Introducing 4D Creation Open Beta: NPCs, 4D Worlds, and the Future of Gaming with Roblox CEO Dave Baszucki

From “virtual doppelgängers” to “real-time dreaming,” online gaming platform Roblox is using AI technology to build the “Holodeck” envisioned in science fiction decades ago. Sarah Guo and Elad Gil sit down with Roblox CEO Dave Baszucki at Roblox headquarters to explore the intersection of AI, physics simulation, and the future of human connection. Dave discusses the evolution of the 4D creation tool in Roblox, a high-fidelity simulation that enables thousands of people to interact in real-time with photo-realistic graphics and acoustic physics. Dave reveals how Roblox is leveraging 13 billion hours of monthly user data to train native AI models that go beyond simple LLMs, enabling NPCs that can navigate and play games with human-like intuition. He also talks about how immersive communication will change video conferencing, how Roblox searches for unlikely talent outside of traditional elite universities, and how he balances rapid weekly iterations with keeping a “long view” on Roblox’s vision. Sign up for new podcasts every week. Email feedback to [email protected] Follow us on Twitter: @NoPriorsPod | @Saranormous | @EladGil | @DavidBaszucki | @Roblox Chapters: 00:00 – Cold Open 00:36 – Dave Baszucki Introduction 01:16 – Realizing Robolox’s 20-Year Vision 05:29 – Using 4D Immersive Simulations in Virtual Interactions 08:22 – Physics Engine vs. Photorealism 11:50 – Storing Roblox History as Vector Data 14:00 – Training NPCs - Moving Beyond LLMs 18:05 – The Future of the Game Designer 19:54 – Video Latent World Models 23:53 – Social Simulation - AI Companions and Virtual Relationships 27:26 – Why Asset Costs Haven’t Changed the Gaming Industry 29:52 – AI Coding in Roblox Studio 31:36 – The Roblox Creator Economy 33:57 – Long-Term Conviction vs. Weekly Iteration 37:50 – Dave’s Hiring Philosophy for Roblox 43:44 – Conclusion

• 0:00Speaker 00

  What proportion of you think your children will have at least one serious relationship with an AI throughout their lifetime?

• 0:05Speaker 01

  I think zero right now. I don't think we're close to crossing the human AI barrier, but it is worth thinking, things are changing every day. We have to incorporate and do that. And then there's also some weird universal truths of things that stick around for 40 years and you kind of have to blend them together. We feel if we build Roblox right, it might be the kind of thing that's kicking around in 40 years.

• 0:36Speaker 02

  Hi listeners, welcome back to No Priors. Today, Alad and I are here with Dave Bazooki, the founder and CEO of Roblox, the 3D immersive world where more than 150 million users come every day to hang out. We talk with Dave about the future of AI in gaming, their investments in AI, how NPCs are going to change and really transform games. Roblox has studio, one of the largest coding platforms out there, and their 20-year mission to build the holodeck. Welcome, Dave. Dave, thanks so much for doing this with us.

• 1:08Speaker 01

  It is so great to be here and thank you for doing this in Roblox headquarters.

• 1:12Speaker 00

  Well, exciting to be here.

• 1:14Speaker 01

  Yeah, great to have you guys. Thanks.

• 1:16Speaker 00

  So I think we want to start with a big picture question. I think you all have been really visionary in terms of where all this is heading and AI is obviously having a big wave and effect on gaming and what's going to happen in the future there. And that's across things like world models. How do you think about NPCs and their evolution? How you think about assets within games? How you think about world creation? So we just love to hear your views of like big picture, 10 years from now, where are we going? Like what's coming?

• 1:41Speaker 01

  Yeah, so two-step on this one. I would say first, I know you both dabble in investing a little. So there is, there is a Roblox business plan PowerPoint deck from almost 20 years ago that we pull out sometimes and we're amazed at the fidelity of it. It pretends a new category. One could call it the category of human co-experience. It's got the sizes of a bunch of companies from 20 years ago. So you can see social networking companies. You can see YouTube. Video, you can see toys and you can see all of that. And what this, what it imagined is really a new category. Some have called it the metaverse, some have called it the, the holodeck,

• 2:22Speaker 01

  which is really the ultimate high fidelity simulation where people can come together and do stuff.

• 2:28Speaker 00

  It's like a Ready Player One, the movie.

• 2:30Speaker 01

  Exactly. And so we do have that business plan slide. And, and what we had always imagined is if this high fidelity space was backed by physics simulation and reality simulation, you'd be able to do stuff. You'd be able to build a car, put wheels on it, drive it around. You'd be able to go to a birthday party and blow out the candles. You'd be able to chop trees down and make a, a house of it. So that, that was literally the genesis of Roblox. And here we are 20 years later, there's so much more to do. But one vision of AI is how do you superpower the creation of that holodeck? And how do you get it photorealistic as soon as possible?

• 3:14Speaker 01

  How do you get 10,000 people in it instead of a hundred? How do you do acoustic simulation that sounds realistic with 10,000 people? So in that sense, AI is super interesting about just trying to get a vision that's been around for 20 years and has arguably been in sci-fi, once again, snow crash holodeck forever. There is another interesting vision, and, and I think it's useful comparing these product things. There's much more, what's the future of just a single person by themselves dreaming what they want to dream? And we sometimes call that real time dreaming category. It's a category that we can see a bit of the trappings of with short form video.

• 3:59Speaker 01

  There's a little bit of that in short form video, you know, someone's 2:00 AM in their bed doom scrolling. It's reacting to dwell time, it's reacting to what you favorite. And you're kind of getting a little bit of a pre-built dream. That I think we sometimes imagine could go all the way to the famous Tom Cruise movie, Vanilla Sky.

• 4:20Speaker 01

  Where he literally realtime dreamed for several years in an imaginary universe, where everyone around him was primarily an NPC and he didn't even know he was doing it. So that

• 4:33Speaker 02

  So that's more world building as content.

• 4:35Speaker 01

  That, that is exactly right. And so if you, if you put those at two extremes, one is communication platform, high fidelity with people you know. The other is realtime dreaming where everyone's an NPC. Everything in between is possible. And I think we see everything in between ultimately as being possible. And we may see weird product categories we never imagined. You know, I, I think people are probably thinking, does the text prompt turn into a voice prompt? And does it automatically turn into a video prompt? If there's like, I want to learn French, video, how you're in a French cafe, start talking to an NPC.

• 5:19Speaker 01

  So I think, I think, I think there's a wide range of products that start to get into this space we're interested in, which is kind of that 3D reality space.

• 5:29Speaker 00

  Do you think that same space will exist for both consumer and business applications? Cause people also talk about this in the context of you're instead of doing a Zoom, why don't you go into a shared space with someone and mute with that?

• 5:38Speaker 01

  I, I think

• 5:39Speaker 00

  What degree of your life becomes virtual?

• 5:40Speaker 01

  Yeah, we have seen, I think on the communication side, people just want more and more fidelity. We can go back to paintings on the wall of the cave. People want more fidelity. We have smoke signals. We have the mail system. We have the Pony Express. We have the telegraph system. We have the phone system. We have the arguably the COVID initiated video. Uh, we also did have the COVID initiated 3D experience on Roblox, actually. And, and we found side by side video. A lot of young people used Roblox as an early test case of how to co-experience. I think as the technology for multiplayer 4D simulation gets better and more photorealistic,

• 6:25Speaker 01

  it's almost going to be like video is the down sampling. And it's like if if we were having this 4D immersive communication, we'd say go to legacy analog mode and make it look like Zoom, basically. And and we could always do that. But we'd be able to do things we could never do on Zoom. You know, you'd say, hey, let's pop up and go walk around my office. And I want to show you something, come and follow me kind of thing. So I do think the 4D simulation, as it gets photorealistic, will be a superset of video. The the other thing for business is there's a few things that 4D, and I use 4D because it's not just 3D shape, but it's function in a way.

• 7:08Speaker 01

  And will provide that video does not provide is acoustical things. And there's some real tricky physics problems when you're having a company meeting of 1,000 people. If we try to do it on Zoom, we have like a bunch of squares. What's who's live, who's not live? How do we mix the sound, which is very tricky? If we do it in a three dimensional simulation, I can hear everyone with attenuation based on how far they are. And it can actually be a much more natural experience. Like I walk closer to you and I I hear you. And that that may actually be a better human interface paradigm. There's one final technical thing that's going to be very, very difficult

• 7:50Speaker 01

  is when people try to sing happy birthday to you together. And because no matter whether you're on Zoom or whether you're in a 3D simulation, you're hearing them like 30 milliseconds later. So what I think we're going to find is people are actually allowing each one of us to sing happy birthday with a time forward extrapolation of each other and then remixing it. So I do to, you know, for the business side, I do think ultimately we're going to see some interesting conferencing solutions.

• 8:23Speaker 02

  Roblox has this amazing wild story with physics simulation or physics engine as like part of the the birth of Roblox. But then today, or for the last decade, I have, I think of this as a immersive world where people have their first coding experiences. They build amazing games, interact with each other. I don't think of that world as being as being super focused on realism.

• 8:46Speaker 01

  That's right.

• 8:47Speaker 02

  So is that something that's changing for you guys? Is it enabled now? Like, how should we think about that?

• 8:52Speaker 01

  I think what you, the way I would think about it is we're pushing really hard on the physics simulation capability. And at the same time, we have this incredible free market that we can't control on content generation. I would say we are just started talking about full acoustic simulation. I just tweeted about it. So actually, you know, approximate, you know, attenuation and all of this kind of things and echoes, that's going to start happening. On the physics simulator, it's behind the scenes gotten better and better and better. It'll keep getting better. But what we have found, I would say, is the diversity of developers are beyond our expectation.

• 9:31Speaker 01

  And I would say some of the things that have gone viral, dress to impress, for example, grow a garden are less physical. But they have leveraged cloud capabilities that we've built side by side. You know, grow a garden was essentially able to build an experience where things keep growing, even when you're not there by leveraging a lot of the cloud persistence we provided. Dress to impress was able to do some very interesting things with how they run contests and layer clothing on. So I think we're generally, you know, the specification for our product could be, we just want to simulate the real world with 10,000 people.

• 10:14Speaker 01

  And the more we go in that direction, the more we'll just see creators using it in different ways.

• 10:20Speaker 00

  How much attention have you been paying to world models and what's been happening in the AI world in terms of how people are shifting how they do physics simulation?

• 10:28Speaker 01

  I think we're really watching it. I think that the biggest thing we have focused on, and I think we'll continue to focus on, is how do we get that 10,000 player multiplayer type experience? And how do we fall back to 100 or 10 or one? We feel it's super robust to build a communication platform. And, you know, that's part of our goal of getting to 10% of gaming. The the technology for this, I think we're going to start unraveling it over the next five to 10 years is what's the most efficient way to synchronize the state of 10,000 people? And and when we-

• 11:06Speaker 00

  It's a deep infrastructure problem versus necessarily a physics simulation.

• 11:09Speaker 01

  How do we synchronize the state and the memory over the last five hours of 10,000 people? Where have they been? Where have they been in 3D space? What have they been doing? The the world model thing is very exciting because they're storing memory in literally video Latent, you know, a minute of frames, it's showing super early promise. I think the technical revolution is going to be what's the ultimate synchronized state for 10,000 people? Is it in video Latent? Is it in much more native 3D format? Is it a hybrid new discovery of some 3D video latent space to be determined?

• 11:49Speaker 01

  I I will share that the format we're using, part of the vision for us is to ultimately store the history of everything on Roblox and to store it, not raster like video, but store it in vector, which means the ability to play back anything that's ever happened on Roblox. That's 13 billion hours a month. And so the ability-

• 12:14Speaker 02

  What do you think that does for you, the history of your world?

• 12:17Speaker 01

  So there's a, there's a sci fi book by Arthur C. Clarke called The Light of Other Days, which talks about the fantasy of what would happen if we had infinite playback in our world. It's like a real social problem, because if everything we've ever done can be played back, actually in that sci fi book, it's not just everything could be played back, but you could look at any playback. And so everything's completely transparent. There's no private conversations. It actually delves into that. I think for us, we would then use it very thoughtfully and privacy compliant and judiciously, but for very simple things, you can imagine, oh, there was a safety incident. Let's go back.

• 13:02Speaker 01

  And put five cameras and listen to the audio and see what happened and make a good judgment call. That's kind of interesting. There's a really special thing that maybe I did with someone in my family. That's really special. Given I have access to my history, go play that back and possibly because it's 3D vector data, not raster data, reshoot it from it. Make a cool video of the best moment. And I think where it ultimately ends up for us is there's a huge push now for people to find kind of hybrid video ASDW, human interaction

• 13:43Speaker 01

  data. People are starting to train on video with keyboard stuff and like, where do I find 200 million hours of this data? The data we have, which is 13 billion hours a month is, can be reproduced from any camera angle and can interact with the 3D space. So it's very powerful data. It leads to a really interesting computer science program idea around how to create great NPCs that are more than just an LLM.

• 14:12Speaker 00

  Hey, tell us more about that. Actually, that sounds really interesting.

• 14:14Speaker 01

  Yeah, go check out on XR X feed. I think the last day or two, we released a bunch of native video demos and maybe we can even play them back here. But what we are starting to train is NPCs based on a combination of all of the data we have in a privacy compliant way that are starting to navigate and starting to play games. The challenge, I think, would be challenge level one, everyone has access to an NPC that can get pretty good at playing any Roblox game. Super cool. Challenge level two is by watching, if I so opt in in a privacy compliant way,

• 14:55Speaker 01

  my behaviors, the way I act, literally my gestures, the way I look at things, the way I talk. Would we allow you to have your own virtual doppelganger if you wanted?

• 15:06Speaker 00

  That's fascinating.

• 15:07Speaker 01

  And then number three would be is if I, if I have a good virtual doppelganger. Just as agentic is all the buzz in other areas, is there a simple user interface for agentic virtual doppelgangers? And that that starts to get interesting because one could imagine sending their virtual self out to go do something. It could be, you know, my, my son or my daughter wants to play with me, I having to do some work. Could my virtual doppelganger fit in for 15 minutes? Sure, that's great. That could be really interesting. And there's a lot of other things, just like we're going to see agentic space in the 2D and the workspace and the productivity

• 15:51Speaker 01

  area. I think it's going to be interesting to imagine.

• 15:53Speaker 00

  So you're basically using these 18 billion hours a month to train?

• 15:56Speaker 01

  13 billion.

• 15:57Speaker 00

  13 billion, exactly, to train.

• 15:59Speaker 01

  Native NPC model, that will be a Roblox privacy compliant model, never share the data. And then over time, no ship date, make that customizable as your own virtual doppelganger, or for creators with the addition of a prompt, you know, upload Benjamin Franklin's history, or it's already embedded in an LLM, and just say, hey, I want Ben Franklin.

• 16:24Speaker 00

  Yeah.

• 16:24Speaker 01

  And I want a happy Ben Franklin.

• 16:27Speaker 02

  Are there experiences that you imagine creators making with these NPCs that are really different from?

• 16:34Speaker 01

  Well, I think there's a lot of standard game NPCs that that I think have done such a good job. I wouldn't claim when these types of NPCs would be good enough, but like Grand Theft Auto, right? They've been working on this forever. That is a very rich world with a lot of very high performance, you know, NPC behavior, imagining someday, many people can create them. I think to make that happen, these platforms have to be very cloud connected platforms rather than running local to spin up inference on all of these things. But you

• 17:14Speaker 01

  could imagine that. You could imagine a very young creator saying, I'd like to build an American history. Thing where you go and talk to Benjamin Franklin. So I think we would want to enable both of those.

• 17:28Speaker 00

  It's really interesting because basically the transition is, this is going to be a poor analog, but there's a transition that happened in self-driving. We went from mapping out heuristics and sub-behaviors and sort of fine detail to capture edge cases to just doing end-to-end models for deep learning. And it sounds like you're basically considering doing the same thing for NPCs.

• 17:47Speaker 01

  I think that's the exciting computer science capability is rather than, you know, all of these wonderful techniques people have used, you know, decision trees and this and this, getting to a purely more general modeling solution would be really fascinating.

• 18:04Speaker 00

  That's super interesting.

• 18:05Speaker 02

  What do you think the role of like a Roblox game designer or any game designer is five, five years out from now? Do you think it's different?

• 18:14Speaker 01

  I think I'm optimistic that typically industrial revolution or whatever revolution, I think we've chatted about stuff like this before. There's a thought that there won't be any jobs after this. I think I'm a little bit more optimistic that a new class of things will show up. And I think there's a huge opportunity for more creators. I think, you know, I'm not to the point where I believe, you know, what we call AI slop today is going to take over. Like, I think the human touch has got a long runway on it. So you could

• 18:53Speaker 01

  imagine more diversity or quality of experiences supplemented by AI. But I do think the role will be more leveraged and the quality coming out of five people will be astounding, you know, but we'll come to expect that quality. So quality will go up. I think iteration on testing will go up. I think where we see platforms like Roblox going is just as people are starting to spin up agents to develop code. I think people, hopefully on the Roblox platform against the Roblox cloud.

• 19:32Speaker 01

  You'll be able to go away for 24 hours and your agents are going to be like tweaking and testing, spinning up a Roblox experience, sending 20 NPCs into them on various client emulators. There's an NPC on a phone, there's an NPC here tuning the game. So hopefully you'll feel more power in what you're building.

• 19:54Speaker 02

  I'm also curious, there is a set of research around world models and I'm sure you've seen that is just trying to directly generate game experiences with like, there's no physics engine underneath, like you don't control the mechanisms, it's just video. What's your view on that?

• 20:10Speaker 01

  I think it's like a super exciting, interesting thing to think about. I think, like I said before, I think one thing is going to be the higher the fidelity is memory stored in video Latent? Or does there need to be another research breakthrough? Where memory is ultimately stored in some new latent space that's 3D or 4D dimension? But the fidelity is amazing. I think that's really beautiful. And I think that the way we would see the ultimate architectures of these platforms is probably for a while, not an all in one thing. We would probably see some hyper efficient thousand person synchronization state engine.

• 20:53Speaker 01

  We might also see some of the photorealism not being done server side, but being done in intermediate spaces or client side as well. And we may see multi-step pipelines where there's the raw framework, there's 3D up sampling, then there's local 2D up sampling. We're actually going, oh my gosh, like people are putting together a bunch of pieces, you know, dedicated NPC capability, dedicated synchronization capability, 3D up sampling, 2D up sampling, and potentially world model stuff. You know, world model, you know, you can imagine that being used even in its early state today, rather than maybe a gameplay state where rather than

• 21:36Speaker 01

  Making a video, you talk about the video while you walk around and you kind of say, put me in a Western world. I want to go left. I want to go right. Add a few horses. And that video actually goes to a multiplayer experience. One could imagine that. I think what a lot of other people are wondering is the short form video, does someone hit it just like they did with short form video into a highly retentive, like local dreaming product that could just react to what you're doing? I think it's still early for both of those, but it's a big possibility. I would say that we are doubled down on a hybrid multi-AI tech stack that supports multiplayer natively.

• 22:17Speaker 02

  Yeah, I think one of the ideas that's like internationally very interesting has been the idea that you can generate like serial micro dramas as a form that is not yet that popular in the West, but seems like a good format for the technology.

• 22:31Speaker 01

  Micro dramas, I've watched a few. I mean, it's really interesting. You know, it's fun to see like some entrepreneurs tried to do that. I won't mention who, like five or 10 years ago, then they naturally emerged kind of thing. It's interesting to think if those micro dramas are easier to produce, they don't need live actors, but you know, what is the plot line that really resonates? You know, we'll, we'll see with that.

• 22:56Speaker 00

  You can mass test that at scale if you're using AI, right? You can actually, you know, a very large number of iterations. This is kind of one of the arguments people make around there. For me, paradox of like, why haven't we seen alien life? And the idea is it's just, there's alien life. They're just stuck in these virtual worlds now where, you know, it's funner than the real world.

• 23:13Speaker 01

  Like what's the famous quote, where is everyone?

• 23:17Speaker 00

  Yeah, exactly. So one option is there in some virtual landscape that's more compelling than the real world. And so therefore there's no reason to travel the universe.

• 23:25Speaker 01

  Or it could be, you know, we could really start conjecturing. What's the future evolution? Is it human or machine life and all of that?

• 23:34Speaker 00

  Yeah.

• 23:35Speaker 01

  And once it gets to maybe human machine life, what happens? So.

• 23:38Speaker 00

  Yeah, this assumption. Yeah.

• 23:40Speaker 01

  Yeah, that's a, yeah, interesting.

• 23:41Speaker 00

  Yeah.

• 23:42Speaker 01

  That's the vanilla sky.

• 23:43Speaker 00

  Yeah, exactly.

• 23:45Speaker 01

  Or the optimistic thing is more the holodeck thing. Like everyone's hanging out with other people rather than a simulation of themselves.

• 23:53Speaker 00

  Yeah.

• 23:53Speaker 02

  Are there experiences that you imagine from a like a communication or a social perspective that you think are going to be just different or new or that you're going to have real insights into from these like really great NPCs, right? Because I'm sure you've seen what has happened with the rise of AI companions and how engaging they are, given they're just text interfaces.

• 24:16Speaker 01

  There's some fun things that like there's a Black Mirror dating episode. I don't know if you've seen it where they have, they have a lot of virtual doppelganger power. And obviously, I'm not so, Roblox is not doing dating. We're just talking about this right now, you know, blah, blah, blah, 18 plus ID verified, all of that. So let's not go crazy, but let's think about a future, I'm in the Black Mirror episode where you literally can spawn the life of a thousand virtual doppelgangers. That thousand virtual doppelgangers, you're on your new dating app in three minutes. You know, someone else's virtual doppelgangers live a thousand virtual lives with your virtual doppelganger. And then they say, what's the success rate? And when the

• 25:00Speaker 01

  like the thousand virtual doppelgangers say 98% of the time, we had a good life, then you say, oh, we should go on a date or something. So there's

• 25:10Speaker 02

  Like social simulation.

• 25:12Speaker 01

  There's a lot of things we I think are going to be really hard to project. I think one of the things we would focus on is the more we can just build raw, high performance infrastructure, have AI as a service, have all of these things capable in a cloud, have the same thing run on a phone as a computer, have it auto translate. Hopefully, we would discover some of these things.

• 25:37Speaker 02

  Yeah, I'm really optimistic about that, given the engagement that people see with these companions as just a text interface with like the beginning of memory, right? Versus if you're Roblox, you have embodiment, you have, as you said, cloud persistence. And I think it's just going to make the characters much more interesting very quickly.

• 25:54Speaker 01

  Thinking through how we would embody memory for NPCs, which is more than just, you know, a prompt. One could imagine any NPC on Roblox, because we have recording of the whole thing, they actually have access to everything they've ever said or done in a very lean format. So they can go back through that, retrain on that. So that hopefully would be a native part of the platform. But I do know, yeah, there's a lot of interesting engagement with these types of companions and imagining fully 3D versions will be like pretty fascinating.

• 26:29Speaker 00

  Yeah, a friend of mine basically asks at dinner parties, he runs a woman AI company and it's sort of controversial question for the group is often what proportion of you think your children will have at least one serious relationship with an AI like throughout their lifetime?

• 26:43Speaker 01

  I think zero right now. I do. I don't think we're close to crossing the human AI barrier. But it is worth thinking.

• 26:53Speaker 00

  And then even in the context of like, even if it's superficial and there isn't true intelligence behind it, just given the behavior with text, once you start having video, virtual worlds.

• 27:02Speaker 01

  I can imagine a real time coach therapist like in your earbud, just hanging out all the time.

• 27:07Speaker 00

  Yeah.

• 27:08Speaker 01

  So I think for that kind of less evasive or over your shoulder coaching, I think that seems like.

• 27:14Speaker 00

  That's where the major use case, it seems for some of the like chat GPT and other platforms where mental health advice or coaching or other things like that are actually one of the major use cases.

• 27:23Speaker 01

  I think that will be arguably very valuable.

• 27:27Speaker 02

  Can I ask like an industry perspective question, just given you've worked in gaming for a very long time now? Why do you think like, there's a lot of discussion about how assets of different types are cheaper to create with generative tools? I don't think that's really changed the dynamics of the gaming industry yet. I don't know if you would agree with that claim or explain it.

• 27:48Speaker 01

  One could say, no matter how cheap it is to create assets, the expectation of quality from consumers goes up at exactly the same velocity.

• 27:57Speaker 02

  Okay.

• 27:58Speaker 01

  And so, you know, what was not possible a while ago, the the expectation of consumers will go up. I think there's a, there's a general notion that the technology for the gaming industry.

• 28:13Speaker 01

  Has to move to cloud vertically integrated and asset management can't just ship on a DVD or be a giant download. I think we're, we're really seeing, we've launched dynamic LOD for textures, meshes is on the way. So you could imagine every single asset in the game, sound, 3D object, image can exist in various formats. Traditional format is it's like a texture or a mesh or an audio file. Future format could be it's an AI prompt and it's generated procedurally on demand.

• 28:55Speaker 01

  So there's these two things. They're both very AI enabled. One can then imagine that you have to be cloud connected because for all of these assets, you know, if a full range of LOD, just like with video, streaming, up sampling on low end devices, you need low LOD, high end devices, you need 4K LOD and ultimately on demand AI generation. That then gets into, once again, one of the demos is probably on 3D up sampling. The notion that if someone makes a very primitive experience on Roblox, they would augment it with a prompt and they just say, I know it's kind of primitive, but make it kind of look medieval and more realistic and have all of

• 29:40Speaker 01

  those assets auto up sample to in 3D in the cloud for free on demand. That, that kind of just in time thing is some of this stuff. We're thinking about how AI can accelerate if you have a cloud platform.

• 29:53Speaker 02

  Roblox is also, from the studio perspective, one of the biggest development environments in the world.

• 29:58Speaker 01

  That's right.

• 29:59Speaker 02

  What are you seeing from the perspective of AI coding there?

• 30:02Speaker 01

  Yeah. So this is great. So once again, industry standard ways of using a different model with studio. So if you're in cloud code, you can control studio is super exciting. And so a lot of our users have started gluing these things together with the hope ultimately, you know, they can build stuff from there. Roblox Studio itself is running a native assistant as well and a native code engine. And I think the the pattern for that is Roblox Studio will more and more for code, follow many of those industry best practices, but in parallel, needs to have kind of an environmental generation kind of component that's very unique to that.

• 30:45Speaker 01

  And I think, I think the future of environmental generation is, once again, any prompt you can imagine, text, image, video, possibly walking around in a world model to define a world, having iterate on that, iterate on that to a primitive 3D skeleton, do I like it? And then iterating on that to a fully functional game. So as that, I think that's our vision is to run those side by side. So first time studio user gets it out of the box. At the same time, we like plug into all of your AI workflow. I think the final thing is because it's so cloud connected, that thing we talked about earlier,

• 31:25Speaker 01

  spinning up jobs, spinning up agents, running test plans, kind of keep grinding away, trying something different, is like you have all of that cloud capability.

• 31:37Speaker 02

  Just out of curiosity, what is the like, cycle of development for a successful Roblox game?

• 31:43Speaker 01

  The, what we have seen

• 31:45Speaker 02

  Yeah.

• 31:46Speaker 01

  is over the years, bigger teams, so that, you know, the top Roblox creators now are well into the 10s and 10s and 30s plus million a year. Like, it's really serious stuff. We have seen in the Roblox creator community, I think, a healthy sign that creator number 1000, and what is their average yearly revenue, is going up faster than creator number one. So that's a bit of a healthy, long tail. And I would say all the way out to creator 1000, there's such a huge community of people making a living. What I think our creators have found is, it's gotten both more healthy, but more competitive. And I

• 32:28Speaker 01

  think we've seen the pattern now of much more live ops, in that because this is cloud, you can update your experience, just like good websites, every day or every week. And that, that is just kind of a constant practice. So I'd say update frequency much more rapid. Which is great for users, because they get

• 32:48Speaker 00

  Updated content at a faster rate.

• 32:50Speaker 01

  That's right. Users love updates. They love new things. It keeps games fresh. So I think the the other thing we have seen, we've on the discovery side, we've tried to and more and more taken the approach. We want discovery to be completely transparent. Like, these are all the things that are going into the discovery algorithm. We we check it. There's a benefit to that transparency and discovery in that it actually keeps a lot of pressure on us, you know? Oh, my gosh, someone's going to game discovery. Well, then we better make it really good, right? And and transparent.

• 33:31Speaker 01

  I actually think transparent discovery and recommendations is an interesting trend for the industry worth talking about. But so so with that, we have seen also the kind of the content distribution get richer and a little bit less spiky. There there's now like the top 20 experiences on Roblox are all really good and all kind of vying for the number one slot, which maybe wasn't so true three or four years ago.

• 33:58Speaker 02

  Zooming all the way out for a second, like I I I think it was true five or 10 years ago when we met. It's true. And I was talking to one of the members of your leadership team today. They will say you're a extremely high conviction, very long term oriented leader, right? So they, you know, 20 year plan for Roblox and we're going to do things the way we have a vision toward. You're obviously also all in on AI, right? Very excited about the technologies and investing in them. How do you like square these things where you have a very long term view when so much is happening in the environment?

• 34:29Speaker 01

  I think we we always want to pair, take the long view, which is one of our four big values with get stuff done. And so

• 34:39Speaker 01

  it's almost like in those old Gartner charts, you know, take the long view, not take the long view. Get stuff done, not take the stuff. Like, the magic quadrant is take the long view, get stuff done. We we typically mean inside the company, rapid iteration. And so if someone has a view of some Exciting new product and we're going to ship it in six months. That can be a very scary thing unless it can be broken down. They're like, we're going to ship this thing every week and iterate towards this. But we have a good target of where we're going. And so I think inside the company, we, I'd say our AI team, our facial age estimation team, our safety team.

• 35:23Speaker 01

  These are literally things working on a weekly basis, which allows a fairly fast reaction time. AI, I think, is arguably, as far as speed right now, just every day.

• 35:36Speaker 02

  But you mean, you make it sound pretty simple in that, like, all the things that you're trying to get done on a week or a six month time scale. They're all aligned toward a vision that like isn't really affected by it.

• 35:46Speaker 01

  If you have your six to 12 month vision.

• 35:49Speaker 02

  Yeah.

• 35:49Speaker 01

  Hopefully that's moving not too fast. But at the same time, the weekly iterations are moving very quickly in that direction. And, and you know, that you could say the Roblox vision is ultimately to build the holodeck. And it's 10,000 people, it's multiplayer, real time modification of the environment, it's NPCs. Photorealistic, that's a pretty stable spec for 20 years. And so then.

• 36:17Speaker 02

  I actually feel like that's the biggest takeaway for me here. Like I spent a lot of time with software CEOs, right? And I think there's a lot of actual concern about like, is the vision we had for the company in terms of its durability or its value. Still as valuable as we thought five years ago. And for Roblox, perhaps because you were, you know, pointing at something so far away to begin with, it's like pretty clear, you know.

• 36:42Speaker 01

  I think.

• 36:43Speaker 02

  AI only moves you toward it.

• 36:44Speaker 01

  We have crisped that up over the years. I would say two years ago, we were a little less crisp. We would say, we're going for a billion DAUs. That's a good spec with this spec I just mentioned. It's a little harder to operationalize around product management teams and that. So we, we took a step, a stepping stone and said, we want to get to 10% of global gaming content on the way to. It's actually remarkably good step. It's about 300 million DAUs. It's about 20 billion. It's about 3X where we are. It can be forensically torn apart. We can see what part is the USA market. We actually have a higher target for US than 10%. And so that has been very clarifying.

• 37:29Speaker 01

  I think for a software CEO, always knowing both 3X and 10X is very helpful. Cause if you don't know 10X, then it's hard to sleep at night. If you don't know 3X, it's hard to plan forensically how we're going to operationalize that, how quickly we're going to get there.

• 37:50Speaker 00

  One other thing that I think that Roblox has done uniquely well is how you've thought about hiring and how you've rethought certain aspects of that.

• 37:59Speaker 01

  Yeah.

• 37:59Speaker 00

  And I think that's really led to people who have more bias to action, who are moving quickly, who are iterating rapidly. Can you talk a little about how you've rethought some aspects of that? If you're willing to share that as part of it.

• 38:08Speaker 01

  Yeah, I mean, one is we've been very public that we really want values aligned people that are good problem solvers, that are very creative. And it's hard to do signal to noise in that kind of a hiring thing. So what? We actually bought a company about five years ago that was doing their best scientifically to build 3D assessment tooling for some big companies to really find people who are problem solvers and creative.

• 38:41Speaker 00

  That's part of what I was referring to.

• 38:42Speaker 01

  Yeah.

• 38:42Speaker 00

  I was facilitating that acquisition.

• 38:44Speaker 01

  So we, we bought Embellis. We got a great team of scientists, people that had been involved with the SAT test and all of that. And we've tuned that to our new college grad intern program so far. We've essentially, I believe, in a very fair meritocracy kind of way, taken the, whatever it is, 10, 50, 60,000 people, run them all through what we've, I think, shown to be somewhat of a very fair, non-weighted by any other social factor assessment.

• 39:18Speaker 01

  And the people that come out of there, we have an incredible confidence that they're problem solvers. They do a bunch of other things really well. So we kind of have this building wave inside the company of three years ago, two years ago, one year ago, these 100, 200, 300, 400 people that hopefully were like mentoring some of the best future leaders in the industry.

• 39:42Speaker 00

  Yeah, it's super interesting because I have a number of friends who always talk about how you discover untapped sources of talent, or how do you find the best people in the world, and are there ways to test for that, or sort of screen for that?

• 39:51Speaker 01

  We've done a lot of testing on this, fairness testing, looking backwards on it. I'm very confident in where it's gonna go for us.

• 40:02Speaker 00

  Super interesting.

• 40:02Speaker 01

  Yeah, it's like a very systems oriented approach. The one thing I we have found, which is a little edgy, might be the correlation between traditional universities and our own testing. And we

• 40:17Speaker 02

  Sorry, what's edgy about that?

• 40:18Speaker 01

  Well, I think people typically think like, I went to Stanford, that's where all the top talent is. We have found that not necessarily to be true. And we have found community college, you know, the small Midwestern engineering school, like because we're assessing our own way, we kind of, we basically ignore the signal of where you went to.

• 40:39Speaker 00

  I've heard that anecdotally from other CEOs now, particularly for certain cohorts, like the cohort that came during COVID. For a lot of the top universities is perceived as like a weaker cohort. And I don't know the reason or the interpretation behind it. But it's interesting to see this shift in terms of how people even think about where is some of the best talent. So it's interesting that you're kind of saying.

• 40:58Speaker 01

  So we get, we get to be to have self determination. We get to say colleges, you do what you want, all of that. We're just going to do it in a fair way that we think is our way. And it's worked out.

• 41:10Speaker 02

  Can you give just, uh, because it's so interesting, like any intuition for something you, some way you test for problem solving capability?

• 41:19Speaker 01

  I think we make public some of the tests that we use. And the, the cool thing is they're all built on Roblox. They're all tested by lots of people. And they're arguably really interesting three dimensional problems like programming a factory or like you're using some new geometric language to program a robot. You know, it's almost like turtle graphics and you can play with that. And see how.

• 41:50Speaker 02

  We're gonna try to get jobs.

• 41:50Speaker 01

  Yeah. I would be, I'm, I'm actually might be a little afraid to take the test myself, given how good some of our candidates are.

• 42:01Speaker 02

  You know, as a long-term predictor of the future, is there something happening in AI that you believe or hold with conviction, like five years out, you think not everybody believes yet?

• 42:10Speaker 01

  That I believe that not, I, I think typically it's the other way. More people believe more stuff than I believe.

• 42:18Speaker 02

  You think of yourself as a skeptic?

• 42:20Speaker 01

  Well, I think of myself as.

• 42:21Speaker 02

  He's building the holodeck, but he's a skeptic.

• 42:24Speaker 01

  I, I think I'm more like a, the power of time and the power of compounding rather than like an amazing invention and just rapid iteration. One, one thing I constantly am amazed by is either the graphical user interface or Microsoft Excel. I think I was playing with Microsoft Excel in like the eighties. And here we are 40 years later. It's not that different. Yeah. So.

• 42:51Speaker 02

  Very useful.

• 42:52Speaker 01

  So it's kind of like this interesting compliment in AI world. Things are changing every day. We have to incorporate and do that. And then there's also some weird universal truths of things that stick around for 40 years. And you kind of have to blend them together. So I think that's how we, I, I think the good news is we feel if we build Roblox right, it might be the kind of thing that's kicking around in 40 years.

• 43:19Speaker 02

  Awesome. Thanks so much, Dave.

• 43:20Speaker 01

  Thank you for joining us. Okay, cool. Cause yeah. That was really good.

• 43:23Speaker 02

  Yeah.

• 43:26Speaker 02

  Find us on Twitter at NoPriorsPod. Subscribe to our YouTube channel if you want to see our faces. Follow the show on Apple Podcasts, Spotify, or wherever you listen. That way you get a new episode every week. And sign up for emails or find transcripts for every episode at no-priors.com.