms":1552500,"appearance_strength":0.203,"shots_appeared_in":["sht_124"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_94f4a7f58a3f24593b2f3e856fdde4a8","appearance_count":1,"total_visible_duration_ms":11769,"first_appearance_ms":48167,"last_appearance_ms":59936,"appearance_strength":0.203,"shots_appeared_in":["sht_4"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_7290ff888c048498e298764f503ec0c5","appearance_count":2,"total_visible_duration_ms":25110,"first_appearance_ms":557917,"last_appearance_ms":610542,"appearance_strength":0.203,"shots_appeared_in":["sht_45","sht_48"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_f6677f64-31a6-4383-b60c-91f0851302c7","appearance_count":5,"total_visible_duration_ms":66640,"first_appearance_ms":801275,"last_appearance_ms":1921998,"appearance_strength":0.016,"shots_appeared_in":["sht_154","sht_155","sht_64","sht_68","sht_69"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_5f383f1fec26b0df6bc815d9aee4cf47","appearance_count":4,"total_visible_duration_ms":51464,"first_appearance_ms":973083,"last_appearance_ms":1791250,"appearance_strength":0.012,"shots_appeared_in":["sht_142","sht_143","sht_144","sht_78"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_0abfc4aa-c8f7-4e9c-babf-662a802b50ef","appearance_count":3,"total_visible_duration_ms":39709,"first_appearance_ms":1087208,"last_appearance_ms":1126917,"appearance_strength":0.009,"shots_appeared_in":["sht_87","sht_88","sht_89"],"filmed_at_score":null,"present_in_score":null,"relation":"mentioned"},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_91b7c497-1d72-471e-8986-108b68b1a8bf","appearance_count":2,"total_visible_duration_ms":27239,"first_appearance_ms":530678,"last_appearance_ms":557917,"appearance_strength":0.006,"shots_appeared_in":["sht_43","sht_44"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_d7804d83-d910-414c-b89e-1f15c7b6503d","appearance_count":2,"total_visible_duration_ms":27330,"first_appearance_ms":863650,"last_appearance_ms":1087208,"appearance_strength":0.006,"shots_appeared_in":["sht_69","sht_86"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_cefdeb8b-eae6-4e47-9e30-0937d7306c09","appearance_count":2,"total_visible_duration_ms":28164,"first_appearance_ms":640203,"last_appearance_ms":668367,"appearance_strength":0.006,"shots_appeared_in":["sht_51","sht_52"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_b29d737e-0ac5-4e9f-9fd2-1c32f379b8cb","appearance_count":2,"total_visible_duration_ms":25878,"first_appearance_ms":1751863,"last_appearance_ms":1777741,"appearance_strength":0.006,"shots_appeared_in":["sht_142","sht_143"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_f65a2fd5-6650-4e71-afa1-6d21f7f4d875","appearance_count":2,"total_visible_duration_ms":24452,"first_appearance_ms":824198,"last_appearance_ms":848650,"appearance_strength":0.006,"shots_appeared_in":["sht_66","sht_67"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_a0462240-fcd3-4af1-9a4b-8329d6a52454","appearance_count":2,"total_visible_duration_ms":27330,"first_appearance_ms":1074878,"last_appearance_ms":1102208,"appearance_strength":0.006,"shots_appeared_in":["sht_86","sht_87"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_2425febecb47ae91ba277939f20b9ef3","appearance_count":2,"total_visible_duration_ms":24275,"first_appearance_ms":985160,"last_appearance_ms":1009435,"appearance_strength":0.006,"shots_appeared_in":["sht_79","sht_80"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_95f91400-3cbe-4754-b031-62b05747dab4","appearance_count":2,"total_visible_duration_ms":30000,"first_appearance_ms":572917,"last_appearance_ms":863650,"appearance_strength":0.006,"shots_appeared_in":["sht_46","sht_68"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_53a106957af99f597540fa15fd63c628","appearance_count":3,"total_visible_duration_ms":40286,"first_appearance_ms":557917,"last_appearance_ms":1618271,"appearance_strength":0.006,"shots_appeared_in":["sht_128","sht_129","sht_45"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_742cfb2c229e3497c79a08f766e6b1f1","appearance_count":2,"total_visible_duration_ms":16979,"first_appearance_ms":1631750,"last_appearance_ms":1648729,"appearance_strength":0.006,"shots_appeared_in":["sht_131","sht_132"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_1003554c9ffa7666633fb3829d422dcf","appearance_count":1,"total_visible_duration_ms":15000,"first_appearance_ms":572917,"last_appearance_ms":587917,"appearance_strength":0.003,"shots_appeared_in":["sht_46"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_a9e5db3c-e768-4e77-ab23-850a134d7b07","appearance_count":1,"total_visible_duration_ms":14485,"first_appearance_ms":985160,"last_appearance_ms":999645,"appearance_strength":0.003,"shots_appeared_in":["sht_79"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_e3c10ccfce46fd7e0ed9304c98cd2977","appearance_count":1,"total_visible_duration_ms":14923,"first_appearance_ms":809275,"last_appearance_ms":824198,"appearance_strength":0.003,"shots_appeared_in":["sht_65"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_c8cb1aeb06fa64ed4294c1279d0c7dca","appearance_count":1,"total_visible_duration_ms":10213,"first_appearance_ms":1648729,"last_appearance_ms":1658942,"appearance_strength":0.003,"shots_appeared_in":["sht_133"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_b99b525cd404f73f2002cdb3fc3cf572","appearance_count":1,"total_visible_duration_ms":13340,"first_appearance_ms":1869875,"last_appearance_ms":1883215,"appearance_strength":0.003,"shots_appeared_in":["sht_152"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_aa5296ff-1580-42d0-87db-52259cc0e257","appearance_count":1,"total_visible_duration_ms":15000,"first_appearance_ms":848650,"last_appearance_ms":863650,"appearance_strength":0.003,"shots_appeared_in":["sht_68"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_33ce2df73e75010d60cb1724225821bc","appearance_count":1,"total_visible_duration_ms":13913,"first_appearance_ms":302919,"last_appearance_ms":316832,"appearance_strength":0.003,"shots_appeared_in":["sht_24"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_8a3691bd61a8532d3aaa3ef9d749e3a9","appearance_count":1,"total_visible_duration_ms":14734,"first_appearance_ms":891549,"last_appearance_ms":906283,"appearance_strength":0.003,"shots_appeared_in":["sht_71"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_4e8516541fed51ef3fa0189dd2eb5f03","appearance_count":1,"total_visible_duration_ms":9724,"first_appearance_ms":263359,"last_appearance_ms":273083,"appearance_strength":0.003,"shots_appeared_in":["sht_21"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_cd4bb989c028399afaf3ca7e59be6d0f","appearance_count":1,"total_visible_duration_ms":13175,"first_appearance_ms":906283,"last_appearance_ms":919458,"appearance_strength":0.003,"shots_appeared_in":["sht_72"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_27473110179dbbafef08367c58bf843b","appearance_count":1,"total_visible_duration_ms":14583,"first_appearance_ms":1763158,"last_appearance_ms":1777741,"appearance_strength":0.003,"shots_appeared_in":["sht_143"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_2b325911-9859-430a-ab1f-c2eee5f11f8e","appearance_count":1,"total_visible_duration_ms":15000,"first_appearance_ms":557917,"last_appearance_ms":572917,"appearance_strength":0.003,"shots_appeared_in":["sht_45"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_a13a7733-0be2-4e51-9318-25622e7b6fbd","appearance_count":1,"total_visible_duration_ms":10143,"first_appearance_ms":1883215,"last_appearance_ms":1893358,"appearance_strength":0.003,"shots_appeared_in":["sht_153"],"filmed_at_score":null,"present_in_score":null},{"video_id":"video_reLhJnQd-Lo","entity_id":"ent_b987acd01400347ba9932dfd412e2c65","appearance_count":1,"total_visible_duration_ms":14666,"first_appearance_ms":610542,"last_appearance_ms":625208,"appearance_strength":0.003,"shots_appeared_in":["sht_49"],"filmed_at_score":null,"present_in_score":null}],"model_provenance":{"transcription":{"provider":"assemblyai","model":"universal-3-5-pro","version":"d1b6b51a-f460-412f-b214-0f7836b2af7f","ran_at":"2026-10-05T09:55:47.124Z"},"shot_detection":{"provider":"uva-cv-lab/omnishotcut","model":"omnishotcut-shot-query-transformer","version":"github@338c0e7","ran_at":"2026-10-05T09:55:47.124Z"},"visual_captions":{"provider":"google","model":"gemini-3.8-flash","version":"cascade-base-window-products-v3-caption-v2","ran_at":"2026-10-05T09:55:47.124Z"},"chapter_summary":{"provider":"google","model":"gemini-3.8-flash","version":"multimodal-chapters-v2","ran_at":"2026-10-09T23:53:39.599Z","methodology_version":"speech+captions+ocr+cuts+audio"},"chapter_visual_description":{"provider":"google","model":"gemini-3.8-flash","version":"chapter-visual-description-v1","ran_at":"2026-10-09T23:54:21.545Z"},"video_facets":{"provider":"google","model":"gemini-3.8-flash","version":"video-insights-v5","ran_at":"2026-10-09T23:55:09.363Z","methodology_version":"chapters+description+ocr;measured:music,speech_pace;evidence"},"shot_facets":{"provider":"google","model":"gemini-3.8-flash","version":"cascade-base-window-products-v3-caption-v2","ran_at":"2026-10-05T09:55:47.124Z"},"entity_extraction":{"provider":"google","model":"gemini-3.8-flash","version":"video-entity-review-v9","ran_at":"2026-10-10T19:14:00.082Z","channels":["window_caption","video_insights","entity_review"]},"entity_canonicalization":{"provider":"google","model":"gemini-3.8-flash","version":"video-entity-link-v4","ran_at":"2026-10-10T19:15:39.713Z","channels":["entity_link"]},"custom_metadata":{"provider":"google","model":"gemini-3.7-flash","version":"video-extract-v1","ran_at":"2026-08-27T09:48:04.884Z"},"embedding":{"provider":"google","model":"gemini-embedding-2","version":"arv-e2-v1","ran_at":"2026-10-05T09:55:47.125Z","channels":["shot_caption_text","shot_mm","video_text","chapter_mm_pieces","chapter_text","window_mm"],"task_hints":{"shot_caption_text":"visual_moment_search","shot_mm":"fine_visual_moment_retrieval","video_text":"video_level_retrieval","chapter_mm_pieces":"fine_visual_moment_retrieval","chapter_text":"topic_search_within_video","window_mm":"fine_visual_moment_retrieval"}},"boundary_snap":{"provider":"agentcdn","model":"boundary-snap-rule-set","version":"v1","ran_at":"2026-10-05T09:55:47.125Z","methodology_version":"shot>sentence>word"},"evidence_grading":{"provider":"agentcdn","model":"evidence-grade-heuristic","version":"v1","ran_at":"2026-10-05T09:55:47.125Z"},"content_hash":{"provider":"agentcdn","model":"sha256","version":"v1","ran_at":"2026-10-05T09:55:47.124Z"}},"model_provenance_status":"recorded","rights":{"policy_id":"rgt_4d2f3416-dfce-4c0b-a4d1-cb77ac6a210b","policy_version":3,"scope":"account","effective_at":"2026-09-30T01:46:50.398Z","permissions":{"search":true,"cite":true,"summarize":true,"quote":{"allowed":true,"max_duration_ms":null},"play":true,"segment_display":true,"source_embed_display":true,"full_video_display":false,"clip":true,"export":false,"train":false},"commercial_use":false,"attribution":{"required":true},"license_url":"https://api.agentcdn.com/public/creators/sandersaar/license.xml"},"transcript_cues":[{"start_ms":0,"text":"We've definitely hit the Apollo landing moment where all eyes are on artificial intelligence and how it will reset everything. My AI buddy always nails the intros, and I am incredibly excited to explore together with you how AI is going to change your and my life, why this is happening now, and what are some of the coolest examples out there. If you asked somebody 5 years ago what would be the first skills to be replaced by artificial intelligence, they would have said the physical labor— warehouse workers, truck drivers— followed by the less demanding cognitive labor like accountants, then the high-demand cognitive labor like computer development, and lastly it's gonna go for creativity because there's something so human about it that cannot be replaced by computers, like creating art and films.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":40000,"text":"Actually, the exact opposite is happening at the moment, where creativity is the first use case for artificial intelligence to go into masses, then followed by high-demand cognitive labor and others. And you can see it in the success of ChatGPT, taking just less than 2 months to reach 100 million users, compared to Facebook taking 4 years, or mobiles taking 12 years to do that. It's one of the fastest platforms to grow, which is not entirely true as Pokémon Go actually reached 100 million users in just less than a month, but that's a topic for another video.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":69000,"text":"Artificial intelligence is top of mind for everybody around the world. That's seen on Google Trends. We're really entering in this new platform era. It started in the 1980s with PCs where we got browsers and search that we're still using. Building on top of that, we got mobile and mobile really brought us those app stores where you can get any service as an app, as well as all the super apps. And social media apps that we're using day to day. That was building on top of the cloud platform which we've been using so far that brought cross-platform usage.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":98000,"text":"It just didn't matter which device you're holding in your hand and collaboration in the cloud. And now we're entering this new phase in artificial intelligence which is bringing first intelligent agents and who knows what in the future that we're going to explore later. But why is this happening now? Let's hear from the co-founder and CEO of OpenAI, Sam Altman.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":118000,"text":"We had the model for ChatGPT in the API for like, I don't know, 10 months or something before we made ChatGPT. And I sort of thought someone was going to just build it or whatever. And that, you know, enough people had played around with it. Definitely if you make a really good user experience on top of something. And like one thing that I very deeply believed was the way people wanted to interact with these models was via dialogue.","speaker_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","speaker_label":"Speaker B","speaker_name":"Sam Altman"},{"start_ms":144000,"text":"And I, you know, we kept telling people this, we kept trying to convince people to build it and people wouldn't quite do it. So we finally said, all right, we're just going to do it.","speaker_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","speaker_label":"Speaker B","speaker_name":"Sam Altman"},{"start_ms":150000,"text":"And they just did it. And really, I think there's more than just the interface, but that's the key part because all the pieces were in place already 10 months before, as you said, but nobody was really using it. I believe the key drivers are the quality of the models themselves, interface, which we're going to talk first, and cost. So let's dive in. On interface, you can see on the left, the ChatGPT, not ChatGPT, GPT-3 playground where you could actually have a chat experience already one and and a half years ago where all the models were pretty much in place with slight tweaks that were made to ChatGPT.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":182000,"text":"So even the slightest changes in the interface and design and the model can make a huge difference. The interface for AI has really changed over time. We started with those base models that are still around like GPT-3, but it's really hard to interact with them because you need to be really precise and a good prompt engineer to get any information out that you need. Then we got those instruct models. That means the input-output models. We had a specific goal in mind, for example, to get weather information or to get a quote from the book so the AI knew what to look for.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":212000,"text":"And now more recently, just a couple of months ago, we've been seeing those exchange interfaces where you can have a 2-way conversation like ChatGPT. On top of that, we now have this retrieval interface model, and that's something you see in the Bing Chat where you put that— take that ChatGPT experience and put that together with the information, the live information that you can retrieve from the real life So you can also quote the results.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":238000,"text":"And that's the stage we're at now. The ultimate version of the interface for AI that everybody's chasing is the action model. That's where everybody has your login details as well as your credit card information. So you can actually do it for you rather than you having to instruct it or have a conversation or try to retrieve information. Now you can just go straight to action and you see it being used everywhere like Replit in the code generator that's built into the tool.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":263000,"text":"They're using definitely like some of the models that we talked about, Or Edge that is using the retrieval model, putting together with a conversational model to optimize the results. So the interface is evolving. The second thing that we talked about, why it's happening in addition to interface, is cost. We talked about in my previous video how the cost of making videos using AI is now almost nothing compared to just 10 years ago where you had to pay several hundred thousand for a camera and tools. If you want to see that video, by the way, click on the video up here and, and check it out, how it's going to change Hollywood.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":293000,"text":"Where you can make videos on your laptop using, you know, GPT as a writer, using Descript or other voice platforms as an audio, and using Synthesia, for example, to create the video, or any other video generation platform. The same thing is about to happen for intelligence. What it used to cost you a quarter of a million dollars to get a degree in an Ivy League college, you could get now an online course, an online certificate for just, you know, $10,000. Or you can just use the ChatGPT interface for $20 a month and have access to all of the intelligence of the world, all of the books that you can imagine, the websites you could never visit.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":328000,"text":"This is happening because number one, we have so much data out there. Everything that is done on computers is now tracked and part of those models. So the amount of data that we could use to model is insane. At the same time, the cost of compute and energy has been coming down rapidly. So building those models and using those models is becoming cheaper and cheaper. And that's why the first to go really are the works that are done on computer. They're using bits that can actually be used to model out.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":354000,"text":"Rather than atoms, the physical elements, the warehouses, which are much harder to track as all of it is not yet connected to the internet. Last but not least is the quality of the models. Why is AI now in a space, or machine learning in a space where we can find it more useful than ever? It really started in the '90s when the algorithmic models were developed. Those were the simple ways of digitizing when we were starting to digitize our assets, like, you know, accounting going to a computer so we can do simple calculations, search and filtering, input-output methods.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":385000,"text":"It then evolved into data science that was driven by big data, where everything started to be connected to the internet and computers. And we started to do more and more of our applications online and on computers. Hence, we started building those larger, more complex algorithms and using data science to resolve them. And then we evolved into machine learning, which is really a subset of artificial intelligence in the middle of 2010s, where we had humans as well as then computers developing the algorithms.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":413000,"text":"And that's where deep learning models were developed and where computers really became better in many things, for example, in understanding image classification, where they surpassed human accuracy in 2015, or even for classifying audio or replicating audio, that was surpassed in the mid-2010s. To now, that has evolved into large generative AI models, which have less human in the loop, which are using completely new methods of artificial intelligence, and the Transformers methods as well as the new attention methods.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":443000,"text":"So it's not somebody guiding the AI to go from one step to another, but we really give free hands to the AIs to figure out themselves where the attention should be in the model and build multiple hundreds of versions of that. And that's why the quality has got so good. And these 3 are the key drivers for making AI paradigm shift. You know, the quality of the models achieving human-level intelligence, the interface and user experience turning into chat, human-like conversations, as well as the tools trending rapidly towards zero.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":470000,"text":"And that's what made ChatGPT a success. Obviously, we as humans have to evolve with it. We started on being, you know, those passive human beings who were foraging around, gathering their own goods, to laborers who started specializing in different skill sets. And that gave a start to communities, cities, and agriculture, to now the stage where we are at, where we started building machines to do the work for us. We started building computers and manufacturing machines, so we became the creators and operating those machines.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":497000,"text":"And the stage we're entering now is becoming narrators so that we don't build the machines, but we just describe what needs to be built and the machines will build themselves. And this does not just apply to language, which we talked about a lot. This also goes to image modeling, audio modeling, video modeling, code modeling, and science modeling. We're gonna dive into each one of them, how they're developed, and some of the coolest examples that you've probably not seen yet. So first of all, what are some of the largest language models out there?","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":522000,"text":"Obviously ChatGPT, we've talked about a lot. In addition to that, we have Bing Chat that has evolved on top of that to get the retrieval model out there. Google has announced Bard, their own chat version of search, as well as Anthropic, which has developed a model called Claude. It's interesting that the talent has really been moving from Google to OpenAI to develop those models, and OpenAI people have now moved to Anthropic to develop new language models. They have hundreds of billions of parameters.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":548000,"text":"Baidu has one of the largest models out there with 260 billion parameters, and many of the other companies that are not on this slide are also developing their own large language models. But how do you actually develop one? Let's use OpenAI as an example. First of all, you need a ton of data that includes in their case, all of the books that you can imagine, Substack, Reddit, Twitter, Wikipedia, and Common Crawl, which really takes almost all of the website data and puts it in a structured form for anybody to access.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":575000,"text":"That's about 45 terabytes of data. And then all of that data is sent for training using the transformer infrastructure to the NVIDIA's most powerful A100 chips, and all of that compute power will cost about $5 million to generate. And then from that, from that 45 terabytes of data and training, you get a GPT-3 model, which is about 800 gigabytes. So you could probably fit it on your computer, but to run it, it still costs quite a bit of money.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":600000,"text":"That's the inference part where you actually get the results. And it costs about anywhere between 2 to 8 cents, depending on the length of the response that the OpenAI ChatGPT is providing. For the server cost. Of course, there's many more models out there. If you want to give it— check them out, go to lifearchitect.ai. And he's also got a great YouTube channel that keeps track of all of the new models coming out, and you can rank them based on different sites. You will find a link for that down below in the description.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":626000,"text":"The uses for language models actually go beyond what people imagine today and what is on this slide. Number one is of course text generation. We talked about that a lot, where you give a prompt and you expect to get a response back. You could also use it for summarization. If you have a large PDF in front of you and you want to just answer your question and find that nugget that will be useful for you, you can use it for that. You can use it for translation. The language models are so much better than Google Translate ever really was and makes the knowledge accessible that is currently only locked to a couple of languages to everybody in any language.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":657000,"text":"Of course, you can use it for search, and that's what Bing reimagines their own experience around, getting more precise and more conversational answers. Straight in the search rather than having to search for them on the websites. And it becomes an intelligence assistant, helping you use applications that maybe you don't even know how to use the user interface, but you can use natural language to give it prompts and guidance. And last but not least, you can use it for analysis, so help to derive data from different websites or from the whole language model to help you make an easier purchase decision, for example.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":688000,"text":"An example of one of them is Jasper, which has built a platform that helps you to do any kind of writing that you can imagine straight on the platform. So let's see how it actually works in Jasper is here to help. It's an app that uses AI to help you create any kind of content you need. It's like having your very own robot sidekick with an infinite supply of ideas. And that's why the perfect pair of socks is like a hug for your feet.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":713000,"text":"Hot damn, that's good! And you can use it really for everything, from generating ideas to generating full scripts. And here's an example how Mint Mobile was using it right, really in practice. So they gave prompts to OpenAI, to use a joke curse word and big wireless and still going titles in the video. And then Ryan Reynolds was seeing if ChatGPT was doing its job and then reading it back and this becoming a real ad that they were running earlier this year.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":738000,"text":"First of all, let me just say Mint Mobile is the shit. But here's the thing. All the big wireless companies out there are ending their holiday promos, but not Mint Mobile. We're keeping the party going because we're just that damn good. Give Mint Mobile a try. And hey, as an added bonus, if you sign up now, you'll get to hear my voice every time you call customer service. Just kidding, that's, that's not really a thing. And that's something where it becomes really practical for them to build content, create awareness by using somebody as a writer that costs them $20 a month.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":771000,"text":"Bing recently launched Bing Chat. As I said, this redesigns and reimagines search. So rather than just giving a 3-word keyword, which most of the people do, and half of the people go straight back to search. You can actually give guidance and then go back and forth to find exactly what you need. This is to get, for example, craft ideas. The other example here is to build a fitness workout program. Again, like for your needs exactly of what you're looking for, and it's gonna try and ask questions in return and then build the perfect program just for you and also referencing the websites where it's getting the information from.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":802000,"text":"So getting, putting the large language model together with a retrieval model of what you can get from the websites And gives you the perfect answer. The second category is image modeling. Some of the largest ones out there that you might've heard of is obviously DALL-E 2 that you can go out there and use, Stable Diffusion, which is open source. You can run it on your, on your own computer and Midjourney, which you can use in Discord, uh, in a chat environment. Google has built some of the strongest models out there as well, but they haven't really made them public yet, but they are one of the largest in terms of parameters.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":831000,"text":"How do you build an image model? Stable Diffusion is a good example here. You use the data. They have large open data like LAION, which has got 5 billion text and image pairs. So you put together the description of the image together with what the image looks like. They also had a deal with Getty Images, which we're going to talk about later, a controversial deal. You put that 120 terabytes of data together. And then based on the CEO of Stability AI, it took about, it took about 150,000 hours of compute for the most powerful Intel A100, NVIDIA A100 chip And cost them about $600,000 to build the model.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":867000,"text":"And then the model itself is just a 2 gigabyte file. So getting from 120 terabytes just down to 2 gigabytes, and you could run it right on your iPhone or right on your Mac or any computer that you imagine, or for free using your own graphics card and your own compute. Or you could go out and obviously use any other service that you pay a few dollars for. What can it be used for? Text to image, of course, we talked about that. Image to image. I'm sure you've seen Lensa AI where everybody's sharing their avatars with those predefined prompts to go from image to image.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":897000,"text":"You can use it for outpainting. So you might have a small image, but you want to expand on that, fill in the space. That's been done for many famous paintings out there. You can use it for inpainting, you know, to replace even somebody like Mona Lisa's hair with mohawk. You can use it for enhancements and editing, for color correcting, or you could also use it for classification and detection of different objects inside the images. So let's look at some of the examples out here now.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":921000,"text":"First of all, DALL-E that can take simple text descriptions like a koala dunking a basketball and turn them into photorealistic images that have never existed before. DALL-E 2 can also realistically edit and retouch photos based on a simple natural language description. It can fill in or replace part of an image with AI-generated imagery that blends seamlessly with the original. It's called inpainting.","speaker_id":null,"speaker_label":"Speaker C","speaker_name":null},{"start_ms":945000,"text":"And moreover, you could also use it not just for generating images, but from something practical point of view. This is an example for designing an interface. Using, using a tool called Galileo where you can say, hey, create me an onboarding screen for this dog walking app, and then also create the settings page with username and photo. And instantly you will have that generated by AI, again using the database that they have. Or you then want to create the profile page for completely something else, another, another app.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":973000,"text":"And by the way, while you generate all of that, this becomes transferable that you can take it out, for example, to Figma, and then from Figma launch it as an actual app. So imagine how far you can take it. Audio is another big area of modeling, specifically in speech modeling and music, where Google has been building large language models like AudioLM, which is building on WaveNet before. Amazon has built Polly, which is the largest— one of the largest voice libraries.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1000000,"text":"OpenAI has got Whisper, the text-to-speech library, Speechify, and many others. And in the music space, we're seeing more recently developments like MusicLM to generate music from text. Or SingSong, which you can use audio to audio, as well as one of the largest open source movements, Uberduck. What are some of the use cases? Of course, number one is text-to-speech. You can just type something up and it'd be spoken back to you. The second thing is taking a library of audio files and turning them into text and maybe making them searchable using AI.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1028000,"text":"You can do speech-to-speech in different languages or even in the same language. You can just change your voice using something like Respeecher. You can do text-to-music. You can just give prompts like, this is the type of vibe I want, this is the type of style I want, and it will generate music for you. Or you can hum some voice and it will fill in the background like a voice-to-music model. Or you can just do enhancements for the audio, you know, make the audio sound better or exactly what you want it to be described using a text model.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1055000,"text":"Here's an example how a company called podcast.ai brought back models from Joe Rogan voice as well as Steve Jobs voice and put them in a podcast format to See what they would say if Steve was still alive.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1068000,"text":"It's always good to see you, buddy. I'm so happy you came on, man.","speaker_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","speaker_label":"Speaker B","speaker_name":"Sam Altman"},{"start_ms":1072000,"text":"Yeah, it's great to be on the show. Your audience is just so different from your normal Apple users, and that's a good thing.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1079000,"text":"It's cool.","speaker_id":null,"speaker_label":"Speaker C","speaker_name":null},{"start_ms":1080000,"text":"Well, you know, I was an Apple user way before I did this show. I've been a fan of yours and Macintosh since the 1980s.","speaker_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","speaker_label":"Speaker B","speaker_name":"Sam Altman"},{"start_ms":1087000,"text":"I think it sounds great, not just from content point of view, but also from the sound of the voice. Of course, it's gonna get better and better over time. And this is an example of somebody who just used ChatGPT to come up with rap rhymes for Eminem and then used Uberduck to integrate, make them into sounds and then integrate into them his song. So this is what you're gonna hear now. This is the future rave sound. I'm getting lost in an army ground.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1112000,"text":"This is the future rave sound. I'm getting lost in an army ground. Eminem, bro. There's something that I made as a joke and it works so good. I could not believe it.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1130000,"text":"I discovered those websites that are about AI.","speaker_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","speaker_label":"Speaker B","speaker_name":"Sam Altman"},{"start_ms":1134000,"text":"And this is how much you can do with just little experiments that we're starting to see happen around us. Moving on to video, and this has really 2 major categories. One is 3D category, which is generating completely new objects that can be editable in any format. And the other one is avatars, which is somebody generating a lookalike like we did in the last video that you saw. In 3D, some of the largest models are Google's Image and Video. Of course, Meta showed their own Make-A-Video platform as well as, uh, KIRKI AI, which we're gonna see an example later.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1164000,"text":"In the avatar space, there's Synthesia, there's D-ID that can take photos to turn them into images and rephrase and many others that don't fit on this slide. What are some of the example use cases for video models? Of course, you can use text-to-video, just give it a prompt and it's gonna generate video out of it. You can do image to video, just give it a photo and then turn that into a moving, moving image. You can personalize video, edit just, you know, some templated objects inside the video to make it very unique.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1190000,"text":"You can do text to 3D, you can describe something and then turn that into a 3D object or a moving object. You can use it for classification, for example, for fitness apps to see how well you're tracking against, against the moves. And you can use it for enhancements and editing, up resolutions, down resolutions, clarifying, etc. So let's see some of the examples. And the first one here is the Runway ML and their new model to stylize any content, generate video in any style, all while retaining quality and flexibility.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1220000,"text":"Gen-1 is able to realistically and consistently apply the composition and style of an image or text prompt to the target video, allowing you to generate new video content using an existing video. We call this approach video-to-video. In addition to video-to-video, we also talked about photo-to-video, and a good example here is the D-ID platform.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1239000,"text":"D-ID's Creative Reality Studio just got even better. Now, as well as animating images from text or audio, you can create avatars and generate scripts using Stable Diffusion and GPT-3. Just describe what you want your avatar to look like and press create. You can make any avatar you want, say, picture of a half-alien goddess. Or a 19th century portrait of a happy Victorian gentleman.","speaker_id":null,"speaker_label":"Speaker C","speaker_name":null},{"start_ms":1263000,"text":"And then the other example here, going beyond just avatars and changing the style of the movies, is also to generate full 3D objects. An example here is the Kerky AI, where you can use text-to-3D prompt.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1291000,"text":"And somebody's using it in real life already is Roblox, where they're using generative AI and prompt guidance to design a game. Of course, they've got a whole host of tools out there that you can use today, but they're still pretty complex and only fraction of the users use them. So here you can just do text guidance of what do you want the car look like, what do you want the paint look like, what do you want the lights to do, and that automatically generates the code for you within the game that you can then launch on their platform.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1316000,"text":"Which makes it so easy and so much more accessible for people using natural language and also in game design or potentially even making movies inside games. Next one is code. And as you saw a little example in Roblox, it takes way beyond that. Some of the largest models out there are Codex by OpenAI as well as Copilot by GitHub, which is built on top of Codex really, and Replit, which built Ghostwriter, and Google, which in secret is about to announce their own coding tool, which they're developing internally, but there's no outside information available yet.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1349000,"text":"What do you use the code models for? Of course, number one is autofilling, helping you get more done faster. The other one is text-to-code, so you can describe something and it will turn into code like we saw with Roblox. Or you can do Q&A and reviews, finding the errors in the code. Or you can do code-to-text— somebody sends you a code but you don't understand what's going on, you can get it translated into human language. Here's an example from Replit. In the first one, you can see, hey, what does this index file do?","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1375000,"text":"And it tells you back in natural language of what the file is about to do. Then it defines an error. Hey, the error is on line 6, and here's how to fix it. So you can copy the code, put it back to your editor, and it fixes the result for you. And then you could use the chat interface to even develop further a new code for it for you, or even run QA on your existing code. This is how powerful it is. And Copilot has shared some results that it makes developers 55% faster, and almost half of the code that they do is now written by artificial intelligence.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1407000,"text":"Last but not least, let's talk about science because that's where some of the biggest breakthroughs might come through. DeepMind and Google have built AlphaFold 2, which is one of the most critical protein folding prediction models out there. That is also something that Meta is working on with ESMFold, as well as many of the other research bodies, including some of the open source ones like the OpenBioML, who's building together with other partners LibraFold project. What are some of the uses for science models? Number one is review and analysis.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1433000,"text":"You can take in all of the science papers and then understand what are the correlations, what others have been working on that humans can't possibly work through hundreds of thousands of papers that are coming out as they're exponentially growing and there's more and more every day. You can use it for predictive modeling, all of the phases, for example, of the development of new pharmaceuticals. You can use it for precision medicine, matching the specific individual and their biology to the specific pharmaceutical.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1459000,"text":"You can use it for genomics in really predicting how different proteins react. You can use it for diagnosis. So for example, taking a image of your retina scanned and then matching it to a diabetes database to be much more accessible and more predictive in your diseases, as well as for monitoring, including in-hospital and at-home care, where you can use connected devices and the behavior models that you can build on top of that to understand if the patients are progressing as well as they could.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1486000,"text":"And here's a quick video to explain how useful and valuable that is from AlphaFold.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1491000,"text":"So we created an AI system to solve this problem. Called AlphaFold. It's trained on the sequences and structures of about 100,000 proteins, painstakingly mapped out by scientists around the world. Today, it can accurately predict a protein shape just from its sequence of amino acids. AlphaFold's predictions could enable progress in all sorts of areas. Imagine a future where we can understand diseases more quickly and develop drugs to fight them, or one where we could use enzymes to break down plastic waste, or even to capture carbon from the atmosphere.","speaker_id":null,"speaker_label":"Speaker C","speaker_name":null},{"start_ms":1525000,"text":"But it's not all rosy for AI and generative AI. There are several challenges, including the hallucinations and the output errors, the high cost of the models, the lack of long-term memory, the slow inference time and data access, as well as the biases in the system. And let's go through each one of them and see what are the potential solutions. Out there you have a lot of articles that show you how the AI models are lying, or they're building a relationship, they're in love, they're sentient.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1550000,"text":"But actually I don't believe that really to be the case. I believe that's very much the kind of reaction the monkey would have with a mirror, you know, trying to fight off somebody that they see that they think is somebody else, but it's actually themselves. And I feel we're in a very similar spot with prompts at the moment. We are prompting the models to say the weird things back to us. We're prompting the model to go into a mode where it would kind of act back to us and what we're expecting from it. And I think this is exactly what's happening today.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1576000,"text":"It's more like a monkey in a mirror effect. Than really the AI models being sentient, because they still need to be prompted. They wouldn't do anything by themselves or on their own. And of course, there are already now more guardrails in place from, for example, Bing, who doesn't even want to talk about it when you ask if it's alive, if it's sentient, does it love you. All of those are now blocked, and it just says it's something that they prefer not to talk about.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1601000,"text":"And the output errors, that's another part of hallucinations, is we saw those errors in the Google demo when they announced Bard, but if you go to Reddit subreddits like ChatGPT or Bing, you'll see a whole lot more on how people are trying to trick the systems. And I think that's very normal to push the boundaries and make them better. And today the generated models are not great being factually correct, but I think in the phase of where Bing is at, where they're putting the large language model together with the, the exchange model together with the retrieval model, that's where you can really get good facts.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1632000,"text":"And that's an opportunity for many other companies like, for example, Wolfram Alpha to put the mathematical models and calculations together with the chat models of ChatGPT and make something that is factually correct but also is something that is more useful and easier for consumers to use in a natural language way. The second thing is high cost. It costs almost 10 times more, from just a fraction of a cent on Google Search to about 2 to 8 cents to run a query on ChatGPT.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1659000,"text":"Of course, doing the same with images today in Stable Diffusion also costs you about 2 cents if you're buying it on their own servers, running it rather than running it on your own. But I'm sure that's something that's going to change in the future as the compute gets again cheaper and we— they get more powerful. The slow inference time is directly connected to the cost. While it takes half a second to search something on Google, it takes about 5 to 20 seconds to get a prompt back in ChatGPT.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1684000,"text":"Of course, I believe that that's also going to change. Just over the last 6 months, Stable Diffusion has managed to decrease and increase the efficiency of the model by about 3 times by not really getting any reductions in the image quality. So instead of taking 6 seconds, it now takes 2 seconds to generate an image. And I think the same thing is going to apply to cost and speed in the future. The third— the fourth area is the lack of long-term memory.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1709000,"text":"Today, similarly, the ChatGPT model can only memorize 4,000 tokens, which is about 3,000 words in human language. And that's obviously a limiting factor if we want those assistants to be personal for everybody. But I'm sure again that's going to change as the compute and the models get more efficient. Data access and copyright. That's a key area. We've seen so many copyright issues now with, you know, Stable Diffusion being sued by Getty Images as they integrate those watermarks and they clearly see that their database has been used to generate the large image models.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1741000,"text":"You can see Elon Musk finding out that OpenAI has turned from a nonprofit to a for-profit organization and they've closed the Twitter access, Twitter data access for OpenAI. As well as many other artist groups and developer groups suing the large language models developers for the access of their data. But I think it's not really that dissimilar from what we used to do before. You know, if imagine if I'm a designer or photographer, I might go to Dribbble and Instagram and Pinterest and Behance to get all the inspiration and ideas in place.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1771000,"text":"I will then use, you know, Canva and Figma as a tool to create it. I will then put it on Behance, for example, as part of my portfolio. And then I'm gonna go and sell it on Shutterstock. I believe that if I then use, you know, copyrighted material, of course I'm going to be penalized and I shouldn't be able to make money with other people's content. That's the same in this flow, even though I took inspiration from them in my research phase. This is now just being done on superpowers, you know.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1796000,"text":"Rather than me being able to go maybe about 50 to 100 pages for inspiration, ChatGPT has done billions of pages for my inspiration. So I can use it as an input to get new ideas and develop new concepts. I can then use a tool like DALL-E or Stable Diffusion or anything else to create that content. I will then put it in my portfolio and then we'll monetize it in Shutterstock. By the way, Shutterstock has got a partnership with OpenAI already to start monetizing AI-generated images because I believe that this is just a super powerful research tool bringing you a lot more in a more simpler way, and it's a super powerful creation tool accelerating the process of creating and making it a lot more accessible for anybody.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1835000,"text":"But the copyright really comes in when you start monetizing it. So I wouldn't really see a version of where you can penalize those models as they're just there for inspiration. That's what we do every day. Artists and musicians take inspiration from others all the time. Now we can just do it on steroids. Last but not least, these are, you know, getting to be closed systems. You can see how OpenAI turned from open source, and that's something that Elon Musk is really concerned about as well, to something that is now just serving Microsoft.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1862000,"text":"And it was really created as like an anti-Google move to be something that is open, as it will be dangerous in his mind if it gets locked to just one organization. But it's never been really different. Take Gutenberg Press. When that was announced, churches went in and they actually created licensing for every Gutenberg Press to be out there. What can you print? How many can you print? Where can you print? They were all regulated by church. Or take even the Locomotive Act in the middle of 19th century in the UK.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1889000,"text":"Just to protect the horses and the industry that was built around it. There was a law where it had to have somebody walking in front of the car going only 4 miles an hour maximum and waving a red flag as they were doing it. Those were rules to protect and build a closed system, but the market has figured it out. In the AI space, there's some company called Stability AI who's behind Stable Diffusion, who is building AI by the people and for the people, aiming to make it open source and accessible for everybody, for any country and any individual to build their own models.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1921000,"text":"Overall, AI will and is about to reset absolutely everything for the 2 reasons.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1928000,"text":"My basic model of the next decade is that the cost of intelligence, the marginal cost of intelligence and the marginal cost of energy, are going to trend rapidly towards zero, like surprisingly far. And those, I think, are 2 of the major inputs into the cost of everything else, except the cost of things we want to be expensive, the status goods, whatever.","speaker_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","speaker_label":"Speaker B","speaker_name":"Sam Altman"},{"start_ms":1949000,"text":"And that's why everything will be about to be reset by AI, because intelligence and energy will be marginal. That's at the moment about 50 to 60% of the cost of the Fortune 500 companies is the payroll, paying for the labor. And if the labor and the intelligence will become essentially free for everybody, then that's going to rewrite how we do, how we get education, how it affects businesses, healthcare, art, software, and gaming.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null},{"start_ms":1974000,"text":"Which we're gonna talk about in the next video. So thank you very much for watching. I hope you enjoyed it. Let me know what you want to hear about first, and see you next time.","speaker_id":null,"speaker_label":"Speaker A","speaker_name":null}],"speakers":[{"speaker_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","entity_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","name":"Sam Altman","speaker_labels":["Speaker B"],"confidence":0.95,"method_version":"speaker-attribution-model-v3"}],"audio_spans":[{"id":"audio_span_d028665eee82c95b3dd0738cd738e1b1","kind":"music","category":"music_instrument","label":"Music","start_ms":698000,"end_ms":700500,"text":"Music","confidence":0.739,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_6f25936d44d227d113d4b6820680afc2","kind":"riff","category":null,"label":"Music","start_ms":698070,"end_ms":700451,"text":"Music notes A2 A4 F3 A2 C4 E4 A2 F2 C4 A3 C3 C3 C#3","confidence":0.4,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_28375414046c1745a9a6e2fc029384f2","kind":"music","category":"music_instrument","label":"Music","start_ms":703000,"end_ms":704500,"text":"Music","confidence":0.651,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_94a242fcba951b6f2ee1344a376dbea6","kind":"riff","category":null,"label":"Music","start_ms":703139,"end_ms":704486,"text":"Music notes C3 G3 B3 E2","confidence":0.364,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_2e5ea2d67e066b668aa20850e8827980","kind":"music","category":"music_instrument","label":"Music","start_ms":705500,"end_ms":707000,"text":"Music","confidence":0.385,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_f9e5368aa49af820a26619c0b43822fc","kind":"riff","category":null,"label":"Music","start_ms":705848,"end_ms":706429,"text":"Music notes A3 E3 A3 A3","confidence":0.372,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_3bdf34092bdeec14b633dfa5e884b489","kind":"music","category":"music_instrument","label":"Music","start_ms":707500,"end_ms":715000,"text":"Music","confidence":0.665,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_ed8505ed11aac134bafe57badb80f118","kind":"riff","category":null,"label":"Music","start_ms":707674,"end_ms":714737,"text":"Music notes C4 E3 G1 G2 B3 D4 B3 D3 E3 G3 E3 G3 C4 E5 G3 F3 E5 F3 A3 F3 A3 E3 G3 E5 and 21 more","confidence":0.427,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_6d6232a22ad7880601acbcec10ef6f39","kind":"music","category":"music_instrument","label":"Music","start_ms":737500,"end_ms":739500,"text":"Music","confidence":0.674,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_43468a94ceb2c8b28b35f5d3d92c50ce","kind":"riff","category":null,"label":"Music","start_ms":737512,"end_ms":738394,"text":"Music notes D#3 F2 A4","confidence":0.45,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_b4767a7e22480c1fed89454b76fe6886","kind":"music","category":"music_instrument","label":"Music","start_ms":742000,"end_ms":744000,"text":"Music","confidence":0.567,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_a9508044835d91effd587ddc7f704c2a","kind":"music","category":"music_instrument","label":"Music","start_ms":746500,"end_ms":748500,"text":"Music","confidence":0.575,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_ad5c7b54dae26887e4882fe2a71e6705","kind":"music","category":"music_instrument","label":"Music","start_ms":750000,"end_ms":752000,"text":"Music","confidence":0.634,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_d13ad48bfff374cfbae123cfcd6c94d1","kind":"riff","category":null,"label":"Music","start_ms":750116,"end_ms":754754,"text":"Music notes F2 F4 A4 E2 C3 C3 F#4 D#4 G4 E4 A2 E4 F4 D4 D4 C3 F#3 F#4 G4 E4","confidence":0.426,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_6ba11c21b1b2976c7c980fc0685a8ba2","kind":"music","category":"music_instrument","label":"Ukulele","start_ms":750500,"end_ms":752000,"text":"Ukulele","confidence":0.443,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_0b8ea7d5d816849836ec904e31b289a2","kind":"music","category":"music_instrument","label":"Music","start_ms":752500,"end_ms":756500,"text":"Music","confidence":0.507,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_41bc66f9808bf6545e82509cc2a72944","kind":"riff","category":null,"label":"Music","start_ms":755415,"end_ms":756158,"text":"Music notes C5 A4 A4","confidence":0.387,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_8f33b79f8d6d917f41bf578d7fb89c30","kind":"music","category":"music_instrument","label":"Music","start_ms":761500,"end_ms":762500,"text":"Music","confidence":0.459,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_1affe73262639b1307f950732951811c","kind":"music","category":"music_instrument","label":"Music","start_ms":926500,"end_ms":928000,"text":"Music","confidence":0.441,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_1b57f55b2390e0d1c1faa494b3ac7f09","kind":"music","category":"music_instrument","label":"Music","start_ms":931000,"end_ms":932000,"text":"Music","confidence":0.51,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_a878d7d0cd1b9031d47f743c47698a8a","kind":"riff","category":null,"label":"Music","start_ms":931070,"end_ms":931964,"text":"Music notes C2 C1 G3","confidence":0.396,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_f9eadd8a810a1c948dc557e161ed7f63","kind":"music","category":"music_instrument","label":"Music","start_ms":934000,"end_ms":935500,"text":"Music","confidence":0.648,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_8aeafca956879c80f06488de704ad236","kind":"riff","category":null,"label":"Music","start_ms":934522,"end_ms":935474,"text":"Music notes A1 D#3 B2 E1 A1","confidence":0.387,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_6ba1eff2481a9e8d9131e2bcf2abcd60","kind":"music","category":"music_instrument","label":"Music","start_ms":944000,"end_ms":945500,"text":"Music","confidence":0.488,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_db3a454e44c33ef5ac8b3389e6ce9cb1","kind":"riff","category":null,"label":"Music","start_ms":944511,"end_ms":945451,"text":"Music notes A1 C3 A1 F#3","confidence":0.516,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_d394134987c07fe6d005431bd0c07c6e","kind":"music","category":"music_instrument","label":"Music","start_ms":1108000,"end_ms":1109500,"text":"Music","confidence":0.74,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_13c7aec51db79171ce5a252bc3adce84","kind":"riff","category":null,"label":"Music","start_ms":1108104,"end_ms":1109091,"text":"Music notes D#3 D3 A1 A1 A1","confidence":0.546,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_79e9a8edae856d9caeebb04fc1516e80","kind":"music","category":"music_instrument","label":"Music","start_ms":1111500,"end_ms":1113500,"text":"Music","confidence":0.578,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_6ff0998be5a71ed02ed837bead5a94a1","kind":"riff","category":null,"label":"Music","start_ms":1112475,"end_ms":1113462,"text":"Music notes A1 A1 A1 D4 D4","confidence":0.494,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_9b4c45c5d635f62ee3627e971f23871a","kind":"music","category":"music_instrument","label":"Music","start_ms":1115500,"end_ms":1124000,"text":"Music","confidence":0.68,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_f97a394a5fb366507c8dc74dc2e4d2d6","kind":"riff","category":null,"label":"Music","start_ms":1119449,"end_ms":1123539,"text":"Music notes A1 G#1 G#1 G#1 G#1 A1 A1 G#1","confidence":0.37,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_4a95b892650b922131b68b086a522ebb","kind":"music","category":"music_instrument","label":"Music","start_ms":1218500,"end_ms":1220500,"text":"Music","confidence":0.71,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_b968b1a06a9d9a1de7eab9db6caf30b7","kind":"music","category":"music_instrument","label":"Music","start_ms":1222000,"end_ms":1223500,"text":"Music","confidence":0.33,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_a0efabb582b0a149b9c9938dc6e90b6a","kind":"music","category":"music_instrument","label":"Music","start_ms":1224500,"end_ms":1228500,"text":"Music","confidence":0.72,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_979081c233082394ad8ff6a21be32a74","kind":"riff","category":null,"label":"Music","start_ms":1227241,"end_ms":1228251,"text":"Music notes G4 G1 D4 C#3 E3","confidence":0.483,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_d5c68cd52eb6f1085bf661da0f378adf","kind":"music","category":"music_instrument","label":"Music","start_ms":1230000,"end_ms":1233000,"text":"Music","confidence":0.67,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_1193fb97fbbf436e9a8fc829dce2b0fc","kind":"riff","category":null,"label":"Music","start_ms":1231579,"end_ms":1232904,"text":"Music notes D4 D3 G3 C#3","confidence":0.498,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_39afde5c44eafc1517eee702ddf04597","kind":"music","category":"music_instrument","label":"Music","start_ms":1242000,"end_ms":1243000,"text":"Music","confidence":0.369,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_e2b08aaef591fdcb5aaffb72f21f030b","kind":"sound_event","category":"household_object","label":"Chopping (food)","start_ms":1243000,"end_ms":1244500,"text":"Chopping (food)","confidence":0.361,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_456aec23624dda974329d14dc0ab9d8f","kind":"music","category":"music_instrument","label":"Music","start_ms":1256500,"end_ms":1260000,"text":"Music","confidence":0.481,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_2909160ecfdef9444bb16c629ca7807e","kind":"riff","category":null,"label":"Music","start_ms":1258858,"end_ms":1259973,"text":"Music notes E3 F#3 B3","confidence":0.525,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_63fea739f7a9968d2d3409ab08992489","kind":"music","category":"music_instrument","label":"Music","start_ms":1262500,"end_ms":1264000,"text":"Music","confidence":0.469,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_2405f3832909e3b4e99c2c2cc28bcbc9","kind":"music","category":"music_instrument","label":"Music","start_ms":1275000,"end_ms":1292000,"text":"Music","confidence":0.676,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_b09fbb653c2930451ad9af14643d3a90","kind":"riff","category":null,"label":"Music","start_ms":1275023,"end_ms":1291601,"text":"Music notes G#3 C3 G2 D2 A#2 G2 G2 G3 A#2 C2 C4 D2 D2 G2 G2 A#2 A#3 C2 C4 F2 G2 G2 G2 G3 and 37 more","confidence":0.487,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_6620538f26db9048f7f8941560b2e5b1","kind":"music","category":"music_instrument","label":"Music","start_ms":1492000,"end_ms":1497000,"text":"Music","confidence":0.737,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_28fbb577ef91d835ddb394bba94b5fd7","kind":"riff","category":null,"label":"Music","start_ms":1492116,"end_ms":1494892,"text":"Music notes E3 G#3 G2 D#3 G3 C#3 F#3 G3","confidence":0.427,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_b2178c7c22e071f442615bdc74c9125a","kind":"riff","category":null,"label":"Music","start_ms":1495635,"end_ms":1496600,"text":"Music notes F1 D2 G3 C3","confidence":0.353,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_e786aa0a370da6946a3cd7fe3aeed474","kind":"music","category":"music_instrument","label":"Music","start_ms":1497500,"end_ms":1505500,"text":"Music","confidence":0.703,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_edfd4fdc20f303926579e2f5a93999b0","kind":"riff","category":null,"label":"Music","start_ms":1498034,"end_ms":1499638,"text":"Music notes D#3 E3 G3 D#3 B1","confidence":0.481,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_db187ac949d1aee401bb2f9deb2562d6","kind":"riff","category":null,"label":"Music","start_ms":1502762,"end_ms":1503727,"text":"Music notes E3 C4 E4","confidence":0.342,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_6afc1d91aeac84e67e9b26715bfff1b6","kind":"riff","category":null,"label":"Music","start_ms":1504458,"end_ms":1505213,"text":"Music notes G3 G3 D2 A3","confidence":0.393,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_d3e97aa7f29a29b7e92bd90e7f36a0f2","kind":"music","category":"music_instrument","label":"Music","start_ms":1506000,"end_ms":1525500,"text":"Music","confidence":0.763,"measured":true,"method_version":"panns-cnn14-dlmax+gate-60dbfs+percat-v2"},{"id":"audio_span_05194cd49c5289fc7d857290e028493e","kind":"riff","category":null,"label":"Music","start_ms":1508637,"end_ms":1509972,"text":"Music notes C2 F3 C3 C5 A#3","confidence":0.432,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_5b895589ebf543d9d6dc37379da9b435","kind":"riff","category":null,"label":"Music","start_ms":1510925,"end_ms":1511541,"text":"Music notes G3 F#3 F#3","confidence":0.491,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_b2c97f564a4279717b577fb47120d53b","kind":"riff","category":null,"label":"Music","start_ms":1513899,"end_ms":1517558,"text":"Music notes C4 A3 F3 G#3 F3 A3 E3 A3 F#3 F3","confidence":0.471,"measured":true,"method_version":"basic-pitch-0.4.0-v1"},{"id":"audio_span_4509200ba609b86bf959e59adbb761d5","kind":"riff","category":null,"label":"Music","start_ms":1518755,"end_ms":1523030,"text":"Music notes C3 G3 A#3 F#3 A#3 C3 F3 F3 A3 C5 C4 F4 A#3","confidence":0.42,"measured":true,"method_version":"basic-pitch-0.4.0-v1"}],"people_spans":[{"id":"person_span_339a8e4ba8949b434f39ee84212d7c86","start_ms":0,"end_ms":7000,"name":null,"description":"AI avatar presenter in a collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"AI avatar presenter in a collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_c4620026a96bf3a2390893065fb7caaa","start_ms":6800,"end_ms":9500,"name":null,"description":"Presenter in a patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter in a patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_b6d814693a55b8cf83d010afda486912","start_ms":9600,"end_ms":18100,"name":null,"description":"Host in a patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_0430cbcf4c074e6f9588a83a0cb827a4","start_ms":18200,"end_ms":33100,"name":null,"description":"Host in a patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_6985413c0a0f95be32078457a74416ec","start_ms":33200,"end_ms":48100,"name":null,"description":"Host in a patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_53ae3d1cf47e4add22b7dd7ea96e67c1","start_ms":48167,"end_ms":59867,"name":null,"description":"Male presenter wearing a collared shirt with floral patch detail","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a collared shirt with floral patch detail (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_54f6263211a33293642c194f7150e928","start_ms":59967,"end_ms":73867,"name":null,"description":"Male presenter wearing a floral-patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a floral-patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_33c40962e8f59435a9d286f97b3b0554","start_ms":74167,"end_ms":89067,"name":null,"description":"Male presenter wearing a floral-patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a floral-patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_7cef29a97e6e9965e0cd8e70f7c2ec62","start_ms":89167,"end_ms":102867,"name":null,"description":"Male presenter wearing a floral-patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a floral-patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_1005e59acf8a08e5817a4a8087d9687f","start_ms":102924,"end_ms":112424,"name":null,"description":"Host wearing a patterned collared shirt speaking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned collared shirt speaking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_23b1699516086702545421aafe5542fb","start_ms":110724,"end_ms":112424,"name":"Sam Altman","description":"Speaker holding a microphone on stage","provenance":"metadata","verified":true,"evidence_quote":"Sam Altman","text":"Sam Altman (Speaker holding a microphone on stage), named in the video's title or description","confidence":0.95,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","relation":"speaks"},{"id":"person_span_ca592ab4b276fa4a6f4774cb35c783c9","start_ms":112524,"end_ms":118124,"name":null,"description":"Host appearing in the bottom left corner while introducing the clip","provenance":null,"verified":false,"evidence_quote":null,"text":"Host appearing in the bottom left corner while introducing the clip (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_783952df25e8b8732dc9a0f518b7a7c3","start_ms":112524,"end_ms":125124,"name":"Sam Altman","description":"Sam Altman seated on stage holding a handheld microphone","provenance":"spoken","verified":true,"evidence_quote":"Sam Altman","text":"Sam Altman (Sam Altman seated on stage holding a handheld microphone), named in speech","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","relation":"speaks"},{"id":"person_span_b45aa62c7f601deb8b0f82fe32276192","start_ms":125224,"end_ms":135924,"name":"Sam Altman","description":"Sam Altman seated on stage wearing a grey sweatshirt speaking into a microphone","provenance":"on_screen_text","verified":true,"evidence_quote":"SAM ALTMAN","text":"Sam Altman (Sam Altman seated on stage wearing a grey sweatshirt speaking into a microphone), named on screen","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","relation":"speaks"},{"id":"person_span_8d0e53bdcf80c433226974b5b7ee5ff1","start_ms":136024,"end_ms":150024,"name":"Sam Altman","description":"Sam Altman seated on stage in a grey crewneck gesturing while speaking","provenance":"on_screen_text","verified":true,"evidence_quote":"SAM ALTMAN","text":"Sam Altman (Sam Altman seated on stage in a grey crewneck gesturing while speaking), named on screen","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","relation":"speaks"},{"id":"person_span_cb9e13c8c5f1c880f8de3d6f94f27480","start_ms":150083,"end_ms":161183,"name":null,"description":"Presenter wearing a light blue shirt with colorful patterned detailing","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter wearing a light blue shirt with colorful patterned detailing (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_97670cb7f7dcf20b6b1c5b1a8935a3cd","start_ms":161283,"end_ms":176183,"name":null,"description":"Presenter wearing a light blue shirt with colorful patterns","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter wearing a light blue shirt with colorful patterns (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_718e06ccaa27c6da7437d10a4221d1da","start_ms":176283,"end_ms":186783,"name":null,"description":"Presenter wearing a light blue shirt with patterned trim","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter wearing a light blue shirt with patterned trim (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_f7cc265377ff631e8eb2e5d8e935f484","start_ms":186883,"end_ms":200683,"name":null,"description":"Presenter wearing a light blue shirt with floral and patterned trim","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter wearing a light blue shirt with floral and patterned trim (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_3751ddb305cbd8886baae76f9ec3dc4f","start_ms":200765,"end_ms":212365,"name":null,"description":"Male host in a patterned collared shirt speaking to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host in a patterned collared shirt speaking to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_8f0fe46ff9c9a5b8cd105fad47b1e4bf","start_ms":212465,"end_ms":221465,"name":null,"description":"Male host in a patterned collared shirt speaking to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host in a patterned collared shirt speaking to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_979772043468014932d10f6cd06f18cb","start_ms":221565,"end_ms":236465,"name":null,"description":"Male host in a patterned collared shirt speaking to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host in a patterned collared shirt speaking to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_9d4485710838864440ab848db98d8b74","start_ms":236565,"end_ms":250065,"name":null,"description":"Male host in a patterned collared shirt speaking to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host in a patterned collared shirt speaking to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_5c5fbb8fc0c0721786e75c5ac483777c","start_ms":250129,"end_ms":263329,"name":null,"description":"Male host wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_65f068403447a7210a7e8e0a8015f0c3","start_ms":263429,"end_ms":273029,"name":null,"description":"Male host wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_294c2c80953987a779ff17efd07c1d44","start_ms":273129,"end_ms":287829,"name":null,"description":"Male host wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_1a5f73b0cc9e17069308063ca2d87378","start_ms":287929,"end_ms":302829,"name":null,"description":"Male host wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_7bfacfcd2146df8e8b23c067c68945a3","start_ms":302919,"end_ms":316819,"name":null,"description":"Male speaker in a patterned shirt talking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male speaker in a patterned shirt talking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_9bf54c10b3234b626049bbd8ce5b7c5c","start_ms":316919,"end_ms":327419,"name":null,"description":"Male speaker wearing a patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male speaker wearing a patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_b86db9dba5a4b7b398f7509366c2e584","start_ms":327519,"end_ms":338819,"name":null,"description":"Male speaker in an artistic collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male speaker in an artistic collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_7a9b28cce571363087a6028744591507","start_ms":338919,"end_ms":346919,"name":null,"description":"Male presenter gesturing with both hands","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter gesturing with both hands (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_27086a39fb6043c19bcbd6f6b5ee1173","start_ms":347000,"end_ms":361700,"name":null,"description":"Presenter wearing a colorful patterned collared shirt speaking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter wearing a colorful patterned collared shirt speaking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_3275784b5d4ae4fc75d31d97c947947e","start_ms":361800,"end_ms":375100,"name":null,"description":"Presenter wearing a colorful patterned collared shirt speaking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter wearing a colorful patterned collared shirt speaking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_faccab9a645e51fe865a7a4168b9e700","start_ms":375200,"end_ms":384200,"name":null,"description":"Presenter wearing a colorful patterned collared shirt speaking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter wearing a colorful patterned collared shirt speaking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_668c9c91e748b8f9b2e4803349b1f87e","start_ms":384300,"end_ms":398000,"name":null,"description":"Presenter wearing a colorful patterned collared shirt speaking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter wearing a colorful patterned collared shirt speaking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_8f5bb9aef31b0b556a90b16846216076","start_ms":398083,"end_ms":411083,"name":null,"description":"Presenter wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_f59438417c878e6f0bcbb2670ad5ae84","start_ms":411083,"end_ms":426083,"name":null,"description":"Presenter wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_d4448291a21192a011464f615b1f4983","start_ms":426083,"end_ms":441083,"name":null,"description":"Presenter wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_fe0fe4a687df080690eee6303a1dbf30","start_ms":441083,"end_ms":455683,"name":null,"description":"Presenter wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_6f7f9f42d14daf244bc55d5050600f8a","start_ms":455750,"end_ms":465250,"name":null,"description":"Male presenter with brown hair wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter with brown hair wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_9a86790b106dde577d68a364225355a0","start_ms":465350,"end_ms":473250,"name":null,"description":"Male presenter in a patterned shirt speaking enthusiastically","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter in a patterned shirt speaking enthusiastically (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_89ce94000935e77e8d69111f754254c4","start_ms":473350,"end_ms":486750,"name":null,"description":"Male presenter gesturing while talking about human history","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter gesturing while talking about human history (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_d4961d6476337c308a32ed57f8136ab1","start_ms":486850,"end_ms":497750,"name":null,"description":"Male presenter gesturing while explaining the narrator concept","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter gesturing while explaining the narrator concept (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_e2819f6c1327b88f3522212738afd208","start_ms":497750,"end_ms":508650,"name":null,"description":"Male presenter speaking with animated hand gestures in front of slides","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter speaking with animated hand gestures in front of slides (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_930884687af1fe7262cf42a559487a52","start_ms":508750,"end_ms":519950,"name":null,"description":"Male presenter speaking enthusiastically in a patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter speaking enthusiastically in a patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_ddb7d174cb42cbbeb0a8f2824383f6d4","start_ms":520050,"end_ms":530650,"name":null,"description":"Male presenter talking in front of an LLM overview graphic","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter talking in front of an LLM overview graphic (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_7faac77a6b9ea92db12c9fcd085af67d","start_ms":530750,"end_ms":545650,"name":null,"description":"Male presenter gesturing with both hands while explaining AI company talent shifts","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter gesturing with both hands while explaining AI company talent shifts (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_54ecc95ccf6ef2b27a08cf5df0e55414","start_ms":545678,"end_ms":557878,"name":null,"description":"Host wearing a patterned shirt speaking directly to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned shirt speaking directly to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_8fadfb8714ffafe5f726ba65f5364da6","start_ms":557978,"end_ms":572878,"name":null,"description":"Host wearing a patterned shirt speaking directly to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned shirt speaking directly to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_8c491c67241ca789bc25e8dd9a3dc7a9","start_ms":572978,"end_ms":587878,"name":null,"description":"Host wearing a patterned shirt speaking directly to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned shirt speaking directly to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_ade88712b8bb568364ab6a1b1bd0089d","start_ms":587978,"end_ms":600378,"name":null,"description":"Host wearing a patterned shirt speaking directly to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned shirt speaking directly to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_70bcf0f3dce319c8fecc18a4d542a147","start_ms":600432,"end_ms":610532,"name":null,"description":"Male host wearing a white patterned shirt speaking on camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host wearing a white patterned shirt speaking on camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_ca2d74fe2b9332d85bb5b164cc060ff7","start_ms":610632,"end_ms":625132,"name":null,"description":"Male host wearing a patterned shirt gesturing toward a spreadsheet slide","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host wearing a patterned shirt gesturing toward a spreadsheet slide (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_bf203449e0a0a505e766ae8ba68f1500","start_ms":625232,"end_ms":640132,"name":null,"description":"Male host in a patterned shirt presenting language model use cases","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host in a patterned shirt presenting language model use cases (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_5fa074e586906ff3a5302b79a9cc1e2b","start_ms":640232,"end_ms":655132,"name":null,"description":"Male host in a white patterned shirt explaining language translation applications","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host in a white patterned shirt explaining language translation applications (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_ef88520fb28d5f60db411fabd7a2e311","start_ms":655203,"end_ms":668303,"name":null,"description":"Presenter in a patterned white and colorful collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter in a patterned white and colorful collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_df0f8a53d19b92b8e456165724a5e4f2","start_ms":668403,"end_ms":678003,"name":null,"description":"Presenter in a patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter in a patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_2763c0f2c4c5a5028ffae6d228a34a13","start_ms":678103,"end_ms":686903,"name":null,"description":"Presenter in a patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter in a patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_fd195e6d46baedc1789e69a8b59ebdc4","start_ms":687003,"end_ms":697403,"name":null,"description":"Presenter in a patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Presenter in a patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_3b3423218ee7afd49b1a26c58f3e5eb8","start_ms":698833,"end_ms":713133,"name":null,"description":"Brunette woman in a reddish collared shirt sitting in an office","provenance":null,"verified":false,"evidence_quote":null,"text":"Brunette woman in a reddish collared shirt sitting in an office (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_1c0122da8fd1e26212fc16b424b34b45","start_ms":714833,"end_ms":725233,"name":null,"description":"Male host in a patterned shirt speaking on camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host in a patterned shirt speaking on camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_2aa653e34baba12c2aa7ad311e87a03d","start_ms":725333,"end_ms":737133,"name":null,"description":"Male host in a patterned shirt speaking on the left","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host in a patterned shirt speaking on the left (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_6ac05d6e12607cdee8e93d0ddc9ab986","start_ms":725333,"end_ms":737133,"name":"Ryan Reynolds","description":"Actor in glasses holding and reading a printed script","provenance":"spoken","verified":true,"evidence_quote":"Ryan Reynolds was seeing if ChatGPT was doing its job","text":"Ryan Reynolds (Actor in glasses holding and reading a printed script), named in speech","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_3e757939-5520-4ae9-967d-eea25453d4db","relation":"appears"},{"id":"person_span_150d6ea62de44c69ef28c8af3c6df03c","start_ms":737233,"end_ms":741833,"name":null,"description":"Male host talking briefly on the left side of the frame","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host talking briefly on the left side of the frame (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_be36163a975b8e21a148ba17339a4fa6","start_ms":737233,"end_ms":751333,"name":"Ryan Reynolds","description":"Actor in glasses and a grey button-up shirt reading from a sheet of paper","provenance":"spoken","verified":true,"evidence_quote":"Ryan Reynolds was seeing if ChatGPT was doing its job","text":"Ryan Reynolds (Actor in glasses and a grey button-up shirt reading from a sheet of paper), named in speech","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_3e757939-5520-4ae9-967d-eea25453d4db","relation":"appears"},{"id":"person_span_108eb906ba5d5b3f6306bebeda75ce2d","start_ms":762417,"end_ms":775817,"name":null,"description":"Host wearing a floral-patterned white collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a floral-patterned white collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_5f3f4f3ab11d627b8cd5921abcdc7b06","start_ms":775917,"end_ms":790717,"name":null,"description":"Host wearing a patterned shirt gesturing next to the graphic","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned shirt gesturing next to the graphic (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_abb9c1a163042cfa80a59bd2d681fae5","start_ms":790817,"end_ms":801217,"name":null,"description":"Host wearing a patterned button-down shirt speaking and gesturing","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned button-down shirt speaking and gesturing (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_4f2928a984169dc3b63b044c758d5044","start_ms":801275,"end_ms":809275,"name":null,"description":"male speaker in a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male speaker in a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_6aa9c4a94f55b372d9252c4bbdcd18c6","start_ms":809275,"end_ms":824175,"name":null,"description":"male presenter talking about AI image generators","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter talking about AI image generators (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_5bafa2cfc31dc7b48fed2296bb5833cb","start_ms":824275,"end_ms":834075,"name":null,"description":"male host talking in a studio setting","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host talking in a studio setting (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_b3483e6e1a7ca7633605d8610b218d6e","start_ms":834175,"end_ms":848575,"name":null,"description":"host talking in front of a digital slide","provenance":null,"verified":false,"evidence_quote":null,"text":"Host talking in front of a digital slide (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_db4d0ed2dc1860349e3f0569e4b0e73a","start_ms":848650,"end_ms":863650,"name":null,"description":"Host in a white patterned button-up shirt speaking to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a white patterned button-up shirt speaking to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_4d7f9870599864202ce8cf5a2a1c7244","start_ms":863650,"end_ms":878650,"name":null,"description":"Host in a patterned shirt speaking and gesturing","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a patterned shirt speaking and gesturing (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_645bc3e50643f22c4307c8084348d5a9","start_ms":878650,"end_ms":891450,"name":null,"description":"Host in a patterned shirt speaking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a patterned shirt speaking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_c070e0bce1e9b1f2377e2d4963a4caba","start_ms":891550,"end_ms":906250,"name":null,"description":"Host in a patterned shirt gesturing with hands while explaining","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a patterned shirt gesturing with hands while explaining (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_a435076caada97482efb32380d47b667","start_ms":906283,"end_ms":919383,"name":null,"description":"Host wearing a colorful patterned collared shirt speaking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a colorful patterned collared shirt speaking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_42b646f93757368535bd0152789d506e","start_ms":919483,"end_ms":922883,"name":null,"description":"Host wearing a patterned shirt speaking","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned shirt speaking (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_bfaa59abb4d7bd7ac8bf94b7f8a58bfd","start_ms":945083,"end_ms":950083,"name":null,"description":"Host wearing a patterned shirt talking in front of a presentation slide","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned shirt talking in front of a presentation slide (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_20e4b59a26ce3a92e6489e265763a844","start_ms":950083,"end_ms":963383,"name":null,"description":"man with brown hair wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Man with brown hair wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_f8d75bcfe4fec77c30166ebc086d35f1","start_ms":963483,"end_ms":973083,"name":null,"description":"man with brown hair wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Man with brown hair wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_45c79491a9ae6989dd06fc0c64934589","start_ms":973083,"end_ms":985083,"name":null,"description":"man with brown hair wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Man with brown hair wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_7b51ff0327cb0995938db493d0c93c65","start_ms":985183,"end_ms":999583,"name":null,"description":"man with brown hair wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Man with brown hair wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_0e4f3a8d08785ee846e5cb0325f2e2e2","start_ms":999645,"end_ms":1009345,"name":null,"description":"Host wearing a patterned short-sleeve shirt speaking to the audience","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned short-sleeve shirt speaking to the audience (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_593c84485bd65fdcefe61c9d68d2c977","start_ms":1009445,"end_ms":1021145,"name":null,"description":"Host wearing a patterned short-sleeve shirt speaking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned short-sleeve shirt speaking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_f7a889942bafea027f13fd837a82f745","start_ms":1021245,"end_ms":1034945,"name":null,"description":"Host wearing a patterned shirt gesturing slightly and speaking","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned shirt gesturing slightly and speaking (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_f72a8a7c9d0ef62b883b0ee35e60b7ef","start_ms":1035045,"end_ms":1047845,"name":null,"description":"Host wearing a patterned shirt speaking directly to the audience","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned shirt speaking directly to the audience (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_8b96b3da23f9f30d0c29014cb7112bb2","start_ms":1047882,"end_ms":1062882,"name":null,"description":"Male host wearing a white shirt with colorful patterned patches","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host wearing a white shirt with colorful patterned patches (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_8efb57f79ef6973a5ec3d8c56e5496e7","start_ms":1056882,"end_ms":1062882,"name":"Joe Rogan","description":"Podcast host mentioned and featured on screen in slide graphic","provenance":"spoken","verified":true,"evidence_quote":"Joe Rogan","text":"Joe Rogan (Podcast host mentioned and featured on screen in slide graphic), named in speech","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_c40b20c6-f989-41d3-8160-0c05014792db","relation":"mentioned"},{"id":"person_span_dbb741d26a52d93fc1be45501e965d0a","start_ms":1058382,"end_ms":1062882,"name":"Steve Jobs","description":"Apple co-founder mentioned and featured on screen in slide graphic","provenance":"spoken","verified":true,"evidence_quote":"Steve Jobs","text":"Steve Jobs (Apple co-founder mentioned and featured on screen in slide graphic), named in speech","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_d139e370-38e7-45c6-b937-d8cd1c341325","relation":"mentioned"},{"id":"person_span_207023bb265d4c3192a5cfc676e5d514","start_ms":1062882,"end_ms":1068182,"name":null,"description":"Male host wearing a white shirt with colorful patterned patches","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host wearing a white shirt with colorful patterned patches (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_ab1fafb67f40fee4620e98ad70f0e889","start_ms":1062882,"end_ms":1071882,"name":"Joe Rogan","description":"Podcast host mentioned and speaking in synthetic voiceover","provenance":"spoken","verified":true,"evidence_quote":"Joe Rogan","text":"Joe Rogan (Podcast host mentioned and speaking in synthetic voiceover), named in speech","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_c40b20c6-f989-41d3-8160-0c05014792db","relation":"mentioned"},{"id":"person_span_7230402f65b39a16e6d4555da05de5ee","start_ms":1062882,"end_ms":1074782,"name":"Steve Jobs","description":"Late Apple co-founder shown in a black-and-white portrait photo with hand on chin","provenance":"spoken","verified":true,"evidence_quote":"Steve Jobs","text":"Steve Jobs (Late Apple co-founder shown in a black-and-white portrait photo with hand on chin), named in speech","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_d139e370-38e7-45c6-b937-d8cd1c341325","relation":"mentioned"},{"id":"person_span_cadb096cdb2ce03597f1d16bc5d19eba","start_ms":1088782,"end_ms":1102182,"name":null,"description":"Male host wearing a white shirt with colorful patterned patches","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host wearing a white shirt with colorful patterned patches (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_553945a5485c918fdd4bb04c3bc27f23","start_ms":1098382,"end_ms":1100882,"name":"Eminem","description":"Rapper mentioned in reference to synthetic AI rhymes","provenance":"spoken","verified":true,"evidence_quote":"Eminem","text":"Eminem (Rapper mentioned in reference to synthetic AI rhymes), named in speech","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_0abfc4aa-c8f7-4e9c-babf-662a802b50ef","relation":"mentioned"},{"id":"person_span_50d1812f1dcad4112b68282f25804759","start_ms":1102208,"end_ms":1104708,"name":"Eminem","description":"Rapper referenced for AI voice cloning","provenance":"spoken","verified":true,"evidence_quote":"rap rhymes for Eminem","text":"Eminem (Rapper referenced for AI voice cloning), named in speech","confidence":0.95,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_0abfc4aa-c8f7-4e9c-babf-662a802b50ef","relation":"mentioned"},{"id":"person_span_39f1ba864d8d62bad2c80b0466cae882","start_ms":1102208,"end_ms":1108408,"name":null,"description":"Host in a patterned light blue shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a patterned light blue shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_82beafdfceb522d0b02460dfd870b7e7","start_ms":1102208,"end_ms":1114108,"name":"David Guetta","description":"DJ performing behind turntables on stage","provenance":"on_screen_text","verified":true,"evidence_quote":"DAVID GUETTA","text":"David Guetta (DJ performing behind turntables on stage), named on screen","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_a43ff92d-60d5-4549-9fff-300b4ee09823","relation":"appears"},{"id":"person_span_b3d2f738a2012d4bbff3dc1457f05a81","start_ms":1114208,"end_ms":1126908,"name":"David Guetta","description":"DJ in green shirt at concert and later in a denim jacket during an interview","provenance":"on_screen_text","verified":true,"evidence_quote":"DAVID GUETTA","text":"David Guetta (DJ in green shirt at concert and later in a denim jacket during an interview), named on screen","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_a43ff92d-60d5-4549-9fff-300b4ee09823","relation":"appears"},{"id":"person_span_e4ede41d997415fa7d5be50ed1920e1d","start_ms":1120608,"end_ms":1121708,"name":"Eminem","description":"Rapper referenced in the interview","provenance":"spoken","verified":true,"evidence_quote":"Eminem, bro","text":"Eminem (Rapper referenced in the interview), named in speech","confidence":0.95,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_0abfc4aa-c8f7-4e9c-babf-662a802b50ef","relation":"mentioned"},{"id":"person_span_5a6691ca96936cd84f8d36e308c018bd","start_ms":1127008,"end_ms":1134008,"name":"David Guetta","description":"DJ speaking during an interview in a dark room","provenance":"on_screen_text","verified":true,"evidence_quote":"DAVID GUETTA","text":"David Guetta (DJ speaking during an interview in a dark room), named on screen","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_a43ff92d-60d5-4549-9fff-300b4ee09823","relation":"appears"},{"id":"person_span_5fc25707e9665b9e6b0d524fc96aad9f","start_ms":1134108,"end_ms":1139408,"name":null,"description":"Host in a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_f34e88c64ec3e4e5e422f5b0d48feb23","start_ms":1139508,"end_ms":1153508,"name":null,"description":"Host in a patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_944d3c532714e5c945c10cf11192919c","start_ms":1153587,"end_ms":1164287,"name":null,"description":"Male presenter with short hair wearing a collared graphic shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter with short hair wearing a collared graphic shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_b23881e6420f152730cf5c8bf863b2a8","start_ms":1164387,"end_ms":1175587,"name":null,"description":"Male presenter with short hair wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter with short hair wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_e064d9012b4e5bb101a41cd936de4f62","start_ms":1175687,"end_ms":1190187,"name":null,"description":"Male presenter with short hair in a collared patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter with short hair in a collared patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_b71c1dc5365c87dbfb86fb2bda6cc8cd","start_ms":1190287,"end_ms":1201987,"name":null,"description":"Male presenter with short hair in a collared graphic shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter with short hair in a collared graphic shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_0e143d178b588e164c2c1425acca69a2","start_ms":1202072,"end_ms":1216972,"name":null,"description":"male host speaking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host speaking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_81b4b99c766887b2cb8db3932b89feea","start_ms":1217072,"end_ms":1221572,"name":null,"description":"man walking on street transformed into claymation character","provenance":null,"verified":false,"evidence_quote":null,"text":"Man walking on street transformed into claymation character (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_716997aed3c8d1d4acfee3ea79111cae","start_ms":1221972,"end_ms":1225572,"name":null,"description":"person walking in snowy mountain landscape","provenance":null,"verified":false,"evidence_quote":null,"text":"Person walking in snowy mountain landscape (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_d7763144a6cb9bc6ed9d1d888ff34d32","start_ms":1226672,"end_ms":1233572,"name":null,"description":"woman transformed into a bronze bust animation","provenance":null,"verified":false,"evidence_quote":null,"text":"Woman transformed into a bronze bust animation (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_b7ba1af20a99d3617e1b2869defdf0b1","start_ms":1235572,"end_ms":1239172,"name":null,"description":"male host speaking beside slide","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host speaking beside slide (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_42da8b954f422c82be37fab7b4906e64","start_ms":1240072,"end_ms":1247572,"name":null,"description":"digital animated female avatar with red hair","provenance":null,"verified":false,"evidence_quote":null,"text":"Digital animated female avatar with red hair (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_480af0ff122a449f054cafd99df317f5","start_ms":1263617,"end_ms":1275417,"name":null,"description":"Host wearing a patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_6e6c9cc142c2ec028eef1bbf423fdad7","start_ms":1289217,"end_ms":1303117,"name":null,"description":"Host in a patterned collared shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a patterned collared shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_d23760700ca67b54e24bc45c19dca62d","start_ms":1303167,"end_ms":1312867,"name":null,"description":"Male presenter wearing a patterned shirt speaking in the lower left corner overlay","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a patterned shirt speaking in the lower left corner overlay (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_a589604f012fc37c23241e95491b6fc1","start_ms":1312967,"end_ms":1326267,"name":null,"description":"Male presenter in patterned shirt talking in the corner inset","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter in patterned shirt talking in the corner inset (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_6f0def3d53c2866504ad80539802f7a1","start_ms":1326367,"end_ms":1341267,"name":null,"description":"Male speaker in a white and floral patterned shirt gesturing towards the slide","provenance":null,"verified":false,"evidence_quote":null,"text":"Male speaker in a white and floral patterned shirt gesturing towards the slide (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_48aa0c871871ba98127b457ffa728c2a","start_ms":1341367,"end_ms":1354767,"name":null,"description":"Male speaker in a patterned shirt positioned on the left side of the presentation screen","provenance":null,"verified":false,"evidence_quote":null,"text":"Male speaker in a patterned shirt positioned on the left side of the presentation screen (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_3a47d049df128955ef7d68ad500a4ace","start_ms":1354788,"end_ms":1367388,"name":null,"description":"Male host wearing a light collared shirt with colorful floral patterns","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host wearing a light collared shirt with colorful floral patterns (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_d9e0d62462699a0a11f14135a3f485c3","start_ms":1367488,"end_ms":1376988,"name":null,"description":"Male host wearing a light collared shirt with colorful floral patterns","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host wearing a light collared shirt with colorful floral patterns (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_fcc95201b6d96b94433bdb59ebd1ad56","start_ms":1377088,"end_ms":1390988,"name":null,"description":"Male host wearing a light collared shirt with colorful floral patterns","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host wearing a light collared shirt with colorful floral patterns (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_112c41ef458b437d1853950e965caff3","start_ms":1391088,"end_ms":1398988,"name":null,"description":"Male host wearing a light collared shirt with colorful floral patterns","provenance":null,"verified":false,"evidence_quote":null,"text":"Male host wearing a light collared shirt with colorful floral patterns (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_d91b985252b005df9dcb86fc2aca8884","start_ms":1399057,"end_ms":1407057,"name":null,"description":"Male speaker wearing a patterned collared shirt in the lower left corner","provenance":null,"verified":false,"evidence_quote":null,"text":"Male speaker wearing a patterned collared shirt in the lower left corner (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_8ac2280d64295ed1e9a21f55889c8a1f","start_ms":1407057,"end_ms":1417957,"name":null,"description":"Male presenter wearing a patterned white shirt talking at the bottom left","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a patterned white shirt talking at the bottom left (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_70eaac1b24983d307660fbbf4c6dd0b4","start_ms":1418057,"end_ms":1429057,"name":null,"description":"Male speaker positioned at the bottom left talking and gesturing","provenance":null,"verified":false,"evidence_quote":null,"text":"Male speaker positioned at the bottom left talking and gesturing (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_2453a065bddb3bab6403c14f24d1df47","start_ms":1429157,"end_ms":1444057,"name":null,"description":"Male speaker in lower left corner gesturing with open hands","provenance":null,"verified":false,"evidence_quote":null,"text":"Male speaker in lower left corner gesturing with open hands (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_7bc17a1acfd336c2d507bc6e1ef32786","start_ms":1444125,"end_ms":1459125,"name":null,"description":"Male presenter with short hair in a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter with short hair in a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_d4e3416f52c8c586ff66c3d71c6d320f","start_ms":1459125,"end_ms":1474125,"name":null,"description":"Male presenter with short hair in a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter with short hair in a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_c10b495bbd03fe2ceaba33fda12adf0b","start_ms":1474125,"end_ms":1485025,"name":null,"description":"Male presenter with short hair in a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter with short hair in a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_5791138bbf05020fdfd98400cfe60492","start_ms":1485125,"end_ms":1490825,"name":null,"description":"Male presenter with short hair in a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter with short hair in a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_ce3a09e69be6f41b5330599037f20806","start_ms":1525383,"end_ms":1538683,"name":null,"description":"Male presenter wearing a patterned collared shirt speaking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a patterned collared shirt speaking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_17fc3459c199d9a046f7ee9b09511a62","start_ms":1538773,"end_ms":1552473,"name":null,"description":"Male presenter wearing a patterned shirt in front of an AI presentation","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a patterned shirt in front of an AI presentation (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_c604a10ab9ea6b547c830487c6cecd71","start_ms":1552573,"end_ms":1565573,"name":null,"description":"Male presenter inset in the bottom left corner talking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter inset in the bottom left corner talking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_7bc7f781f76764c4a947922b216cd58c","start_ms":1565673,"end_ms":1577973,"name":null,"description":"Male presenter speaking in an inset frame on the bottom-left","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter speaking in an inset frame on the bottom-left (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_793c04b68855d11d10a297b50d2dd1ed","start_ms":1578073,"end_ms":1592973,"name":null,"description":"Male presenter speaking to camera while positioned inset before switching to presentation backdrop","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter speaking to camera while positioned inset before switching to presentation backdrop (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_05c22ba67a8f226bf9b2b4ad390780cb","start_ms":1592985,"end_ms":1607985,"name":null,"description":"Host wearing a white patterned short-sleeve shirt speaking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a white patterned short-sleeve shirt speaking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_cca80a20bea2b111569378284840d328","start_ms":1607985,"end_ms":1618185,"name":null,"description":"Host wearing a white patterned shirt talking to the viewer","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a white patterned shirt talking to the viewer (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_e14382faa81d21fdd3078313e8e89b3a","start_ms":1618285,"end_ms":1631685,"name":null,"description":"Host talking and gesturing in front of the slide display","provenance":null,"verified":false,"evidence_quote":null,"text":"Host talking and gesturing in front of the slide display (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_89af66ddc07c758323ca87cebf6c78c3","start_ms":1631785,"end_ms":1640685,"name":null,"description":"Host wearing a patterned shirt speaking with animated hand gestures","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned shirt speaking with animated hand gestures (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_b5e2ba2a905ef42322eaf90b9b2d3038","start_ms":1640729,"end_ms":1648729,"name":null,"description":"Host wearing a patterned shirt speaking on camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned shirt speaking on camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_cefaf94164eacf215edbc31f5d185c53","start_ms":1648729,"end_ms":1658929,"name":null,"description":"Host gesturing next to a presentation slide","provenance":null,"verified":false,"evidence_quote":null,"text":"Host gesturing next to a presentation slide (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_50f61bb80bf5207906e0018e0d214f80","start_ms":1659029,"end_ms":1673929,"name":null,"description":"Host talking beside a graphical slide","provenance":null,"verified":false,"evidence_quote":null,"text":"Host talking beside a graphical slide (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_3255d13f5e70f355ec2fd31cdaf8b4c5","start_ms":1674029,"end_ms":1682929,"name":null,"description":"Host gesturing towards an inference time comparison slide","provenance":null,"verified":false,"evidence_quote":null,"text":"Host gesturing towards an inference time comparison slide (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_8b69ab27ab94c16f113c98b175f41ec7","start_ms":1683000,"end_ms":1698000,"name":null,"description":"Male presenter wearing a patterned collared shirt talking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a patterned collared shirt talking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_981539919cf74396b949b8e612819280","start_ms":1698000,"end_ms":1709400,"name":null,"description":"Male presenter wearing a patterned collared shirt speaking to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a patterned collared shirt speaking to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_0d407d925ef0debd0adb96894a2886d4","start_ms":1709500,"end_ms":1717500,"name":null,"description":"Male presenter wearing a patterned collared shirt speaking to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a patterned collared shirt speaking to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_2388899c962e6383f32ec165c936fa2f","start_ms":1717600,"end_ms":1725500,"name":null,"description":"Male presenter wearing a patterned collared shirt speaking to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a patterned collared shirt speaking to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_ba68d1bb94dc09c787a25f61edc7a9c4","start_ms":1725523,"end_ms":1740523,"name":null,"description":"Male presenter wearing a patterned collared shirt speaking to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a patterned collared shirt speaking to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_ef20e2b99e79f354e2252ae83a0f14ef","start_ms":1740523,"end_ms":1751823,"name":null,"description":"Male presenter wearing a patterned collared shirt speaking to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a patterned collared shirt speaking to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_aebf161c8fd64219740360c0255d2381","start_ms":1745523,"end_ms":1751823,"name":"Elon Musk","description":"Public figure and CEO of Twitter mentioned in news headline and speech","provenance":"spoken","verified":true,"evidence_quote":"You can see Elon Musk finding out that OpenAI has turned from a nonprofit to a for-profit organization","text":"Elon Musk (Public figure and CEO of Twitter mentioned in news headline and speech), named in speech","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_1e21f75d-cb8e-4cd4-9c15-c19108075d5c","relation":"mentioned"},{"id":"person_span_0be5684a1c60ddaf3d2264a8fb61d972","start_ms":1751923,"end_ms":1763123,"name":null,"description":"Male presenter wearing a patterned collared shirt speaking to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a patterned collared shirt speaking to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_afd14c73fb63162221fdc4c5e2f14a46","start_ms":1763223,"end_ms":1777723,"name":null,"description":"Male presenter wearing a patterned collared shirt speaking to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter wearing a patterned collared shirt speaking to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_60277a7c20bb3ec11f32479fe2f9b1de","start_ms":1777741,"end_ms":1791241,"name":null,"description":"Male speaker wearing a patterned collared shirt talking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male speaker wearing a patterned collared shirt talking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_a694018c02d0334305aee06e65ec3b6e","start_ms":1791341,"end_ms":1804541,"name":null,"description":"Male speaker wearing a patterned collared shirt presenting to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male speaker wearing a patterned collared shirt presenting to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_7f175684b5c24970fc70737308e598bd","start_ms":1804641,"end_ms":1816241,"name":null,"description":"Male speaker in a patterned shirt speaking directly to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Male speaker in a patterned shirt speaking directly to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_068c3316794d76349cf17e9ef5aab5ab","start_ms":1816341,"end_ms":1831241,"name":null,"description":"Male speaker in a patterned shirt gesturing with both hands while presenting","provenance":null,"verified":false,"evidence_quote":null,"text":"Male speaker in a patterned shirt gesturing with both hands while presenting (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_ed0388f424768ac72f29b2278d26cd81","start_ms":1831325,"end_ms":1843125,"name":null,"description":"Host wearing a patterned button-down shirt gesturing as he presents","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned button-down shirt gesturing as he presents (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_d9b4a752a8da0a3055d6a799b162852a","start_ms":1843225,"end_ms":1851125,"name":null,"description":"Host wearing a patterned shirt presenting to camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned shirt presenting to camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_e80ae1d9d6493b2c4bcd0b4cc65496ff","start_ms":1849125,"end_ms":1851125,"name":"Elon Musk","description":"Public figure shown in a displayed tweet profile","provenance":"on_screen_text","verified":true,"evidence_quote":"Elon Musk","text":"Elon Musk (Public figure shown in a displayed tweet profile), named on screen","confidence":0.98,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_1e21f75d-cb8e-4cd4-9c15-c19108075d5c","relation":"mentioned"},{"id":"person_span_53e398ab3992946c4f118c14711870c2","start_ms":1851225,"end_ms":1861525,"name":"Elon Musk","description":"Tech executive featured in a cited tweet and discussed by host","provenance":"spoken","verified":true,"evidence_quote":"Elon Musk","text":"Elon Musk (Tech executive featured in a cited tweet and discussed by host), named in speech","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_1e21f75d-cb8e-4cd4-9c15-c19108075d5c","relation":"mentioned"},{"id":"person_span_a4266dc2307ccccbe9c53bb90bd4952d","start_ms":1851225,"end_ms":1861525,"name":null,"description":"Host wearing a patterned shirt presenting commentary","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a patterned shirt presenting commentary (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_dee3c6b39ecc7d5ce8a662789cd39b80","start_ms":1861625,"end_ms":1869825,"name":"Elon Musk","description":"Subject of the on-screen tweet card","provenance":"on_screen_text","verified":true,"evidence_quote":"Elon Musk","text":"Elon Musk (Subject of the on-screen tweet card), named on screen","confidence":0.98,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_1e21f75d-cb8e-4cd4-9c15-c19108075d5c","relation":"mentioned"},{"id":"person_span_ef7b5d4260141e4a39e360e1dda8afa2","start_ms":1861625,"end_ms":1869825,"name":null,"description":"Host speaking directly to the audience","provenance":null,"verified":false,"evidence_quote":null,"text":"Host speaking directly to the audience (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_e0251121edd9c3575c7452acb09c47d6","start_ms":1869875,"end_ms":1883175,"name":null,"description":"Male presenter with short brown hair wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter with short brown hair wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_ebcc46fdcfd27a4edf7227e186e587ee","start_ms":1883275,"end_ms":1893275,"name":null,"description":"Male presenter with short brown hair wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter with short brown hair wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_81331794ddeba94807b02eac067ede76","start_ms":1893375,"end_ms":1907575,"name":null,"description":"Male presenter with short brown hair wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter with short brown hair wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_bfbf11e742303ac7a9d8e4184b5d25bd","start_ms":1907675,"end_ms":1921975,"name":null,"description":"Male presenter with short brown hair wearing a patterned shirt","provenance":null,"verified":false,"evidence_quote":null,"text":"Male presenter with short brown hair wearing a patterned shirt (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_3eb629dc25b3bcba264d44e0afd8fefe","start_ms":1921998,"end_ms":1927098,"name":null,"description":"Host in a patterned collared shirt speaking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a patterned collared shirt speaking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_0e5bce4131c37233add011940ea5f19b","start_ms":1921998,"end_ms":1936998,"name":"Sam Altman","description":"Speaker wearing a grey hoodie and white t-shirt on a stage","provenance":"on_screen_text","verified":true,"evidence_quote":"SAM ALTMAN","text":"Sam Altman (Speaker wearing a grey hoodie and white t-shirt on a stage), named on screen","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","relation":"speaks"},{"id":"person_span_c5f14f4b8883fbeea1206f2fe36e5015","start_ms":1936998,"end_ms":1948898,"name":"Sam Altman","description":"Speaker in a grey hoodie gesturing while seated on stage","provenance":"on_screen_text","verified":true,"evidence_quote":"SAM ALTMAN","text":"Sam Altman (Speaker in a grey hoodie gesturing while seated on stage), named on screen","confidence":0.99,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a","entity_id":"ent_7ec30179-e819-470f-a3e4-9917fb57fca9","relation":"speaks"},{"id":"person_span_884e7e9256bb7db4eff8996c74504b9d","start_ms":1944098,"end_ms":1948898,"name":null,"description":"Interviewer seated on stage opposite Sam Altman","provenance":null,"verified":false,"evidence_quote":null,"text":"Interviewer seated on stage opposite Sam Altman (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_17839182532636d27301e1f510c816f5","start_ms":1948998,"end_ms":1962898,"name":null,"description":"Host in a patterned shirt speaking and gesturing in the lower left","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a patterned shirt speaking and gesturing in the lower left (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_a6c8b4ab5ed6ac46ff6565fcad5eae96","start_ms":1962998,"end_ms":1973198,"name":null,"description":"Host in a patterned shirt speaking and gesturing","provenance":null,"verified":false,"evidence_quote":null,"text":"Host in a patterned shirt speaking and gesturing (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"},{"id":"person_span_9ff0972bb5549774583359615afaef73","start_ms":1973297,"end_ms":1981297,"name":null,"description":"Host wearing a colorful patterned shirt, gesturing and speaking to the camera","provenance":null,"verified":false,"evidence_quote":null,"text":"Host wearing a colorful patterned shirt, gesturing and speaking to the camera (unnamed)","confidence":0.4868654966809702,"method_version":"window-people-v1+ver_2c7e57c8-4d55-4054-9999-56ecf5309b8a"}],"products":[{"id":"vprod_11909d126e93eef94cae63a7a87c6798","relation":"shown","brand":null,"name":"Synthesia","kind":"online_service","url":"https://bit.ly/synthesialinks","url_host":"bit.ly","sku":null,"link_verified":true,"position":0,"method_version":"description-products-v5","spans":[{"id":"product_span_606bf3bb880bf94041ffec84740a0aef","relation":"mentioned","evidence":["description"],"evidence_quote":null,"first_seen_ms":0,"end_ms":9542,"confidence":0.9,"text":"Synthesia, listed in the video description","method_version":"window-products-v10"},{"id":"product_span_a4eb45801fefd630a61a3020855755db","relation":"shown","evidence":["spoken","on_screen","description"],"evidence_quote":"Synthesia","first_seen_ms":293729,"end_ms":316832,"confidence":0.99,"text":"Synthesia, named in speech, shown on screen and listed in the video description","method_version":"window-products-v10"},{"id":"product_span_263b34642a8d2881b3e06b8cabebd805","relation":"shown","evidence":["spoken","on_screen","description"],"evidence_quote":"Synthesia","first_seen_ms":1134108,"end_ms":1175617,"confidence":0.95,"text":"Synthesia, named in speech, shown on screen and listed in the video description","method_version":"window-products-v10"}]},{"id":"vprod_c01f1ea28de4631ec6e730d589a1e08f","relation":"shown","brand":null,"name":"Descript","kind":"software","url":"https://bit.ly/descriptlinks","url_host":"bit.ly","sku":null,"link_verified":true,"position":1,"method_version":"description-products-v5","spans":[{"id":"product_span_b8274626e8d0bb9e0715248603a42e55","relation":"shown","evidence":["spoken","on_screen","description"],"evidence_quote":"using Descript","first_seen_ms":293729,"end_ms":316832,"confidence":0.99,"text":"Descript, named in speech, shown on screen and listed in the video description","method_version":"window-products-v10"}]},{"id":"vprod_cf9e80f651c52645b7a5a5042e5e41f7","relation":"shown","brand":null,"name":"Jasper AI","kind":"software","url":"https://bit.ly/jasperailinks","url_host":"bit.ly","sku":null,"link_verified":true,"position":2,"method_version":"description-products-v5","spans":[{"id":"product_span_b51554a5188b9d926ef44d73c9b7910b","relation":"shown","evidence":["spoken","on_screen","description"],"evidence_quote":"Jasper is here to help","first_seen_ms":688403,"end_ms":713167,"confidence":0.99,"text":"Jasper, named in speech, shown on screen and listed in the video description","method_version":"window-products-v10"}]},{"id":"vprod_2d5ecb003ae16a479231bab9c11d24f3","relation":"shown","brand":null,"name":"Dall-E 2","kind":"software","url":"https://bit.ly/dalle2links","url_host":"bit.ly","sku":null,"link_verified":true,"position":3,"method_version":"description-products-v5","spans":[{"id":"product_span_426482c35454ca0beee3d1d4534fa2a0","relation":"shown","evidence":["spoken","on_screen","description"],"evidence_quote":"DALL-E 2","first_seen_ms":809275,"end_ms":824198,"confidence":0.99,"text":"DALL-E 2, named in speech, shown on screen and listed in the video description","method_version":"window-products-v10"},{"id":"product_span_dfa235014368b104955217b36870c9be","relation":"shown","evidence":["spoken","on_screen","description"],"evidence_quote":"DALL-E 2","first_seen_ms":922983,"end_ms":950083,"confidence":0.99,"text":"DALL-E 2, named in speech, shown on screen and listed in the video description","method_version":"window-products-v10"},{"id":"product_span_7d6130c9fe52c5a64579116f69c82973","relation":"shown","evidence":["spoken","on_screen","description"],"evidence_quote":"I can then use a tool like DALL-E or Stable Diffusion or anything else to create that content.","first_seen_ms":1791341,"end_ms":1843223,"confidence":0.99,"text":"DALL-E 2, named in speech, shown on screen and listed in the video description","method_version":"window-products-v10"}]},{"id":"vprod_5e741af1a7486bcc562eeb162e0f03f4","relation":"shown","brand":null,"name":"Galileo","kind":"software","url":"https://bit.ly/usegalileolinks","url_host":"bit.ly","sku":null,"link_verified":true,"position":4,"method_version":"description-products-v5","spans":[{"id":"product_span_33fe27c5cdf8ea7a73b0affefac132df","relation":"shown","evidence":["spoken","on_screen","description"],"evidence_quote":"Using, using a tool called Galileo","first_seen_ms":950083,"end_ms":973083,"confidence":0.95,"text":"Galileo AI, named in speech, shown on screen and listed in the video description","method_version":"window-products-v10"}]},{"id":"vprod_c95476123202b7acc28db8afde8d714d","relation":"shown","brand":null,"name":"RunwayML","kind":"software","url":"https://bit.ly/runwaymllinks","url_host":"bit.ly","sku":null,"link_verified":true,"position":5,"method_version":"description-products-v5","spans":[{"id":"product_span_34fe4c1888a8d92b08391ce9b857692b","relation":"shown","evidence":["on_screen","description"],"evidence_quote":null,"first_seen_ms":1134108,"end_ms":1153587,"confidence":0.9,"text":"Runway, shown on screen and listed in the video description","method_version":"window-products-v10"},{"id":"product_span_0af28be2a583ec8feced269098fc2e65","relation":"shown","evidence":["spoken","on_screen","description"],"evidence_quote":"Runway ML and their new model to stylize any content","first_seen_ms":1212572,"end_ms":1226625,"confidence":0.99,"text":"Runway, named in speech, shown on screen and listed in the video description","method_version":"window-products-v10"}]},{"id":"vprod_1c934da47631a16d256fda81f17d2433","relation":"shown","brand":null,"name":"D-ID","kind":"software","url":"https://bit.ly/didlinks","url_host":"bit.ly","sku":null,"link_verified":true,"position":6,"method_version":"description-products-v5","spans":[{"id":"product_span_a699be73c9edbd9e60432feb1d1748f7","relation":"shown","evidence":["spoken","on_screen","description"],"evidence_quote":"D-ID","first_seen_ms":1134108,"end_ms":1175617,"confidence":0.95,"text":"D-ID Creative Reality Studio, named in speech, shown on screen and listed in the video description","method_version":"window-products-v10"},{"id":"product_span_22e031e23278b19ed9719c46d3ea4425","relation":"shown","evidence":["spoken","on_screen","description"],"evidence_quote":"D-ID platform","first_seen_ms":1238072,"end_ms":1263542,"confidence":0.99,"text":"D-ID Creative Reality Studio, named in speech, shown on screen and listed in the video description","method_version":"window-products-v10"}]},{"id":"vprod_afd3c25d2dc24a104acbde1c46398e25","relation":"shown","brand":null,"name":"Krikey AI","kind":"software","url":"https://bit.ly/krikeyailinks","url_host":"bit.ly","sku":null,"link_verified":true,"position":7,"method_version":"description-products-v5","spans":[{"id":"product_span_9596025f4fec9b2728209494ed53240a","relation":"shown","evidence":["on_screen","description"],"evidence_quote":null,"first_seen_ms":1134108,"end_ms":1164305,"confidence":0.95,"text":"Krikey AI, shown on screen and listed in the video description","method_version":"window-products-v10"},{"id":"product_span_35a1f0aca0753b54d3c0619d65a2937b","relation":"shown","evidence":["on_screen","description"],"evidence_quote":null,"first_seen_ms":1263617,"end_ms":1289208,"confidence":0.95,"text":"Krikey AI, shown on screen and listed in the video description","method_version":"window-products-v10"}]},{"id":"vprod_15028701e5d08bef0739f03eef64041b","relation":"shown","brand":"Replit","name":"Replit Ghostwriter","kind":"software","url":"https://bit.ly/replitghostwriterlinks","url_host":"bit.ly","sku":null,"link_verified":true,"position":8,"method_version":"description-products-v5","spans":[{"id":"product_span_3e7c17914e507a635d91f0e514f44239","relation":"shown","evidence":["on_screen","description"],"evidence_quote":null,"first_seen_ms":257429,"end_ms":263359,"confidence":0.95,"text":"Replit Ghostwriter, shown on screen and listed in the video description","method_version":"window-products-v10"},{"id":"product_span_3d9a5ad20910ec5f6c6795b3b4a253b2","relation":"shown","evidence":["spoken","on_screen","description"],"evidence_quote":"Replit, which built Ghostwriter","first_seen_ms":1326367,"end_ms":1341350,"confidence":0.98,"text":"Replit Ghostwriter, named in speech, shown on screen and listed in the video description","method_version":"window-products-v10"},{"id":"product_span_f597a250db5963ff3a6cc1d9bb282a3f","relation":"shown","evidence":["on_screen","description"],"evidence_quote":null,"first_seen_ms":1368288,"end_ms":1391042,"confidence":0.98,"text":"Replit Ghostwriter, shown on screen and listed in the video description","method_version":"window-products-v10"}]},{"id":"vprod_45c80d45a25dd40d5e83f51f493129f6","relation":"shown","brand":"Github","name":"Github CoPilot","kind":"software","url":"https://bit.ly/githubcopilotlinks","url_host":"bit.ly","sku":null,"link_verified":true,"position":9,"method_version":"description-products-v5","spans":[{"id":"product_span_3970925f37ad30c0271267a7b6f4c031","relation":"shown","evidence":["spoken","on_screen","description"],"evidence_quote":"Copilot by GitHub","first_seen_ms":1326367,"end_ms":1341350,"confidence":0.98,"text":"GitHub Copilot, named in speech, shown on screen and listed in the video description","method_version":"window-products-v10"},{"id":"product_span_85d8308e2d05318c555a94e264922e96","relation":"shown","evidence":["on_screen","description"],"evidence_quote":null,"first_seen_ms":1391988,"end_ms":1407057,"confidence":0.99,"text":"GitHub Copilot, shown on screen and listed in the video description","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"OpenAI","name":"ChatGPT","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_d55682725a694f4f98be1bd72037c940","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"And you can see it in the success of ChatGPT","first_seen_ms":50067,"end_ms":59936,"confidence":0.99,"text":"ChatGPT, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_5b9605366bf57027369167e4c41d0859","relation":"mentioned","evidence":["spoken"],"evidence_quote":"ChatGPT","first_seen_ms":120924,"end_ms":125198,"confidence":0.99,"text":"ChatGPT, named in speech","method_version":"window-products-v10"},{"id":"product_span_1af3643cf5980dbec50d7f4abd8ecfd7","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"with slight tweaks that were made to ChatGPT","first_seen_ms":167783,"end_ms":186833,"confidence":1,"text":"ChatGPT, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_4bf85fcf22ed12872a4d364099695717","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"like ChatGPT","first_seen_ms":212465,"end_ms":250129,"confidence":0.99,"text":"ChatGPT, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_235e5b5c9ca69e6b2d142c5953960a2f","relation":"mentioned","evidence":["spoken"],"evidence_quote":"Or you can just use the ChatGPT interface","first_seen_ms":316919,"end_ms":327458,"confidence":0.99,"text":"ChatGPT, named in speech","method_version":"window-products-v10"},{"id":"product_span_5cb3bcb2750cb546e4f5c3391bb659da","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"And that's what made ChatGPT a success.","first_seen_ms":455750,"end_ms":473255,"confidence":0.99,"text":"ChatGPT, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_139110fafe115c7089cb015f6ab71f4c","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Obviously ChatGPT, we've talked about a lot.","first_seen_ms":520050,"end_ms":557917,"confidence":0.98,"text":"ChatGPT, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_be20f3541ee7c6b2b22fafc66d083d73","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"OpenAI ChatGPT is providing","first_seen_ms":601132,"end_ms":610542,"confidence":0.98,"text":"ChatGPT, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_25d1289412593c226448ce86d4955126","relation":"mentioned","evidence":["spoken"],"evidence_quote":"Ryan Reynolds was seeing if ChatGPT was doing its job","first_seen_ms":725333,"end_ms":737167,"confidence":0.99,"text":"ChatGPT, named in speech","method_version":"window-products-v10"},{"id":"product_span_9017ea350f454eece300a1ee1d2e1c5f","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"used ChatGPT to come up with rap rhymes","first_seen_ms":1088782,"end_ms":1114141,"confidence":0.99,"text":"ChatGPT, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_d8183095ee3f5009cd752c51cb78c1fe","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"ChatGPT","first_seen_ms":1599585,"end_ms":1717523,"confidence":0.98,"text":"ChatGPT, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_084458ff26e0f663c99f811d3a8f42f0","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"ChatGPT has done billions of pages for my inspiration.","first_seen_ms":1791341,"end_ms":1804583,"confidence":0.99,"text":"ChatGPT, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_0f7aafd1b1b34669efd59462c3ee36ad","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":1831325,"end_ms":1843223,"confidence":0.95,"text":"ChatGPT, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Meta","name":"Facebook","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_4d3e93ebac8bbbe75cae0088dabeb11b","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"compared to Facebook taking 4 years","first_seen_ms":50067,"end_ms":59936,"confidence":0.98,"text":"Facebook, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"ByteDance","name":"TikTok","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_38c39994d709cc4f6d7f4760983b5704","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":50067,"end_ms":59936,"confidence":0.95,"text":"TikTok, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Niantic","name":"Pokémon GO","kind":"game","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_ef3f47dd5eff352b328bf57f5d365a70","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Pokémon Go actually reached 100 million users in just less than a month","first_seen_ms":59967,"end_ms":74091,"confidence":0.99,"text":"Pokémon GO, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Google","name":"Google Trends","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_dd9abd286560cf05ddd4ebe2aecb4e50","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"That's seen on Google Trends.","first_seen_ms":69267,"end_ms":74091,"confidence":0.98,"text":"Google Trends, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"OpenAI","name":"GPT-3","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_ef044660e84fa9c3d256ce06a2df5328","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"GPT-3 playground","first_seen_ms":167783,"end_ms":212426,"confidence":1,"text":"GPT-3, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_cb8e1436bdd62c4f47910d6c6dc1218a","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":302919,"end_ms":316832,"confidence":0.95,"text":"GPT-3, shown on screen","method_version":"window-products-v10"},{"id":"product_span_7d6868f0a6621970e793be38e822d2f8","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":430483,"end_ms":455750,"confidence":0.9,"text":"GPT-3, shown on screen","method_version":"window-products-v10"},{"id":"product_span_3bc52075eff2aadef6de5b161a01694b","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"GPT-3 model","first_seen_ms":593878,"end_ms":610542,"confidence":0.99,"text":"GPT-3, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_7fbf2e3b57cbd1ea09af0938b55b87fe","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"GPT-3","first_seen_ms":1251072,"end_ms":1251417,"confidence":0.97,"text":"GPT-3, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"OpenAI","name":"InstructGPT","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_ca51069b02e8d10c40a5006882e47614","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":198083,"end_ms":212426,"confidence":0.95,"text":"InstructGPT, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Microsoft","name":"Bing Chat","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_74662b0d5d1beecae7e1b3e884f93cc3","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"in the Bing Chat","first_seen_ms":221565,"end_ms":250129,"confidence":0.98,"text":"Bing Chat, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_50cef8562e208216d42c38e8ae3467ef","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"we have Bing Chat that has evolved on top of that","first_seen_ms":525750,"end_ms":557917,"confidence":0.98,"text":"Bing Chat, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_49b4f98928a600fb4df4f6a105190b5c","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":1540573,"end_ms":1552500,"confidence":0.95,"text":"Bing Chat, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Google","name":"Google Bard","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_b68f213004b7c149abee31a5911bad61","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Google has announced Bard, their own chat version of search","first_seen_ms":530750,"end_ms":557917,"confidence":0.98,"text":"Google Bard, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_ff6caf2b6b1083e9e5c57a27cfbfa7fd","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Bard","first_seen_ms":1599585,"end_ms":1618271,"confidence":0.95,"text":"Google Bard, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Anthropic","name":"Claude","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_d9a069564ad08c533d9fa66e9c23596e","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Anthropic, which has developed a model called Claude.","first_seen_ms":533750,"end_ms":557917,"confidence":0.98,"text":"Claude, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Baidu","name":"Ernie Bot","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_762f8f37993b501b16e163b216071351","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":545678,"end_ms":557917,"confidence":0.95,"text":"Ernie Bot, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"mentioned","brand":"Substack","name":"Substack","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_72d74e19ff1ac9901a2c09a2e270b898","relation":"mentioned","evidence":["spoken"],"evidence_quote":"Substack","first_seen_ms":567178,"end_ms":572917,"confidence":0.9,"text":"Substack, named in speech","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Reddit","name":"Reddit","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_a87239b56f7e5b13a3a8388952882cc7","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Reddit","first_seen_ms":567478,"end_ms":572917,"confidence":0.95,"text":"Reddit, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Twitter","name":"Twitter","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_3d3332d347ba717380a6035c03577dec","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Twitter","first_seen_ms":567878,"end_ms":572917,"confidence":0.95,"text":"Twitter, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_db40b743d9aad9a1128e5e5ec8dcffbe","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"they've closed the Twitter access, Twitter data access for OpenAI.","first_seen_ms":1747523,"end_ms":1751863,"confidence":0.95,"text":"Twitter, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Wikimedia Foundation","name":"Wikipedia","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_750f0799bc7d4fd8a6d3260531abe23e","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Wikipedia","first_seen_ms":568278,"end_ms":572917,"confidence":0.95,"text":"Wikipedia, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Microsoft","name":"Microsoft Azure","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_5edeb7da39013ec69c2432c935a25051","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":577578,"end_ms":610542,"confidence":0.95,"text":"Microsoft Azure, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"NVIDIA","name":"NVIDIA A100","kind":"physical_product","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_637a8f9e3b81b6a7e9ceda182210cb9b","relation":"mentioned","evidence":["spoken"],"evidence_quote":"NVIDIA's most powerful A100 chips","first_seen_ms":585778,"end_ms":587917,"confidence":0.95,"text":"NVIDIA A100, named in speech","method_version":"window-products-v10"},{"id":"product_span_c1250503fdb77fe095ad8bb51d99c1f7","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"NVIDIA A100 chip","first_seen_ms":860150,"end_ms":863650,"confidence":0.95,"text":"NVIDIA A100, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"mentioned","brand":"Google","name":"Google Translate","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_ef122759aaf301c25e460c1af2fd876d","relation":"mentioned","evidence":["spoken"],"evidence_quote":"Google Translate","first_seen_ms":647632,"end_ms":668367,"confidence":0.98,"text":"Google Translate, named in speech","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Microsoft","name":"Microsoft Bing","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_8d08d56961cd20b5b41dd80f1df2c3e6","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Bing","first_seen_ms":660003,"end_ms":668367,"confidence":0.99,"text":"Microsoft Bing, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_f1af676725561120add4e1e6fb35338f","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Bing recently launched Bing Chat.","first_seen_ms":762417,"end_ms":775890,"confidence":0.98,"text":"Microsoft Bing, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_573980e4a92b816d2dfdee568f7d472d","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Bing, who doesn't even want to talk about it","first_seen_ms":1584973,"end_ms":1631750,"confidence":0.95,"text":"Microsoft Bing, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Midjourney","name":"Midjourney","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_441740ba0c1584d1f2f401cd27d5cc48","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Midjourney","first_seen_ms":809275,"end_ms":824198,"confidence":0.99,"text":"Midjourney, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Stability AI","name":"Stable Diffusion","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_fd092711ae99769e3d8c9191b3c50939","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Stable Diffusion","first_seen_ms":809275,"end_ms":848650,"confidence":0.99,"text":"Stable Diffusion, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_f6206c0e3920fd890eafd5b96894f38d","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Stable Diffusion","first_seen_ms":1250572,"end_ms":1251417,"confidence":0.97,"text":"Stable Diffusion, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_72ddfb3b3fa34091111d7b76f5980bee","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Stable Diffusion has managed to decrease","first_seen_ms":1690800,"end_ms":1709466,"confidence":0.98,"text":"Stable Diffusion, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_7f1dd13afa6087d383a7f40ecff564f5","relation":"mentioned","evidence":["spoken"],"evidence_quote":"I can then use a tool like DALL-E or Stable Diffusion or anything else to create that content.","first_seen_ms":1807741,"end_ms":1816325,"confidence":0.99,"text":"Stable Diffusion, named in speech","method_version":"window-products-v10"},{"id":"product_span_9483b3b7ce7c2364f2b11852c59ea6a4","relation":"mentioned","evidence":["spoken"],"evidence_quote":"Stable Diffusion","first_seen_ms":1910375,"end_ms":1921998,"confidence":0.98,"text":"Stable Diffusion, named in speech","method_version":"window-products-v10"}]},{"id":null,"relation":"mentioned","brand":"Discord","name":"Discord","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_af16b7cbbe298766b80ed53bb6b04cf5","relation":"mentioned","evidence":["spoken"],"evidence_quote":"Discord","first_seen_ms":821775,"end_ms":824198,"confidence":0.95,"text":"Discord, named in speech","method_version":"window-products-v10"}]},{"id":null,"relation":"mentioned","brand":"Apple","name":"Apple iPhone","kind":"physical_product","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_ae54acaaa948ef57266b9eab184b7416","relation":"mentioned","evidence":["spoken"],"evidence_quote":"right on your iPhone","first_seen_ms":874150,"end_ms":891549,"confidence":0.95,"text":"Apple iPhone, named in speech","method_version":"window-products-v10"}]},{"id":null,"relation":"mentioned","brand":"Apple","name":"Apple Mac","kind":"physical_product","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_ba482dd318ebedeebd648171c7d9fa1a","relation":"mentioned","evidence":["spoken"],"evidence_quote":"right on your Mac","first_seen_ms":875650,"end_ms":891549,"confidence":0.95,"text":"Apple Mac, named in speech","method_version":"window-products-v10"},{"id":"product_span_c8f4dbf620fff5bf9c43309879a89d12","relation":"mentioned","evidence":["spoken"],"evidence_quote":"I've been a fan of yours and Macintosh since the 1980s.","first_seen_ms":1085082,"end_ms":1102208,"confidence":0.99,"text":"Apple Mac, named in speech","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":null,"name":"DiffusionBee","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_ff09ed325d85b9ddaedb8f2d128d0fe3","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":881650,"end_ms":891549,"confidence":0.9,"text":"DiffusionBee, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":null,"name":"Draw Things","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_63e3366fd285708f1d231702a31ec018","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":881650,"end_ms":891549,"confidence":0.9,"text":"Draw Things, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"mentioned","brand":null,"name":"Lensa AI","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_1c44b40799bc973c83d45787d4c2011b","relation":"mentioned","evidence":["spoken"],"evidence_quote":"Lensa AI","first_seen_ms":893150,"end_ms":906283,"confidence":0.95,"text":"Lensa AI, named in speech","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Figma","name":"Figma","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_ee5c2dde20ddd498abc4d18684d103b0","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"you can take it out, for example, to Figma","first_seen_ms":975583,"end_ms":985160,"confidence":0.95,"text":"Figma, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_65a71101794f81957bb0f9925d2ad189","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"I will then use, you know, Canva and Figma as a tool to create it.","first_seen_ms":1773123,"end_ms":1777741,"confidence":0.99,"text":"Figma, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Amazon","name":"Amazon Polly","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_f2c4477f60fca91209b857e5decc2f69","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Amazon has built Polly","first_seen_ms":985183,"end_ms":999645,"confidence":0.95,"text":"Amazon Polly, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Google","name":"AudioLM","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_739b85d2dcdb1c78c7dbdf44513f2451","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"models like AudioLM","first_seen_ms":985183,"end_ms":999645,"confidence":0.95,"text":"AudioLM, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"OpenAI","name":"MuseNet","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_3e1a682da2f42f11bc73393ab9a00b5e","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":985183,"end_ms":999645,"confidence":0.9,"text":"MuseNet, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Google","name":"MusicLM","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_f9bfdcf7d52feed2e36fc0de6f063085","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"MusicLM to generate music from text","first_seen_ms":985183,"end_ms":1009435,"confidence":0.98,"text":"MusicLM, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Google","name":"SingSong","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_4737461f17544a1a882a7e7b53a39d05","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Or SingSong, which you can use audio to audio","first_seen_ms":985183,"end_ms":1021189,"confidence":0.98,"text":"SingSong, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Speechify","name":"Speechify","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_d73555c2e02e70dc8681ce9bff9fbedc","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"the text-to-speech library, Speechify","first_seen_ms":985183,"end_ms":1009435,"confidence":0.98,"text":"Speechify, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Uberduck","name":"Uberduck","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_ed8fe2892f86c6c684604bd461d0cbab","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"as well as one of the largest open source movements, Uberduck.","first_seen_ms":985183,"end_ms":1021189,"confidence":0.98,"text":"Uberduck, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_66793a4fcfaba6d4961dadb03662336b","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"then used Uberduck to integrate","first_seen_ms":1088782,"end_ms":1126917,"confidence":0.99,"text":"Uberduck, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Google","name":"WaveNet","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_bb2720de651d53ea5b18510dbc24ddc6","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"which is building on WaveNet before","first_seen_ms":985183,"end_ms":999645,"confidence":0.95,"text":"WaveNet, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"OpenAI","name":"Whisper","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_e3b8f2c7e5531fab02f83d0369c66ff0","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"OpenAI has got Whisper","first_seen_ms":985183,"end_ms":1009435,"confidence":0.99,"text":"Whisper, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_01d8d8c538a1218c3ca11bae5c7d9c9e","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":1047882,"end_ms":1062882,"confidence":0.95,"text":"Whisper, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"mentioned","brand":"Respeecher","name":"Respeecher","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_ef616b2b4c04d665261eb5a8d9669493","relation":"mentioned","evidence":["spoken"],"evidence_quote":"You can just change your voice using something like Respeecher.","first_seen_ms":1032445,"end_ms":1047882,"confidence":0.95,"text":"Respeecher, named in speech","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"PlayHT","name":"PlayHT","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_54326a7fc4923e69e05db900abda8aec","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":1068282,"end_ms":1087208,"confidence":0.9,"text":"PlayHT, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Google","name":"Imagen Video","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_5f61f56308aad7c612af66ac0f93ba20","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":1153587,"end_ms":1164305,"confidence":0.9,"text":"Imagen Video, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Meta","name":"Make-A-Video","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_4f7eef6524b7f60be8542e823852db95","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Make-A-Video","first_seen_ms":1153587,"end_ms":1164305,"confidence":0.95,"text":"Make-A-Video, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Rephrase.ai","name":"Rephrase.ai","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_2d8bb31f2700d195a0c0b8ad6b0c0d0e","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":1164387,"end_ms":1175617,"confidence":0.9,"text":"Rephrase.ai, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Roblox Corporation","name":"Roblox","kind":"game","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_3fe8a03654dee27c7068ca4830e3639a","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"using it in real life already is Roblox","first_seen_ms":1289217,"end_ms":1303167,"confidence":0.95,"text":"Roblox, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_18033c84d47f16aa552f39f3786fa9fd","relation":"mentioned","evidence":["spoken"],"evidence_quote":"like we saw with Roblox.","first_seen_ms":1357988,"end_ms":1367458,"confidence":0.9,"text":"Roblox, named in speech","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"OpenAI","name":"OpenAI Codex","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_844a763d27fdac852960082fb7b96886","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Codex by OpenAI","first_seen_ms":1326367,"end_ms":1341350,"confidence":0.95,"text":"OpenAI Codex, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"DeepMind","name":"AlphaFold","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_da035f55ecba7e4e23384f7a3d37293b","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"DeepMind and Google have built AlphaFold 2","first_seen_ms":1407057,"end_ms":1418005,"confidence":0.99,"text":"AlphaFold, named in speech and shown on screen","method_version":"window-products-v10"},{"id":"product_span_6c099f087cb88db54032acd657c4d156","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"And here's a quick video to explain how useful and valuable that is from AlphaFold.","first_seen_ms":1488925,"end_ms":1525340,"confidence":0.99,"text":"AlphaFold, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Meta","name":"ESMFold","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_579093e631ada25d7faaf9eb67998aa3","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Meta is working on with ESMFold","first_seen_ms":1418057,"end_ms":1429125,"confidence":0.95,"text":"ESMFold, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"OpenBioML","name":"LibreFold","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_93d441f198a765f8f1272cbed31ea00b","relation":"shown","evidence":["on_screen"],"evidence_quote":null,"first_seen_ms":1422057,"end_ms":1429125,"confidence":0.9,"text":"LibreFold, shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Wolfram","name":"Wolfram Alpha","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_703bc7a65f49ef2b835faebbd4193d4e","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Wolfram Alpha","first_seen_ms":1631785,"end_ms":1648729,"confidence":0.95,"text":"Wolfram Alpha, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Google","name":"Google Search","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_0933e99cbcd8d7591f4fe50261dd0612","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"Google Search","first_seen_ms":1652729,"end_ms":1698000,"confidence":0.95,"text":"Google Search, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Dribbble Holdings Ltd.","name":"Dribbble","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_36dd101a14ca697389453b4cbe9bbbc1","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"I might go to Dribbble and Instagram and Pinterest and Behance","first_seen_ms":1766023,"end_ms":1777741,"confidence":0.99,"text":"Dribbble, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Meta","name":"Instagram","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_e2b0df035e964244032468a8c21cff2d","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"I might go to Dribbble and Instagram and Pinterest and Behance","first_seen_ms":1766723,"end_ms":1777741,"confidence":0.99,"text":"Instagram, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Pinterest","name":"Pinterest","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_0355fbd8de119d0df871e83839f30b42","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"I might go to Dribbble and Instagram and Pinterest and Behance","first_seen_ms":1767323,"end_ms":1777741,"confidence":0.99,"text":"Pinterest, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Adobe","name":"Behance","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_68fe729f475b4ed689fbd54cb8f2ca8f","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"I might go to Dribbble and Instagram and Pinterest and Behance","first_seen_ms":1767823,"end_ms":1777741,"confidence":0.99,"text":"Behance, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Canva","name":"Canva","kind":"software","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_4a313e8576fb935acf61052a5e991db9","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"I will then use, you know, Canva and Figma as a tool to create it.","first_seen_ms":1772523,"end_ms":1777741,"confidence":0.99,"text":"Canva, named in speech and shown on screen","method_version":"window-products-v10"}]},{"id":null,"relation":"shown","brand":"Shutterstock","name":"Shutterstock","kind":"online_service","url":null,"url_host":null,"sku":null,"link_verified":false,"position":null,"method_version":null,"spans":[{"id":"product_span_bce3a3899e2c3b309dd1ba5a6bcd00e4","relation":"shown","evidence":["spoken","on_screen"],"evidence_quote":"And then I'm gonna go and sell it on Shutterstock.","first_seen_ms":1777741,"end_ms":1831325,"confidence":0.99,"text":"Shutterstock, named in speech and shown on screen","method_version":"window-products-v10"}]}],"places":[{"id":"place_0f645c3055d2b79a5f520c1e2bdf7685","name":"Hollywood","kind":"region","relation":"mentioned","country":"United States","basis":["spoken"],"wikidata_qid":null,"entity_id":"ent_a1fc8b5cbebc767088b4a24c69533339","label":null,"coordinates":null,"first_ms":287967,"spans":[{"start_ms":287967,"end_ms":292899,"via":"spoken","relation":"mentioned"}],"evidence":[{"source":"spoken","quote":"how it's going to change Hollywood","start_ms":287967}],"method_version":"video-places-v5"},{"id":"place_9f81aeb1b03766a476453d5dca7addd6","name":"Times Square","kind":"landmark","relation":"mentioned","country":"United States","basis":["on_screen_text"],"wikidata_qid":null,"entity_id":"ent_020e6de8011774c4a3bc625581c1b366","label":null,"coordinates":null,"first_ms":1202072,"spans":[{"start_ms":1202072,"end_ms":1217000,"via":"on_screen","relation":"mentioned"}],"evidence":[{"source":"on_screen_text","quote":"Robot dancing in times square","start_ms":1202072}],"method_version":"video-places-v5"},{"id":"place_f72a930f546b0ba07c10dd1d0624e928","name":"San Francisco","kind":"city","relation":"mentioned","country":"United States","basis":["on_screen_text"],"wikidata_qid":"Q62","entity_id":"ent_6fffde63-5390-4610-9a57-ce9cca2de6be","external_ids":{"osm":"relation/111968"},"label":"San Francisco","coordinates":{"latitude":37.775,"longitude":-122.419444444},"first_ms":1717523,"spans":[{"start_ms":1717523,"end_ms":1725523,"via":"on_screen","relation":"mentioned"}],"evidence":[{"source":"on_screen_text","quote":"IANS I San Francisco","start_ms":1717523}],"method_version":"video-places-v5"},{"id":"place_1959a940c423cca18ce77fca2204bc28","name":"Port of Bristol","kind":"landmark","relation":"shown","country":"United Kingdom","basis":["on_screen_text"],"wikidata_qid":"Q7231091","entity_id":"ent_303e846641375e9063f0a0c65e00cfe6","label":"Port of Bristol","coordinates":{"latitude":51.5033,"longitude":-2.71621},"first_ms":1869875,"spans":[{"start_ms":1869875,"end_ms":1907623,"via":"on_screen","relation":"shown"}],"evidence":[{"source":"on_screen_text","quote":"PORT OF BRISTOL","start_ms":1869875}],"method_version":"video-places-v5"},{"id":"place_e58af6bb6cbcb5b105837d7b7eb63a8d","name":"UK","kind":"country","relation":"mentioned","country":"United Kingdom","basis":["spoken"],"wikidata_qid":"Q145","entity_id":"ent_a13a7733-0be2-4e51-9318-25622e7b6fbd","external_ids":{"osm":"relation/62149"},"label":"United Kingdom","coordinates":{"latitude":54.6,"longitude":-2},"first_ms":1885430,"spans":[{"start_ms":1885430,"end_ms":1888721,"via":"spoken","relation":"mentioned"}],"evidence":[{"source":"spoken","quote":"Or take even the Locomotive Act in the middle of 19th century in the UK.","start_ms":1885430}],"method_version":"video-places-v5"},{"id":"place_1f297a61d672531131e7cab74fb850ac","name":"Los Angeles","kind":"city","relation":"filmed_at","country":"United States","basis":["description"],"wikidata_qid":"Q65","entity_id":"ent_9ac7ddfb-b7db-40b2-b456-604d03112fdb","external_ids":{"osm":"relation/207359"},"label":"Los Angeles","coordinates":{"latitude":34.05223,"longitude":-118.24368},"first_ms":null,"spans":[{"start_ms":0,"end_ms":125198,"via":"on_screen","relation":"filmed_at"},{"start_ms":150083,"end_ms":698833,"via":"on_screen","relation":"filmed_at"},{"start_ms":713167,"end_ms":737167,"via":"on_screen","relation":"filmed_at"},{"start_ms":762417,"end_ms":927500,"via":"on_screen","relation":"filmed_at"},{"start_ms":939083,"end_ms":1074878,"via":"on_screen","relation":"filmed_at"},{"start_ms":1087208,"end_ms":1114141,"via":"on_screen","relation":"filmed_at"},{"start_ms":1126917,"end_ms":1217000,"via":"on_screen","relation":"filmed_at"},{"start_ms":1263542,"end_ms":1275417,"via":"on_screen","relation":"filmed_at"},{"start_ms":1289208,"end_ms":1494583,"via":"on_screen","relation":"filmed_at"},{"start_ms":1525340,"end_ms":1921998,"via":"on_screen","relation":"filmed_at"},{"start_ms":1948917,"end_ms":1981297,"via":"on_screen","relation":"filmed_at"}],"evidence":[{"source":"description","quote":"February 2023, Los Angeles","start_ms":null},{"source":"description","quote":"February 2023, Los Angeles","start_ms":null}],"method_version":"video-places-v5"}],"topics":[{"topic_id":"topic_2d1a26daafb7b60f18c0","label":"Human AI Prompting Roles","takes_url":"https://api.agentcdn.com/public/creators/sandersaar/topics/topic_2d1a26daafb7b60f18c0/takes.json"},{"topic_id":"topic_9675670c47825a3c1c04","label":"Synthetic Voice Quality","takes_url":"https://api.agentcdn.com/public/creators/sandersaar/topics/topic_9675670c47825a3c1c04/takes.json"},{"topic_id":"topic_d85aae62fd88207a3f46","label":"AI Coding Assistants","takes_url":"https://api.agentcdn.com/public/creators/sandersaar/topics/topic_d85aae62fd88207a3f46/takes.json"},{"topic_id":"topic_e764bcdb18170a9aa15d","label":"Declining Cost of Intelligence","takes_url":"https://api.agentcdn.com/public/creators/sandersaar/topics/topic_e764bcdb18170a9aa15d/takes.json"},{"topic_id":"topic_e83142f429a910f914db","label":"AlphaFold Protein Structure Prediction","takes_url":"https://api.agentcdn.com/public/creators/sandersaar/topics/topic_e83142f429a910f914db/takes.json"}],"labels":[{"kind":"sound","label":"Music","count":54,"first_ms":698000,"last_ms":1525500},{"kind":"action","label":"speaking","count":42,"first_ms":0,"last_ms":1981097},{"kind":"action","label":"gesturing","count":37,"first_ms":9600,"last_ms":1981097},{"kind":"action","label":"talking","count":24,"first_ms":316919,"last_ms":1962898},{"kind":"object","label":"presentation slide","count":20,"first_ms":557978,"last_ms":1883175},{"kind":"object","label":"man","count":15,"first_ms":347000,"last_ms":1981097},{"kind":"action","label":"explaining","count":13,"first_ms":293729,"last_ms":1658929},{"kind":"product","label":"ChatGPT","count":13,"first_ms":50067,"last_ms":1843223},{"kind":"object","label":"diagram","count":11,"first_ms":102924,"last_ms":1791241},{"kind":"object","label":"infographic","count":11,"first_ms":273129,"last_ms":1816241},{"kind":"place","label":"Los Angeles","count":11,"first_ms":0,"last_ms":1981297},{"kind":"action","label":"presenting","count":9,"first_ms":33200,"last_ms":1921975},{"kind":"action","label":"talking to camera","count":9,"first_ms":48167,"last_ms":1851125},{"kind":"object","label":"infographic slide","count":9,"first_ms":600432,"last_ms":1709400},{"kind":"object","label":"slide presentation","count":9,"first_ms":655203,"last_ms":1831241},{"kind":"action","label":"smiling","count":6,"first_ms":319919,"last_ms":1981097},{"kind":"action","label":"speaking to camera","count":6,"first_ms":302919,"last_ms":1883175},{"kind":"object","label":"shirt","count":6,"first_ms":762417,"last_ms":1981097},{"kind":"person","label":"Sam Altman","count":6,"first_ms":110724,"last_ms":1948898},{"kind":"action","label":"gesturing with hands","count":5,"first_ms":221565,"last_ms":1102182},{"kind":"action","label":"hand gesturing","count":5,"first_ms":93167,"last_ms":1831241},{"kind":"product","label":"GPT-3","count":5,"first_ms":167783,"last_ms":1251417},{"kind":"product","label":"Stable Diffusion","count":5,"first_ms":809275,"last_ms":1921998},{"kind":"object","label":"infographic diagram","count":4,"first_ms":302919,"last_ms":1843125},{"kind":"object","label":"logo","count":4,"first_ms":1278917,"last_ms":1640685},{"kind":"object","label":"microphone","count":4,"first_ms":111224,"last_ms":150024},{"kind":"object","label":"paper sheet","count":4,"first_ms":714833,"last_ms":762417},{"kind":"person","label":"Elon Musk","count":4,"first_ms":1745523,"last_ms":1869825},{"kind":"action","label":"explaining concept","count":3,"first_ms":89167,"last_ms":1592973},{"kind":"action","label":"gesturing with hand","count":3,"first_ms":136924,"last_ms":1190187},{"kind":"action","label":"pointing","count":3,"first_ms":287929,"last_ms":570678},{"kind":"action","label":"speaking into microphone","count":3,"first_ms":118224,"last_ms":150024},{"kind":"object","label":"black and white photograph","count":3,"first_ms":1068282,"last_ms":1088682},{"kind":"object","label":"chart","count":3,"first_ms":411083,"last_ms":1673929},{"kind":"object","label":"chimpanzee","count":3,"first_ms":1552573,"last_ms":1584873},{"kind":"object","label":"code editor","count":3,"first_ms":1303167,"last_ms":1390988},{"kind":"object","label":"smartphone mockup","count":3,"first_ms":600432,"last_ms":973083},{"kind":"person","label":"David Guetta","count":3,"first_ms":1102208,"last_ms":1134008},{"kind":"person","label":"Eminem","count":3,"first_ms":1098382,"last_ms":1121708},{"kind":"product","label":"Bing Chat","count":3,"first_ms":221565,"last_ms":1552500},{"kind":"product","label":"DALL-E 2","count":3,"first_ms":809275,"last_ms":1843223},{"kind":"product","label":"Microsoft Bing","count":3,"first_ms":660003,"last_ms":1631750},{"kind":"product","label":"Replit Ghostwriter","count":3,"first_ms":257429,"last_ms":1391042},{"kind":"product","label":"Synthesia","count":3,"first_ms":0,"last_ms":1175617},{"kind":"action","label":"animating graphics","count":2,"first_ms":1494583,"last_ms":1517283},{"kind":"action","label":"gesturing while speaking","count":2,"first_ms":863650,"last_ms":1709400},{"kind":"action","label":"presenting information","count":2,"first_ms":59967,"last_ms":1843125},{"kind":"action","label":"presenting slide","count":2,"first_ms":473350,"last_ms":1175587},{"kind":"action","label":"reading paper","count":2,"first_ms":725333,"last_ms":752817},{"kind":"action","label":"typing prompt","count":2,"first_ms":951383,"last_ms":971083},{"kind":"object","label":"DJ mixer","count":2,"first_ms":1088782,"last_ms":1114108},{"kind":"object","label":"glasses","count":2,"first_ms":737233,"last_ms":752817},{"kind":"object","label":"mirror","count":2,"first_ms":1552573,"last_ms":1577973},{"kind":"object","label":"podcast interface","count":2,"first_ms":1055482,"last_ms":1068182},{"kind":"object","label":"presentation graphic","count":2,"first_ms":18200,"last_ms":48100},{"kind":"object","label":"screen recording","count":2,"first_ms":1202072,"last_ms":1239172},{"kind":"object","label":"screenshot","count":2,"first_ms":167783,"last_ms":186783},{"kind":"object","label":"spectacles","count":2,"first_ms":1068282,"last_ms":1087182},{"kind":"object","label":"text box","count":2,"first_ms":950083,"last_ms":1216972},{"kind":"person","label":"Joe Rogan","count":2,"first_ms":1056882,"last_ms":1071882},{"kind":"person","label":"Ryan Reynolds","count":2,"first_ms":725333,"last_ms":751333},{"kind":"person","label":"Steve Jobs","count":2,"first_ms":1058382,"last_ms":1074782},{"kind":"product","label":"AlphaFold","count":2,"first_ms":1407057,"last_ms":1525340},{"kind":"product","label":"Apple Mac","count":2,"first_ms":875650,"last_ms":1102208},{"kind":"product","label":"D-ID Creative Reality Studio","count":2,"first_ms":1134108,"last_ms":1263542},{"kind":"product","label":"Figma","count":2,"first_ms":975583,"last_ms":1777741},{"kind":"product","label":"GitHub Copilot","count":2,"first_ms":1326367,"last_ms":1407057},{"kind":"product","label":"Google Bard","count":2,"first_ms":530750,"last_ms":1618271},{"kind":"product","label":"Krikey AI","count":2,"first_ms":1134108,"last_ms":1289208},{"kind":"product","label":"NVIDIA A100","count":2,"first_ms":585778,"last_ms":863650},{"kind":"product","label":"Roblox","count":2,"first_ms":1289217,"last_ms":1367458},{"kind":"product","label":"Runway","count":2,"first_ms":1134108,"last_ms":1226625},{"kind":"product","label":"Twitter","count":2,"first_ms":567878,"last_ms":1751863},{"kind":"product","label":"Uberduck","count":2,"first_ms":985183,"last_ms":1126917},{"kind":"product","label":"Whisper","count":2,"first_ms":985183,"last_ms":1062882},{"kind":"action","label":"DJing","count":1,"first_ms":1108608,"last_ms":1114108},{"kind":"action","label":"applying texture","count":1,"first_ms":1291717,"last_ms":1303117},{"kind":"action","label":"attacking mirror","count":1,"first_ms":1556473,"last_ms":1564173},{"kind":"action","label":"cheering","count":1,"first_ms":1114208,"last_ms":1119908},{"kind":"action","label":"critiquing platform strategy","count":1,"first_ms":1851225,"last_ms":1861525},{"kind":"action","label":"dancing","count":1,"first_ms":1275417,"last_ms":1276917},{"kind":"action","label":"driving car","count":1,"first_ms":1309067,"last_ms":1312867},{"kind":"action","label":"explaining historical context","count":1,"first_ms":1861625,"last_ms":1869825},{"kind":"action","label":"explaining interface","count":1,"first_ms":775917,"last_ms":790717},{"kind":"action","label":"explaining presentation","count":1,"first_ms":18200,"last_ms":33100},{"kind":"action","label":"floating","count":1,"first_ms":1517383,"last_ms":1522283},{"kind":"action","label":"floating car","count":1,"first_ms":1319467,"last_ms":1324767},{"kind":"action","label":"generating avatar","count":1,"first_ms":1253417,"last_ms":1263517},{"kind":"action","label":"generating image","count":1,"first_ms":927583,"last_ms":932283},{"kind":"action","label":"gesturing while explaining","count":1,"first_ms":200765,"last_ms":212365},{"kind":"action","label":"gesturing while talking","count":1,"first_ms":1047882,"last_ms":1062882},{"kind":"action","label":"gesturing with open hands","count":1,"first_ms":384300,"last_ms":398000},{"kind":"action","label":"hand gesture","count":1,"first_ms":329919,"last_ms":335919},{"kind":"action","label":"performing","count":1,"first_ms":1114208,"last_ms":1119908},{"kind":"action","label":"pointing at diagram","count":1,"first_ms":212465,"last_ms":221465},{"kind":"action","label":"pointing to eye","count":1,"first_ms":1469125,"last_ms":1472625},{"kind":"action","label":"pointing upward","count":1,"first_ms":60167,"last_ms":63167},{"kind":"action","label":"presenting to camera","count":1,"first_ms":1538773,"last_ms":1552373},{"kind":"action","label":"reading aloud","count":1,"first_ms":741933,"last_ms":751333},{"kind":"action","label":"speaking with animated expressions","count":1,"first_ms":236565,"last_ms":250065},{"kind":"action","label":"swinging tree branch","count":1,"first_ms":1565673,"last_ms":1573773},{"kind":"action","label":"talking head animation","count":1,"first_ms":1240572,"last_ms":1247572},{"kind":"action","label":"text animation","count":1,"first_ms":1491925,"last_ms":1494525},{"kind":"action","label":"typing","count":1,"first_ms":705833,"last_ms":708133},{"kind":"action","label":"walking in snow","count":1,"first_ms":1222072,"last_ms":1224572},{"kind":"action","label":"waving hand","count":1,"first_ms":1109208,"last_ms":1114108},{"kind":"object","label":"3D character","count":1,"first_ms":1275417,"last_ms":1276917},{"kind":"object","label":"3D model","count":1,"first_ms":1286317,"last_ms":1289117},{"kind":"object","label":"Venn diagram","count":1,"first_ms":455750,"last_ms":465250},{"kind":"object","label":"armchair","count":1,"first_ms":933283,"last_ms":939083},{"kind":"object","label":"arrow","count":1,"first_ms":1496583,"last_ms":1504483},{"kind":"object","label":"astronaut","count":1,"first_ms":1224572,"last_ms":1226572},{"kind":"object","label":"astronaut graphic","count":1,"first_ms":0,"last_ms":9500},{"kind":"object","label":"bar chart","count":1,"first_ms":1646829,"last_ms":1648729},{"kind":"object","label":"bust sculpture","count":1,"first_ms":1229572,"last_ms":1234072},{"kind":"object","label":"car","count":1,"first_ms":1289417,"last_ms":1303117},{"kind":"object","label":"cat","count":1,"first_ms":939083,"last_ms":944983},{"kind":"object","label":"chair","count":1,"first_ms":1921998,"last_ms":1936998},{"kind":"object","label":"chat interface","count":1,"first_ms":1584973,"last_ms":1592973},{"kind":"object","label":"clay figure","count":1,"first_ms":1219572,"last_ms":1221572},{"kind":"object","label":"cloud","count":1,"first_ms":1522583,"last_ms":1524883},{"kind":"object","label":"code editor screenshot","count":1,"first_ms":1347967,"last_ms":1354767},{"kind":"object","label":"completed pyramid diagram","count":1,"first_ms":236565,"last_ms":250065},{"kind":"object","label":"computer interface","count":1,"first_ms":687003,"last_ms":694703},{"kind":"object","label":"computer monitor","count":1,"first_ms":698833,"last_ms":700633},{"kind":"object","label":"computer screen","count":1,"first_ms":1251417,"last_ms":1263517},{"kind":"object","label":"cost comparison chart","count":1,"first_ms":316919,"last_ms":327419},{"kind":"object","label":"data versus cost graph","count":1,"first_ms":327519,"last_ms":338819},{"kind":"object","label":"diagram graphic","count":1,"first_ms":74167,"last_ms":89067},{"kind":"object","label":"digital avatar","count":1,"first_ms":1240072,"last_ms":1249572},{"kind":"object","label":"digital diagram","count":1,"first_ms":1759323,"last_ms":1763123},{"kind":"object","label":"display slide","count":1,"first_ms":1175687,"last_ms":1190187},{"kind":"object","label":"dog","count":1,"first_ms":933283,"last_ms":939083},{"kind":"object","label":"educational graphic","count":1,"first_ms":587978,"last_ms":600378},{"kind":"object","label":"flowchart","count":1,"first_ms":362400,"last_ms":375100},{"kind":"object","label":"graphic chart","count":1,"first_ms":545678,"last_ms":557878},{"kind":"object","label":"graphic display","count":1,"first_ms":678103,"last_ms":686903},{"kind":"object","label":"graphic slide","count":1,"first_ms":625232,"last_ms":640132},{"kind":"object","label":"grid presentation slide","count":1,"first_ms":891550,"last_ms":906250},{"kind":"object","label":"historical plaque photo","count":1,"first_ms":1883275,"last_ms":1893275},{"kind":"object","label":"infographic card","count":1,"first_ms":473350,"last_ms":486750},{"kind":"object","label":"infographic chart","count":1,"first_ms":50067,"last_ms":59867},{"kind":"object","label":"infographic layout","count":1,"first_ms":1429157,"last_ms":1444057},{"kind":"object","label":"infographic pyramid","count":1,"first_ms":212465,"last_ms":221465},{"kind":"object","label":"laptop screen","count":1,"first_ms":697503,"last_ms":698803},{"kind":"object","label":"layer diagram","count":1,"first_ms":221565,"last_ms":236465},{"kind":"object","label":"line chart","count":1,"first_ms":69267,"last_ms":73867},{"kind":"object","label":"microscope","count":1,"first_ms":1507583,"last_ms":1517283},{"kind":"object","label":"mobile app interface","count":1,"first_ms":945083,"last_ms":950083},{"kind":"object","label":"news headline","count":1,"first_ms":1540573,"last_ms":1552373},{"kind":"object","label":"plastic bottle","count":1,"first_ms":1517383,"last_ms":1522283},{"kind":"object","label":"pose tracking diagram","count":1,"first_ms":1190287,"last_ms":1201987},{"kind":"object","label":"pyramid diagram","count":1,"first_ms":200765,"last_ms":212365},{"kind":"object","label":"red car","count":1,"first_ms":1312967,"last_ms":1324767},{"kind":"object","label":"robot avatar","count":1,"first_ms":694803,"last_ms":697403},{"kind":"object","label":"robot toy","count":1,"first_ms":703933,"last_ms":705833},{"kind":"object","label":"slide","count":1,"first_ms":292129,"last_ms":302829},{"kind":"object","label":"slide graphic","count":1,"first_ms":486850,"last_ms":497750},{"kind":"object","label":"social media post graphic","count":1,"first_ms":1851225,"last_ms":1861525},{"kind":"object","label":"social media tweet display","count":1,"first_ms":1861625,"last_ms":1869825},{"kind":"object","label":"software dashboard","count":1,"first_ms":700733,"last_ms":703833},{"kind":"object","label":"software interface","count":1,"first_ms":1239272,"last_ms":1251372},{"kind":"object","label":"software logo","count":1,"first_ms":922983,"last_ms":923983},{"kind":"object","label":"sports car","count":1,"first_ms":1303167,"last_ms":1312867},{"kind":"object","label":"spreadsheet","count":1,"first_ms":610632,"last_ms":625132},{"kind":"object","label":"stage photo","count":1,"first_ms":1920075,"last_ms":1921975},{"kind":"object","label":"table","count":1,"first_ms":1944098,"last_ms":1948898},{"kind":"object","label":"text prompt","count":1,"first_ms":924083,"last_ms":927483},{"kind":"object","label":"timeline diagram","count":1,"first_ms":89167,"last_ms":102867},{"kind":"object","label":"title card","count":1,"first_ms":1490925,"last_ms":1494525},{"kind":"object","label":"trend graph","count":1,"first_ms":338919,"last_ms":346919},{"kind":"object","label":"tweet graphic","count":1,"first_ms":1849125,"last_ms":1851125},{"kind":"object","label":"webpage screenshot","count":1,"first_ms":1709500,"last_ms":1717500},{"kind":"object","label":"website screenshot","count":1,"first_ms":1906175,"last_ms":1907575},{"kind":"object","label":"wind turbine","count":1,"first_ms":1507583,"last_ms":1517283},{"kind":"object","label":"workflow diagram","count":1,"first_ms":1763223,"last_ms":1777723},{"kind":"place","label":"Hollywood","count":1,"first_ms":287967,"last_ms":292899},{"kind":"place","label":"Port of Bristol","count":1,"first_ms":1869875,"last_ms":1907623},{"kind":"place","label":"San Francisco","count":1,"first_ms":1717523,"last_ms":1725523},{"kind":"place","label":"Times Square","count":1,"first_ms":1202072,"last_ms":1217000},{"kind":"place","label":"UK","count":1,"first_ms":1885430,"last_ms":1888721},{"kind":"product","label":"Amazon Polly","count":1,"first_ms":985183,"last_ms":999645},{"kind":"product","label":"Apple iPhone","count":1,"first_ms":874150,"last_ms":891549},{"kind":"product","label":"AudioLM","count":1,"first_ms":985183,"last_ms":999645},{"kind":"product","label":"Behance","count":1,"first_ms":1767823,"last_ms":1777741},{"kind":"product","label":"Canva","count":1,"first_ms":1772523,"last_ms":1777741},{"kind":"product","label":"Claude","count":1,"first_ms":533750,"last_ms":557917},{"kind":"product","label":"Descript","count":1,"first_ms":293729,"last_ms":316832},{"kind":"product","label":"DiffusionBee","count":1,"first_ms":881650,"last_ms":891549},{"kind":"product","label":"Discord","count":1,"first_ms":821775,"last_ms":824198},{"kind":"product","label":"Draw Things","count":1,"first_ms":881650,"last_ms":891549},{"kind":"product","label":"Dribbble","count":1,"first_ms":1766023,"last_ms":1777741},{"kind":"product","label":"ESMFold","count":1,"first_ms":1418057,"last_ms":1429125},{"kind":"product","label":"Ernie Bot","count":1,"first_ms":545678,"last_ms":557917},{"kind":"product","label":"Facebook","count":1,"first_ms":50067,"last_ms":59936},{"kind":"product","label":"Galileo AI","count":1,"first_ms":950083,"last_ms":973083},{"kind":"product","label":"Google Search","count":1,"first_ms":1652729,"last_ms":1698000},{"kind":"product","label":"Google Translate","count":1,"first_ms":647632,"last_ms":668367},{"kind":"product","label":"Google Trends","count":1,"first_ms":69267,"last_ms":74091},{"kind":"product","label":"Imagen Video","count":1,"first_ms":1153587,"last_ms":1164305},{"kind":"product","label":"Instagram","count":1,"first_ms":1766723,"last_ms":1777741},{"kind":"product","label":"InstructGPT","count":1,"first_ms":198083,"last_ms":212426},{"kind":"product","label":"Jasper","count":1,"first_ms":688403,"last_ms":713167},{"kind":"product","label":"Lensa AI","count":1,"first_ms":893150,"last_ms":906283},{"kind":"product","label":"LibreFold","count":1,"first_ms":1422057,"last_ms":1429125},{"kind":"product","label":"Make-A-Video","count":1,"first_ms":1153587,"last_ms":1164305},{"kind":"product","label":"Microsoft Azure","count":1,"first_ms":577578,"last_ms":610542},{"kind":"product","label":"Midjourney","count":1,"first_ms":809275,"last_ms":824198},{"kind":"product","label":"MuseNet","count":1,"first_ms":985183,"last_ms":999645},{"kind":"product","label":"MusicLM","count":1,"first_ms":985183,"last_ms":1009435},{"kind":"product","label":"OpenAI Codex","count":1,"first_ms":1326367,"last_ms":1341350},{"kind":"product","label":"Pinterest","count":1,"first_ms":1767323,"last_ms":1777741},{"kind":"product","label":"PlayHT","count":1,"first_ms":1068282,"last_ms":1087208},{"kind":"product","label":"Pokémon GO","count":1,"first_ms":59967,"last_ms":74091},{"kind":"product","label":"Reddit","count":1,"first_ms":567478,"last_ms":572917},{"kind":"product","label":"Rephrase.ai","count":1,"first_ms":1164387,"last_ms":1175617},{"kind":"product","label":"Respeecher","count":1,"first_ms":1032445,"last_ms":1047882},{"kind":"product","label":"Shutterstock","count":1,"first_ms":1777741,"last_ms":1831325},{"kind":"product","label":"SingSong","count":1,"first_ms":985183,"last_ms":1021189},{"kind":"product","label":"Speechify","count":1,"first_ms":985183,"last_ms":1009435},{"kind":"product","label":"Substack","count":1,"first_ms":567178,"last_ms":572917},{"kind":"product","label":"TikTok","count":1,"first_ms":50067,"last_ms":59936},{"kind":"product","label":"WaveNet","count":1,"first_ms":985183,"last_ms":999645},{"kind":"product","label":"Wikipedia","count":1,"first_ms":568278,"last_ms":572917},{"kind":"product","label":"Wolfram Alpha","count":1,"first_ms":1631785,"last_ms":1648729},{"kind":"sound","label":"Chopping (food)","count":1,"first_ms":1243000,"last_ms":1244500},{"kind":"sound","label":"Ukulele","count":1,"first_ms":750500,"last_ms":752000}]},"errors":null}