World Models & General Intuition: Khosla's largest bet since LLMs & OpenAI

Introduction

0:00Hi, listeners. As you may know, I recently wrapped up the AIE Code Conference in New York, and while I'm

Introduction

0:01traveling, I do like to visit top AI startups in person to bring you interviews that you don't find on any other podcast that just does a Zoom call. General Intuition, or GI for short, is a spin-out of a 10-year-old game clipping company called Metal, which has 12 million users, but in comparison, Twitch only has 7 million monthly active streamers. Metal collects this data by building the best retroactive clipping software in the world. In other words, you don't need to be consciously recording. You actually just have Metal on in the background while you're playing, and you hit a button to clip the

0:34last 30 seconds after something interesting happens. It's very similar to how Tesla and self-driving does bug reporting if you've ever done a self-driving bug report in Teslas. The result is that Metal has accumulated 3.8 billion clips of the

Peak Human Behavior

0:47best moments and actions in games, resulting in one of the most unique and diverse data sets of peak human behavior actively mining for the interesting moments. They were also very prescient in navigating privacy and data collection concerns by mapping actions to these visual inputs and game outcomes. As you saw on our Fei-Fei Li and Justin Johnson episode with World Labs, and with the recent departure of Yan LeCun from Meta, there's a lot of interest in world models as the next frontier after LLMs to improve on spatial intelligence and to work on embodied robotics use cases.

1:19DeepMind has been working on this with Genie 1, 2, and 3 and SEMA 1, and 2. And this year, OpenAI still finally agree because they have been depending on LLMs a lot, and they made the news by offering $500 million for Metal's video game clip data. Our guest today, Pim, turned down that money and instead chose to build an independent world model lab instead. Kostler Ventures led the $134 million seed round, which is Vinod Kostler's largest single seed bet since OpenAI. We were able to get an exclusive preview of GI's models, which unfortunately we cannot show you directly.

1:52But I can confirm they're incredibly human-like, and we chose to include the first 11 minutes of the demo discussion, even though I couldn't show it to you. It may be hard to follow, but I tried to call out what was noteworthy for you to know as your likely reaction if you were watching along with us. Now enjoy the world's first look at my first look at Genial Intuition. So what I'm about to show you is a completely vision-based agent that's just seeing pixels and predicting actions the exact same way a human would. And so yeah, what I'll show you here is what this looks like four

2:28months ago. So this was, so again, this is just an agent that's seeing, that's receiving frames, and it's just predicting action. So you can see it has like a decent sense of, of being able to, you know, navigate around. It tabs a scoreboard, just like gamers always tab the scoreboard. So these are purely, these are pure imitation learning. I see. So the C is slicing the knife. Yeah, exactly. So it's doing everything that like humans would. In this case, here's, here was the first interesting part that we saw, like it gets stuck and then it

3:01has, they have memory as well. So you see, it can get unstuck. How long is the memory? Four seconds. Yeah. Four seconds. Okay. So this was four months ago. This was maybe a few weeks after that. So you can, you can see there's like, it's still doing the scoreboard thing, but it's, they're still, they're still quite like, and these are bots too. So you can see that. It's very human. Let's just say that. Yeah. And then, right. So this was really like the early days of research where you can see, right.

3:31It does one thing and then it goes for another. And then we've been scaling right on, on data and compute. And also we've just been making the models better. And this is where we are now. So what you're seeing is pure, like I said, pure imitation learning. This is just a base model. There's no RL, no fine tuning. This model sees no game states. It is purely capable, not sequence, acceptance. It's purely predicting the actions from the frames. That's it. And if this is playing against real humans, just like, like a human would play.

4:03And it's also, it's running completely in real time. So there's absolutely, everything here plays exactly like human. Do you give it a goal? No. It just figures out it's a goal because obviously it's trained on the same. Yes. Um, and I, I, I picked, right. I picked the sequence where also it doesn't do well initially. So you can see like, this is just, this is just like a sequence, a random sequence. But this is the, I mean, it looks like it's very well. So, um, oh, okay. Yeah. Watch. Yeah, that's pretty good.

4:35Maybe too good. Um, this is my favorite part. So you can see it does something that like, um, here, like human would never do this, then gets unstuck, then has four realizes, which, and then in the distance. So you're saying one, it makes a mistake that a human would never make, but it unstacks itself. And two, what we just saw is it is doing superhuman things. Yeah. Okay. Yeah. Um, I mean, there are things that, that, that demons said, obviously. Um, but because it is trained on, on the highlights of things that all the exceptional things, it's inherited those.

5:08Yeah. So it's not like move 37 where we are all their way into something, but it's yeah. We're replicating a superhuman. Yeah, exactly. Or like peak human.

Metal and Data

5:18The baseline of our data set is peak human performance. Yes. Yeah. Um, okay. So that, that's the agent. Uh, so now I'm going to show you is we then are able to take those action predictions and we're able to label any video on the internet using those actions. So, um, um, and so this is, this is just frames in actions out. Yellow is the, uh, model prediction or sorry. Yellow is ground truth. Purple is the model prediction. And then bottom left is a compound error over the entire sequence.

5:49And then this is reset per prediction. Reset meaning every now and then you reset. Yeah. So this just means it resets the baseline. Um, and so this basically a single error in the entire sequence compounds here, but it doesn't compound here. If that makes sense. Yeah. Um, so, and again, this is just seeing frames, right? It's, it's not, it's not seeing any of the, any of the actions. Um, and so, you know, so what we did, right, is we, we, we trained it on less realistic games

6:20and we transferred it over to a more realistic game. And then, and this is where it gets really exciting. We transferred it over to a real world video, which means that you can use any video on the internet as free training. What was it predicting? Um, it's predicting it as if you were controlling and using keyboard and mouse. So if you were, if you were basically playing this sequence as the human. Is there some sense of error or? Uh, so that's why you, you transfer it to more realistic games first.

6:52Yeah. And then you transfer it to real world video because you can't get a sense from ground truth from, from the real world video yet. Um, let's see. And then, um, so we don't, they don't, so I'll show you here. Uh, this one is also, um, this is the same, uh, uh, agents that I just showed you. This is playing against other AIs. This one's playing against bots. Yeah. Um, the previous one was against players. Uh, but with the sniper, it doesn't really matter that much, as

7:22you'll say. It's like, uh, so one, one thing that's really interesting is you notice that it behaves differently as it has like different items, right? That makes sense. Yeah. I think there's also a question about egocentricity versus like, so the third person. Yeah. Doesn't matter. Um, the third person I think will be very, very helpful if you're, for instance, trying to control multiple objects in an environment later on. Uh, right now, I think having fully in perception, uh, first person is quite helpful. Um, this one's also, this is the policy itself.

7:54What do you mean this is the tallest? The agent. Yeah. Same for the strengths that I just told you about. Yeah. Like this where, right, where it hides, that to me was just incredible. Like just from, from knowing, being able to, to predict. The appearance also hides when you see it. Yeah, exactly. Yeah. Yeah. Um, and it needs the spatial intuition to go, well, this is hiding and that's not hiding. Exactly. And, and, and right while it was reloading. Yeah. Um, okay. So that, uh, so those are, uh, that, that's the policy and this is a completely general recipe,

8:29meaning we can scale this to any environment. Uh, is this work closest? Okay. Now let's, let's keep going on demos until, um, I was going to go into research. Yeah. Yeah. That sounds good. Um, okay. So, and then this is, this is these, so what I'm about to show you are the world models. Um, there's a few really, really interesting parts about our world models. So the first is, uh, we actually made the decision to, uh, transfer. Um, sorry, we made the decision to, um, pre-train world models from scratch, but also we've actually been able to

9:03fine tune open source video models to get a better sense of physical transfer. Um, and so one of the things that you'll notice here is like our world models have mass sensitivity, which is something that like gamers absolutely want. Right. So you can have these like very rapid movements, which you couldn't do in any other world model. Um, and so this is a holdout set. So this clip was never seen before, um, at training time. Um, as you can see, it has, it has a spatial memory. This is, this is about a 22nd ish, uh, generation.

9:34And here's what's fascinating. This is an explosion that occurs and you can see that in the, um, in the physical world, right, the camera would shake and in the game that would never happen. So you see, you see the world model inherits the physical world camera shake, but the, the, the actual, um, uh, uh, game never does that. Uh, which, which is, which is sort of that, that to us was quite fascinating. Right. Also did the models that I just showed you that we used to transfer over from video. The two of those combined will allow us to like push way beyond games in terms of, in terms of

10:08training. Um, this is another interesting, so this is a world model. This is rapid camera motion. So like, again, this is stuff that we're literally just taking one second from here in the context and the actions and replaying it here. Right. Um, and so you, you'll, you, you never essentially have, um, uh, like what we're saying is the skill that you see in the clips that like that speed and the movement that also pays off at training time when you're doing world models. Um, this is my favorite example. Uh, so this shows that the world model is capable, um, of performing with partial observability.

10:40So what you're going to see is, um, uh, again, you're replaying the actions from here in here, just using a one second of video context, everything after that is completely generated. Um, so what you're going to see is the model is going to encounter in this case smoke. Normally now models break down, but you actually see is comes out in the same place. Um, and so it's capable of, um, uh, even with partial observability, still maintaining, um, uh, its position in the world. Um, and then here it is also interesting. So this is, uh, this is sniping.

11:10So this is give, this gives you like a, um, reaction time. Uh, like the fact that it can do depths and like sequences in completely different views. Right. So this is a completely different view than if you were to be outside of that view. Right. And so it's, it's, it's able to maintain consistency. Um, while zooming in. Yeah, exactly. Um, uh, and so, um, yeah, so you can see, uh, so even while this goes out of scope, right, watch, and then it can, and then it comes back and you'll see it's still, it's still there. Yeah. Um, and so, uh, yeah, this is the work that, that Anthony, who has been

11:44working on. I'm just wondering how much game footage you have to watch in order to find these things. Um, we can ask Anthony, you know, it's, it's, uh, I'm, I'm sure he's not going to be too excited to play these games afterwards. Uh, you're not playing it, right? You're just watching. Yeah. Yeah. Yeah. Um, great. Okay. So those were the models. Um, let's see. These are interesting. So we also were able to distill into like really, really tiny models. Um, so this is for instance, uh, um, a long sequence on a very, very tiny one. You can see it makes like a bit more stupid mistakes.

12:17Uh, like it, it does things that are not as optimal. Um, but I haven't seen anything yet. Uh, at the beginning it was running into a wall for free. Exactly. Um, uh, I mean, I do that too. Yeah. Yeah. Um, it's looked, I mean, it's doing pretty well. Yeah. And, and again, all these models are running completely in a real time. There's no, uh, I was thinking your main model does real time anyway. What's the goal of distilling? Is it cost or? Uh, yeah. Parameters. Yeah. Yeah. Yeah. Yeah. This is the industry one. It peaks the corner. That's what we mean by like the space and the poor reasoning aspect is that humans

12:52actually, they, they sort of simulate the optical dynamics of their eyes and how to actually space. It's interesting. All the data. Yeah. Right. You've seen all this. Yep. Um, exactly. And so, uh, like even in like real, this is kind of interesting, even in like the real world, um, with, uh, for instance, YouTube data, right? You have to first solve for pose estimation. Then once you have pose estimation, maybe you do something like inverse dynamics, right? Where you basically are able to like somehow label some of the actions that you're seeing. And then you still have to account for optical dynamics of like where your eyes actually looking before the decision.

13:26Cause like there's just three levels of information loss. Or when you're playing video games, you're actually simulating the optical dynamics with your hand. Right. And I think that like, that's why I think why games are a better representation of switch support reasoning initially than, um, uh, than YouTube videos, for instance. Okay. We're in the GI offices with the CEO and the way. Welcome. Oh, thank you. Thanks for having us in your office. Yeah. Excited to be here. If I'm in New York and you're one of the hottest races of the year, I have to come and visit and, uh, thanks for taking some time on the weekends or.

13:58Yeah. Yeah. So you've raised 133 million C, uh, so general intuition, most people don't fear about you. Like, I guess this GI is new, but more, uh, more gamers would have been in the middle. Indeed. And before that you ran, I mean, probably the largest depth, we say, um, what's your reflection on just that, that journey of life now you're an AI founder and you started off like Rootskate. Yeah. I think, um. So I grew up with Tourette's. Uh, I, uh, spend most of my time as a teenager coding and playing video games. Uh, so in that sense, it doesn't feel that much different.

14:29Um, but I think for, uh, so yeah, so I started the largest privacy of Road RuneScape, worked at Dr. Subwardess for three years, first in Ebola, and then on like satellite, satellite based map generation for disaster response, um, uh, which was already like very AI related adjacent. I built some models back then and then started, uh, metal, which became one of the largest social networks and video games. I've always been kind of like AI, like adjacent. You know, I, I, I, I, I'm a self-taught engineer. Uh, so for me, the modeling itself always felt a little foreign. I actually had to, uh, take a ton of tons of classes over the summer and, and, and early this

15:03year to get better at it. Uh, because I, it still felt like, like I was really, really good at the infrastructure side and I had written like our, our, our transcoders for metal myself. So I was very, very familiar with CUDA and like the GPU side and all the video infrastructure that we were using, uh, for this stuff. But the, the modeling side itself was, was still quite foreign. Um, luckily, obviously we have, we have really, really good co-founders, but they, they, they essentially put a bunch of coursework together for me to, to go complete, to get really, really good at understanding the fundamentals better.

15:34I think for me, I had seen inside of the labs that I had really, really good, uh, leadership with fundamentals on top and also the ones that didn't. And I think the ones that did were just like much better. Um, and so for me, um, yeah, I want it to be more like that. So in that sense, it was a bit, it was first very foreign. And then now I feel pretty comfortable with everything. And, but yeah, like the thing for, um, uh, there's a lot to be explored starting in video games and also reverse engine. Like, I think the interesting thing about reverse engineering is it kind of teaches you to look at problems very

16:09differently. It's like the ultimate form of deductive reasoning in a way. Uh, and so, um, uh, so this is just how I think, how I operate. And so for me, it's, it's been a really, really interesting journey. Uh, you know, I don't claim to have any of the credentials or, or skills that some of the other guests do have had on, but hopefully it will make for a good time. Yeah. Well, your co-founders, uh, definitely bring a lot of that definability and you bring a lot of the, I guess, gaining expertise, mostly what I bring to the table. Just, just a little bit of history of metal. Let's establish metal for those who don't know.

16:43Uh, the lady, uh, Twix, yeah. The year, um, that's, you have more active users, concurrent users in Twitch, something like that. Yeah. On the creator side, I think. And the reason is because metal is a lot more like Instagram than it is like Twitch. So people, um, so the way, the way you think about metal is it's, it's a native video recorder, like unlike something like Twitch, where you actually have to use other software to record and stream to Twitch. Um, it's not a streaming software. It's actually a video recording software. And a lot of gamers love to put things like overlays on top of

17:13their videos. Um, and as a result of that, we have sort of the largest data set of ground truth, action labeled video footage on the internet by maybe one or two orders of magnitude. Yeah. Well, what was an example of an overlay play? The only overlay I usually think of is like the case CAD. Yeah. Yeah. Also, um, controller overlays, for instance, if you're playing, um, like, let's say you're playing, uh, console. Yeah. Like flight simulator, you get like, you know, the joystick and all the things. So you get the actual actions that people take inside the games as well as the frames of the games themselves, which is a loop, right? Because it's essentially you perceive, then you act, then there's a state update, and

17:47then you perceive again, you act, state update, which is like roughly precisely what you use in order to trace, to train these agents. Yeah. It's, it's, uh, almost perfect training data. We were showing, you were showing me in the demo and we've shown some B-roll here on, uh, how you don't love Q. Yeah. It's very important to use a lot of action. Yeah. When did you figure this out? Oh, um, maybe starting a year and a half ago. Yeah. And, and we realized that like figuring out the side of the research for us was we very much never wanted to be in a position where we eroded privacy or something like that.

18:17So we never wanted to actually log like a W or A or S and a D, which for researchers, the fact that we don't do that, like often it sounds strange. Like why wouldn't you do that? But I think for us, the privacy is going to get the data. Yeah. I think, you know, a lot of, a lot of the, the, um, the researchers that did, they hadn't quite understood yet that you can actually just get away with just doing the actions. Um, and the reason is like at training time, having the actual keys is always anyways. Like if there is sex in the screen and you would want to, in theory, uh, make that, um, part

18:49of the training, then like reading texts from a frame is like really easy. And so for us, if we actually can, so we convert, basically hit, you hit the, uh, the input, we convert it to the actual action. So we had thousands of humans label every single action you can take in every single video game over the past year and a half, uh, which is an enormous amount of action labels. Um, yeah. So when you act, you, we, we get the actual, um, action itself. And then it being said at training time, you, you can for like, uh, the general set of that, of that game, uh, convert back into computer inputs if you want to, but you can never do it for any

19:22individual person. And so that for us from, from like a design perspective was, was important. So we, we, we figured all that stuff out. Then we actually started pushing, um, like we already had features as well with this. So for instance, like gamers already love to be able to navigate their clips by like things that happened. So we have an events capture system and then we also have the overlays where you actually just want to overlay and render the actions on top of your clip. We developed kind of in tandem with the feature set itself. And then obviously when, when world bottles became a thing and it's very, very clear that all the, all the

19:53data for this was precisely like that sequence. Yeah. We were able to sort of be first to market, recruit the best researchers and start a lab. Yeah. And that's a, that's terrible. So, uh, one more question on metal before I remove photos of the DI. It's been 10 years. Yeah. What is the, I don't even know how you bro something like this. Do you know what I'm just kind of curious and like the opportunity to ask you, what really worked? Yeah. Um. That you became so, so huge because I'm, you're not the only one. Yeah. But, uh, I'm sure it's performance and everything, but. A few things that really worked. I think the first was a lot of our competitors were focused on solving

20:27the social network and a recorder at the same time and that never like our bet was really that we could get so many people to record with us that we could bootstrap the network on top of that. And that worked. So while everyone was sort of distracted trying to bootstrap a social network, we were just focused on building a really, really good capture tool. And then we got tens of millions of people to use that, which then we are with a bootstrap a network on top of the share behaviors. We already had like the profile behaviors and the share behaviors, obviously, but the actual content consumption piece and the sharing piece really only came after we hit critical mass.

20:59It was actually early days during COVID when like the network really accelerated. Fortnite happened, which was really important. And I think also the fact that discord existed, um, uh, made it quite a different time than, uh, when other types of networks of these types had launched because discord essentially was like the connective tissue already between gamers that like never really existed before. And so I think those combinations of things really, really made it. I think we also built a product that for instance, with, with most video recorders, you have to remember to start and stop the recorder. So you have to go into the application, then hit start, then start your game.

21:31And then, um, you know, maybe you'll play games for three hours and you'll close the game. Then you have to close your video application. Then you, well, then you have to process like a multi gigabyte file. Uh, then you have to upload those somewhere. And so like, this was a pain for people. And so what we did is we just ran this kind of recorder. When you hit that button, it does a retroactive video record. So all the recording initially is in memory. And then when you hit that button, it exports only that sequence to disk and syncs it to your phone. And so that, that became super popular.

22:03It also was interesting about it also means that you're not sort of behaving or acting differently because it's always there and you can just export whatever happens, which is also very, very helpful for, for trading, obviously. Um, the thing, you weren't the first to do that. Yeah. The thing you were explaining just before this was, it's similar to how Tesla does the bug reports, right? You're driving and from the having disengaged autopilot, they're like, well, tell us what happened. Exactly. Exactly. See, see, you're driving. Tesla doesn't want to train on the like 10 hours of you driving through a desert where nothing interesting happens. You have the clip button on the steering wheel.

22:36Something interesting happens either while FSD is engaged. And I'm not sure if you can use it without FSD as well, but you hit the clip button and it basically uses that precise sequence to mark, which is then more helpful for training because it's more unique as a training time. Yeah. Yeah. I mean, so one thing we're going to get to this on the inside, one thing that I, that does, that does pop up as well. A lot of life is boring. A lot of life is going from a, a lot of playing games is doing the boring stuff that is not capable. Yeah. Somehow you see the generalized fight. Yeah.

23:08Yeah. Yeah. It makes you think, right? It makes you think. Yeah. Yeah. It's also quite interesting. Like I showed you the models, like what happens when you increase the size of the context window and how behaviors actually are largely shaped by the size of the context window. Yeah. That, that to me was like one of the most interesting parts about the research made me think about our own behaviors in a way. Yeah. Let's talk about also the like forming a chain. On your website, you have 12. I don't know if that's changed now. I meet four, three co-founders. Yeah. And just let's talk about how this team comes together, because you may not,

23:41if it's your self-taught, you don't have that at the end of the network. Yeah. What do you manage the elements, people? Yeah. I started reading all the research papers. By that time, I was already pretty deep into like having a decent understanding of not world models, in particular, in particular LMs and transformer based models. And so there was Genie, there was Sima. Those two were really, really interesting. And Sima in particular was interesting because what they do is they basically take 10 games and then they, they have a graphic in Sima where you can see kind of the precise actions that are inside of those games that they mapped. And I believe they found something like a hundred, which are

24:16actually actions that also exist in the real world. And what they did was they then, I believe it was specifically for navigation. They did a 9-1 holdout set. So they, they, they trained an agent on the nine games and then they had to play the 10th game, the holdout game. But then they also trained a specialized agent just on a 10th game and they compared how good they did. And if I recall correctly, it did roughly as well playing the 10th game on navigation specifically on the holdout on the nine game agent than it did on the one game agent. And that's where it was really interesting because that's precisely the type of data that we had.

24:51Right. And so for us, the thinking was, okay, what if we did exactly what LLMs did? What if we use, right, this, right. So LLMs were trained on predicting like text tokens on words on the internet. What if we predict action tokens on essentially what is the equivalent of the common crawl data set, but for interactivity. Vision input. Yeah. Action output. Correct. That's it. Well, I think we'll actually, I'm going to double back a little bit to like a question I hand, which is one of the, one of the reasons why I thought you would want to prefer keyboard and mouse over actions is the action space is potentially unbounded, right? You can jump, walk left, walk right, but then also look

25:26up, look left to the bench. It's unbounded. So it's huge, isn't it? Yeah. I think. Yeah. There's, there's benefits to the action space being small to start with. So I think we're going to start with anything that you can control using a game controller. But yeah, long-term, we want to actually predict maybe like action embeddings and have models sit inside a general action space to be able to transfer out to other inputs as well. Yeah. Okay. And then let's see going on the, on the research side. So Gini, Simba. Yeah. And then the co-farmers. Yeah. So there was the diamond paper, there was Gini, and then

25:57there was Simba. The diamond paper for me was really interesting because they had actually managed to get this world model called diamond running on a consumer GPU. I believe it was a 4090 at 10 FPS and you could play it. And they did that on like 90 hours of data, like 95 hours. I think it was 87 hours. And I think eight and the whole that set or something like that. That was just incredible, right? That they had something playable on that little data. So I actually cold emailed the entire group of students and I was, and I told them, Hey, I think

26:28we have this thing. And then it was pretty interesting. So like right when that happened, a lot of the labs also started understanding what we had. And so we started very aggressively, multiple labs tried to bring us in in various ways. And they were part of that. Like they basically were seeing that happen. And I think for them that also kind of like solidified how real it was. And then when we chose to do our own thing, you know, initially we thought that we were going to have to just work on world models, right? So we thought, okay, the main benefit of this data set is

26:59like Gini is world models. What we didn't realize at the time is that we have so much of this data is that we can essentially do these world models in parallel and take the equivalent of like the LLM bet mostly on imitation learning and then use the world models after that to get into like our all stage. Right. And so for us... And eventually get rid of the world models. This is something that you can... I mean, ideally you get rid of the imitation. Yeah. The imitation learning. But yeah, we essentially realized that we could get so far on just imitation learning. The way to look at it is we essentially... Like let's take the LLM analogy.

27:33We essentially have sort of the internet or like common crawl, if you will. And every single lab is trying to simulate that, right? In order to get similar data, in order to train their agents. And so for us, the reason why we say independent and we just said our own thing was we think we can essentially leap every single company that's forced to either be consumers of world models or build world models and take this foundation model bet for spatial temporal agents and be in a place where we have a lot of customers years before any of the labs even get there. And maybe the most similar comparison is like what Anthropic did with code, right?

28:07Anthropic just focused really, really hard on nailing the code use case. Their models are incredible for it. A lot of their customers use it for it. So we just want to become incredible at this spatial temporal agent use case. And likely that starts in like game simulation and then using world models, we can then start expanding out to other areas. So would you show me a little bit of how you think it does generalize out? Yeah. Things. But although games has come to come in prayer. Yeah. Games and simulation. I would specify it as game engines in VertiKiller. So even if you're, for instance, simulating human behavior in Omniverse because you're trying to create better training data for

28:41factory floors, you can use it. Yeah. Maybe Meta has a similar data set because of the quest. I never really asked them. I never really looked into the Meta quest specifically. So you need a few things. You can't just, like there's lots of companies that have like maybe recorders, but you also need the public graph. Otherwise you can't train on the data, right? You can't train on people's like private videos that they have saved somewhere. Right. And so I think you need the social network graph components because these videos need to be on the internet. To rank? No, to train on them. Yeah. I mean, I think generally people don't want to train on

29:15like, because these things, they live on your device usually, right? Yeah. And you can't train on anything that lives on your device. Like you actually need to go and upload and do your thing, right? For Meta specifically, I think also VR, the scale of VR is still pretty small. The amount of environments in VR that have like consumption at scale is probably in like the hundreds. Whereas on PC, it's probably in the tens of thousands, right? And so you get a lot less diversity. The three-dimensional input space of VR is pretty interesting. We see some of this too, obviously. And so, yeah, I do suspect, you know, Meta starts using these types

29:49of things, but it's unclear to me whether they can get to like a similar scale of data or diversity on the environments as we can. Yeah. There's a lot of challenges there. Yeah. Okay. I want to take this and make a few different things. But I guess let's fill up the papers. Maybe one more to mention is Tyre. Yeah. Which I actually interviewed at Tyre offers, but that too seems like the particular insight that brought it overseas. Yeah. So Anthony Tu, who led the research on Gaia Tu, is also one of the engineers that joined our team. So it's all the Diamond, the core contributors for Diamond and then Anthony.

30:23And we just had three more researchers join this week. It's been a good week. And yes, I think a lot of the approaches in Gaia Tu were heavily inspired by Diamond. And then Vin Sa, who was one of the authors of Diamond, also already was at Wave by the time that I emailed them. Anthony also realized what this was and realized that, you know, you could scale world models to a much larger scale and decided just to make the leap as well. So I think everybody that sees the dataset makes the leap because it takes a while to wrap your head

30:54around it because it's like, oh, it's video games, right? Like intuitively, it doesn't make sense. And then when you actually understand and you see, right, how we've been able to transfer it to a physical world video and things like that, then it makes sense. And then everybody tends to jump with that. Don't call it video games. And then, yeah, if I lived in San Francisco, maybe I would. Yeah. Just a quick note, because we actually cover all these papers in the latest day Super Club. SEMA 2 did not seem to have as much impact on SEMA 1. And I don't really know why. They did it a lot more.

31:28G3 had a ton of impact. And, but I also felt like because you couldn't play with the model or people, it just seems they have an extension of all those days. I guess any quick takes on SEMA 2, GNA 3, which will report this year as well. Yeah. I'll talk about SEMA 2. The steerability of SEMA 2 was to me the most impressive part because lighting up the action sequences and the text conditioning is quite hard to do, right? And so that, and the fact that they were, like, it's also quite interesting that it means that they can

31:58sort of use Gemini as part of the flywheel, right? Where you can sort of scale this orchestrator as, like, an independent, almost like a puppet master, if you will. And then, like, in theory, Gemini could orchestrate many instances of SEMA, right? That to me is the most interesting part is where I tend to agree with this, where, like, I think our models will initially be used as, like, you'll have, like, an orchestrator VLM of sorts that's kind of, like, managing instances and instructing them. And I think sort of SEMA showing that you can do this was fascinating. Also, the fact that you could, they didn't just have text conditioning, but they also were able to do, like,

32:34drawings and markings of where to go. They really took an interesting end-to-end approach to me that I look forward to seeing a lot more of. Are you talking to them like you said it? Is that the one collaborative? Yeah, I think that, yeah, we're very friendly with DeepMind. We like them a lot. I just saw the team not too long ago. And I think, you know, big fans of their work. The fin line that I kind of shade from Alice Heath's coverage of you. Yeah. Is, uh, you are the biggest bet that Vino Kostler has made since OpenAI.

33:04Yeah. How did that conversation start? Okay. So, for now it's style, and maybe I'll get slapped in the fingers for revealing this or whatever, but, uh... Forgive me if I forgot. Um, is, he asked you to, like, draw a 2030 picture of your company. And I think he just picks N plus five years. Whatever. I don't know. I did the same to you. Yeah. Um, he asks you to, like, walk that back from first principles all the way from today. And, and, and he asks, he expects you to do that flawlessly, where he can challenge any assumption, any part of the vision that, that, and he asks you questions, right?

33:36He has a very technical background. He also has a bunch of technical people on his team. And he truly backs people that have these, like, very large visions on that vision and the ability to defend it alone. Um, and that's what he did for us. Um, and I think that's why I made that bet. So, I think also through this, uh, through, through this question, he, he, he gets to know a lot of things about how technical you are. He gets to know how well you think from first principles, because if that, if that vision is not connected to something real, it's very easy to suss it out by asking good questions.

34:09Um, and then, and then he just backs fully. I think like he, he really gets in your corner, um, if it's the right fit. And yeah, they've, they've been incredible partners. They, they, they've opened so many doors for us. I had to ask the question. I think it's just like, it's a, it's a very notable story, uh, obviously in a lot of work went into it and, and, but it's also worth it and come out of side. For sure. One of the things I also wanted to, I think I kind of asked this question out of sequence, but, um, one of the things that's exciting about talking to you is there are a lot of people

34:43like you who are founders of business and businesses that along the way have a ton of data and yours happens to be highly valuable. You pursue it before deciding to do an independence journey. You can also talk to other companies about potential licensing or acquisition and stuff like that. What is your learnings from those periods? And so also like one, one version of this is very simply, how do you value data? Yeah. I don't think you can value it unless you actually model it yourself and see what the capabilities are. That's my, that's my real outcome. You say model, but change the model. Yeah. But that's obviously like not doable for, for everyone.

35:16Um, and also I think my general advice would be as model capabilities increase, you, and models are also like, you know, these VLMs out, they're all very, very good at labeling as well, generally. Right. What I was afraid of when I was having some of these conversations was okay. Like, you know, as the capabilities increase, you're just going to need less, uh, ground truth data. And like, you can do more model based data generation or synthetic data generation. I would recommend if you're going to do large data deals, like just try to get like a large chunk of equity in the company that you're doing it with.

35:47Um, if you can now, a lot of them won't do this, but I think, uh, that to me would, or just go do the research, figure out what's actually possible. Well, in our case, we were quite lucky in the sense that this is actually the foundation data. Right. And I think, right. Like that's not true for, for every data set. I think, you know, we just happened to hit a particular gold mine. And, but you also did, you read Cooperative, you did the action thing one for five years ago. Yeah. So you, you did work. Yeah. That's the thing. Like you, you have to be grounded. Right. And I think a lot of the, um, and, and I think

36:21that's the hard part. And I think a lot of what's interesting is you can also kind of look for if like scaling laws already exist on your data type, which like for video, there were some, but for these like input action labeled, uh, sets there, there really wasn't any. The other question is like, does it go into LMS? Does it go into a world models? Does it go into like, what type of model is it going to be used for? And I think that's an important thing to know. And so I just want to, you know, if you, if you're having these conversations with labs about data, just

36:53like make sure that you actually understand like what it's going to be used for. Cause that's a very, very good way for you to like make the decision yourself about what are you want to pursue that. Now, a lot of them won't tell you that. And I think, you know, I think in, in, in that case, you generally just don't want to do it because like, I think, I think for our case, like we really cared that like, for instance, there weren't going to be competing products with game developers built, right? Cause we didn't want to like bite the hand that feeds us. And I think we are part of the games industry.

37:25So those questions I think are normal. And then we eventually decided, you know, you just have the data. We're just going to go do it ourselves. And that's when the rest happened. Yeah. And he assembled the team. I can, uh, think about it. I feel like that's, you've aligned a lot of stars in order to make GI happen. Yeah. That other data founders, they're at the beginning of the training. Yes. One data founder. Founders who happen to have data, but they have a main business, right? I don't know if they're there. There's two sides to this, right? There it's really easy to be super naive about it. And like, I had a lot of people tell me initially, oh, it's not that valuable.

37:59You're just like making this up. And, and, and so for me, like doing the work and actually understanding it myself was a really, really big part of, of building that confidence and go start the company. But a lot of times it is true that like model capabilities increase so quickly that like the certain data you just don't need anymore. Yeah. Um, and so I think it is, it's really important to like get people to do the work such that you can make these types of distinctions. Yeah. And, and, and so, so my recommendation would be go build models with your data, see if you can create any sort of capabilities that, that aren't clearly already there or on path to being there and then figure

38:33out, um, where you go. Yeah. I did want to ask this earlier, but he gave me the opportunity to, uh, we say do the learn thing, do coursework and all that. And your co-founders gave you some homework. Yeah. Uh, is this like some books? I mean, Coursera. No, this was, um, Francois, uh, Fleurais, uh, Fleurais. So he has a little book of deep learning and then he also has a full course that he's published, um, uh, on his website. I went through the entire course, uh, over the summer. I believe it's like something like 30 or 40 lectures, which also take home projects and things like that. Um, and I would recommend anybody, uh, uh, does this and it goes through right history of deep learning, like

39:07the topology. It takes you through, um, the literary algebra, the calculus eventually end up with like chain rule. And by this time you've, you've done like all the, the, the more important concepts, it takes you through how do you create neural networks using, uh, using these concepts that you've learned. Wow. This is super first principles. This guy. And I've, I've, I've had the, the, uh, opportunity to spend some time with him as well. He is one of the most first principles people I've met in my entire life. I'm convinced, like I actually asked him, why did you do this course? He said, Oh, cause I thought all the other courses weren't right.

39:38And because, because he is so first principles and he can only explain things from like everything you see and how he explains the saying it's everything is from first principles, including like the history of deep learning itself was part of, of the course. And, um, yes, it goes, uh, um, so all, so he goes through everything and then, uh, and by the end of it, I think you're like, I now have like a pretty good intuition. Understanding of how everything works, but obviously still, right. Like I like to describe it as, um, I'm like the, the guy who just got his driver's license. I can drive the car. And like my co-founders are like the F1 drivers that like have done this for years.

40:12They know where all the, um, uh, where all the, the gaps are. And, and so I, I enjoy getting to learn from them. The cool thing is also that work models is just like a very, very new space. And so, you know, I, I got to bring ideas to the table that like, you know, one thought of, and not because I'm great at this, just because it's such a new space that like people just haven't tried it yet. Um, so let's get a hit on definition. Yeah. What are world models to you? You know, in a video model, you might predict the next likely sequence or the next most entertaining, the next most entertaining frame.

40:44Um, what world models do is they actually have to understand the, the full range of possibilities and outcomes, um, from the current states and based on the action that you take generates the next states, right? So the next, the next frame. And so it is a, it is a much more sort of complex problem than, than traditional video models. So to me, it is, it is a world that is accurately generated based on the actions that you take as a result of what's already been generated. But just to fact check, uh, that is, it needs to understand physics. It needs to understand if I'm building a type of material, you need the power, it interacts with some type

41:17of material. Yeah. I think the interactions is the most important part. I think the reasons why world models are so fascinating. One of the things that I did when I was studying over the summer was I tried to actually build a super rudimentary, um, PyTorch based physics engine, which I would not recommend writing a physics engine in PyTorch for obvious reasons, but I wanted to be able to, um, cause it's, it's differential. So you can, uh, you can, you can generate the, uh, yeah, exactly. You can, and then you can, um, uh, uh, train. And so I wanted to, you know, I got so many people ask me about, you know, why aren't you

41:51just using, uh, uh, why aren't you just simulating or generating this data? Um, and I really wanted to understand from first principles why. And I think the most important thing that I figured out was the compute complexity of simulation goes up really, really rapidly with three variables. First, the numbers of agents in an environment. Second, uh, they're DOF. So, uh, they're individuals of freedom. Yeah. And then third, the information that each action reveals. Um, so like, um, for instance, if you, if you have a tech, if you have a text action or a speech action, the environment can change so much based on whether you say, right, water or fire, that the

42:25outcomes are going to be completely different of like how a human would behave in that type of situation. And so it goes up so quickly with those three variables that at some point you just hit a point where you just want to maximally bet on either video transfer or generation of these environments using world models. Because that type of stochasticity is just incredibly difficult, but it's already very, very present in a lot of the video pre-training, uh, that goes into, into these world models. Right. And so I think for, for us, it is more so about making a maximal bet on video transfer

42:56and interacting with things that are difficult to simulate. And the steerability is also really interesting with text, uh, than it is on betting against simulation or something like that. And so I think there's still a large market for, for traditional simulation engines, specifically in areas where video is really hard to get. Is this exactly what the big labs are also saying when they're talking to that? I honestly haven't talked about the big, to the big labs. Like since we started working on them ourselves, I think people are more reserved with what they share with us. Yeah. Of course. It can make sense. That's a formula question.

43:28How would you contrast your version of world models with, with the lead? Yeah. Yeah. I'm a fluent. Yeah. So I don't know exactly what Yankun is doing today. My understanding, it's based on the Fijepa, like the Jepa approach, which is, so I'll start with Fei-Fei Li. I think what's really interesting about Fei-Fei Li's approach is that you in some way are able to reuse the, the, um, the spots, right. In game engines and in things that let you stay in verifiable domain, um, which I think is a really interesting approach. Um, however, my understanding is they're currently not interactive, which in my opinion

43:59is like the whole point of, of world models, right. It's, it's environments, they're great environments. And I think from a business perspective, I think they, they, they picked a really important part of the tool chain, but to me, that's not really, uh, a world model, but I, my, my guess is they'll get there, right. They'll, they'll start generating. Yeah. They just haven't been reused. Yeah, exactly. Exactly. And I think, right. Fei-Fei is one of the like founders of the entire space. Um, uh, so I think it's going to be really interesting to me on, on, on what maybe that interactive

44:29piece looks like for me to really judge their approach. I, I, I think we interviewed just before we moved to Yann, uh, we interviewed her with Justin Johnson, uh, her co-founder. He was, he was more focused on the physics side of things and the interactivity. Uh, they just haven't finished it yet, but I, I, I, I do think that basically that the splats, if you just add more dimensions on, I guess, the forces acting on them, then, then you get to interactivity out of the box. Cause you have basically, these are virtual atoms that then has all the global physics applied to

45:00them. Yeah. I'm, uh, I'm excited to see what that looks like when they actually release it. It's really hard to really hard for me to comment on anything. I really like the, um, uh, the, the, the frame based approach, um, because all of our video or all of our training data is in this format. Yes. Oh yeah. So there we, we actually asked them about this and they were like, yeah, it's possible, but we're choosing the spider for, yeah. Yeah. And you can also go from splat to frames, right? I'm sure you can write like at some, it's, it wouldn't be easy. Like you'd have to actually render out the environment, do the, so sure it's not, it's not going to be

45:36a simple problem, but like in theory, it has to be something that you can do if you really want it to. So I could, cause it's almost like having a more sort of grounds for three dimensional representation of the underlying world. Yeah. Right. So I think it's an interesting approach. Um, it might be overkill, right. Uh, uh, you're also dealing with like a much larger, like degrees of freedom on the output space. Right. So, so who knows how well it scales. I like the fact that like, I think these video models also use things like auto encoders, right. You can actually have the world models predict like much smaller, um, uh, maybe like a Rhythm of the Student

46:10or Saris. Yeah, exactly. And then you can use like diffusion upscaling or methods like this to actually, um, uh, enrich. And so I think that world models just allow a much more, or world models in my sense for a much more like controlled space that, that, that we know really well. Um, I'm not suggesting their approach is wrong. I'm just, you know, like, this is, I think what we really like about it. Honestly, Jan's podcast that he did, I don't remember which one it was, but a long time ago where he, where he basically proclaimed LLMs to be a dead end, um, uh, was one of the things

46:41that inspired me to do this. I think this is very consensus around world models. We will basically, everyone that has this is like stops with the LLMs and just goes through to world models. I would say that the main perspective, I asked this exact question to Nolan Brown from OpenAI and he was like, well, they learn implicit world models, right? So there's basically that we didn't see an inclusive and there's this and B, uh, or what are you on and put it down here, everyone? Yes. So yeah, I, I, I'm not one to proclaim LLMs or dead ends personally. I think, um, I think they're actually quite useful in particularly as orchestrators.

47:13Like the way I think about it is as humans, right? We had sort of a three-dimensional worlds. Then we invented text as like, uh, in a way in compression method, right? So you had, we invented text in order to communicate with each other in a, in a common way, uh, in a, in a way that actually compresses all this information that we are perceiving in three dimensional space into just like a single sequence. And I think that allowed science to emerge, right? It allowed so many literature, like so many, uh, parts of the world that we, that we charge. So I think it's a critical part of, uh, of the whole picture.

47:45I also agree that, that, uh, it's very, very clear that they do build sort of the internal implicit world models, uh, inside LLMs. Um, and so I think there'll be very helpful as things like orchestrators. Um, the, the problem is when it comes to the generalization, I think text as a generalization backbone, when most of the, the, uh, um, when most of the, the, the, the pre-training is, is, is, is text, right? Or, or, or, or largely text sequences, then I think you want that backbone to be kind of more spish temporal in nature. And then also just have text, like as one of the, as, as part of that.

48:17And I think the actual argument of, um, of LLMs is also, for instance, the autoregressive nature of the prediction itself. So the, um, the fact that it's running the entire output, right, through the transformer and then in order to predict the next token, which doesn't like the environment in the real world is continuous, right? It's always, it's always changing and LLMs kind of just forget about that, right? I think a lot of the, the, the argument isn't the first, right? So I think the, the fact, the fact that like text doesn't necessarily generalize well to sufficient temporal, um, context and then the autoregressive nature of the prediction and using text for that, right?

48:50So I think those are the, those are the two main arguments. Um, I think, I think text prediction is just one of the actions that is going to come out of, of these, you know, these, these policies and world models. I think speech and text generation will just be one of the actions that, that can, that can be a part of that. I think that there will just be labs coming at this problem from both sides. Um, and everyone ends up in roughly the same place and the same place will be whatever people think is cool, uh, right? Like whatever the consumer, whatever is closest to EGI.

49:21Yeah. And so I don't think there's like a clear answer. I think it's really interesting to come, to come at it from the world modeling side, but it's also because we have to, right? Cause like text is largely commoditized. We can import all the text. I think it's interesting and tempting. Yeah. Like I'm attempting, it makes sense that you can probably recover. It's sort of like you, you're taking a step back. You're starting your branch of the ML research sheet, but you might actually just end up recovering all the other text stuff emergingly. Yeah. Yeah. We can import a lot of that research, right?

49:51It's a lot of that. That's really cool on the, on the research side. Let's talk about the stuff that GI is producing more like, I guess the sort of research and products output. You mentioned the word customers. What are your turn customers? Yeah. So we're already working with some of the largest game developers in the world. Yeah. Uh, we're also working with game engines directly. And so really what we're doing at the moment is replacing essentially the player controller inside of a game engine. So anything that you're currently, that maybe like behavior trees or things that you're deterministically coding, we hope to replace with a single API, which is just, you stream us frames

50:25and we predict actions and that can be inside an engine or it can be, um, eventually even inside the real world. Hopefully those are then also steerable. So the models that you saw weren't text steerable yet, but I think we want to get to a point where they're fully text steerable. Well, it is steerable and it's like, well, I want you to both to share or figure out anything else. I don't agree. Yeah. I think it's, it's sex conditioning on the generation. So yeah, the ability to, to, you're right. We want to get to a point where you can generally, and that's why it's called general intuition, where we can sort of can mimic the intuition of all these gamers into

51:00human-like behaviors in any situation. Um, as I mentioned also, the lab is named after the, the, the, the, the code from alpha fold, which is, wouldn't it be amazing if we could mimic the intuition of these gamers who are, by the way, only amateur biologists, um, on his path to, um, he tried to get an AI to train, fold it, to generate a lot of data for, for alpha fold. And so for us, really the, the North star, right, what we hope to get to one day is being able to represent scientific problems in three-dimensional space and then have a space in the world agent capable of perceiving

51:32that space and using hopefully also the, the, right, the text reasoning capabilities that LLAMs have today, in addition to the space in the world capabilities to be able to work on the other side of that problem. So that for us is, is sort of the North star. That's why, you know, we're, we're sort of trying to be hyper-focused space in the world workloads, the same way that Anthropic was hyper-focused code. And use that to then get into organizations and expand from there. Yeah. Just as a side note, since you mentioned Anthropic, um, any idea what they did on this, uh, to, to solve for India? No. Out of any lab, I probably know Anthropic the least.

52:06Yeah. I admired him though. Yeah. Well, the, the, the current working theory is that they had a super lucky, um, roll of the ducks, but we'll, and then he compounds from there. That sounds like a nice story. I'm sure it's not that. Yeah. Okay. So, um, why do the game developers want this? So if you're a game developer, how well you're actually retaining players is like, um, if you have a game that's already at scale, it's like decently dependent on how good your bots are. So if you're logging in at an obscure time, let's say 3am in America and your player liquidity is low,

52:37then you need really, really good bots to keep those players engaged. Is this known? Is this a thing? Yeah. For sure. For like 4day and whatever. A lot of good members do this. Yeah. Um, and so if, if you're like, as a human, do I want to play against bots? Usually it's not just bots. It's like players can extend with bots because you don't want to play just against bots, but it's better to have a full game than to have like an empty game. Yeah. Um, and so I think as long as it's part of the environment, I think it's okay. That means you also have to sort of grade that skill level.

53:10Yeah. Yeah. Which we can do. Um, cause we have, we know exactly how good people are at these games. Yeah. Yeah. I think for us, um, bots is kind of like step one, uh, right? So what I, what I was showing you is we're building a general agent that can sort of play any game, um, in real time, but really that extends into all of simulation, right? Like in GTA 5, for instance, people are genuinely role-playing real life, right? And so they're actually behaving in quite aligned ways with, with the goals they set for themselves. So you have all these examples represented in video games, right? You have truck simulator, power wash simulator, power wash simulator, where like actually the behaviors that you'd want, uh, an

53:45agent to be able to perceive, they're all there. Okay. Yeah. It's really, it's really like how seriously some gamers take truck simulator. Um, did he haven't seen these tips? You should watch it. Yeah. Uh, they buy the whole like truck driving set and they're doing the job of a truck driver. Yeah. What I mentioned to you, we have more people at any given time on metal playing with steering wheels and like truck simulator and these types of games than Waymo has cars on the road. Um, it's a ridiculous stat, but it's true. Yeah. I mean, it's, so, you know, I, I used to think that while it's a soft, soft driving, you kind of just need to play along the GTA 5.

54:19Um, yeah, I mean, it's not, it's not bad for this. Yeah. Yeah. Our bet is not that we can zero shot any of these things. It's just that like the next self-driving company can maybe have collect 1% of the data because right. Also, for instance, clips already self-select into negative events and adversity. Right. And so like a lot of our data set, cause already the highlights is, is really, um, precisely what a lot of these companies spend like their last 20% doing. Right. And I think that's the main argument. If you're, if you're, if you're another company that's looking at what we're doing, I think the thing that people are not, that people won't understand is that anything that you're currently

54:52doing in pre-training, as long as your robot can be controlled using a game controller, we hope that we can move that to post-training for you. So our bet is not that we can create the next self-driving car company. It's just that the next self-driving car company hopefully only needs 1% of the data or maybe 10% of the data. I don't know. Right. To be able to deliver a really good product. Yeah. It's also the, the term that comes to mind a lot is active learning. I don't know if you've, but you said identify with that. It's got less cool for a bit and now it seems like the only uptrend, uh, which, which obviously you

55:24have the best data set for the supply intensity or you said negative, but feeling for negatively, it could be a negative ballpark of it. Yeah, for sure. I think negative events is just because it's the most common term that people use for like, if you're, if you're Tesla, you want the crashes, you want like, right. Um, yeah, right, right, right. But, but it's only gaming. It's both. Yeah. So, you know, the model that you saw obviously had really, really incredible moments and that was largely, yeah, that, um, uh, that it had a large representation of people at their best. Yes. Yeah. And worst. Yeah. Yeah. Amazing. Okay, cool. Uh, and you have anything else on the customer development side

55:57that you want to sort of flinch off? Yeah. Um, uh, we're also already working with robotics companies, but again, that, and manufacturing, but the key is that the robot has to have gaming inputs. So we're like, our bet is not that we can transfer over to like higher DOF robots and the keyboard and mouse. It's, it's really just that we can move the hard work of, of, of pre-training, hopefully to post-training. Yeah. It's kind of like the foundation model that is a very good basis to start. Yeah. You're going to straight, you're going to give us frames and, and likely some text. Or you'll license the model to because they've been a wonderful training.

56:29Yeah. Our, our business model is initially going to be an API. I got like the Anthropic API. Um, but you also saw, for instance, some of the video labeling models that we've been able to develop. So, um, the goal is for any company to be able to take in their, uh, their video data as well. And we can create first, obviously custom versions of the policy for you at the agent. Um, if that doesn't work, then, um, we, we've already working with a customer that, that is doing, we distill a model and they, uh, turn that into a product for themselves. So people can engage with you on the agent level, API level, people can engage with you on the sort

57:02of model level. Can you also buy data? No. All right. Yeah. We don't sell data. Okay, cool. So that's the, that's the business. Um, and is there a world in which, I mean, I, I think this is on the landing page. If you are, you know, Frontier Labs for, for world models, is there a world in which there is a more sort of application layer thing that you, that comes out, like a chat GPT for whatever. Yeah. You're going to see us launch a few things on, on metal itself that are going to blow your mind, uh, as a result of this, this, um, this agent. So I'll, I'll leave to the imagination for now. If people took a grade out, you know, and yeah, on

57:36the world modeling side, like I think one people underestimate is that metal is already one of the largest, you know, video consumption platforms as well. People watch millions and millions of videos a day. Um, so, um, world model based entertainment and things like that. Well, it's not like a focus for us right now. I think we'll be like on the consumer side. We have the ability to move very, very quickly here, um, and, and get it integrated in a way that I don't, I don't think anyone else can. Yeah. You could theoretically do a video gen, like a Sora, like a, and what is, what is that Isabelle? What's the meta one? And meta meals? Not, not Reyes.

58:08Um, yeah. You could theoretically generate clips that nobody play, but you know, it's a device. I think for us, the games being so human centric is like a really big part of what makes it special. Like I actually, I actually just don't think that would work. Like one thing that we are really excited about though, I'll give you one sneak peek of what we're thinking about is what if you could literally replay any of the clips that you have inside a world model or your friends can play them. Like I showed you a model that already took part of your clip as a contact. Since the replay entered that world. But it's also how we go from imitation learning to RL, right?

58:41Cause like it's part of our research rope app anyways, to make every single, every single clip on metal playable. Um, so, uh, yeah. Who's, who is to say that that doesn't apply to just the actual clips that you take? Yeah. Yeah. Can you say more about the RL potential? We describe metal as, as the episodic memory of humanity and simulation. So when you take a clip, really the way to think about it is you get the highlight of what is maybe three hours of playtime, right? You maybe get like two to three minutes of the things that were the most out of distribution, right? It is genuinely your episodic memory, um, of that playtime and simulation, the things

59:13that you most want to remember and share. We want to be able to load, uh, and this is the work that Anthony who is doing. The reason why we built world models is every crash that you run into and you're a truck simulator or American truck simulator or a driving game. We want to be able, right. And again, these are ground truth labels. So we know precisely the actions that lead up to the negative events. Um, they're also title labeled when people upload it onto their platform, they say, oh good, it's a crash. Right. And so we can select all these events and if we can put them inside a world model, we

59:44can go into, right. We can, um, uh, we can train reward models to then, uh, reward based on how you perform in clips that actually contain negative events, for example. And so for us, it's, it's very much about, um, uh, right. We can, we can create this, this, this, this like LLM moment on, I've been going to invitation learning, but actually making every single clip on the platform playable, um, at billions of clips scale is how we go from invitation learning to RL. Cool. Uh, we covered a lot of it. Uh, is there anything else that you want to do before we to the grab all the, the, the long term vision stuff? Yeah. Yeah. I think, I think for us, um, this is a very, very ambitious long-term vet.

1:00:19We need the best researchers in the world that, that, that, that want to work on this stuff. Um, it's really exciting not being extremely data constrained. Um, like we really get to, like we get so many learnings every week that we didn't think were possible and it makes it for a, for a joy working here. Also, the other thing is because we have such a large data modes, we don't have to be as concerned as the LLM companies about publishing because we don't need. The ones that would be able to exactly, no one can replicate the models. Right. And so for us, um, we really want to bring back like the original culture of, of open research.

1:00:51This is why we did the partnership with Qtai in France. Um, I actually didn't, didn't, yeah, we just did a, um, we just announced our partnership with Qtai in France, which is, which is an, an open science lab in Paris, one of the best research labs in the world. Um, Eric Schmidt, I believe funded in addition to some, some French people, they are essentially acting as the partner that's currently doing a lot of open research on the data. We also want to partner with universities who, um, because like, we do believe this is the frontier, but it's so data constraint that really no, everyone has their hands tied behind their back right now.

1:01:22And so we want to help fix that. So for instance, um, we want to work with universities to build like negative event prediction models for maybe like trucks in India on all the truck data where all these crashes occur. We have all these things that we know we can do that we just have it at the time to do. Um, and so if, if you're listening to this and you're, uh, maybe an academic institution or something, and you want access to some of this data and a research, um, in an educational research fashion, I think we're, we're quite open to doing that because we want to educate people. And, uh, yeah. And other than that, we just want to work with the best infrastructure and research engineers on

1:01:56the planet as we're going into scaling, you know, runs that have thousands, tens of thousands, eventually hundreds of thousands of GPUs. Yeah. Yeah. Amazing. Uh, I primed you this as like the closing question. Yeah. Um, like it's a little bit of a no cost that three, three, 30 percent. I didn't know. Yeah. So what does GI become by the risk? Yeah. In 2030, we want to be the gold standard, um, of intelligence, uh, and any sequence, uh, long enough is fundamentally spatial and temporal, right? Which I think is, um, so by nailing spatial and temporal reasoning, you go after the root killer problem of intelligence itself. What the world looks like is we want to have eight.

1:02:28So I sort of group, um, the sequences of AI in three stages. And I credit Andre Kaparthi for, for teaching us, um, bits to bits, atoms to bits and bits to atoms, and then atoms to atoms in the atoms to atoms stage. I want like, I want GI models to be responsible for 80% of all the atoms to atoms interactions driven by AI models. Uh, uh, and, and, and the reason, the reason for that is because we were able to unblock intelligence so quickly in robotics, like intelligence is the bottleneck that supply chains actually converged on gaming inputs as their, as their primary input methods. And, and they converged on essentially simpler systems that let us do a lot

1:03:00more, a lot quicker. So we are essentially the, the 80% market approach. And then you have lots of companies that have kind of like specialized, maybe humanoid robot OS stacks that are, that are the other 20. And then, so I want to be responsible for 80% of all the atoms to atoms interactions driven by, uh, by these models and be the goal center for intelligence and maybe a hundred X more in simulation. Cause I think simulation will actually be the larger market initially. So I think in simulation, um, because you have very little constraints, uh, also from a safety perspective, simulation is much easier. So I think a lot of the takeoff initially sits in simulation.

1:03:32So a lot of the simulation use cases, like what I mentioned, scientific use cases, I'm really, really excited about. And so, um, yeah, 80% of atoms to atoms interactions, uh, coming downstream from these types of spatial and the Borel foundation models, and then a hundred X more in simulation. Yeah, that, yeah, it reminds me a lot of that, uh, what Mark and Priscilla from the Chad Zuckerberg Institute are doing with virtual biology, cause you can do a lot of pre-simulation and you can do, yeah. Or you can do it a lot faster, uh, with interest. Um, amazing. Thank you for inviting us to your office. Yeah. And thank you for sharing a little bit about your training.

1:04:04Thank you. Yeah. Yeah.