When 0:00:00 - 0:00:05 I saw a black background with a white logo on it. I found a white and black street sign, the word samsung on a picture, a black background with white lettering, writing on the picture, the text says east. I heard someone say " ♪ " When 0:00:05 - 0:00:06 I saw a black background with a white logo on it. I found a white and black street sign, the word samsung on a picture, a black background with white lettering, writing on the bottom of the picture, the text says east. I heard someone say " ♪ " When 0:00:06 - 0:00:33 I saw a man standing in front of a laptop computer with his hands out. I found a man wearing an orange shirt, a plant in a white pot, a green plant in a blue bowl, the back of a laptop, a long blue table, a black framed print, a big blue bowl, a book on the table. I heard someone say " ♪ To the GPT-4 developer demo livestream. Honestly, it's kind of hard for me to believe that this day is here. OpenAI has been building this technology really since we started the company, but for the past two years, we've been really focused on delivering GPT-4. That started with rebuilding our entire training stack, actually training the model, and then seeing what it was capable of, trying to figure out its capabilities, its risks, " When 0:00:33 - 0:00:56 I saw a man in a blue shirt is making a face while talking. I found the man is wearing an orange shirt, the man is wearing a brown shirt, a green house plant, the head of a man. I heard someone say " trying to figure out its capabilities, its risks, working with partners in order to test it in real-world scenarios, really tuning its behavior, optimizing the model, getting it available so that you can use it. And so today, our goal is to show you a little bit of how to make GPT-4 shine, how to really get the most out of it, you know, where its kind of, you know, weaknesses are, where it's—we're still working on it, and just how to really use it as a good tool, a good partner. " When 0:00:56 - 0:01:43 I saw a man standing in front of a laptop computer on a table. I found a man wearing a yellow shirt, a plant in a white pot, a green plant in a blue bowl, the back of a laptop, a long blue table, a black framed print, a big blue bowl, a black coffee mug. I heard someone say " and just how to really use it as a good tool, a good partner. So if you're interested in participating in the stream, if you go to our Discord, so that's discord.gg slash openai, there's comments in there, and we'll take a couple of audience suggestions. So the first thing I want to show you is the first task that GPT-4 could do that we never really got 3.5 to do. And the way to think about this is all throughout training that, you know, you're constantly doing all this work, it's 2 a.m., the pager goes off, you fix the model, and you're always wondering, is it going to work? Is all this effort actually going to pan out? And so we all had a pet task that we really liked, and that we would all individually be trying to see, is the model capable of it now? And I'm going to show you the first one that we had a success for 4, but never really got there for 3.5. " When 0:01:43 - 0:01:47 I saw a man is using a laptop computer in front of a green screen. I found man is looking at the screen, a green and black box, a black wall with green trim, a laptop computer, a white apple logo. I heard someone say " but never really got there for 3.5. So I'm just going to copy the top of our blog post from today, going to paste it into our playground. " When 0:01:47 - 0:01:48 I saw a man is speaking on a video call with a laptop. I found man is wearing a shirt, a white page, a black wall, a white apple logo. I heard someone say " going to paste it into our playground. " When 0:01:48 - 0:01:52 I saw a man is speaking on a video call with a laptop. I found man on television screen, a white page, a black laptop, a white apple logo. I heard someone say " going to paste it into our playground. Now, this is our new chat completions playground " When 0:01:52 - 0:01:55 I saw a man is giving a presentation on a laptop computer. I found man is wearing a yellow shirt, a white page, a black laptop, a white apple logo. I heard someone say " Now, this is our new chat completions playground that came out two weeks ago. I'm going to show you first with GPT-3.5. " When 0:01:55 - 0:02:28 I saw a man is standing in front of a computer screen. I found a man is on the screen, a white page, a black laptop, a white apple logo. I heard someone say " I'm going to show you first with GPT-3.5. 4 has the same API to it, the same playground. The way that it works is you have a system message where you explain to the model what it's supposed to do, and we've made these models very steerable. So you can provide it with really any instruction you want, whatever you dream up. And the model will adhere to it pretty well. And in the future, it will get increasingly, increasingly powerful at steering the model very reliably. You can then paste whatever you want as a user. The model will return messages as an assistant. And the way to think of it is that we're moving away from sort of just raw text in, raw text out, " When 0:02:28 - 0:02:32 I saw a computer screen with a document open on it. I found a white computer screen, black text on white background, the black bottom of the screen. I heard someone say " from sort of just raw text in, raw text out, where you can't tell where different parts of the conversation come from, " When 0:02:32 - 0:02:55 I saw a computer screen with a document open on it. I found a computer monitor screen, a window for the microwave. I heard someone say " of the conversation come from, but towards this much more structured format that gives the model the opportunity to know, well, this is the user asking me to do something that the developer didn't intend. I should listen to the developer here. All right, so now time to actually show you the task that I'm referring to. So everyone's familiar with summarize. This is an article into a sentence, okay? Getting a little more specific, but where every word begins with G. " When 0:02:55 - 0:03:15 I saw a man is giving a presentation on a computer screen. I found a man on a screen, a white page, a black laptop, a logo on the computer. I heard someone say " but where every word begins with G. So this is 3.5. Let's see what it does. Yeah, it kind of didn't even try. Just gave up on the task. This is pretty typical for 3.5 trying to do this particular kind of task. If it's, you know, sort of a very kind of stilted article or something like that, maybe it can succeed. " When 0:03:15 - 0:03:33 I saw a man is using a laptop computer in a dark room. I found man on the screen, a white and black calendar, a black wall, a white apple logo, a black laptop computer. I heard someone say " or something like that, maybe it can succeed. But for the most part, 3.5 just gives up. But let's try the exact same prompt, the exact same system message in GPT-4. So kind of borderline whether you want to count AI or not, but so let's say AI doesn't count. " When 0:03:33 - 0:03:42 I saw a computer screen with a message that says "please wait". I found a computer monitor screen, two rows of horizontal lines, a black background, the black bottom of the screen. I heard someone say " but so let's say AI doesn't count. That's cheating. So fair enough. " When 0:03:42 - 0:03:55 I saw a man is giving a thumbs up in front of a computer screen. I found a man wearing a yellow shirt, a white page, a black laptop, a white apple logo. I heard someone say " So fair enough. The model happily accepts my feedback. So now to make sure it's not just good for Gs, I'd like to turn this over to the audience. I'll take a suggestion on what letter to try next. In the meanwhile, while I'm waiting for our moderators to pick the lucky, lucky letter, " When 0:03:55 - 0:04:07 I saw a man is using a laptop computer in a dark room. I found man is looking at the screen, a white page, a black laptop, a white apple logo. I heard someone say " to pick the lucky, lucky letter, I will give a try with A. But in this case, I'll say GPT-4 is fine. Why not? " When 0:04:07 - 0:04:27 I saw a man is using a laptop computer in a dark room. I found man on the screen, the computer monitor is on, the computer is on, a white apple logo. I heard someone say " Why not? Also pretty good summary. So I'll hop over to our Discord. All right. Wow, people are being a little ambitious here. I'm really trying to put the model through the paces. We're going to try Q, which if you think about this for a moment, I want the audience to really think about how would you do a summary of this article that all starts with Q. " When 0:04:27 - 0:04:37 I saw a computer screen with a message that says "you have a new message". I found a computer monitor screen, a black background, the black bottom of the screen. I heard someone say " that all starts with Q. It's not easy. " When 0:04:37 - 0:04:49 I saw a computer screen with a message displayed on it. I found a computer monitor screen, a black background. I heard someone say " It's pretty good. That's pretty good. All right, so I've shown you summarizing an existing article. I want to show you how you can flexibly combine ideas between different articles. " When 0:04:49 - 0:04:50 I saw a man standing in front of a laptop computer with a bonsai tree in the background. I found man wearing orange shirt, a large green plant, plant in white pot, the laptop is black, blue tablecloth on table, a black framed picture, a big blue bowl, a white coffee mug. I heard someone say " between different articles. So I'm going to take this article " When 0:04:50 - 0:05:05 I saw a man is sitting in front of a computer screen. I found man on television screen, a large flat screen tv, the logo on the tv, a plant in a pot, the counter is blue. I heard someone say " So I'm going to take this article that was on Hacker News yesterday, copy-paste it into the same conversation so it has all the context of what we were just doing. I'm going to say, find one common theme between this article and the GPT-4 blog. " When 0:05:05 - 0:05:17 I saw a computer screen with a document open on it. I found the screen is on, logo on the screen, black text on a screen. I heard someone say " between this article and the GPT-4 blog. So this is an article about Pinecone, which is a Python web app development framework, and it's making the technology more accessible, user-friendly. " When 0:05:17 - 0:05:18 I saw a computer screen with a message on it that says "please wait". I found a computer monitor screen, a message with black letters, a black background. I heard someone say " and it's making the technology more accessible, user-friendly. " When 0:05:18 - 0:05:32 I saw a man is using a laptop computer in a dark room. I found man on a screen, a white page, a black laptop, a white apple logo. I heard someone say " and it's making the technology more accessible, user-friendly. If you don't think that was insightful enough, you can always give some feedback and say, that was not insightful enough. Please, no, I'll just even just leave it there, leave it up to the model to decide. So bridging the gap between powerful technology " When 0:05:32 - 0:05:42 I saw a man is giving a presentation on a laptop screen. I found a man wearing a yellow shirt, a white page, a black laptop, a white apple logo. I heard someone say " So bridging the gap between powerful technology and practical applications. Seems not bad. And of course, you can ask for any other kind of task you want. Using its flexible language understanding and synthesis, you can ask for something like, " When 0:05:42 - 0:05:51 I saw a man is using a laptop computer in a dark room. I found a man on a screen, a white page, a black laptop, a white apple logo. I heard someone say " you can ask for something like, now turn the GPT-4 blog post into a rhyming poem. " When 0:05:51 - 0:06:17 I saw a man is making a video on a computer screen. I found man on television screen, a white page on a computer screen, a computer monitor, a logo on the computer. I heard someone say " Picked up on OpenAI evals, open source for all, helping to guide answering the call. Which, by the way, if you'd like to contribute to this model, please give us evals. We have an open source evaluation framework that will help us guide and all of our users understand what the model's capable of and to take it to the next level. So there we go. This is consuming existing content using GPT-4 " When 0:06:17 - 0:06:27 I saw a man is standing in front of a laptop computer. I found a man wearing an orange shirt, a plant in a white pot, a green leafy plant, the laptop is black, blue tablecloth on table, a doorway with a black frame, a big blue bowl. I heard someone say " This is consuming existing content using GPT-4 with a little bit of creativity on top. But next, I want to show you how to build with GPT-4, " When 0:06:27 - 0:06:36 I saw a man is using a laptop computer in a dark room. I found man is on a screen, a picture on a wall, a logo on the tv. I heard someone say " But next, I want to show you how to build with GPT-4, what it's like to create with it as a partner. And so the thing we're going to do is we're going to actually build a Discord bot. " When 0:06:36 - 0:06:41 I saw a man is using a laptop computer with a video screen. I found a man on a screen, a white page, a black wall, a logo on the tv. I heard someone say " is we're going to actually build a Discord bot. I'll build it live and show you the process, show you debugging, show you what the model can do, " When 0:06:41 - 0:06:47 I saw a man is standing in front of a laptop computer. I found a man is on the screen, a white page, a black wall, a logo on the computer. I heard someone say " show you debugging, show you what the model can do, where its limitations are, and how to work with them in order to sort of achieve new heights. So the first thing I'll do is tell the model " When 0:06:47 - 0:06:57 I saw a man is standing in front of a laptop computer. I found a man is on the screen, a white page on a poster, a black wall, a logo on the tv. I heard someone say " So the first thing I'll do is tell the model that this time it's supposed to be an AI programming assistant. Its job is to write things out in pseudocode first and then actually write the code. And this approach is very helpful " When 0:06:57 - 0:06:58 I saw a man is talking on a video call with a person on a laptop. I found a man on a screen, a white page, a black wall, a white apple logo. I heard someone say " And this approach is very helpful " When 0:06:58 - 0:07:14 I saw a man is standing in front of a computer screen with a message on it. I found a man is on the screen, a white page, a black wall, the shirt is orange, a logo on the tv. I heard someone say " And this approach is very helpful so that the model break down the problem into smaller pieces. And then that way, you're not kind of asking it to just come up with a super hard solution to a problem all in one go. It also makes it very interpretable because you can see exactly what the model was thinking and you can even provide corrections if you'd like. " When 0:07:14 - 0:07:37 I saw a computer screen with a few different windows open. I found the screen is on, a window open on the door. I heard someone say " and you can even provide corrections if you'd like. So here is the prompt that we're going to ask it. This is the kind of thing that 3.5 would totally choke on if you've tried anything like it. But so we're going to ask for a Discord bot that uses the GPT-4 API to read images and text. Now, there's one problem here, which is this model's training cutoff is in 2021, " When 0:07:37 - 0:07:50 I saw a man is speaking on a computer screen with a picture of a man in a blue shirt. I found a man is on the screen, a white page, a black laptop, a white apple logo. I heard someone say " which is this model's training cutoff is in 2021, which means it has not seen our new chat completions format. So I literally just went to the blog post from two weeks ago, copy pasted from the blog post, including the response format. It has not seen the new image extension to that. And so I just kind of wrote that up in, you know, " When 0:07:50 - 0:08:04 I saw a man is using a laptop computer with a video screen. I found a man on a screen, a white page on a computer screen, a black laptop, a logo on the computer. I heard someone say " And so I just kind of wrote that up in, you know, just very minimal detail about how to include images. And now the model can actually leverage that documentation that it did not have memorized, that it does not know. " When 0:08:04 - 0:08:05 I saw a man is talking on a phone while looking at a computer screen. I found man in orange shirt talking on phone, a white page on a desktop, a black laptop, a white apple logo. When 0:08:05 - 0:08:29 I saw a computer screen with a document open on it. I found the screen is on. I heard someone say " In general, these models are very good at using information that it's been trained on in new ways and synthesizing new content. And you can see that right here, that it actually wrote an entirely new bot. Now, let's actually see if this bot is going to work in practice. So you should always look through the code to get a sense of what it does. " When 0:08:29 - 0:08:38 I saw a man is working on a laptop computer in front of a screen. I found man on the screen, a web page on a monitor, a black and white image of a man and a computer monitor, a white apple logo, a small black box. I heard someone say " to get a sense of what it does. Don't run untrusted code from humans or from AIs. And one thing to note is that the Discord API has changed a lot over time, " When 0:08:38 - 0:08:39 I saw a man is standing in front of a computer screen. I found a man on a screen, a web page on a monitor, a black and white image of a man and a computer monitor, a logo on the computer. I heard someone say " has changed a lot over time, and particularly that there's one feature " When 0:08:39 - 0:08:53 I saw a man is using a laptop computer in front of a screen. I found a man on a screen, a white display screen, a black and white image of a man and a computer monitor, a white apple logo. I heard someone say " and particularly that there's one feature that has changed a lot since this model was trained. Give it a try. In fact, yes, we are missing the intense keyword. This is something that came out in 2020. So the model does know it exists, " When 0:08:53 - 0:08:54 I saw a man is giving a presentation on a laptop computer. I found man is wearing a white shirt, a webpage on the screen, a black and white image of a man and a computer monitor, a white apple logo. I heard someone say " So the model does know it exists, but it doesn't know which version of the Discord API we're using. " When 0:08:54 - 0:08:59 I saw a man is giving a presentation on a laptop computer. I found a man on a screen, a webpage on the screen, a black and white image of a man and a computer monitor, a white apple logo. I heard someone say " but it doesn't know which version of the Discord API we're using. So are we out of luck? Well, not quite. " When 0:08:59 - 0:09:01 I saw a man is using a laptop computer in a dark room. I found man is wearing a yellow shirt, a white page, a black laptop, a logo on the computer. I heard someone say " Well, not quite. We can just simply paste to the model exactly the error message, " When 0:09:01 - 0:09:34 I saw a computer screen with a message that says "you have a new message". I found a white computer screen, text on the screen. I heard someone say " We can just simply paste to the model exactly the error message, not even going to say, hey, this is from running your code. Can you please fix it? We'll just let it run. And the model says, oh, yeah, whoops, the intense argument. Here's the correct code. Now let's give this a try, once again, kind of making sure that we understand what the code is doing. Now, a second issue that can come up is it doesn't know what environment I'm running in. And if you notice, it says, hey, here's this inscrutable error message, " When 0:09:34 - 0:09:47 I saw a man is standing in front of a laptop computer. I found a man on a screen, a web page on a laptop, a black wall, a logo on the computer. I heard someone say " here's this inscrutable error message, which if you've not used Jupyter Notebook a lot with async IO before, you probably have no idea what this means. But fortunately, once again, you can just sort of say to the model, " When 0:09:47 - 0:09:48 I saw a man is using a laptop computer in a dark room. I found a man on a screen, a white page, a black laptop, a white apple logo. I heard someone say " But fortunately, once again, you can just sort of say to the model, hey, I'm using Jupyter and would like to make this work. " When 0:09:48 - 0:10:04 I saw a computer screen with a message that says "you have a new message". I found a white computer screen. I heard someone say " hey, I'm using Jupyter and would like to make this work. Can you fix it? And the specific problem is that there's already an event loop running, so you need to use this nest async IO library. " When 0:10:04 - 0:10:05 I saw a computer screen with a menu on it and a few buttons. I found the screen is on, list of ingredients, a black background. I heard someone say " so you need to use this nest async IO library. " When 0:10:05 - 0:10:22 I saw a man is using a laptop computer in a dark room. I found a man on a screen, a white page on a desktop, a black wall, a logo on the computer. I heard someone say " so you need to use this nest async IO library. You need to call nest async IO.apply. The model knows all of this, correctly instantiates all of these pieces into the bot. It even helpfully tells you, oh, you're running in Jupyter. Well, you can do this bang pip install in order to install the package if you don't already have it. That was very helpful. " When 0:10:22 - 0:10:36 I saw a man is using a laptop computer in a dark room. I found man on a screen, the computer monitor is on, the computer is on, a logo on the computer. I heard someone say " That was very helpful. Now we'll run, and it looks like something happened. So the first thing I'll do is go over to our Discord, and I will paste in a screenshot of our Discord itself. " When 0:10:36 - 0:10:47 I saw a man is giving a presentation on a laptop screen. I found a man wearing a yellow shirt, the microwave is on the wall, the scene is in the kitchen, a white apple logo. I heard someone say " and I will paste in a screenshot of our Discord itself. So remember, GPT-4 is not just a language model. It's also a vision model. In fact, it can flexibly accept inputs that intersperse images and text arbitrarily, " When 0:10:47 - 0:11:03 I saw a man standing in front of a laptop computer with a bonsai plant in the background. I found man wearing a brown shirt, a plant in a white pot, a green plant in a blue bowl, the laptop is black, a white table, a black framed picture, a big blue bowl. I heard someone say " In fact, it can flexibly accept inputs that intersperse images and text arbitrarily, kind of like a document. Now, the image feature is in preview. So this is going to be a little sneak peek. It's not yet publicly available. It's something we're working with one partner called Be My Eyes in order to really start to develop it and get it ready for primetime. " When 0:11:03 - 0:11:10 I saw a man is using a laptop computer in a dark room. I found man on television screen, the computer monitor is on, the scene is in the kitchen, a logo on the tv. I heard someone say " in order to really start to develop it and get it ready for primetime. But you can ask anything you like. For example, I'll say, GPT-4, hello world. " When 0:11:10 - 0:11:21 I saw a man is using a laptop computer in a dark room. I found a man on a screen, the microwave is on the wall, the scene is in the kitchen, a logo on the computer. I heard someone say " For example, I'll say, GPT-4, hello world. Can you describe this image in painstaking detail? " When 0:11:21 - 0:11:26 I saw a computer screen with a dark background and a purple background. I found the microwave is silver. I heard someone say " Can you describe this image in painstaking detail? All right, which first of all, think of how you would do this yourself. There's a lot of different things you could latch onto, " When 0:11:26 - 0:11:27 I saw a computer screen with a dark background and a purple background. I found the microwave is red, microwave oven is black. I heard someone say " There's a lot of different things you could latch onto, a lot of different pieces of the system you could describe. " When 0:11:27 - 0:11:30 I saw a computer screen with a dark background and a purple background. I found a red microwave, microwave oven is yellow. I heard someone say " a lot of different pieces of the system you could describe. And we can go over to the actual code, and we can see that, yep, " When 0:11:30 - 0:11:37 I saw a computer screen showing a computer program. I found a white display screen, a black background. I heard someone say " And we can go over to the actual code, and we can see that, yep, we in fact received the message, have formatted an appropriate request for our API. " When 0:11:37 - 0:11:50 I saw a man is standing in front of a laptop computer. I found a man is on the screen, the microwave is on the wall, the scene is in the kitchen, a white apple logo. I heard someone say " for our API. And now we wait, because one of the things we have to do is we have to make the system faster. That's one of the things that we're working on optimizing. In the meanwhile, I just want to say to the audience that's watching, we'll take an audience request next. So if you have an image and a task you'd like to accomplish, " When 0:11:50 - 0:11:51 I saw a man is giving a speech in a dark room. I found a man is on the screen, the microwave is on the wall, the scene is in the kitchen, a white apple logo. I heard someone say " So if you have an image and a task you'd like to accomplish, " When 0:11:51 - 0:12:06 I saw a man in a blue shirt is looking down at his cell phone. I found man wearing a tshirt, a large green plant, white blinds behind man, a mans face, a white door. I heard someone say " So if you have an image and a task you'd like to accomplish, please submit that to the Discord. Our moderators will pick one that we'll run. So we can see that the Discord, oh, it looks like we have a response. Perfect. " When 0:12:06 - 0:12:11 I saw a man is using a laptop computer with a video screen. I found a man is on the screen, the laptop is on, the computer is on, a white apple logo. I heard someone say " Perfect. So it's a screenshot of a Discord application interface. Pretty good. Did not even describe it. It knows that it's Discord. " When 0:12:11 - 0:12:16 I saw a man is using a laptop computer in a dark room. I found the laptop is on, man is on a screen, the microwave is black, a white apple logo. I heard someone say " It knows that it's Discord. It's probably Discord written there somewhere, or it just kind of knows this from prior experience. " When 0:12:16 - 0:12:25 I saw a computer screen with a dark background and a purple background. I found the laptop is on, the bottom of the microwave is black, white spacebar on the microwave. I heard someone say " or it just kind of knows this from prior experience. Server icon labeled GPT-4 describes the interface in great detail. Talks about all the people telling me that I'm supposed to do Q. " When 0:12:25 - 0:12:41 I saw a computer screen with a dark background and a purple background. I found the laptop is on, white writing on the microwave. I heard someone say " Talks about all the people telling me that I'm supposed to do Q. Very, very kind audience. And describes much of the notification messages and the users that are in the channel. And so there you go. That's some pretty good understanding. Now, this next one, if you notice, first of all, we got a post, but the model did not actually see the message. " When 0:12:41 - 0:12:42 I saw a man is sitting at a desk with a laptop and a mouse. I found a silver tv set, man on a screen, picture of a bear on a television, a white apple logo. I heard someone say " but the model did not actually see the message. " When 0:12:42 - 0:12:45 I saw a man is standing in front of a laptop computer. I found a silver tv set, a man is on the screen, picture of a bear on a television. I heard someone say " but the model did not actually see the message. So is this a failure of the model or of the system around the model? " When 0:12:45 - 0:12:51 I saw a man is using a laptop computer and a video camera. I found a man on a screen, a web page on a computer screen, a black and white image of a computer monitor, a logo on the computer. I heard someone say " So is this a failure of the model or of the system around the model? Well, we can take a look. If you notice here, content is an empty string. " When 0:12:51 - 0:12:55 I saw a man is giving a presentation on a laptop computer. I found a man is on the screen, a web page on a laptop, a black and white image of a computer monitor, a logo on the computer. I heard someone say " If you notice here, content is an empty string. We received a blank message contents. " When 0:12:55 - 0:13:09 I saw a computer screen with a list of items on it. I found a white and orange box, a yellow and blue stripe, the bottom half of the computer. I heard someone say " We received a blank message contents. The reason for this is a dirty trick that we played on the AI. So if you go to the Discord documentation, and you scroll through it, all the way down to, I can see it hard for me to even find, honestly, " When 0:13:09 - 0:13:12 I saw a computer screen with a message that says "please check your email". I found the screen is on. I heard someone say " and you scroll through it, all the way down to, I can see it hard for me to even find, honestly, to the message content intent. " When 0:13:12 - 0:13:26 I saw a man is giving a presentation on a laptop computer. I found a black computer monitor, a man on a screen, a logo on the tv. I heard someone say " to the message content intent. You'll see this was added as of September 2022 as a required field. So in order to receive a message that does not explicitly tag you, you now have to include this new intent in your code. Remember I said intents have changed a lot over time. " When 0:13:26 - 0:13:27 I saw a man is on a laptop computer with a video screen. I found a man wearing a yellow shirt, a white page, a black laptop, a white apple logo. I heard someone say " you now have to include this new intent in your code. Remember I said intents have changed a lot over time. " When 0:13:27 - 0:13:41 I saw a screen shot of a video of a man speaking on a laptop. I found man is wearing a yellow shirt, a white page, a black laptop, a white apple logo. I heard someone say " Remember I said intents have changed a lot over time. This is much newer than the model is possibly able to know. So maybe we're out of luck. We have to debug this by hand. But once again, we can try to use GPT-4's language understanding capabilities to solve this. " When 0:13:41 - 0:13:54 I saw a man is standing in front of a computer screen. I found a man on a screen, a white page on a computer screen, a black laptop, a logo on the computer. I heard someone say " But once again, we can try to use GPT-4's language understanding capabilities to solve this. Now, keep in mind, this is a document of like, I think this is like 10,000, 15,000 words, something like that. It's not formatted very well. This is literally a command a copy paste. Like this is what it's supposed to parse through to find in the middle of that document that, " When 0:13:54 - 0:14:06 I saw a man is sitting at a computer with a laptop. I found man on a screen, a white page, a black wall, a white apple logo. I heard someone say " Like this is what it's supposed to parse through to find in the middle of that document that, oh yeah, message contents. That's required now. But let's see if we can do it. So we will ask for, I am receiving blank message contents. " When 0:14:06 - 0:14:14 I saw a computer screen with a black background and a white screen. I found the screen is white, instructions on the screen. I heard someone say " But let's see if we can do it. So we will ask for, I am receiving blank message contents. Can you, why could this be happening? How do I fix it? " When 0:14:14 - 0:14:15 I saw a computer screen with a black background and a white screen. I found the screen is white, list of ingredients, a black background. I heard someone say " Can you, why could this be happening? How do I fix it? " When 0:14:15 - 0:14:34 I saw a man is standing in front of a computer screen. I found a man on a screen, a white page on a computer screen, a black laptop, a white apple logo. I heard someone say " Can you, why could this be happening? How do I fix it? So one thing that's new about GPT-4 is context length. 32,000 tokens is kind of the upper limit that we support right now. And the model is able to flexibly use long documents. It's something we're still optimizing. So we recommend trying it out, but not necessarily sort of really scaling it up just yet, " When 0:14:34 - 0:15:02 I saw a man in a blue shirt is looking at something on a table. I found man wearing a brown shirt, a mans face, white blinds in the window, a white door, a large green plant. I heard someone say " So we recommend trying it out, but not necessarily sort of really scaling it up just yet, unless you have an application that really benefits from it. So if you're really interested in long context, please let us know. We want to see what kinds of applications it unlocks. But if you see, it says, oh yeah, message content intent was not enabled. And so you can either ask the model to write some code for you, or you could actually just do it the old-fashioned way. Either way is fine. I think this is an augmenting tool, makes you much more productive. " When 0:15:02 - 0:15:03 I saw a man is talking on a laptop computer screen. I found man is wearing a white shirt, a white page on a computer screen, a black laptop, a logo on the tv. I heard someone say " I think this is an augmenting tool, makes you much more productive. " When 0:15:03 - 0:15:17 I saw a man is sitting at a computer with a laptop and a webcam. I found a man on a screen, a web page on a computer screen, a black and white image of a man and a computer monitor, a logo on the black wall. I heard someone say " I think this is an augmenting tool, makes you much more productive. But it's still important that you are in the driver's seat and are the manager and knows what's going on. So now we're connected once again. And Boris, would you like to rerun the message? " When 0:15:17 - 0:15:29 I saw a man is using a laptop computer with a video screen. I found a silver microwave, a man is on the screen, a red and white poster on the wall, a white apple logo. I heard someone say " So now we're connected once again. And Boris, would you like to rerun the message? Once again, we can see that we have received it, even though the bot was not explicitly tagged. " When 0:15:29 - 0:15:45 I saw a computer screen with a picture of a person on it. I found laptop screen is on, a dog on the screen, the laptops keyboard, the screen of the laptop. I heard someone say " Once again, we can see that we have received it, even though the bot was not explicitly tagged. Seems like a pretty good description. Interesting. This is an interesting image, actually. Looks like it's a dolly-generated one. And let's actually try this one as well. " When 0:15:45 - 0:15:46 I saw a man is using a laptop computer with a picture of a cat on it. I found a silver tv set, man on a screen, picture of a cow on the television, the scene is in the kitchen. I heard someone say " And let's actually try this one as well. " When 0:15:46 - 0:15:51 I saw a man is using a laptop computer with a video screen. I found man on a screen, a silver microwave, the scene is in the kitchen, a white apple logo, picture of two people on television. I heard someone say " And let's actually try this one as well. What's funny about this image? Oh, it's already been submitted. " When 0:15:51 - 0:16:15 I saw a man is using a laptop computer with a mouse and keyboard. I found a silver tv set, man on a screen, picture of two bears, a white apple logo. I heard someone say " What's funny about this image? Oh, it's already been submitted. So once again, we can verify that it's making the right API calls. Squirrels do typically eat nuts. We don't expect them to use a camera or act like a human. So I think that's a pretty good explanation of why that image is funny. So I'm going to show you one more example of what you can do with this model. " When 0:16:15 - 0:16:22 I saw a man is using a laptop computer with a mouse and keyboard. I found a silver tv set, man on tv screen, picture of a cow on the television, a picture of a man, a white apple logo. I heard someone say " So I'm going to show you one more example of what you can do with this model. So I have here a nice hand-drawn mock-up of a joke website. " When 0:16:22 - 0:16:31 I saw a man in a blue shirt is looking at something on his phone. I found man wearing a yellow shirt, the man is wearing a yellow shirt, the face of a young boy, a green plant in the corner. I heard someone say " So I have here a nice hand-drawn mock-up of a joke website. Definitely worthy of being put up on my refrigerator. So I'm just going to take out my phone, literally take a photo of this mock-up. " When 0:16:31 - 0:16:32 I saw a man standing at a table with a laptop computer on it. I found a man in an orange sweater, a green plant in a blue bowl, the laptop is black, a long blue table, plant in a white pot, a framed blackboard, a blue bowl on the table, a white piece of paper, the vase is white, man using a laptop. I heard someone say " So I'm just going to take out my phone, literally take a photo of this mock-up. " When 0:16:32 - 0:16:34 I saw a man is using a cell phone on a table with a book. I found a person is using a cell phone, a white piece of paper, hands holding a remote control, a grey laptop, a green plant in the background, a black cell phone. I heard someone say " So I'm just going to take out my phone, literally take a photo of this mock-up. " When 0:16:34 - 0:16:46 I saw a man standing at a table with a laptop and a cell phone. I found a man wearing an orange shirt, a plant in a white pot, a green plant in a blue bowl, a long blue table, a laptop on the table, a black framed picture, a blue bowl on the table, a small jar on the table, the walls are white. I heard someone say " So I'm just going to take out my phone, literally take a photo of this mock-up. And then I'm going to send it to our Discord. " When 0:16:46 - 0:16:48 I saw a man standing at a table with a laptop and a cell phone. I found man wearing orange shirt, a plant in a white pot, a green plant in a blue bowl, a long white table, a laptop on the table, a framed picture on the wall, a big blue bowl, the walls are white, a black coffee mug. I heard someone say " And then I'm going to send it to our Discord. " When 0:16:48 - 0:16:50 I saw a man standing in front of a laptop and a table. I found a man wearing an orange shirt, a plant in a white pot, a green plant in a blue bowl, a long white table, a grey laptop, a framed picture on the wall, a blue bowl on the table, two people sitting at a table, a black coffee mug. I heard someone say " And then I'm going to send it to our Discord. All right. Going to send it to our Discord. " When 0:16:50 - 0:17:02 I saw a man standing in front of a laptop and a cell phone. I found man wearing a yellow shirt, a plant in a white pot, a green plant in a blue bowl, laptop on the table, a long blue table, a black framed picture, a big blue bowl. I heard someone say " All right. Going to send it to our Discord. So this is, of course, the rockiest part, making sure that we actually send it to the right channel. " When 0:17:02 - 0:17:05 I saw a man is using a laptop computer and a phone. I found a silver tv set, man wearing yellow shirt, picture of a bear. I heard someone say " So this is, of course, the rockiest part, making sure that we actually send it to the right channel. " When 0:17:05 - 0:17:12 I saw a man is using a laptop computer with a picture of a cat on it. I found man on a screen, a silver tv set, picture of a bear. I heard someone say " So this is, of course, the rockiest part, making sure that we actually send it to the right channel. Which, in fact, I think maybe I did not. " When 0:17:12 - 0:17:21 I saw a man is using a laptop computer and a phone. I found a silver tv set, man is wearing a yellow shirt, picture of a bear, the scene is in the kitchen. I heard someone say " Which, in fact, I think maybe I did not. Sent it to the wrong channel. It's funny. It's always the sort of non-AI parts of these demos that are the hardest part to do. " When 0:17:21 - 0:17:27 I saw a man is using a laptop computer with a picture of a cat on it. I found a silver tv base, man is wearing a yellow shirt, picture of two dogs, the logo of the company that made the display, a poster on the wall. I heard someone say " It's funny. It's always the sort of non-AI parts of these demos that are the hardest part to do. " When 0:17:27 - 0:17:30 I saw a man standing at a table with a laptop computer on it. I found man sitting in front of computer, a plant in a white pot, a green plant in a blue bowl, the laptop is black, a long table, a black framed picture, a big blue bowl, the man is in front of the window. I heard someone say " It's funny. It's always the sort of non-AI parts of these demos that are the hardest part to do. And here we go. Technology is now solved. " When 0:17:30 - 0:17:31 I saw a man is standing at a desk with a laptop computer. I found man wearing a yellow shirt, a large green plant, a plant on a windowsill, a long blue table, black laptop on the counter, a blue bowl on the window sill, a doorway to the building, a black coffee mug. I heard someone say " And here we go. Technology is now solved. " When 0:17:31 - 0:17:35 I saw a man is using a laptop computer with a video screen. I found the computer monitor is on, a man is on the screen, a white paper with writing on it, the computer is on, a white apple logo. I heard someone say " And here we go. Technology is now solved. " When 0:17:35 - 0:17:42 I saw a man is standing in front of a laptop computer. I found man is looking at the screen, a white and black calendar, a black wall, a logo on the black wall, a black laptop. I heard someone say " And here we go. Technology is now solved. And now we wait. So the thing that's amazing in my mind is that what's going on here is we're talking to a neural network. " When 0:17:42 - 0:17:45 I saw a man is speaking on a laptop computer screen. I found man is wearing a white shirt, a white and black calendar, a black and white image of a man and a computer monitor, a white apple logo. I heard someone say " So the thing that's amazing in my mind is that what's going on here is we're talking to a neural network. " When 0:17:45 - 0:17:47 I saw a man is standing in front of a computer screen. I found a man is on the screen, a white and black calendar, a black and white image of a man and a computer monitor, a white apple logo. I heard someone say " So the thing that's amazing in my mind is that what's going on here is we're talking to a neural network. And this neural network was trained to predict what comes next. " When 0:17:47 - 0:18:00 I saw a man in a blue shirt is making a face while talking. I found man wearing a yellow shirt, a man with a beard, a green houseplant. I heard someone say " And this neural network was trained to predict what comes next. It played this game of being shown a partial document and then predicted what comes next across an unimaginably large amount of content. And from there, it learns all of these skills that you can apply in all of these very flexible ways. " When 0:18:00 - 0:18:03 I saw a man standing in front of a laptop computer on a table. I found a man wearing an orange shirt, a plant in a white pot, a green leafy plant, a laptop on the table, a long rectangular table, a black framed picture, a big blue bowl. I heard someone say " And from there, it learns all of these skills that you can apply in all of these very flexible ways. " When 0:18:03 - 0:18:07 I saw a man is using a laptop computer in front of a screen. I found man is on a screen, a white and black picture, a black and white image of a man and a computer monitor, a logo on the photo. I heard someone say " And from there, it learns all of these skills that you can apply in all of these very flexible ways. And so we can actually take now this output. So literally we just said to output the HTML from that picture. " When 0:18:07 - 0:18:11 I saw a man is using a laptop computer and a video screen. I found man is on a screen, a white and black picture, a black and white image of a man and a computer monitor, a white apple logo. I heard someone say " So literally we just said to output the HTML from that picture. " When 0:18:11 - 0:18:37 I saw a computer screen with a message on it that says "i am not a robot". I found a red wall, a white sign on a shelf. I heard someone say " So literally we just said to output the HTML from that picture. And here we go. Actual working JavaScript filled in the jokes. And in comparison, this was the original of our mockup. And so there you go. Going from hand-drawn beautiful art, if I do say so myself, to working website. " When 0:18:37 - 0:18:38 I saw a man is standing in front of a laptop computer. I found a man is on the screen, the top of a laptop, the scene is indoors, a logo on the black wall. I heard someone say " Going from hand-drawn beautiful art, if I do say so myself, to working website. " When 0:18:38 - 0:18:44 I saw a man standing in front of a laptop computer with a bonsai plant in the background. I found a man wearing a brown shirt, plant in white pot, a green plant in a blue bowl, a long blue table, the back of a laptop, a big blue bowl, a black framed picture, a black coffee mug. I heard someone say " Going from hand-drawn beautiful art, if I do say so myself, to working website. And this is all just potential, right? You can see lots of different applications. We ourselves are still figuring out new ways to use this. " When 0:18:44 - 0:18:49 I saw a man standing in front of a laptop computer on a table. I found a man wearing a sweater, a plant in a white pot, a green plant in a blue bowl, a long white table, the back of a laptop, a framed picture on the wall, a black coffee mug, a big blue bowl. I heard someone say " We ourselves are still figuring out new ways to use this. So we're going to work with our partner. We're going to scale it from there. But please be patient because it's going to take us some time to really make this available for everyone. " When 0:18:49 - 0:19:06 I saw a man standing in front of a laptop computer on a table. I found plant in white pot, a green leafy plant, man wearing an orange shirt, a long blue table, the back of a laptop, a black framed print, a big blue bowl, a black coffee mug. I heard someone say " But please be patient because it's going to take us some time to really make this available for everyone. So I have one last thing to show you. I've shown you reading existing content. I've shown you how to build with the system as a partner. The last thing I'm going to show is how to work with the system to accomplish a task that none of us like to do, but we all have to. " When 0:19:06 - 0:19:08 I saw a man is sitting at a table with a laptop computer. I found man wearing a yellow shirt, plant in a white pot, the tv is on a table, a door with a window, a long blue table, a palm tree is painted on the wall, a black box on a counter, a man in a yellow shirt, a circular design on the wall. I heard someone say " The last thing I'm going to show is how to work with the system to accomplish a task that none of us like to do, but we all have to. " When 0:19:08 - 0:19:14 I saw a man is standing in front of a laptop computer. I found a man on a screen, a white projector screen, a black wall, a logo on the black wall. I heard someone say " The last thing I'm going to show is how to work with the system to accomplish a task that none of us like to do, but we all have to. So you may have guessed. The thing we're going to do is taxes. " When 0:19:14 - 0:19:19 I saw a man is using a laptop computer with a video screen. I found a man is on the screen, a white page, a black wall, a logo on the tv. I heard someone say " The thing we're going to do is taxes. Now, note that GPT is not a certified tax professional, nor am I, so you should always check with your tax advisor. " When 0:19:19 - 0:19:20 I saw a man is talking on a laptop computer screen with a picture of a spiral. I found man is wearing a white shirt, a white page, a black wall. I heard someone say " Now, note that GPT is not a certified tax professional, nor am I, so you should always check with your tax advisor. " When 0:19:20 - 0:19:25 I saw a man is standing in front of a laptop computer. I found a man wearing a yellow shirt, a white page, a black laptop, a logo on the computer. I heard someone say " Now, note that GPT is not a certified tax professional, nor am I, so you should always check with your tax advisor. But it can be helpful to understand some dense content to just be able to empower yourself to be able to sort of solve problems and get a handle on what's happening when you could not otherwise. " When 0:19:25 - 0:19:32 I saw a man is speaking on a video call with a laptop. I found man is wearing a white hat, a white page, a black laptop, a white apple logo. I heard someone say " But it can be helpful to understand some dense content to just be able to empower yourself to be able to sort of solve problems and get a handle on what's happening when you could not otherwise. " When 0:19:32 - 0:19:33 I saw a man is talking on a laptop computer screen. I found a white page, a man on a screen, a black laptop, a logo on the computer. I heard someone say " But it can be helpful to understand some dense content to just be able to empower yourself to be able to sort of solve problems and get a handle on what's happening when you could not otherwise. " When 0:19:33 - 0:19:38 I saw a man is talking on a laptop computer screen. I found man is wearing a yellow shirt, a white and red calendar, the words are brown, a black laptop, a white apple logo. I heard someone say " But it can be helpful to understand some dense content to just be able to empower yourself to be able to sort of solve problems and get a handle on what's happening when you could not otherwise. So once again, I'll do a system message. In this case, I'm going to tell it that it's tax GPT, which is not a specific thing that we've trained into this model. " When 0:19:38 - 0:19:41 I saw a man is standing in front of a laptop computer. I found man on television screen, a white page on a computer screen, a black laptop, black words on white, a white apple logo. I heard someone say " In this case, I'm going to tell it that it's tax GPT, which is not a specific thing that we've trained into this model. " When 0:19:41 - 0:19:47 I saw a man is speaking on a video call with a laptop. I found a man is on the screen, a white page, a black laptop, a white apple logo. I heard someone say " In this case, I'm going to tell it that it's tax GPT, which is not a specific thing that we've trained into this model. You can be very creative if you want with the system message to really get the model in the mood of what is your job, what are you supposed to do. " When 0:19:47 - 0:19:53 I saw a man is using a laptop computer with a video screen. I found man is on a screen, a white page, a black laptop, a white apple logo. I heard someone say " You can be very creative if you want with the system message to really get the model in the mood of what is your job, what are you supposed to do. So I pasted in the tax code. This is about 16 pages worth of tax code, and there's this question about Alice and Bob. " When 0:19:53 - 0:19:58 I saw a man is using a laptop computer with a video screen. I found man on television screen, a white page, a black laptop, a white apple logo. I heard someone say " This is about 16 pages worth of tax code, and there's this question about Alice and Bob. They got married at one point, and here are their incomes, and they take a standard deduction. " When 0:19:58 - 0:19:59 I saw a computer screen with a video of a man on it. I found a silver computer monitor, a man is on the screen, the bottom of the screen is black, logo on the screen. I heard someone say " They got married at one point, and here are their incomes, and they take a standard deduction. " When 0:19:59 - 0:20:27 I saw a computer screen with a few windows open on it. I found a computer monitor screen, a black background, three windows on the screen. I heard someone say " They got married at one point, and here are their incomes, and they take a standard deduction. They're filing jointly. So first question, what is their standard deduction for 2018? So while the model is chugging, I'm going to solve this problem by hand to show you what's involved. So the standard deduction is the basic standard deduction plus the additional. The basic one is 200% for joint return of subparagraph C, which is here. Okay, so additional doesn't apply. The limitation doesn't apply. Okay, none of these apply. " When 0:20:27 - 0:20:28 I saw a man is using a laptop computer with a video screen. I found a man on a screen, a white page, a black laptop, a logo on the tv. I heard someone say " Okay, none of these apply. " When 0:20:28 - 0:20:34 I saw a man is speaking on a video call with a laptop. I found man on television screen, a white page, a black laptop, a white apple logo. I heard someone say " Okay, none of these apply. Oh, wait, special rules for taxable year 2018, which is the one we care about, through 2025, you have to substitute 12,000 for 3,000. " When 0:20:34 - 0:20:48 I saw a man is using a laptop computer in a dark room. I found man is wearing a yellow shirt, a white page, a black laptop, a logo on the black wall. I heard someone say " Oh, wait, special rules for taxable year 2018, which is the one we care about, through 2025, you have to substitute 12,000 for 3,000. So 200% of 12,000, 24,000 is the final answer. If you notice, the model got to the same conclusion, and you can actually read through its explanation. " When 0:20:48 - 0:21:22 I saw a man in a blue shirt standing in a room with a potted plant. I found the man is wearing an orange shirt, a green house plant, the man is looking at the monitor, the head of a man. I heard someone say " If you notice, the model got to the same conclusion, and you can actually read through its explanation. And to tell you the truth, the first time I tried to approach this problem myself, I could not figure it out. I spent half an hour reading through the tax code, trying to figure out this back reference and why there's subparagraph, just what's even going on. It was only by asking the model to spell out its reasoning, and then I followed along, that I was like, oh, I get it now. I understand how this works. And so that, I think, is where the power of this system lies. It's not perfect, but neither are you. And together, it's this amplifying tool that lets you just reach new heights. " When 0:21:22 - 0:21:36 I saw a computer screen with a picture of a person on it. I found silver and white laptop, man is looking at the camera, logo on the screen. I heard someone say " And together, it's this amplifying tool that lets you just reach new heights. And you can go further. You can say, okay, now calculate their total liability. And here we go. It's doing the calculation. " When 0:21:36 - 0:21:42 I saw a computer screen with a message that says "please wait". I found the screen is on, text on the screen, a black background, the black bottom of the screen, the list of the same direction. I heard someone say " It's doing the calculation. " When 0:21:42 - 0:22:13 I saw a man standing in front of a laptop computer on a table. I found a green plant in a blue bowl, plant in white pot, man wearing a brown jacket, the back of a laptop, a long rectangular table, a big blue bowl, a framed blackboard. I heard someone say " It's doing the calculation. Every time it does it, it's just amazing. This model is so good at mental math. It's way, way better than I am at mental math. It's not hooked up to a calculator. That's another way that you could really try to enhance these systems. But it has these raw capabilities that are so flexible. It doesn't care if it's code. It doesn't care if it's language. It doesn't care if it's tax. All of these capabilities in one system that can be applied towards the problem that you care about, towards your application, towards whatever you build. " When 0:22:13 - 0:22:21 I saw a man in a blue shirt is making a face while talking. I found man wearing a yellow shirt, two men in a room, man holding a wii controller, the head of a man, dark green leaves. I heard someone say " All of these capabilities in one system that can be applied towards the problem that you care about, towards your application, towards whatever you build. And so to end it, the final thing that I will show is a little other dose of creativity, which is now summarize this problem into a rhyming poem. " When 0:22:21 - 0:22:31 I saw a man is using a laptop computer in a dark room. I found man is wearing a brown shirt, a white and black note, a black laptop, a white apple logo. I heard someone say " And so to end it, the final thing that I will show is a little other dose of creativity, which is now summarize this problem into a rhyming poem. " When 0:22:31 - 0:22:41 I saw a computer screen with a menu on it and a few buttons. I found a white and black laptop, text on the screen, a black background. I heard someone say " And so to end it, the final thing that I will show is a little other dose of creativity, which is now summarize this problem into a rhyming poem. And there we go. A beautiful, beautiful poem about doing your taxes. " When 0:22:41 - 0:22:51 I saw a man standing in front of a laptop computer on a table. I found a man wearing an orange sweater, plant in white pot, a green plant in a blue bowl, the back of a laptop, a long blue table, a big blue bowl, a black framed picture, a black coffee mug. I heard someone say " A beautiful, beautiful poem about doing your taxes. So thank you, everyone, for tuning in. I hope you learned something about what the model can do, how to work with it. And honestly, we're just really excited to see what you're going to build. " When 0:22:51 - 0:22:52 I saw a man standing in front of a laptop computer with his hands up. I found plant in white pot, a green leafy plant, a man wearing an orange shirt, a long white table, back of a laptop computer, a big blue bowl, a framed picture on the wall. I heard someone say " And honestly, we're just really excited to see what you're going to build. I've talked about OpenAI evals. " When 0:22:52 - 0:22:59 I saw a man standing in front of a laptop computer on a table. I found plant in white pot, a green plant in a blue bowl, a man wearing an orange sweater, a long blue table, a grey pot on the counter, a small jar on the table, a big blue bowl, a framed picture on the wall. I heard someone say " I've talked about OpenAI evals. Please contribute. We think that this model, improving it, bringing it to the next level, is something that everyone can contribute to. " When 0:22:59 - 0:23:12 I saw a man standing in front of a laptop computer on a table. I found a plant is in the background, a green plant in a blue bowl, a man wearing an orange sweater, a long white table, a grey box on the table, a framed picture on the wall, a big blue bowl. I heard someone say " We think that this model, improving it, bringing it to the next level, is something that everyone can contribute to. And we think it can really benefit a lot of people, and we want your help to do that. So thank you very much. We're so excited to see what you're going to build. " When 0:23:12 - 0:23:28 I saw a plane flying in the dark with a light on the side. I found photo was taken at night, the sky is black. I heard someone say " OpenAI. " When 0:23:28 - 0:23:30 I saw a plane flying in the dark with a light on the side. I found photo was taken at night, the sky is black. When 0:23:30 - 0:23:50 I saw a plane flying in the dark with a light on the side. I found photo was taken at night, the sky is black. I heard someone say " OpenAI. " When 0:23:50 - 0:23:52 I saw a plane flying in the dark with a light on the side. I found photo was taken at night, the sky is black. I heard someone say " OpenAI. " When 0:23:52 - 0:23:55 I saw a plane flying in the dark with a light on the side. I found photo was taken at night, the sky is black. When 0:23:55 - 0:24:05 I saw a plane flying in the dark with a light on the side. I found photo was taken at night, the sky is black. When 0:24:05 - 0:24:15 I saw a plane flying in the dark with a light on the side. I found photo was taken at night, the sky is black. When 0:24:15 - 0:24:19 I saw a plane flying in the dark with a light on the side. I found photo was taken at night, the sky is black. I heard someone say " OpenAI. " When 0:24:19 - 0:24:20 I saw a plane flying in the dark with a light on the side. I found photo was taken at night, the sky is black. I heard someone say " OpenAI. "