7. Hyper-scalable simulations and MMOs (Rashid Mansoor, Metagravity)

My name is Michelangelo, and I'm an investor at Stride. Welcome to Chaotica, land of chaos. Our aim is to explore the future of human experience as we dive into the virtual age. My guest today is Rachid Mansour, founder and CEO of MetaGravity. He went from being a doctor of research in topology of complex and chaotic systems to starting AdBrain, which he then exited to The Trade Desk, after which he went on and co-founded Hadean, the world's most advanced hyperscale real-time compute and distributed cloud game engine, head-to-head against Improbable. His pedigree in building hyperscalable simulated virtual worlds brings him to his latest quest, MetaGravity, which just closed its seed fundraise at around $10 million. With MetaGravity, he claims that it will be possible to have thousands, hundreds of thousands, if not millions or an unlimited number of concurrent players interacting with each other and with AI objects, Something that was thought of impossible as of today. For complete transparency, I am a tiny angel investor into MetaGravity, but of course, I'll try my best to stay inquisitive and unbiased. Thank you to the underground community of thinkers and sponsors of the show, Citizen of Tajikistan, for the continuous support. Follow the signal. Thank you very much for coming on. In the meantime, do we want to— I think a good place to start would be to start from your story, and you are a fanatic, if we can say, of complex systems. I would be extremely interested in understanding how this interest developed and where it led you today.

Yeah, that's a great question, and thanks for having me on board, Michelangelo. So regarding complex systems, I think this goes right back to my earliest attempts also writing programs. So we had a 386 computer at home, That was sort of one of my early encounters with computers.

Mm-hmm.

I started learning how to program initially to build games. And I don't know if you're much at all familiar with the 8086 architecture, modern PC architecture, but it was never built for games. There used to be other architectures like the Amiga, for example, which was much better at games. But the one thing that this processor was good at is raw instructions, so MIPS, millions of instructions per second. And the limitation was, how do you do graphics at a very fast pace? Like, how do you output graphics you have as animation? It was not designed to do that for gaming. It was designed more for user interfaces and simple things like that for business applications. So learning how to write these applications to be fast required delving deeper into the hardware. And that's something that got me very interested in how the hardware, the internals work, because you had to kind of hack your way in order to make real-time graphics appear on the screen on these machines. At around the same time, I read Jurassic Park because I'd always had an interest in dinosaurs. And in that book, say, unlike the movie, like Ian Malcolm, Jeff Goldblum's character, gets a little bit of time to be philosophic about the future prospects of Jurassic Park. But in the book, he goes a lot deeper into it. And then the second book goes much more into complex systems. That's kind of what sparked my interest. I was about 10 years old when I read Jurassic Park. But that meant that I, started reading loads of other books on the topic. And one of the books is Mitchell Waldrop's Complexity. It was a book that clearly even Crichton had read when he wrote Jurassic Park, because a lot of the ideas in the book directly come from this. So combined with my forays into game programming, these two kind of combined. I started imagining creating simulations of ecosystems and understanding how complex systems in nature might work using computers.

Mm-hmm.

And that got me really, really excited about the idea of computer technology over time being able to simulate reality, to compute reality, essentially, to understand it, to predict it. That was a fascination of mine, but it also made me more interested in the underlying mathematics of it. And it's something that I had wanted to study at some point in the future. How do you understand complex systems? How do you predict nature? How do you compute nature? Because nature is a complex system at the end of the day. We ourselves are complex systems. We build ourselves, for example, from simple local rules that come together in how proteins interact, how cells organize, but you get an individual, a human being, and you get some— Or a dinosaur. Or a dinosaur, yes, absolutely. And the brain, this again is an example of a complex system. I had around this time also read books on neural networks. So this is Stephen Wellstead's book on neural nets, which is quite a mathsy book on neural nets. And it had a lot of interesting kind of ideas And I began to understand, like, if the future of AI was ever going to happen, it would have to come from local simplicity going to global complexity with something like machine intelligence. Whereas if you go back to the '80s and '90s, the AI topics that were sort of front and center were not neural networks at the time. They were like expert systems and so on, because these were more useful in pragmatic cases. Neural nets were a very interesting curiosity, but they were still very much a toy.

Hmm.

But the fact that they produce complexity from local rules, to me, kind of spoke deeply to me. And so it was something just I was very fascinated by as a consequence.

Complexity from local rules, you mean, what do you mean by local rules?

So this is the hallmark of a complex system. A complex system isn't necessarily a complicated system. Often a complex system is very simple. It's how when you have local rules that come together and interact in a very local manner, you have global behaviors that you could never predict by looking at the local rules. So the age-old example given, like, from the '70s to '90s is taking the watch apart and seeing how it works doesn't work for complex systems because usually it's the interactions of these components that create complexity. If you look at a neuron, and if you didn't know anything about the brain, you would not be able to predict that the neuron would produce cognitive capabilities of the type it has.

Right.

If you look at an individual ant, You wouldn't know that when you put many, you know, thousands of ants together, you would get the equivalent behavior, the global behavior of an ant colony.

Right.

So they're able to solve problems that are very complex, but if you take them apart, very simple individual components are very simple and they're behaving very locally and not globally.

Interesting. And so reality around us is full of complex systems, but What are complicated complex systems if complex systems that are made by individual atoms that are very simple are still called complex?

Yeah, so complicated would mean that the system itself is very complicated. And if you look at an expert system, that's an example of perhaps a complicated system that's not a complex system. Whereas if you look at a neural net, it's a complex system that, yes, arguably is complicated if you look at a modern large language model. But that also probably suggests to us why we haven't quite cracked it yet. Because if you look at the brain, there are neurons that themselves are quite complex and complicated. But if you look at how a lot of cognition and intelligence arises, it's collections of large numbers of neurons that they themselves can do none of the work that the brain can do, or a complex neural circuit, like say the visual cortex can do. So, but if you look at how humans build things, we complicate things quite a bit. almost every other attempt at AI. Like, let's take IBM's Watson. It was a very complicated system to play Jeopardy. And if you look at modern LLMs, that's a simpler system, especially like some of the newer models like Alpaca and LLaMA that have been coming out. They are relatively simple, but they're far more powerful than some of the complicated AI systems that we were building in the '90s and the 2000s and so on. Right? So it's complexity that comes from simplicity is what makes that a hallmark. And it's a very interesting phenomenon. Like in nature, we find much of that. Often when we look at a system that looks complicated, it's often not. Like starling swarms are a good example. You look at starlings in the sky and the way they swarm, they produce such intricate patterns. And for a long time, we didn't understand how they worked. Philosophers even thought that maybe there's some sort of concept of a group soul that Wow. Makes them behave in such synchrony. But we realized '70s, '80s during the study of complex systems that actually each individual starling, the bird, is following a simple set of rules in how it steers based on its neighbors. And then you get these beautiful complex patterns that arise, as well as the ability to make these wind gusts that can knock predators on their back as a result of their group behavior.

Although what we are observing really is the output, and we're kind of deducting what the basic rules of engagement are, right?

Well, the reason we know that this is how it works is because we're able to simulate it on a computer. So Craig Reynolds wrote a paper, I think it was late 1980s, called Boids, where instead of, like, you know, people looking at bird flocking behavior or insect swarms assumed that they were very complicated in the rules they follow. But Craig Reynolds was able to break it down into 3 simple rules that you could essentially write in 3 lines of code, which is around steering or alignment, rather cohesion and separation. And when he implemented these 3 rules, you have these dots that come together and suddenly start to swarm and fly around the screen like a swarm of birds.

Interesting.

And to do that in the '80s was a remarkable step forward when everyone who'd looked at birds before and insect swarming had assumed there's some sort of complicated magic going on here behind the scenes.

Interesting, because now we're looking at simulation as a proxy or as a mean of understanding the world around us, which also leads us to what you're building today with MetaGravity, which is essentially a hyperscale simulation engine. But I'll let you explain it and let you make the bridge between these complex systems and MetaGravity.

Yeah. So I wouldn't quite describe it necessarily as a simulation engine. What we are doing is we're working— so today, if you look at game engines, they've done a good job of building a lot of the features you need to build an interesting, entertaining simulation. And that the games have tried to approximate nature to an extent.

Right.

So what MetaGravity is really doing is looking at the limitations of game engines. And today, individual game engines, you have single-player games and you have a story, but then you have multiplayer games and you're interacting as part of a world. And that can kind of be pushed to the extreme. We have games like Fortnite, which have about 100 players in a shared experience where you're playing competitively for a period of time. There have been games from the late '90s and early 2000s that have tried to push this even beyond and have these big immersive worlds that you're all part of. But there's a limitation to how many players you can have in any single kind of causally self-consistent reality. You can only have about 100 players. And by causally self-consistent, what I mean is, so if I move this cup, then for 100 other— for 99 other players, this cup should have moved.

Yes.

As far as their reality is concerned. You can't continue to extend that arbitrarily because at some point your CPU can't compute that world without slowing down. And if you are a game, you have to do it in real time, so you can't slow down. So you set a limit on how big your world can be.

Yes.

Your alternative is to find a way to parallelize that, but that's a very difficult, I would argue, open problem.

Parallelize that? What do you mean by parallelize?

You need to distribute that compute. Like, instead of having a single world where that reality is self-consistent and all running on one CPU, you need to be able to do it on multiple CPUs, multiple server nodes, and so on.

Oh, so then it becomes hard to align and make sure that everything runs smoothly at the same time in a coordinated manner.

Yeah. So it's not even just about it running smoothly, but how do you really take a simulation of reality and break it down into multiple smaller subprograms? And then if you do that, how do you make sure the subprograms are self-consistent, that they are representing the same view of reality in their sum rather than incongruent realities across them?

Right. And the reason that is important is because we want the experience to be consistent across gamers. And you can see that in the pinnacle of gaming esports. You can't have 2 different professional gamers have slightly different experiences of the same reality. Because that can lead to an advantage or disadvantage.

Yeah, even the slightly different advantages, let's say, even if they were forgivable, you'd have much bigger breaks in reality. You would have a wall that's broken for one person, and for another person, it's not. And therefore, you're seeing someone pass through a wall from your perspective, or you'd see an object, you'd stumble into a chair that someone else moved because it's not moved from your perspective. So the fabric of virtual reality would break down as we scale out these worlds if you're not able to do it self-consistently. So what we're doing with MetaGravity is finding a way to scale those experiences without causing those breaks. We're not doing that by building a completely new engine, essentially. We're not building our own simulation engine. We've found ways to make that work, that parallelization of these worlds work with existing engines. And that's actually where a lot of the power comes from. Because if you look at these game engines today, like Unreal Engine and Unity, they've multi-deca-billion-dollar companies, and they've invested an enormous amount of effort in building these very rich engines that are very graphically sophisticated, but also have all of the library tools to build gameplay and interactions, scoring and AI and all of this. And what we're really doing is how do we take those engines and empower those engines to build very, very large worlds? Previously, I did build a startup focused on building that simulation engine from the ground up and had a lot of learnings doing that and realized what works, what doesn't, both theoretically and practically. What we're doing with MetaGravity is somewhat different. We want not just Unreal and Unity in the future, not only kind of custom enterprise engines, like different AAA studios have their own, or scientific labs might have their own simulation frameworks. What we're developing is a general-purpose ability to parallelize these simulations that are already being handled by those client-side engines and helping those to scale to be much bigger.

So those engines will then take care of the rendering process?

Not just the rendering, but the gameplay, the logic. So if you think of rendering, rendering is just what you're viewing on the screen, and that's itself very sophisticated. But at the end of the day, the simulation itself has to be also built in a game engine today. So if you look at games and how they look today, they look dramatically different to how they looked 20 years ago. But if you look at the gameplay from 20 years ago, it has not changed much in the last 20 years.

Mm-hmm.

And this is because while the graphics, it's somewhat straightforward to parallelize the graphics, the rendering, and that has developed very rapidly, it's not quite so straightforward to do that for the simulation. It's not easy to parallelize a simulation because you have to maintain this causal self-consistency. And so these game engines, they do have rich libraries for building these kinds of interactions, like different types of mechanics, whether you're driving a car or shooting at everyone, or whether it's a corporation between other players in, say, building, I don't know, a farm or a community or something, you're still— the simulation is being handled by the game engine. And then at the end of the day, based on what you can see, based on the angle of your camera looking at that scene, the rendering has to give you as nice a picture of what's happening as possible. Now, this latter, we've improved significantly over the last 20 years. But the former, which is the simulation behind it, has not really improved much in the last 20 years because CPUs haven't improved very much in the last 20 years, even though GPUs have. Graphics processing has improved a lot.

Right. So the entire backend system that supports the simulations themselves directly.

Yes. But in terms of actually creating things at scale, that's what needs to improve, right? Because otherwise, all you're doing is making things look better and shinier. But behind the scenes, it's still a simpler model. A modern first-person shooter game is not dramatically different to one from the early to mid-2000s, but graphically it's very different, right? Graphically, it looks completely different. It looks amazing.

But you think that the gameplay hasn't changed because of the limits themselves or because it's already kind of enjoyable? Like, these gameplays, people get so involved and they spend so much of their life. So you think it's because of the limits or just the enjoyable?

To an extent, it is. You know, you don't fix what's not broken. And there are lots of fun game experiences you can absolutely build with existing, within those limits. And game designers and game developers have gotten really, really good at working within those constraints. But at the same time, since forever, there has been a drive to create bigger, richer worlds. And that has been limited by our ability to deal with the backend scaling. Like if you look at, virtual worlds like MMO games, for example. From the beginning, the promise was always, here's an immersive world, and you can go in there and you can make an impact. But the reality of it is that they're not. They're static experiences that are very scripted. The content is designed to just be recycled and repeated by an endless number of players, including any given player. You can just repeat the same formula. But if you really go back to the origin of the MMO game, these were— I mean, there were early games like Ultima Online and EVE Online, for example.

Yeah.

They tried to be different. They were building more of a world that you can go into, where you could move an object and have an impact, where you could experience the butterfly effect, again, a concept from complex systems, where maybe you go and destroy a particular ship, and that has ripple effects through the economy, which eventually, much later, causes a war to break out between 2 factions. These things can't happen in modern MMO games. But if you go even further back from the earlier MMOs like Archival Online and EVE to the earlier ones, the MUDs, multi-user dungeons, the idea behind multi-user dungeons, which were text-based, was that you're part of this rich, immersive world. You were part of a world that was alive, where you could have actions that have consequences. They were not constrained by the limits of a real-time, graphically graphical simulation, because it was just text.

Right.

It could take seconds or minutes sometimes for certain things to happen. You could have much greater richness in these games. And for those who played tabletop RPGs and such, these games had much greater richness. You know, there were actions that could have sweeping consequences because they were both controlled in terms of the size of the game. And there was a lot of imagination involved, but also because, you know, they're not real-time graphical simulations. There was a richness. Modern MMOs can't do that anymore. These games are limited. Sandbox games can't do that. So instead, the gameplay is— there's a simplified formula that everyone follows, and you know how to build it, and there isn't enormous technical risk for building it. But, you know, you need the means, the tools to build something bigger before you'll even see designers and developers exploring that design space.

Okay, and when we say big, what kind of size are we talking about? Where do we start hitting a limit where it starts becoming kind of really hard to scale?

So today, really, on game engines and the network tech behind it, it's about 100 players that you can't really exceed for a third-person style game. If it's a strategy game, then the number is much lower, maybe in the tens of players you can't, because each player controls an army or other units.

Right.

And that adds more complexity. In an FPS game, you probably don't even want 100. And that's maybe one of the extreme niche examples, because FPS games are about, like you said, it's very esportsy, quite a few of these. plays Counter-Strike to play with 1,000 players. It's about precision. Yes, exactly. And it's about fairness, and it's about who gets the kill shot. But in terms of the bigger open-world games, you do have games that say, oh, there are thousands, millions, tens of millions of players. Like World of Warcraft, at its peak, had 13 million players. But how did they achieve that? You shard the world. There are many, many copies of this world. And then within one of those worlds, realms, as they call them, you have all these zones, and no single zone is really designed to support more than 100 players.

Right.

No player experiences would allow for more than 100 players. Like, why do you not have— I mean, it would be completely sensible in a game like World of Warcraft, say, to have faction versus faction combat in the world where you can build cities and maybe fight others and take over cities. You can't do that because inevitably it'll lead to overcrowding.

Right.

The moment you have more than 100, 50-50 on each side, the server is going to crash.

Yeah. So there is a sort of engineered bottleneck between the interactions between one shard and the other so that the engine doesn't just break.

Yes. Or rather within zones in a shard. And then you're replicating the shards many times where each shard is pretty much identical.

But this is very different from how we live our lives in real life, where those interactions are able to happen.

And it's about, if you think about it in real life, you know, you could go out and if you look at human civilization and how it's evolved as well, if you look at history, people go and occupy new terrain and build communities there and develop over time and start to expand into new territories. But say, take the survival genre, right? Games like Conan Exiles and so on. There are games that have Valheim. These are again limited by this. So you have this 100 player shards usually, which you can go into and play with other players. But that's sort of a contained experience. Whereas really, the intuition here is that it should be one big world that has millions of people, right? And if you— it's not to say that all million of those players have to be interacting with each other at the same time. But what it does mean is that, say, if I want to go and interact with a set of players, there are no restrictions. There are no artificial restrictions, right? An arbitrary subset of that million-player say, player base would, might want to go build a city somewhere or found a new kingdom or a new country, right? That freedom doesn't exist today. And if you were to do that, you know, we're talking about a real economy here, like the real world. You can't build a true economy in the virtual sphere without having to encounter all these limits. There are games like EVE Online that have it to an extent, but there are serious limitations at the end of the day because I can't build a kingdom here and then have trade with another kingdom or another corporation or something somewhere else in an unrestricted way because there are these false barriers erected by technology limits.

And you think that this allows greater sophistication of experiences because we're not artificially— well, artificially, we're not engineering a limit. We just have the limit that then we replicate into games with these artificially engineered limits so that So we don't break the entire experience. And I think this higher level of sophistication of experiences allow us to replicate some economical models that we may have in the real world. What else other than the economical aspect of it?

So social models, I think, are the most important, right? For example, one of the games that would benefit from this, if you look at most MMOs, they're sort of fantasy-themed or medieval-themed, but an aspect of of medieval social development was the beginnings and the evolution of feudalism, right?

Mm.

But how could you create a feudalistic society in the virtual sphere? You couldn't do that today, right? You can do that as a canned simulation for single-player games or small multiplayer games, but you couldn't really create an ongoing sort of situation where you have hierarchies within the world. You have everyone from kings and emperors all the way down to the peasants and the citizens of that world. You couldn't have lords and knights, and you couldn't have a Game of Thrones, if you will, within these realms. You couldn't have those social structures. The economy is— it underpins a lot of these social structures and the drive for them. But it's really about social interaction. But it's not just about that either. I mean, EVE has done a really good job. And again, using that as an example, because there aren't very many sandbox games out there because of all The limitations, the few that have done it have done so with serious technical limits in the experience, making them non-intuitive, right? You know, they are very artificial constrained mechanics and limitations. But, you know, having, you know, there are other games that, you know, there's one in particular we're working on that we can't announce. It's a third party who's doing some really, really interesting work in building the space economy. You couldn't today have this world where, you know, suddenly the space race is on and you're trying to colonize distant planets and you're building these spacefaring organizations or spacefaring factions that might be trading, that might be warring, that might be coming together and interacting with one another. You can't create that today in the virtual sphere. And you make that accessible and intuitive for the average gamer. You can't do that today. But being able to deal with simulation scale makes that possible, right? It's really about, we should really create worlds where the players, are free to self-organize within those worlds, because they will. Because that's what reality is like, not the designers imposing artificial constraints on how they should self-organize.

Although arguably the limitations of the experience is what makes the game itself fun, in the sense that partially what is a game is an alternate reality with different rules of engagement that we have. So even when we're playing we're playing football. It's like, oh, for a little while, let's pretend that life is about who puts the ball into the other person's goals, right? So these are made-up rules of interaction and their limits, because that's your purpose. It's sort of saying that you as a player, your purpose is to put the ball in the other person's net. And because of these limits, then the fun experience is born. If you had a limitless Are we sure we would be able to organize anything out of it, or would we be lost because we wouldn't know where to start looking?

Yeah, so that's a great question. And again, it's not— football is a good example, and these are games, right? But take the game of life. Football is a mini game within it, right? And we're not going to build a million-player football. That doesn't make any sense. And it's not about that. that. But as we speak, there are many football sessions happening around the world, I'm certain, right? This capability is not useful for very targeted, specific, small games, 100%, right? That's not what it's about. But it's about if you're trying to build open worlds, or if you're trying to build real alternative existences, then you can't be constrained by our current abilities. Yeah. All you can do today I'll build those minigames. That's the reality. And how do we shift to we can also create alternate realities? You can't do that today, right?

So I understand intuitively the need to support these kind of scaled experiences. But let me nail down specifically why you think it's really important that we are able to sustain this. Why do you think that the core reason for Why does the— now I'm going to put in the word metaverse, if you're ready for it. But why do we need to support these scaled experiences without the limits when it comes to— so what kind of social structure will we want to replicate in the virtual world? What kind of economical structure? Why do we need to transport it into the virtual world in the first place? Why do we have this need?

So, yeah, that's a great question. And I think you're kind of leading on a fairly controversial word today, like the metaverse. What does it mean? And everyone's talking about it. So there we are moving from games to not only games. I think games are an important part of this shift in what we call the metaverse, but let's talk about that for a second, right? So the metaverse is really about, we've come from a world where the web is 2-dimensional, right? But we've also come from a world where the web is very centralized. If we look at the first wave of the web, it was actually increasingly decentralized. Because people were running servers all around the world. Web 1.0 was more decentralized, but it was less functional. And Web 2.0 was about ISPs, well, service providers, cloud providers, data center providers providing this capability. The next shift is about moving from flat web to moving to things that are much more intuitive because the web, we've learned how to live with it. it, right? But it's not immediately intuitive. People had to learn, and they've done a good job. But what is intuitive to humans is the spatial experiences and the spatial connections. And we've already seen this with games today. Gen Z, they're, they're a 3D-first generation, and they're growing up with these 3D spatial experiences as being the primary experience in games like Roblox and Minecraft, right? And that's evolving into What do the future social networks look like? What does collaboration and coworking look like in the future, right? How do we actually make problem solving more intuitive? If I can visualize proteins and their interactions, maybe I can understand how they fold in an intuitive way that I won't today just by looking at equations and diagrams on a flat screen or page.

So the claim is that we've limited our ability to represent, I mean, we haven't really cracked abstract a way to represent our ways of interaction, the interface that we've utilized to interact with each other on the internet and with the internet itself has been very lacking. Sort of, you know, mouse, computer, a 2D screen, and then these social media that are text-based kind of, or image-based. They're not immersive. Would immersive be the keyword here, or is it something else?

I think immersion is a part of it. But that said, you don't always want to increase immersion. So that's also a thing, right? Sometimes it's better to have Like, if you look back at, say, the '70s, we thought that we'd be talking to everyone on our watches with a face on it. But that's actually not the best way to communicate with someone. You often just want to send a text message. But there are instances where you do want to immerse yourself. So certainly, immersion is a part of it. But the other side of things is what's intuitive? The problem with flat interfaces is we've learned to work with them, but they're not intrinsically intuitive.

So this is interesting. I have a problem with intuition in the sense that intuitive is not necessarily the best, most effective. I can learn something that maybe is not as intuitive, but then it turns out to be the best methodology and the fastest one to do something.

Yeah.

So I wonder whether intuitive is the thing that we should be optimizing for.

Yeah. So that's a good question. Sometimes the intuitive solution to something is not the best solution. But here what I mean more is, as far as our experiences go, what's more intuitive and what's less intuitive. And the reason I think that's important is because our ability to connect with something or to adopt something rapidly increases if it's easier, if there's less we need to learn to do it, if there are fewer models in our way.

Right.

But that said, you're right that sometimes the intuitive sounding solution is not the best. Sometimes the best solution is a counterintuitive one. But here I'm talking more about usage or the user experience.

Adoption. This is an adoption question almost. We want to make it as simple as possible, as intuitive, which is different from simple, but as intuitive as possible for someone to get up to speed onto an experience so that there is no friction of interaction. So that the interface is almost one-to-one Yes. Of how you interact with the real life and virtual life.

And it introduces something, another dimension to it, which is, for example, something that we've seen with players and avatars is that you connect to your avatar in an intuitive way that you don't a chat app.

Right.

So people form relationships in these online worlds that you wouldn't if you were just using WhatsApp, for example.

Yeah, this is really interesting because I feel I've had one experience with one of our startups here, and what they do is they try to really create immersive VR experiences by linking real and digital world in a really kind of interesting way. And for the first time, I realized, oh, my body here is my joystick. Like, I don't need to have a third-party interface to interact with this world. I myself am the interface, just like I'm the interface in my real life, because right now, this is my joystick.

Yeah.

My body is my joystick in the same way it's translated. So this is about VR/AR, though, right? Or is there a bigger point for you?

There absolutely is. But even just to touch on what you just said about VR/AR, right? Seeing an avatar on the screen that you're controlling with your keyboard and mouse already creates a deeper connection between the player and, say, just looking at a chat app. But when you get to VR/AR, your entire body and your being reacts in a very profound way. Right? Like, you just throw someone at a wall in VR, and their body's going to just tense up. Because even if we know that it's not real, it feels real, right, at such a deeper level. So when we have experiences around connection and interaction, they're so much more meaningful the more we relate the person to that experience. Now, the step above that, I mean, the end goal for all of this, right, if you look at the metaverse, the end goal for all of this is at some point, our interaction with these systems, it's not gonna be through a mouse or a keyboard or a VR controller. It's going to be at the level of our neural systems interface with it, right?

Right.

And there's work happening in this direction. At some point in the future, how long it may take, virtual realities will be indistinguishable from physical realities. When that happens, Right? It alters our definition of what reality means, right? You could have already, you can have very meaningful experiences. There was this World of Warcraft player who died in Norway, and his, he was disabled, and his parents had no idea. I can't remember his name now, but his parents had no idea that he had all these rich relationships online. They thought he was just just on his own in the basement until people started showing up to his funeral. Right. But how much more meaningful does that get when you're in VR and AR? And how much more meaningful when these are relationships that are happening in an entirely different virtual sphere where you're interacting directly through your brain, right? Where it feels real. This could feel real to us, and it inevitably will eventually. That changes the game.

I think in a lot of the listeners' mind, probably a question is going through, and is, is this sad or is it not sad? Because partially, like, is this idea that we find so meaning into virtual realities, is this idea something that should lead us to feel almost a bit awkward about the fact that we're escaping from our real life? Or, or should it be enriching? Oh, look at the possibilities. How should we feel about this?

I'm definitely of the camp that thinks that this is an exciting new frontier and probably the most important frontier, because there are many problems we will solve in this sphere that we can't in the physical world or won't in the timelines we're talking about. I actually think that this shift— and I gave a talk on this recently at the Royal Institution for Twin Global— um, there is a shift to the virtual sphere that's going to create for us an opportunity to improve the efficiency of our civilization by not 2x or 10x, but by much greater orders of magnitude— 100x, 1,000x. The upper bound being actually 100 quadrillion.

The efficiency.

The efficiency.

Why efficiency?

And what I'm talking about here is today our existence is very matter, costly in terms of matter. It's very matter intensive. And we are exploiting the environment, the planet at a rate that is unchecked. There are no natural predators to humans. There are no natural checks and balances. And our base physical needs that have been around for about 300,000 years as anatomically modern humans, those haven't really changed. changed, but we've become so effective at taking ourselves outside of that equation that we can exploit matter and nature to serve our psychological needs to a very profound degree. That's not efficient at all. If we can shift those needs, because those needs don't necessarily need to be matter, if we can shift those needs to the digital realm, and that requires simulation scale, that requires these next-generation interfaces, that requires digital scarcity and decentralization. to happen. But if you're able to do that, then we are moving from matter to pure energy as the means with which we satisfy the craving of mankind. And if we do that, the efficiency gain for pure energy is significantly greater than anything we can possibly achieve when we have to serve those needs with matter.

You said psychological needs specifically. What do you define as psychological needs?

Yeah, so I mean, one model to look at this is Maslow's hierarchy of needs, right? We have base needs around food, shelter, the need to replicate, the need to feel safe. The amount of food we need is, if anything, there's an overabundance of food and calories in much of the world today, especially the developed world, but I would argue much of the world. And we're not really serving ourselves physically in terms of our physiology as effectively as we should, right? We're overindulging because our base nature nature, the human animal needs more, right? And it drives us. But if you look at Maslow's hierarchy, as you climb up that pyramid, self-esteem needs, self-actualization needs are best served in a different realm altogether. Because, you know, for example, you don't need a Ferrari is one of the examples I've used in the past, right? We don't need it for physical reasons. You need it to satisfy a higher order demand, a psychological demand.

So we don't need to produce a Ferrari to satisfy that desire of yours. You can just have your digital Ferrari, sort of, or the counterpart of whatever satisfied that little piece of your puzzle that really makes you fulfilled and happy about yourself.

Yes. Scarcity, I think, is important. The idea of having a Ferrari is to have something that nobody else has. Well, for a few people, maybe it's really about driving the Ferrari, right? But it's largely about having something that few others do.

Right, right. The feeling to stand out. Yes.

And we've already seen sort of in the last year or two sort of a little bit of a quirk in the world economy on that front. We've seen sort of the NFT craze, for example, right? People paying insane amounts of money, including a lot of celebrities, for what's essentially a JPEG which tied to a unique ID on the blockchain, right? And why were Why did that make sense? It only makes sense if you think about it from the perspective of those psychological needs. You know, it's about having the one thing that's extremely scarce and nobody else has, right? I mean, why is gold valuable? There's nothing intrinsic about gold that makes it valuable other than it's a time-honored asset and a resource, right? Gold has 4 times the market cap of copper, but copper is actually valuable. Right? So really, like, I mean, human— the human drive is to have what others don't. We can create scarcity in the digital realm. We can also have post-scarcity in the digital realm if you want to, but still have scarcity, right?

But, and so do you think scarcity is the only thing that leads to psychological sort of psychological alignment of our needs that are In real life, we have specific needs and we're able to replicate them just if scarcity is there, or is there something else that needs to happen for that to be fully fulfilled? Because I think I'd rather have a Ferrari than the NFT still today. Obviously also because of the amount of people that accept it on one side, but I would like to understand what else needs to happen for really have this transition of psychological needs transferred to the virtual world, which makes complete sense if I think about it.

Yeah. And I have to caveat this by saying that we're very much in the hypothetical realm, quite a few of the things we're talking about here. And these things are still yet to be explored and understood. But let's talk about a stable point in the future, right? And at some point in the future, let's say virtual reality is indistinguishable from physical reality.

Right.

And let's say that you could spend a lot of money and buy a spaceship in that virtual reality. and fly to distant stars and planets, so through Jupiter, through the, through the, through the, what do you call it, the gaseous surface of Jupiter and experience what that feels like down to, you know, your nerve endings in your fingertips. If it feels real, would you spend a certain amount of money on that? Or would you rather spend it on a Ferrari in the physical world to drive on a road?

The cost of production of virtual, of just photons, coming up with photons, and also the energy that I need to have for my neural interface to respond to that, it's probably much lower compared to the real, if we have to create it in real life, because we haven't mastered real life, but we can master virtual life.

Absolutely, because we're not bound by the same laws of physics, like teleportation, right? I hate flying, for example. I hate taking the plane wherever I go, and it's a real drawback. At some point, you won't need to. You can instantly feel like you're in the Maldives, Right. And it would feel no different at some point in the future to actually going there, right?

Right.

What would you pay more for, right? If your answer is this because it's real, I think that there'll always be that. There are still people who, you know, hold on to some antique classical ideas because of the quirkiness, but eventually it would make very little sense, right? There are people who like to be out in the wilderness, you know, hunting and farming their your own food. But the current mindset of, hey, everything has to be physical that I interact with, I believe at least will disappear or diminish to the bare minimum. Because the experiences that you want to have, you can have in the virtual and even beyond. Like, imagine being able to swim through the ocean like a dolphin and experience what that feels like down to your nerve endings. Like, that experience, you can't get that in the physical world, but we will be able to in the virtual.

Yes, because you're not bound to your physical body. Like a physical constraint. Yes. And so this is interesting because we just actually had the argument that scarcity is something that drives satisfaction of the physical needs. But this is not a scarce experience.

It's potentially—

It's potentially limitless and accessible to everybody. So does that contaminate the quality of the experience? After I go into the dolphin's body a few times, does it just everything become completely meaningless? If I don't have the limits, what even are we like? Are we able to grasp onto anything?

If you can indulge yourself any way you like, is there any reason to not? Yeah, that's a good question. And we've seen social media, for example, hack the brain chemistry to quite an extent with notifications and with this sort of needing to give apps the immediate attention. And so it is a double-edged sword a little bit, and it's hard to answer. Some of the social and ethical questions around it are harder to answer here and now. We have to see how it evolves. Regarding scarcity though, just to go back to that, we're talking about a world where a lot of the experiences that are difficult today will become easier, right? But we're still talking about, like, human beings are hierarchical by nature. We see hierarchies, we believe in hierarchies. That is how we evolved to act. And therefore, there's a need for humans to feel like they're able to go up and down that, or rather up that ladder of hierarchy.

Right.

And where scarcity comes in is giving us the ability to do that. Is that meaningful long-term? I don't know, but I do think that it's fundamentally biological. So we're not gonna outthink our way when it comes to these things. So there is gonna need to be some sense of scarcity within these worlds. Maybe the people who build the experience of swimming like a dolphin will gatekeep that and make money off of it, and not everyone gets the best experience. I actually don't believe that everything will be available to everyone for free.

Which is crazy thinking that the cost of offering that experience is almost nothing. One would think, well, we, we should have a free, open, almost non-economy at that point. We could abstract. If everything is extremely easy to create, we should almost not have a cost of access. And then we're adding the cost of access artificially. I don't know if that makes sense, if it's just to provide an additional level of pleasure to the people that are experiencing that. I would lean more on the sad argumentation that we were talking about earlier.

Yeah. Yeah. So also a little bit like NFTs, I guess, then, because they're not hugely— it's not about creating a set of NFTs is not particularly difficult, but it's about what those represent to people. And that's very much about the network it creates. Right. And regarding the cost of doing these things, I mean, you could argue to an extent that, yes, it does take you— like today, games aren't free, right? Even though it's a lot cheaper to have an experience that you couldn't imagine in a game than you could doing it in real life. Someone has to do the work. When things like generative AI come into the picture, then you have to question whether when you extrapolate that forward, whether people will even have to do any work.

Right.

Right? And when that's the case, should you still give it away for free? I think we're at something that feels a bit like a singularity. There is going to be a great wave of rebalancing of things economically, both in the shift to the digital and both through the power of things like AI.

I mean, also, if you can satisfy all of your psychological needs already, why would you need to charge? What do you need the money for? What could you possibly buy that you can just have in the virtual world?

Because I think the most profound psychological needs humans have is the need to be better than the next person. And that's the bit that we have to— that people will always need satisfied. So money becomes ultimately the driver.

So money becomes just a ladder. It's not functional anymore because we can just satisfy all of our needs at scale independently. And it becomes just a representation of where we're at.

Yeah, but arguably that's sort of where we already are in a sense. I mean, you know, for 300,000 years anatomically modern humans have been fighting for survival. We're not really doing that anymore, right? I mean, yes, you could argue that we're striving for greater comfort. But really, I mean, there are other ways of achieving comfort, right? We're really striving for how we view ourselves relative to others and how we are perceived. That psychological need is what's driving civilization towards more. That's what's unsustainable, right? That's not efficient in terms of matter. It's matter-intensive. When we move to the digital realm, that drive will remain. Like, you know, individual people will see themselves relative to others, and there'll be a need to still have those hierarchies to fulfill that need.

I think that the need to feel superior to the next person, however, kind of— I see it fading away with the amount of experience that I personally accumulate in my life. And so I wonder whether it kind of slims and the fact that we can have this many experiences kind of changes my, my need to be better than the next person because I've already experienced it. I could have tried it. And so I know what it feels like and I can put everything into perspective because I want to be better than the next person. I don't want to miss out on life. I don't want to live this, this 100 years of my life, if I'm lucky, in a way that's not the best experience that I could have on earth, right? If I get to experience that, And arguably, there is no reason really for me to still have that drive to be better than you.

Yes. So I think there are people who are able to stop at some point and say, actually, I'm satisfied. Do I think that the human species will do that at any point? I'm skeptical. And maybe my viewpoint sounds a bit cynical, but I think ultimately humans are motivated by the desire and the hunger. for more. And that is also what we have. If you look at history and look at civilization today and the trajectory, the overall kind of the arrow, the direction of the arrow is up and driving for more. I think on the whole, as a population, that is very much the direction. I think that's very intrinsic, and I don't think we're going to overcome that. But, you know, there will always be exceptions. I just think they'll be the exception rather than the rule. I might sound cynical, but that's— I think the evidence supports that more strongly than the counterpoint, the idea that humans are moving towards being egalitarian. I think we are to an extent, but ultimately even within that, the base human drive, and it's not just humans, it's any species, But with humans in particular, because we are a social species, and it's a stronger driver in social species to sort of be ahead relative to others, to fight for dominance is in our genes.

Right. I mean, my argument is once we satiate the psychological need with these unlimited experience, there is no need maybe to evolve as a species anymore that we feel intrinsically need.

I just don't think it'll ever— I don't think we'll ever reach the limit point. Let me say, Jim Carrey, like, he was one of my favorite actors, and he did this speech at the Golden Globes that I really like. He was introduced as, I think I might be paraphrasing here, 2-time Golden Globe winner Jim Carrey. And he comes up to the stage and he goes, I am 2-time Golden Globe winner Jim Carrey, and when I go to sleep I dream, I don't dream any old dream. I dream of being 3-time Golden Globe winner Jim Carrey because then I'll be enough.

Right.

Right.

Right.

I think we always think that here's the next, this next mountain will be enough. And I think some people are sad. I just think that it's intrinsic to our species to be driven to go further.

And that'll never stop unless we're able Unless maybe once we feel comfortable that someone is doing the work to go forward and we get to experience all of these unlimited experiences on one side. So if we delegate innovation sort of to an AI that you were talking about, where the AI takes care of making sure that we're okay physically, but we, on the other hand, we get to have these unlimited, beautiful experiences, maybe we can tone down and we won't have Maybe the desire is fulfilled. I just don't have to work for it, right? So, I won't have a relentless existence trying to look for something else, and I can just have my existence fully immersed into this dopamine-producing or other whatever chemical we're really trying to chase-producing system. And, at that point, why?

So, I think that if we got to that point, that would sound sad.

Fair enough, we got that.

Because it's the equivalent of disconnecting from, from the rat race of life. And I think what we're really saying is that the appearance of disconnecting from that rat race is sad. Maybe we have our priorities backwards, but if you really think about it, today you could stop and say, that's enough, and I'm going to go off the grid and live this existence. You could do that, and you could have things that for 99 or 97% of human, modern human life on Earth, humans have craved for. But we're not really doing it. We're doing it at a local level, but not as a population. I don't believe that we have an ability to say, let's stop now. I think there'll always be a few who do. But I think we will always be driven by what's over that next mountain. Right.

I lean on the other side. I feel like we may reach a saturation point where we're okay with being lazy. It's fine. We have so much that we can experience, and we'll just tame our spirit of innovation and hunger.

I think at that point, we will start to value things that are scarce rather than necessarily things that are, I don't know, exciting.

And so at that point, we can expect a full circle back to our real life scarce existence, possibly. Okay, well, since we're dreaming and in this realm already, I wanted to ask about generative AI, as you spoke about it, and I know you've looked at AI in great depths. So tying it back to metagravity and AI, how do the two come together within these experiences and simulations of our world? Where does the AI come in, and how can it really enrich our experiences?

Yeah, I think that's a really good question because we are very much at the start of exploring what AI will mean for the world, right? We're sort of at the point where the world has kind of woken up and thanks to things like large language models realized that maybe general intelligence isn't a pipe dream and maybe it's on the horizon and how is that gonna change the world? And we're really seeing, like we saw people on ArtStation protesting the use of generative AI for art because they're afraid of their jobs. And that's happening to an extent with writers and so on, programmers. But increasingly, the current wave at least is being seen as AI, the AI tools are an extension of the user and you still need to have a degree of expertise. But at some point, it's conceivable that AI will outperform us on every front. And, you know, human work will start to be devalued. What happens then? I think it's very hard to tell. For the simple reason that that point in time has been termed the singularity, and it's very hard to look beyond the singularity. But that said, in terms of certain avenues for this, right, I do think that there will be a sudden upsurge in abundance, not just in the physical world, but if you look at the virtual and our shift towards the virtual, it's much easier for AI to be creating If we think about these worlds that we'd want to inhabit, you know, in the immediate timeframe, that's the games and the social networks, but in the longer-term timeframe, that's, say, most of human experience, we are not going to be able to craft these experiences by hand. It would simply take too much effort, right? We've seen this already with MMO games. It's an enormous amount of effort to create the content to populate these worlds. So then these worlds need to be self-generating. And some systems we can already build that are self-generating without really invoking AI. Like, procedural generation is a topic that's been around in the games industry for a very long time. And in simulation, we understand how erosion happens and how rivers and valleys are formed, right? But we also understand how plants might spread and how ecosystems might happen.

So these are just some foundational understanding of how things work. work, which allows us to perf quickly, some basic building blocks that allows us to replicate and build quickly worlds at scale.

Yes. But often what these systems tend to struggle with is that when it's a person experiencing content, there's a need for content that feels more crafted and special often, right? Rather than feeling generic. Like you create a forest and every part of the forest looks similar to every other part because the procedural generation algorithms tend to produce more of the same often. This is one of the kind of first areas where AI comes in, right? Being able to create things that feel handcrafted, like there's a story behind it to an extent, but still being at a scale large enough that, you know, an enormous player base can populate, an enormous user base. Building world-building tools, therefore, with AI is, I think, a major next step for us, both for MetaGravity and for the industry as a whole, because Creating the experiences in which humans will want to spend increasing amounts of time is important. We saw attempts to be purely generative with procedural algorithms and create a large world. No Man's Sky tried this. And what they actually discovered was that a lot of what the players experienced—

Repetitive.

Yeah, it was repetitive.

Interesting.

Right? But AI has the opportunity to make it feel intelligent, like there's a guiding hand behind it. And not just the content, but the story. Creating unique stories that are tailored to individual players, but created in real time. I think we're driving to a world where a lot of the content is generated in real time and is designed to optimize the experience of the user.

So how could that tailoring to a specific user differ from another one?

So you might have certain preferences. and certain proclivities and so on in the type of content you consume, that information, that metadata, could be used by the AI to inform the AI. Again, it might be that the type of challenge that you can encounter and experience and work with is different to what someone else can. So you could have, because this is one of the things even about game design, if the game is too easy, you get bored. If the game is too difficult, you get frustrated and give up. And there's a sweet spot. And there's a nice little curve that you need to kind of ramp the player through, right? With AI, this could become completely adaptive in terms of how you— not just the content and the setting and the story that you experience, but how you experience it.

Interesting.

But you could also, in real time, tie all those stories together into a cohesive whole rather than having completely separate stories. Because at the end of the day, we don't just want stories. We want to be at the center of a story, every individual human. Again, perhaps I sound cynical, but that's the reality of, I think, the human experience, that we want to be at the center. How do you make every person, every user feel like they're at the center of their world and there are other players around, but they're very much at the center and experiencing an optimal trajectory?

Interesting, because this is the opposite of what we were speaking in the sense that on one side you want to give a shared experience that everybody can agree on and it's going on at the same time for everybody precisely at the same time. Same for everyone. On the other side, you want extremely tailored experiences that are really adaptable to the single player. So that seems like a— is it a balancing act or is it a trade-off?

So I think there will be— it will be a balancing act with some trade-offs, but I think those are actually self-consistent. Right. Again, going back to perhaps my view that the human ambition is to rise above others, and that is inescapable. We want to feel like there is struggle and that we are moving up. And we don't necessarily even need to feel like we are at the top, but that there's always more to endeavor towards. Taking that, but creating a story for yourself where you get to feel like it's your own path, it's your own experience, but putting that in the context of other real players, I think, is what makes it the most meaningful.

Is that also demanding from a compute standpoint?

Yes, this is demanding. The scaling of simulation is demanding. Being able to create independent story trajectories within a world is demanding because it will require generative AI that can focus on multiple different viewpoints. And then being able to tie those stories together in a cohesive whole, that is demanding because you have to reconcile many different events that can happen. The way to do that is really— the simple way to do that is to think about the base world and its base base kind of laws as being immutable. Everyone should abide by it. But then individual players are given trajectories that they can follow and they can engage with that takes them down a particular path within those rules. So what you end up with is no 2 trajectories can really contradict because they ultimately have to be agreed upon by that base rule set.

Who decides the base rule set? Is it the game creator? And who's the game creator in when GenAI is here?

At this stage, it would be humans, but eventually, perhaps those rules could be governed by AI. For the time being, though, it's human designers who have to design them. And the next step for generative AI is very much, how do we create unique experience? For example, let's say you start your character out in a little village, and then the generative AI produces a series of interesting events that happen that take you out of there, right? This just sort of goes back to, the age-old story, the hero with a thousand faces sort of thing, right? You know, you go on this journey, you maybe reach some level of learning and self-actualization, and you return to vanquish your foe and be successful. These kinds of stories, the AI can weave through this world, right? I think it's more creating these kinds of experiences, but again, tailored to individuals that are the more immediate interest and excitement. Brilliant.

And would users want to or need to know about these rules? Would we even have to expect them to know about these rules to enjoy the experience? Or do you imagine a future where they don't need to know and they're already automatically collaborating or interacting with those users?

I think some of the rules are going to be immediately intuitive because base spatial reasoning is something intuitive to us, right? 3-dimensional Euclidean space is intuitive, right? Things pass behind things and come out the other way. This is also why magic is interesting, because magic magic, when we see magic tricks, it breaks causality and therefore it like surprises us, right?

Right.

And it's not just purely human. There are really interesting videos of magic tricks being shown to other animals. And I find primates especially interesting because they have reactions that look like human reactions. So there are these agreed-upon rules that we all agree on. But then when we open storybooks, you know, to varying degrees, they often break reality. variety, right? Especially, say, fantasy story, which is the most popular genre when it comes to open-world MMO-type games. But having a magic system that behaves coherently and self-consistently, right, which players can learn, that's also interesting. Imagine if in the real world, this is why perhaps Harry Potter is so interesting, the idea that it's sort of kind of real. Yeah, kind of real. It's kind of like London, but, you know, and Oxford, but It's, um, it's got laws that are— that break physical law, right? And you have to learn about, which is exciting for people. You could create that in a virtual world where you do have to learn the rules of magic by experimenting with it, by studying it.

Although it does feel like because we're so known in living in one perspective, breaking it is so exciting for now for us. I wonder whether a generation that grows up fully immersed into an environment where the rules are very, very different, if almost nonexistent, because you are creating all sorts of experiences. I wonder if they'll have the same level of appreciation for magic, specifically magic. But we can talk about any really experience where there is this breaking. I wonder whether they'll be as appreciative of that.

Yeah, I mean, that's a good question. If you grew up in— let's say we've transitioned to where a lot of our existence is in the virtual sphere. then if you're encountering, if the world you're inhabiting has a lot of magic, it might become commonplace to you, right? It's not special anymore. So yes, that's entirely possible. Then it might become very frustrating to leave it because—

Or extremely interesting.

Yeah.

Oh, look at all these rules that I can't break.

Yeah, yeah, it could, it could go the other way. Yeah, absolutely. But equally, the imagination can create many different alternate existences. So even if you got really good and had an intuition for one system, you may not for another. Like, this is like the, maybe the Harry Potter versus Lord of the Rings versus Star Wars sort of debate, right? Like, you know, these different worlds have different systems that function very differently. Just because you know one, it doesn't mean that you can understand the other, but it might actually be exciting. So EVE Online, again, is an interesting game because there are people who've spent so much time learning how to build corporations corporations within EVE and running them, it's almost like they've got their MBA within EVE because it's actually really, really hard to build a successful corporation in this game. And whenever there's like, you know, you hear about a guy who got into a corporation, worked his way up for 2 years, and then switched sides, and he was always planning to do that, that gets reported in the media, in the games media, right? When you create these worlds that are completely unimagined, there are an infinity of possible rules and systems that could be created that you could engage with in a very Deeply meaningful way, right?

Yeah. At that point, the real scarce thing is going to be human attention, which already kind of is. But if I have the possibility of all of these worlds that can be so incredibly complex, where do I choose to allocate my time? I can spend 2 years playing EVE Online, but maybe I could have done something else with it. So I wonder how we'll even be able to choose in the sea of complexity and choice that we have. have.

And maybe this is where the scarcity question comes back. Like, if you can create anything easily, and let's say we've solved fusion, and even without solving fusion, fission is extremely, extremely energy— extremely efficient as an energy production process. If we simply fix the policies and make that transition where we are actually leveraging modern physics in order to generate energy rather than outdated physics, then we would be very efficient in our ability to create these experiences, perhaps at at that point, the most valuable experiences will be not necessarily the most valuable experiences in a kind of a pure clinical sense, but rather the ones which have the greatest investment of human attention. So I would rather— like, this is— you already see this in games. Players want to play games that other players are playing. And sometimes you have these games that come out that are technically very good but aren't that exciting because there aren't enough players.

Yes. But that break If we can replicate players, if we have NPCs, non-playing characters, that fully meet every single requirement, at that point, we don't need network effects anymore within a system because the network effects are fulfilled by AI agents at that point.

Yeah, no, I agree with that. The counterargument to this is if we have AI that feels so real, you can't even tell it's a player. So the Turing test, they pass the Turing test completely. and permanently, then what does it even mean?

So if it does, it kind of feels like we're diverging wherever each one of us really goes into properly tailored experience. If it doesn't reach there and we still have the feel for human connection, it's probably going to converge in shared world experiences.

Yeah.

Um, or it could end up in a combination of both, as you were offering. But it does seem that metagravity especially thrives if there is a shared reality. of some sort. Is that right?

So I think what we are solving is essential regardless. People, humans aren't going to disappear. And the next phase is the internet over the next decade becoming increasingly 3-dimensional and spatially organized. The technology to support that, to power that, doesn't exist. Like, if you go back 20 years ago, web servers were the backbone and web browsers were the client and the interface. If we look at the next 10, 20 is game engines are the client, but the backbone to build these large experiences doesn't exist. And MetaGravity is building essentially the web server. Unlike the web server, which is just an engineering problem, building the backend for virtual worlds and games of arbitrary scale is not just an engineering problem. It requires a new way of thinking about the fundamental principles. Solving that is going to be important to power that next phase, Internet 2.0, Let's call it, right? If you want to call it the metaverse, that's certainly a term that's been around for a while. But beyond that, what happens in the long term for us to satisfy our needs increasingly? And it'll— this virtual sphere, it's already competing very effectively with physical reality where we're more and more immersed in spending much of— most humans in the developed world probably spend about a third of their time in some way online or plugged into a digital system.

Looking at the viewers. Exactly.

Um, and that's only going to get better in terms of experience. And why would you think that it wouldn't therefore suck up more of our time and be more demanding? The challenge then is to make that meaningful. But as we go down that path, so MetaGravity, the key thing we're doing is making that possible and also powering that long term. When AI comes into the picture and replaces the need to work, that's a harder thing to maybe predict, like what happens economically and socially. I mean, Elon Musk, for example, has talked about, you know, universal credit being important, and so have others. In a post sort of intelligence world, it may be that, yes, we do need to think about other economic models. Maybe we're experiencing the end of capitalism as we know it. It's had a good 400-year run. But maybe that's starting to reach an endpoint. It's harder to say, maybe predict what the right social and economic organizing—

Yeah, it could be an interesting exercise to see how each variable could affect the capitalistic structure as a whole.

And this is again where even with decentralization, say from the world of blockchain, explorations of things like decentralized autonomous organizations and other types of structures, do national boundaries mean as much if you no longer need, if spatial constraints don't exist as much anymore. Metagravity as an organization, we have people from all over the world, from various different countries, from all the different continents except Antarctica working on Metagravity. So what does that mean for an organization going forward? It's certainly a new world. But the efficiency gain and the value that it brings is very interesting. Here's the thing, here's an idea, right? So we often talk, I mean, there's a lot of talk about quantum computing and there are a lot of VCs investing in that. I think the most important problem that we can solve, say, in the next 10 years is quantum communication, right? Communicating from A to B through entangled particles. Because when you solve that problem and you develop that technology and you can scale it up efficiently, you have zero latency communication between 2 points, any 2 points. Right? Zero latency.

Because the 2 particles are kind of connecting.

Yeah, they're entangled. They're entangled, right? So when you manipulate one, you're manipulating the other instantaneously without needing to—

Zero latency.

Zero latency. With the talk I recently gave, the interesting— the underlying theme of that was proximity. When we achieve this kind of digital connectivity and the virtual work scalability and communication, we're talking about everything becomes well, we bring everything into close proximity with everything else. And that's a really, really interesting world, right? You can instantaneously be connected to someone else without latency, to any information without latency. And as we expand outside of Earth as a civilization, that introduces interesting things. As we travel to distant stars and planets, for example, perhaps it's not our physical bodies that go there. But it's our consciousness. Today, if you're doing telepresence on Mars, you still have significant latency because of speed of light.

What do you mean telepresence?

Like, if you're controlling a robot, it takes tens of minutes, I believe. I'm not sure of the exact number. You do an action on Earth, and for that to actually— the electromagnetic radiation to travel to Mars. But once you've solved this problem, you can instantaneously be there, and you can be sitting on Earth but experiencing things in real time. time, wherever we can send our technology, we can be there. Well, that's meaningful.

It kind of feels like at that point we have mastered the rules of the real world as well. And it seems like we don't need quantum to do that in a virtual environment. We do need quantum to do that in the real environment, in the physical.

Or we need— we sort of do. I mean, we don't 100% do need— uh, we don't 100% need it to do it in a virtual environment within some limits, right? So for example, if I'm interacting with someone in Japan, right, in the virtual sphere, there's still latency. And there's a, you know, there's a limit to how instantaneous that interaction can be today on current hardware because electromagnetic waves have to travel from London to Japan, right? And that takes time, right? It takes a couple of hundred milliseconds. And that's still a small amount of latency and sort of it's okay. But when we talk about other planets and so on, that's tens of minutes, days, weeks, months, years to the nearest star system, right? But when we solve the problem of efficient instantaneous communication, all the technology we're building, this virtual world technology, AI technology, it just changes the game. We could have at some point, like, you know, neural interfaces with which we control the virtual, but we could also have neural interfaces into data center-scale neural networks that are enhancing our intelligence. Now, if we could also solve the communication problem to those data centers, then you've just expanded your brain. You've doubled its size, you've quadrupled it, you've made it 1,000x bigger.

I'd have a lot of questions about how the interfacing between the brain, my brain, and then the extended brain would work, whether it would feel smooth, because I still feel like I have a limit. It's really hard for me to envision that there can be a communication pathway that's so expanded between me and a computer. It feels like my limits are my limits. It's just my brain limits. So you think that there is a possibility of a world where that bottleneck is not there?

Not only is it a possibility, I think it's inevitable. And it's not even as speculative as terraforming a distant planet and settling on it. It's a known physics. It's theoretically well understood. It's a lot of engineering, and there are unknowns in the engineering, but it's theoretically possible and therefore inevitable. We'll get there. It's just a matter of time, and we'll probably get there much sooner than we will colonizing another planet.

When you say, I love saying theoretically possible and therefore inevitable, I really like that sentence. I wonder what you mean by theoretically possible. Are we confident?

When I say theoretically possible, the theoretical physics for it is understood. Not just the theoretical physics, it's experimentally possible, right? For example, we know how to manipulate an entangled particle and, you know, send that across a lab, and I believe even into space, the more recent experiments. There are papers on this. The papers are titled Quantum Teleportation, because you're teleporting quantum states across a lab or into space. space. One mechanism for that is eventually we'll be able to teleport people. But theoretically, if we're doing that, all we're doing is we're replicating the quantum information at the quantum level, information that describes you as you elsewhere. So we would be creating a replica of you. But if we are sending information, that's not a limitation.

So sending information, but yes, and then the question would be, how, where does my consciousness come into play? What, how Is it part of the information so I get to actually be transferred there, or do I stay here controlling almost as a Mars rover, as you described earlier? So how would that come into play is probably an unknown.

But think about it this way, right? Right now, your brain is a computer with neurons connected, and there are interconnections between all the different neurons. If we were able to create a data center with 1,000x the power of your brain and connect you to One problem might be the communication latency between your real brain and the data center. But if we've solved the communication problem where we can have that cluster of compute nodes all effectively equidistant from one another, meaning instantaneous communication in this case as well, if each node in a data center has a distance of 1 to every other node, that's not the case today because you have wires or fiber optics and 2 servers on 2 extreme ends are further away than 2 that are nearby.

Right.

Right. If you can have instantaneous communication, quantum communication, and then if you could have that interface with your brain, your brain and any neuron in your brain, depending on how we work out how to do the interfaces, is connecting to that off-platform, if you're the platform computer. Now your brain isn't just the brain in your head. Right. Your consciousness is the consciousness across all of those compute nodes.

And it feels like you're—

And yes. And at that point, who are you? Right. Your body is 1% of who you really are.

Yeah. Which we're already kind of starting to experiment with our phone in the sense that if I deleted all the pictures of my phone, I would feel like part of me is taken away.

But here, literally part of you would be—

That's even more.

I mean, of course, there are open problems and there is still, you know, there are agreements to be reached on the philosophy of sort of the mind versus the body. And, you know, sorry, the mind versus the brain rather, and so on. But if you really think about what we do know, we know that your brain exists and does computation. And we know that you're a conscious being. And perhaps the most parsimonious position to take is to say that your mind is in your brain, and therefore who you think you are is what's contained in your brain, that consciousness. That consciousness exists in space and time, and therefore, say, we can take a step further forward and say your consciousness is a dynamical system. It is a system that is time-dependent, and it exists in space and And the dynamics arise from the interaction of electrical signals and chemical signals in your brain. Now, if you're able to somehow take that and expand that out with similar architecture, but significantly larger, you have far more neurons to play with. Your consciousness occupies a much bigger realm, but again, it's a dynamical system, but significantly expanded out.

So the only constant is that the dynamical— it remains a dynamical system, and then the body is a variable. My brain is also a variable.

variable.

And so as long as there is the consciousness, which seems to be an everlasting, or you call it dynamical system, as long as there is a consciousness, it doesn't really matter in which form it manifests itself slash compute power leverages. But there has to be that constant. And so I wonder how to define that constant. For me, it's not obvious.

It's not been like— we haven't quite defined consciousness in precise, or at least we haven't agreed as a civilization, we haven't agreed upon a definition, right? It's an open problem from both the science standpoint and the philosophy standpoint. For example, one problem with all of this is the strong problem of consciousness, right? It's one thing to say that if we build neurons and interconnect them or create a similar structure even on a computer, you get intelligent behavior. And we've seen that with neural nets evolving over time for the last few decades. decades. But that feeling that you have, right? Right. Like when you touch, you feel something. How do we quantify that, right? Qualia, as the philosophers call it, the strong problem of consciousness.

Because it feels like my consciousness is directed here when I feel.

Yes. We don't— we haven't worked this out yet, I guess. So maybe it's not— maybe it's not even important. Maybe like one viewpoint is that when information is being processed, that feeling naturally arises. We can't— maybe computers are already feeling things as they process information. Maybe that's an easy answer. Maybe that's not correct, but that's one suggestion. And so when we connect up artificial neurons, those neurons are already feeling things, right? Maybe, right? We don't know. But, you know, we still can't answer where does that feeling come from?

Yeah.

We've demonstrated we can build by following the patterns in the brain. We can build intelligent things, but we don't know where the feeling of consciousness comes from. But this is an interesting open problem. If we do scale your consciousness up, I think it is reasonable inductively to assume that that has then expanded your consciousness up, say, 1,000 equivalent of your brain, x 1,000, is you have the power of 1,000 brains, and now you're on another realm altogether, sphere altogether. together in terms of you as an organism, as a conscious being, right? Imagine having the consciousness of 1,000 beings, but it's you, right? At the heart of it is still your consciousness.

Right, because at that point, we could, if we found a way to organize it, at that point, if we have our brains that are expanded throughout with external compute and we're able to connect among each other and find a way of coordinating, we are sort of reaching an intelligence level, or I'm not even sure if intelligence is the key element here, a consciousness level that is above each one of us. And so we would become really one. And then the question is, how will it feel? Because where am I at an atom level versus as a system level? Is there even a difference?

Yeah. So does— what does individuality mean in that realm, right? Is it as meaningful? Because individuals are necessary for the species to propagate. Is the convergence of this ultimately some sort of unity? I don't know. Or is it, you know, maybe at the heart of it, your current consciousness values individuality, so maybe your expanded one does, right? These are hard questions. And it potentially could also be how we decide to build things, how we design it could be the answer, right? But you could be having, you know, conceivably simultaneous experiences. You could both be your consciousness traveling in a shuttle, traveling across deep space, and you could be on Earth interacting with friends and family.

So it does feel like directionally we're going towards oneness, which is a very philosophical— I mean, it's something that you kind of hear and you really understand. But it's funny how we got here thinking about technology and then its implications. We're just seeing Oh, these are the patterns, this is where we're going, and this is where we could possibly end up. And it feels— and almost, you claim, it's almost inevitable. And where we can project it, it's a oneness, a one element of all together, which is kind of crazy.

Yeah, I mean, it's— you expand consciousness and then potentially there's a convergence. I'm not certain that oneness will be the consequence, but I can see why it would be given that that, I mean, we should think about it from this perspective, right? As we become transhuman in a sense, right? As we go beyond our base human sort of underpinning, what's the right experience for us? What's the right way of connecting? What's the right way of moving forward? It's easy to think, oh, you know, adopting and immersing yourself in technology is sad because it's not where we've come from.

Mm-hmm.

But why do we that way. We think that way because we are evolved to look favorably at things that have an adaptive advantage, right? It's Darwinian because we have very recently walked out of the plains and forests and built civilization. Very, very recently. We haven't evolved genetically in that time to a significant enough degree because for almost 97% of modern human existence, we have been hunter-gatherers. And so we're still constrained by those genetics. The question of how much of that is still relevant as we evolve as a species is an open question. And how much of that will technology enable? How much of that will we need to direct, right? For example, if you take individuality, which is at the core of humans, and extrapolate consciousness out, it may be reasonable to assume that we will continue to remain individual. And you might be super intelligent, a super intelligence amongst other super intelligences, all of which have their individuality. Equally, you could say, is individuality as meaningful when self-replication is not the problem, when we are truly immortal, or does it make sense for us to be connected? Regardless of whether or not we become one mind, our minds are likely to be connected in a far deeper, frictionless way than it's ever been when we look back at history, where there's always been friction in the way that we connect and communicate. When we're in that realm, for example, even if we are separate minds, perhaps our communication will be more telepathic. I don't want to say telepathic, but if you're communicating at the level of, you know, you're not speaking to communicate anymore.

You're probably just feeling or being.

Exactly, yeah. You could provide profounder insights by inviting the other person to experience a thing that you are imagining. And a lot of science fiction has explored this, right?

Right.

that this is perhaps the inevitable consequence of the evolution of, say, alien species. Often alien species are depicted as occupying a different realm of intelligence and awareness. And perhaps that's not so bizarre.

Right. Interesting how also you kind of put into perspective now, drawing it a little bit back to some of the topics that you touched upon during this session. Interesting how you refer to AGI only really— and now I'm putting words in your mouth, but AGI only really being possible once we find communication pathways that is more efficient. Is that a pun at silicon?

So I wouldn't say that AGI in the sense of general intelligence is only possible once we solve some of these problems around communication. I think there theories, but there may be a need to develop more efficient compute substrates than even silicon in order for us to achieve the efficiency we need to truly achieve general intelligence. But I'm actually quite surprised positively about the degree of progress LLMs have made, right, on existing. And also very rapidly after things like ChatGPT came out and how rapidly open-source models that are much smaller, have managed to perform quite so well, right? That's also surprising. Um, so I think that potentially general intelligence is possible on the existing substrates that we have, which is silicon. Um, but, uh, and by that I mean, you know, new semiconductor technologies, etc., like carbon-based transistors and the evolution of the technology may not be required for general intelligence. But I think, I think one of the most important advances we can make is zero latency communication, because I think it's underappreciated but would solve many of the problems looking forward in how we build not only, say, the metaverse infrastructure, how the infrastructure for the metaverse evolves, but also how intelligence evolves, how human interaction with these technologies evolve, right? Like VR headsets, One of the big problems with VR headsets is latency, eye tracking and latency, because you get that wrong, we have very negative physiological responses to getting that wrong because we think we've eaten mushrooms and need to vomit. Right.

Interesting. And how does that come into play for metagravity specifically? Would that solve your problems partially?

I mean, I think what we're solving is a more important problem in some senses, because it's one that's not appreciated at all. There is an utter lack of awareness for how difficult it is to scale a virtual world. Amongst almost everyone, even many in the games industry, there's an expectation that we can just scale up a virtual world simulator. And those in multiplayer games in particular, and especially open world large games, understand that's a lot harder.

Yeah.

So it's overlooked, and therefore it's important to focus on, and we're focusing on that. And our solution, our approach is also, I mean, if we keep reducing latencies, we are agnostic of that. If we eliminate latencies, our approach and solution is agnostic of that. It just gets enhanced by it. And what we're doing will accelerate and drive the need for lower and lower latency communications, media, and interfaces and technologies. So I think it'll demonstrate the value in that much more. But I do think it's an important technology that, again, is also overlooked a little bit because we're very— like, if you look at, you know, I know a few VCs in London, but elsewhere as well, there's a lot of talk about funding quantum computing. And here's the thing about quantum computing. Quantum computing isn't automatically going to be faster because there are certain problems with certain symmetry properties that quantum mechanically you can exploit in order to exponentially speed up algorithms. That's not universally the case. There are many algorithms that are not gonna, well, much of compute in general is not really gonna benefit from it. And the technology is much more primitive. So quantum circuits are sort of where silicon was back in the 1950s or '60s today, right? And silicon has had all of that development. So even once we've developed quantum computers, what we're really gonna use it for is like an accelerator in your computer that speeds up certain computations. And probably initially, those will be mostly simulations of quantum phenomena are gonna be faster on a quantum computer than on a classical computer. That's sort of the result we're gonna see. But quantum communication is useful for everyone, from a VR headset to data centers, to telepresence, to the future of AI and neural interfaces and our ability to scale content. All these things are bound by physical limits around the speed of light, and we exceed that.

But do you think that quantum communications progress is contingent or dependent on quantum, our understanding and ability to master quantum itself?

Quantum phenomena, yes. But it's a simpler problem than mastering quantum compute, figuring out quantum computing.

Oh, so you think it will— but will it happen before quantum?

Well, it ought to, right? It should and it ought to if we have our priorities straight. I think it should, yes.

Okay.

And I think it will, yes. But it just doesn't get as much screen time, let's say.

Well, all the VCs in London that are listening now. Okay, cool. It's crazy to me how we started speaking about simulation and we ended up speaking through a flow, ended up speaking about the future of ourselves as a species, which is fantastic, and I really enjoyed it. Another thing that's crazy is that everything seems to start from gaming. And I know that you're a gamer, and I know that the first solution that you're really working towards is for gaming. And so I wonder, why do you think that it's always about gaming that tends to be right at the beginning of innovation? What is it about gamers that's so special?

Yeah, it's an interesting question. It's an important question. And I didn't know if I'll describe myself as a gamer because It's very rare that I play games. And even as a child, when I play a game, I had a brother who loved to play games, and he's older than me, and he'd play and I'd watch him. But then when I'd start playing, I would start getting ideas and I'd go write code, trying to recreate mechanics on the basis of the idea, trying to explore some technology. So I've always been biased in that direction. But I'm interested in games because I think what's interesting about games is they're models of reality. on a computer that haven't the need to be precise and accurate. And that is a superpower. Because in science, for you to be able to make any use of a model of a natural phenomenon, it has to be accurate. Otherwise, it's useless. A game doesn't need that, which means that game simulations in games can iteratively improve over time. time in a manner that you can't in science. You either meet the bar of it's precise enough, accurate enough, scaled enough, and you can run it efficiently and yield a result, or you can't and it's completely useless. Very binary. Games can be, you know, the technologies can improve iteratively because at the end of the day, the goal is entertainment, right? And every little improvement and quirk can be exploited ultimately for commercial value. At the heart of what I believe, I believe that if you want to make a certain long-term macroeconomic shift happen, you've got to build a ramp of economic incentives along that path, right? So for example, like let's take games already. Games have been instrumental in establishing the modern computer architectures and technologies. Like, you know, I talked maybe about how in the early days of programming, say on a 386, the hardware was designed do arithmetic faster than older architectures, but they were much worse at doing real-time graphics. Unless you worked out how to manipulate the hardware, you had to write into the memory address region, which was being output to the monitor. And you had to write in a particular way to create pixels and circles and lines and images. And that was not, it's not like standard video BIOS stuff. You had to learn, figure out how to do So it gets you very low level, and the operating system isn't abstracting over any of that. You're directly manipulating the hardware. But then architecture, the hardware evolved to support games increasingly because people were building and playing games on these machines, and they're getting more and more exciting and being a big market. And it led to much of the success of Windows, and then later with things like DirectX, the API coming in to make graphics easier to do. Until Windows 98, you were strictly, the games were running strictly slower on Windows than if you just ran it directly on the them on DOS, which meant you were just directly running on the hardware. Because DOS did almost nothing as an operating system to really help you run, build and run a game. Like, there was very little abstraction. But this led to things like 3D accelerators, which initially were just to make graphics faster. And over time, you had more general purpose 3D accelerators, because you not only wanted to accelerate, say, 3D and the texture mapping and all that, you had a unified unified model for how the GPU is working, doing compute, which applied for all of those functions and were very programmable. And then later over time, those 3D accelerators became so general purpose that modern AI as an industry wouldn't exist without the availability of graphics processing units that were originally designed for games, which then became general purpose, right? Researchers would not have been able to train the neural nets just by buying a GPU off the shelf. They would have needed specialized hardware. Very few people would have had access to any specialized hardware. The exploration, the investigation would have been very localized and very limited. And therefore, we would not have made the discoveries that we have thanks to the broad-based distribution of this research. So if you kind of look back, that need to only provide some cool, new, unique feature that then could be entertaining was the only requirement which meant that technology could evolve. Today, if you look at what Unreal and Unity are used for, they're used in a lot of industrial simulation use cases. Right? If you look back 10 years, there were specialized engines for those things that are now dead as a business because they were too niche. If you look back 20 years, supercomputer architectures from 20 years ago were very different. Modern supercomputers are lots of GPUs and x86 processors, x86-64 processors, or ARM processors now as well. If you look back, so if you look back at history, you have so many examples of industries that had highly specialized solutions that were very niche to that industry, that gaming has gulped up, or games-related technology has gulped up over time. And that's the ability to iterate rapidly and make incremental improvements and still deliver economic value that given a few years, like the span of a decade or 2, completely outcompetes niche technologies. And I think that's why gaming is so interesting interesting. We made a deliberate decision with MetaGravity, having founded another company previously, Hadean, where we were sort of being steered towards prematurely going into non-gaming industries. I knew this was strategically a mistake, that you had to remain in gaming for a period of time. It's a necessary rite of passage. Companies like Epic and Unity who built Unreal and Unity, yes, today you use those engines to simulate factories and train driverless cars. and DeepMind uses it to build environments to train AI, but they started out purely for the purpose of delivering features for entertainment value. That was so important. That rite of passage was so important for them to become mature enough to then spill over into other industries. And I don't think there's a way to short-circuit that.

Amazing. Wow, it's a really comprehensive answer. You touched upon the takeaways of, sorry, the the value that we can take of the ability of simulate that goes beyond metaverse that we've spoken about and simulating virtual worlds. And I think to conclude, I would like to explore with the last point on all the other things that are very practical and that do, and that are helpful for us in the real world, not just the virtual one that we're able to really leverage its technology for, which I think is important for a lot of people to resonate with. So could you give us a sense of all the sorts of areas which we can leverage some of this engine for?

Yeah, I'm sure I won't be able to give you all the areas, given that there'll be so many yet to be discovered. But I think some interesting ones, like if you look at Facebook, I mean, for, I guess, the millennial generation in particular, and maybe the one slightly before it, it's been such an important way of everyone connecting, coming together, and immersing ourselves. But Facebook is a billion-user web website. And the equivalent of, so what does the next social network look like? It's probably not a website, right? It's probably an immersive 3D experience. And this is something, so social experiences are an important one that this technology will underpin. If you take entertainment, you have a limit in how many people you can meaningfully pack in a stadium and give a real kind of engaged experience to, but that's still in the thousands or perhaps tens of But when you move to the virtual sphere, you can have entertainment experiences, concerts, and so on that have millions of users, and not just locally in one region, but around the world being able to participate and enjoy an event. And that's like an immediate— these are all low-hanging fruit, immediate application areas where the experience can be enriched. If you think about e-commerce, That e-commerce has sort of— it was initially sort of the hallmark of an overambitious business idea if you go back to the late '90s, early 2000s. But today it has become so foundational to how commerce is done for the internet. But the next phase is even deeper, like knowing exactly what a product looks like, knowing exactly how your clothing might look like when you wear it. Right? This sort of experience has much deeper meaning when we talk about metaverse experiences, right? But also just connecting the world, like shared work, shared collaboration, overcoming geographic boundaries. Like you could be someone in some rural part of China or India and be interacting and working with, or collaborating with someone across the world, whether it's in a university department or a corporation, or simply finding like-minded collaborators and being able to work in a manner that enriches that collaboration, that experience, makes it feel local, you know, through telepresence, for example. That starts to become increasingly possible with this kind of technology.

Wow. Okay, interesting. Amazing. Well, thank you very much. I really appreciate it. We went over the, the virtual side of things, but also the real practical implications that we can already have today. And we also took quite a deep dive into the future and where it really goes. But I appreciate the intellectual honesty and willingness to explore these things in a public environment. And it's really hard to do. And I know it takes a lot of courage. So I appreciate that a lot. And thank you so much for choosing to come on.

Absolutely. Thanks for having me on. It's a really great experience and it's really thought-provoking. So I'm sure some questions will fester and I'll find better answers for them later and think, oh, I should have said that. But that's the great thing about this conversation, is thought-provoking.

Absolutely. And the reason of existence for Chaotica is to give a platform, and the platform will always be there. So if in 2 years we want to record the follow-up, I would be thrilled to.

That's awesome. Yeah, I'm looking forward to it.

Thank you so much.