Full Episode: The AI Industrial Revolution

Welcome. You're listening to the Naval Podcast, your authoritative source for new knowledge. We're trying something new today. Uh, I have 3 frontier founders with us, 3 good looking guys actually, and a 4th good looking guy, Naval. And let me just introduce everybody. Guillermo, the G, Rauch. Um, he's building Vercel into an AI cloud for the world of agents and whatever comes after that.

Good to be here.

Blake Scholl, he's building supersonic aircraft in his own factory and jet engines as well. Blake's company, Boom Supersonic. And then Max Hodak from Science, he's building a biohybrid brain interface that grows living neurons on silicon to restore sensory functions like sight, but then eventually to explore new parts of the brain and new senses. All three of these guys are not composing their products with off-the-shelf parts. They're building their own factories. And, you know, we don't care as much about what they're building exactly as we do about what they're learning about how they're building. What's the new knowledge they're generating? What's their alpha? What principles are they discovering that other founders can learn from? What are they trying to figure out right now? And also, what are the cutting edge or crazy ideas that they haven't even talked about yet and they're still forming in their brains? Naval, do you have any reactions to any of that before I jump into Guillermo Rauch?

Yeah, let's just have fun.

Yeah, you guys should just jump in.

Yeah, so I can't remember my exact quote, by the way, but I've been really pilled with this idea of software factories and the job of the engineer being something that you just show up to work. You used to, used to ship the output directly and everything inside the company was you know, how good is person A at shipping output B? And now what's happening is the way that I'm judging you as an engineer is like, are you producing the factory that will produce multiplicative outputs B through Z, right? And that's a pretty significant change because basically like we used to believe and it used to be somewhat controversial that there's 10x engineers. Like now clearly there's 100x or 1,000x engineers and the world hasn't fully adjusted to this.

I used to get flamed on Twitter for saying they're 10x engineers. Yeah, you should. Because it flies in the face of so much like equality philosophy that everyone's equal. But the reality is when you're operating in idea domains, when you're operating intellectual domains and virtual digital domains, it's not even 10x, it's 100x or 1000x. And it always has been. Satoshi, Notch, you know, the guy who invented JavaScript, the Brendan Ikes of the world, John Carmack. I mean, these are 1000x programmers, not to even mention if you choose the right thing to work on versus the wrong thing to work on, that's an infinity difference. And it could just be not necessarily even a better programmer, just one who had a better judgment on what to work on in the first place. And now obviously it's less controversial because of AI leverage.

What's controversial is that the token leaderboards, right? Like people are still getting a little confused because now they think, well, I have a bunch of 100x engineers, look at all these tokens that I'm paying for. I'm curious if you guys have seen the same, like, How do you measure ROI?

It's like the old measuring lines of code. It's, you know, token consumption, lines of code feel like similarly not direct paradigms.

I mean, my observation has been that Claude or ChatGPT, um, or GPT is about, is basically as good as you are in a domain. And so, uh, if you're, if you're a really capable developer, then these things are really powerful. And if you're a junior developer, then you'll kind of find it to be like more of a junior developer. Like on the one hand, these models are incredibly capable. On the other hand, the feedback that you give them sporadically seems to be incredibly important. And these little updates seem to totally determine the types of, uh, performance you get out of them.

There's a new kind of support that I give, which is you come to me and like you didn't get good output out of the model, and I tell you what to prompt the model with. So like the idea of like the quality of the reprompting, which I think you're alluding to, is, is extremely important.

But I mean, and to be clear, I think that this will become less important over time. Like, as the models get much, much smarter, then you'll be able to put in less and get more out. Um, but at least at this stage, it really seems to kind of reflect back the judgment that the user brings in, in my experience.

I've kind of resisted learning all the tricks and tips. Like, you know, there was a, oh, use Ralph Wiggum, use OpenClaw, use Hermes, use this prompt engine, use this scaffolding. Plug in this piece, you know, always use plan mode. I just ignored all of that. I just assumed the model is just going to get better faster than I would figure out how to use it. It would figure out how to use me faster than I would figure out how to use it. And so I've just been completely ham-fisted with them and I get frustrated at them and just sort of, I found myself typing less and less information and doing less and less work as time goes on with the models because I just assume I can brute force my way through it., and I'll throw Codex, Claude, and Gemini at the same problem over and over and just waste tokens to save time. And I think no matter how expensive these models might seem, they're still way cheaper than a human. So I would say just waste tokens, save time. Don't look at the tokens either as inputs or outputs. Just look at your time and look at the final output. And even if they're writing low-quality code, which I know in many cases they are, it's not necessarily production quality or scalable code. When the time comes and I want to ship it to production, I'll just throw more tokens at it. I'll say, okay, now go through, look at it, rewrite it. And they're just going to get better every generation. So yeah, I don't see where this necessarily stops. As long as we have verifiable domains and solve problems, they're going to resolve those problems. Now it's in the unsolved problems domain where maybe you're Terence Tao, you're the cutting edge of creativity that you need to be, you know, working very collaboratively and carefully and closely with the model., but I'm not in that, I'm not at that level in software engineering. And then Guillermo, you're probably the most extreme software engineer in the team, right? Like out of this set, you're probably the one who most hardcore came up from a software background. Like how are you finding these models at the edge of their capability?

Well, there's one thing that's happened recently that what you're saying resonates strongly with, which is it used to be that you would give a prompt to the model and it kind of does it like classic, like next token prediction thing and it like runs away with your idea. And models now have been doing this like intuitive planning mode without, to your point, not even having to plan where it comes back to you and says, look, what you're asking me for, there's these 3 routes we can take. There's this set of trade-offs that we're gonna go down. That's a moment where like, you know, people do the whole thing on X. It's like, oh, now we have a, PhD-level engineer model, like that's very clear that the models at some point graduated. They used to be junior engineers. Now they're principal engineers because they come back to you with a set of trade-offs. And obviously sometimes they bullshit, which is hilarious. It tells you this one is going to take 3 weeks and this many tokens. It tries to make really bad predictions, but clearly it's now this, like I respect the models a lot more as a peer, like that I'm going back and forth intellectually with, but there are a lot of gaps still. So like if you're a really, really proficient engineer or architect, I think you're still extracting more juice. So the question sort of that Max was positing of like, if you're junior, do you get junior back? Well, clearly not because a junior gets more advanced knowledge in code that they would've never been able to write by themselves. But doesn't an experienced architect get 10x, whereas a junior engineer gets 2x? That's what I'm kind of trying to figure out still.

Yeah, but I mean, I think there's, there's architectural decisions. So when you think about the development, I'm seeing this now with some of our, the junior software engineers on the team of like, what is the next step in their career progression? It's going from like writing implementation for a feature to picking technologies, like choosing between Postgres versus some other database or picking between ZMQ versus some other message queue or like some other queuing system. And those, I mean, the models can suggest them, but that's the thing where like you'll see it and you'll be like, no, no, no, I wanna use this other thing. That's the type of little feedback that I'm saying really matters and the types of output that you seem to get at this point.

Taste and judgment, right? Taste and judgment. That said, you can ask them, which one should I use and why? And they know everything. They'll give you really good trade-offs.

That's the change that I was saying has happened recently where you would say, hey, go and put this super high cardinality telemetry data into Postgres. And it's like, no, no, no, bro. Like, we don't put that kind of data into Postgres. Like, you should consider ClickHouse or Athena or whatever. Like that's happened to me a lot, which is really impressive. But I, the thing I'm still like kind of struggling with is clearly the human is still completing the model. Like at one point, is it the other way about? Like the human is the one sort of getting the instructions back on like, go get me this API key because it's something that only you can do. Or get me this amount of capital for my next set of investments that I need to make. Yeah, you just watch. Like clearly we're still not there yet.

That's a temporary aberration. Pretty soon every good SaaS company or hosting provider will have a CLI and API interface that the models can communicate directly. They don't even necessarily need an API. Like as long as it's like text-based, Unix-based, the agent can hack its own API. And then the money part, you insert crypto tokens, you know, put in Bitcoin, put in whatever, and the model goes and just pays for whatever it needs. And I think like, yeah, You know, there are people working on this, but the thing I am now thinking through is, is pure software dead? Like, is pure software engineering like an obsolete thing? It's like saying speaking English, right? The models now speak English. We had to learn code to communicate with the models. Now the models speak English and they speak fuzzy, sloppy English like a human and they understand things. So where's the moat, like for a founder?

Hardware?

It's a boon, you know, like now you had to build hardware, it was hard to build a software company alongside, like Patrick Collison says, software is art and it's hard to hire artists. So now as a hardware founder, great. You can have really good software developed fairly quickly. If you're creating models, maybe that's the new software engineering, training models and tweaking models and post-training and fine-tuning models. But classic software engineering, is that dead? Is pure software investable? Is pure software something and organize a company, a team around and try to get some leverage.

Did you guys see the, uh, there was an article on X by Mitchell Hashimoto called The Block Economy or the Building Block Economy, something like that. Like his argument is that the most useful thing for agents to have now is really powerful reusable building blocks. Because to Max's example, you wouldn't expect your clanker to reinvent a queue infrastructure system. Every time it needs to send an email. It needs to bring in the right building block that's right size for the task that you're asking for and say, well, okay, for this one it's BoltMQ. I challenge the notion that I would want the agent to reinvent the entire universe from first principles in a way that's incompatible with the rest of society and civilization. Like it's almost like reinventing highways, laws, policies, et cetera, just for you. Even if there is a potential for extra optimization and extra juice that you can get out of it, there's still a sort of like cooperation at large scale value of saying we're both depending on Postgres 13.2. And so that's still really, really, really valuable. I would say like the category of infrastructure software and building blocks that these agents are gonna use, obviously in bias is this world building, seems extremely valuable. And I don't see the agent anytime soon. And by the way, you could even, another metaphor I've been using is like, anything that's already been created that the models can reuse as like a token cache. Because you don't want to churn through a trillion tokens to reproduce what's already existing. And so there's always starting points that the model can fork off from, but it's going to change things quite profoundly.

So these are like libraries and dependencies, but for models?

Yes, for agents specifically.

To Naval's question though, I mean, I learned to program when I was really little. And I, like, that was the thing that through all of, like, being a teenager and in my 20s, like, I'd get, like, sucked into it and just, like, code for, like, 20 hours and it was super fun and I knew all this stuff about programming languages. I haven't written a single line of code in quite a while now. And I mean, partly that's because my job is different, but also since December I've built a huge amount of software that I now use every day. There's all these projects that I've kind of fantasized about for years that now I'm, like, using, um, that I've actually built. And I didn't write any of that. And I just can't imagine going back to like actually writing code by hand anytime. Like, I mean, I'm unlikely to do that anyway, but just like in general, I see that I have a hard time seeing that as part of the future.

Yeah. There's something really cool is that you understand how the pieces click together. Like, I feel like anyone that understands what an API is and how data flows, inputs and outputs, performance, because you kind of you have to orient the model around like, this is a certain level of expectation that I have out of this operation. Like, that's always been infinitely more useful than writing code. Like, I feel like a really good, proficient engineering leader has been quote unquote like vibe coding through people on Slack or one-on-ones, because you're transmitting your will, your intent, your experience, and you're letting others run with it. It's just that now we do the same, but with agents. And so I think that's why you've been successful with it, but I don't know that everyone sees the same level of success.

I mean, I went from not having written code in 20 years to I'm coding all the time now, but through agents and I'm building tons of software. And it turns out that just understanding the basic principles of software engineering and algorithms actually gets you a long ways because the reason I stopped coding was because I didn't have time to figure out the latest language, latest architecture, infrastructure pieces to plug into. And I know Vercel makes it a lot easier, but even then just getting started was a bear, like just plugging pieces together, assembling infrastructure was just so annoying.

The thing that really changed is, I mean, it used to be that you could build a lot, like you, like there's a lot that was straightforward, but then you would hit some random thing and then you could spend kind of some indefinite period of time debugging, some narrow thing. And now with the agents, what happens is you just don't get stuck anymore, which is pretty amazing.

Or they get stuck.

It's removed. Well, no, I mean, like relatively quickly they can find like the right way to do things. And it used to be that like, I remember when other friends would learn to program and be like, nope, it's just like intrinsically frustrating. Like if like that's part of the deal, that's how you learn. And that just isn't true anymore.

Blake, how are you applying all this stuff at, uh, Boom Supersonic?

Yeah, what I found is it completely changes the role of software and hardware developers. The thing that we did from day Day 1 was try to take a lot of traditional engineering workflows, and I mean hardware engineering workflows, and turn them into software. And so if you haven't been around hardware engineering, let me see if I can make this more clear. There's a lot of engineering, hardware engineering that happens in Excel spreadsheets on engineers' laptops in a silo. And it's very complex spreadsheets, sometimes like VBScript code. And all of this is actually software. But it's treated as if it's not software. There's no, there's no source control, there's no automated testing. If you want to hand something off from like an aerodynamicist to a structures engineer, that's done manually with like a spreadsheet over email, like it's the 1990s. It's terrible. And so we started building these kind of like software frameworks that can automate and make repeatable hardware engineering flows with the idea we could reduce the cost of iteration. But it was, it was slow going. 'cause we can never get enough, we can never like afford enough software engineers. And what we've gotten into is this mind-blowingly different model where the software engineers actually create the architectures 'cause they understand systems, they understand algorithms, they understand, you know, division of concerns. And then the hardware engineers can vibe code their pieces because what they know about hardware engineering. And the result is just like mind-blowingly different productivity for small teams. I give an example, like if you're designing a turbine blade, blade, like classically, so a turbine blade starts like cold, but when it runs, it's hot, so it gets bigger. And so you have to design both the aerodynamics and the structural design of the thing to work with its cold shape and this hot shape. And so you have to convert between cold and hot, and you convert between structures and aerodynamics. And this takes like one engineer, one day, for one blade, for one piece of the analysis. And there are like 1,000 blades in a jet engine. And so you can't do much. And we literally now, with a combination of software and hardware people creating the solution, you can change blade geometry. You can see in real time the structures and aerodynamics results. And so it allows two engineers to design an entire jet engine, which is just wildly different.

One of the things you mentioned is that you have software engineers creating the tools and architectures for the rest of the engineers. That to me is the biggest the cataclysm of enterprise software is that there is no startup that builds hardware collaboration tools that can sell you anything anymore. Because internally you're just coding the right things that you need at any given time. Even spreadsheets are kind of cooked, right? Because the reason spreadsheets were successful is that no one could build custom software. So the thing that approximates custom software the most is a spreadsheet with a bunch of VBScript functions.

I personally have moved almost entirely from Excel to Python models, where I can actually like get like believable simulations of things.

Yeah.

I mean, the thing that AI hasn't come to yet that I think it will within the next year, like probably within 26, that will be very, very exciting, is right now it can generate software, but soon it'll be able to generate STEP files and PCB layouts. And when it comes for mechanical and electrical engineering, that will be a whole other thing that we haven't seen yet. That'll be very, very cool. Yeah.

On the hardware side, I think it's really a boon for like all these little gadget companies and part companies that write really bad software because they can't make great software. And now they're going to be able to make good enough software, or it may not even software that has a human front end. It might just be completely agentic for an agent to access and you just talk to it through voice and control hardware. And this is why one of the reasons why I think, for example, China is big into open source models, right? They're basically going all in on it because they have hardware superiority. They have these very complex supply chains and component chains. And they're basically saying, hey, if I can just generate software on demand, then I don't have this disadvantage anymore against Silicon Valley. So that's not the only reason why they're doing open source. I think they're also behind their distilling models. They're catching, you know, they're collaborating on resources. But I think the Chinese government has a history of funding efforts that then sort of help their entire ecosystem along, especially in network effect businesses. And so I think they want to like pool all their resources, catch up on AI and use it to give their hardware stuff an advantage. And ironically, they're doing all the open source stuff because OpenAI is not open. You know, Groq publishes models, but I think they're a model or two behind. Google has some local models, but nothing really that competitive. Anthropic, to my knowledge, I don't even I don't know of any open source models from them. So all the open source heft is coming from China. It helps all our hardware founders, but it helps their hardware founders and factories and so on that much more. But all the crappy little software that goes with all the little random knickknacks and thingamajigs that you buy off Amazon to tinker with a lazy Saturday afternoon, that software is getting a lot better very quickly.

I think everyone's had the wake-up call that without great frontier coding models, you don't have self-improvement. And so imagine China as a whole not having the ability to produce frontier everything, right? It's not just producing software, it's in any piece of this hardware pipeline, like Blake was saying, like you need to generate software. If you fall behind in your ability to generate software, you fall behind in the ability to generate everything. One thing I'm curious about from you guys is like, because everyone loves to talk about Chinese models, like, Do you use Chinese models? Do you know anybody that uses Chinese models?

This is an argument I had yesterday, actually, which is one person at the table dinner was claiming that, you know, you just use Deepseek for 97% of things because it's so cheap. And if you need more intelligence, you'll just run it over and over again, the same problem. And you'll only use the OpenAI, Anthropic, et cetera models for the most advanced tasks. And I was kind of like, I don't know. I think intelligence is an unalloyed good. You always want more intelligence. And when these models make a mistake, you don't know it. And it's always cheaper than a real person and real time. So you'll just use the most intelligent model available, which isn't great news necessarily because it means that, you know, you're going to end up creating a monopoly or oligopoly kind of situation in AI. But I always want the most intelligent programmer. I always want the most correct answer. I always want the best judgment. And given the amount of leverage that I'm going to pour into it through capital and code and people and, you know, marketing, I want to make the right decision every time. And often when between two models, let's say like I have one model that I know is a little smarter than the next one and they both give me answers, often I actually don't know which is the correct answer, right? So if I know one model's a little smarter, I'm going to go with that answer and eventually I'm going to stop asking the model that I think is less intelligent. But I don't know, have you guys found a use for these so-called less intelligent models?

We see uses so that, so we have the AI gateways, data that basically like every application agent center goes through. And so there's definitely usage of open models, but the top is like heavily dominated by the frontier intelligence. And there's a subcategory or there's like a caveat to that, which is that frontier intelligence at reasonable cost and performance like slaps at scale. So like people don't get really excited about Gemini, but they put out these models that are like super smart, at the right performance-cost combination. And for a lot of tasks, other than coding, actually, interestingly enough, they're the best models. They're like the best, like, industrial production models. You can throw them at, like, support tasks or browser automation. Like, I would always put a Gemini model there. And I would look to Chinese models for those kinds of things. But anytime I'm working to push the frontier, You need the best possible coding model. And that's basically now like 2 or 3 models. And, uh, and the Chinese are not, certainly not in it.

Hey Max, you're pushing pretty hard into vertical integration and extreme urgency. Do you want to talk about that?

Yeah, I mean, for many things we, um, I mean, you can't buy it, so you got to make it somehow. Our preference would always be to buy something. Um, like if there's a vendor that offers a service at a great price, that's like, for example, like PCBs, like we don't make PCBs, like those are, they're basically free. You can buy them in unlimited quantity from Asia. But the, the closer that our products get to being like a single block of covalently bonded matter, the better they'll be. Lower power, smaller, higher performance, last longer. And, um, there's just like, they're like, the components aren't available. And in order to do that type of integration, be able to actually innovate beyond things, just piecing together things that you can buy off the shelf, which really is, is very, very limiting. I guess you have to like learn to do it yourself and that shows up as vertical integration. So we own a captive MEMS foundry on the East Coast, which we bought because there was really no other way to do the type of packaging and assembly stuff that we wanted to do. And I think that all of this is going to be affected heavily by AI over the next few years. It's not quite there yet. In fact, ironically, one of the biggest impacts that we've seen of AI inside the company is in regulatory interactions. Because if we can do things like generate documentation, or if we can ask, like, we want to change, we want to evolve this product, like, there's thousands of ISO standards that might apply. Which ones do we have to comply with? And like trace this through. This used to be like you're, like you're following a whole regulatory and quality team for several months as they trace this. And now the AI just kind of knows. Um, but when I think about stuff like the, the Surgical Program or the MEMS fab, I think ultimately the software still needs hands. Like it's going to be smarter than us, but if it can't make things, then like those are real, real boundaries. And so we've instrumented our foundry as well as many other parts of the company in ways where as these models get better, that should show up pretty immediately in things like the cell engineering that we're doing and the material science that we're developing.

It sort of makes me realize that like it's been a while since I've generated a basic legal document using a lawyer, right? I stopped asking lawyers for NDAs and, you know, agreement for this and sign that and research this. And like all the basic legal tasks are gone too, because, you know, there's the old joke that law is like spaghetti code. You know, they have this very complicated code that they try to put in English and it contradicts this code over here and has to fit into that code over here. And there are no real APIs for it. But for just like junior engineers and junior engineering, I should say, junior engineers basically got a promotion to senior engineers. And junior engineering got taken over by agents. And so the same way, I think in a way, the downside is you can look at law and say, you know, paralegals just got fired, or you could say paralegals just got promoted to senior lawyers and now they can spend their time thinking about the law.

It's actually kind of interesting to think about the parallels of how software engineering is evolving with lawyers, because lawyers, you never know what they put into these documents exactly. You just trust them. Like, hey lawyer, can you look at this document? Can you tell me if it's legit? Can you do red lines, whatever? Like at the end of the day, you're, what you're valuing in the relationship with a lawyer is that, is that they're a trusted authority. They went to law school and they're putting their reputation on the line. I think there's a parallel with like the biggest problem in software engineering today is these mountains of slop that end up as a PR. And then people are saying like, there's all these memes on Twitter, like, Way back in the day, we used to read every line of code of a PR. Well, in my world, infrastructure, I want engineers to be able to say, I understand. Doesn't necessarily mean that you've read every line of the PR. You need to be able to say, I am signing off on understanding the consequences of this PR, or I wrote the test harness, the simulations, the proofs, the type checkers, et cetera. To be able to say, even without reading this, I have confidence I can sign off on it's gonna be safe in production. And so it's, it's kind of interesting cuz there's a world in which we embrace that everything's gonna be spaghetti code and that we don't fully understand it, but we write the basically evaluators that give us confidence. And then we rely on like people, like the infrastructure production engineers to say, okay, I'm fine. Sending this into prod, at the end of the day, someone's going to get paged if your systems go down. And I think another thing that people are underestimating is that creating software is really easy, zero to one, but think about 1,000 days from now. What does your software look like? Is it secure? Is it tested? Is it production grade? Is it performant? And are you still motivated to invest all of those tokens in maintaining it in prod?

I mean, humans are becoming verifiers, right? And that's kind of how we train these models with good verification data, and now we need human verifiers. So yeah, I think a lot of the old function of people, lawyers, engineers, operations people moved to verifying the stack and saying, yeah, this is roughly correct, and I'll roughly stand behind it, and I'll support you if it goes wrong.

One of the things we've seen related to the regulatory is it massively reduces change aversion. And improves iteration. So to give you an example, like let's, let's say you're going to go certify an airplane. One of the zillions of things you have to do is prove that it can withstand a lightning strike and that the regulatory documentation for the test plan for such a thing stretches on for, say, 200 pages. And what you classically do is hire a, let's be honest, not super bright engineer who's willing to be there monkey at keyboard writing 200 pages of regulatory compliance documentation. And it takes a couple months. And by the way, if you change the airplane, now you want to cry because there's another like 2 months of rework of this like rote kind of regulatory compliance documentation. And what we found is, you know, we can build a RAG that will enable us to basically prompt our way through all of that work, you know, in let's call it minutes. The first order of fact is, oh, you save a lot of time. The second order of fact is if you change the specification of the airplane,, it now takes, you know, minutes, not months. So you can actually be willing to change. And the third-order effect is you can now, you know, basically get rid of the not-very-great engineers and have a small number of really creative ones. They can iterate rapidly because the cost of change goes down. And in a certain sense, like, the entire regulatory burden, which really hurts the ability to iterate, drops away.

I think that this is a really undersold story in AI right now. I think the consensus in Silicon Valley is that, like, regulation sucks. Like, any— like, we want to go faster. We want to realize this amazing future. We want abundance. We want, just like prosperity and stuff that slows down that future is just kind of to be avoided. And certainly I think we've overregulated. We've made it impossible to build stuff. It's just like, it's totally crazy what goes into getting, building any type of thing in a lot of places, either physical or otherwise. But, you know, like a lot of the regulations themselves are not the problem. Like if you've actually read a lot of these things, like having non-smog choked cities is great. Being able to swim in like many rivers is great. Like having like a lot of these things were progress. The problem is that it is really difficult for humans to deal with understanding and complying with this. And that every time you have to exchange a letter with the government, you wait months. And if you could take a lot of the things that we've learned and kind of make them like totally frictionless, that would actually be pretty cool. And I think that, um, that I think is an under, an undersold story in, in AI right now.

Yeah, until the regulators start spewing tokens back at us. And then you start getting huge amounts of documents from the regulators that you have to comply. And it's agent-on-agent wars.

But— But that's basically what we have now.

Yeah. But, but there is a fair fight.

Yeah.

I'd argue that's an improvement from where we are now. Like, one of the terrible things right now is if you're going to build anything physical, you have to get a building permit. It's like you're guilty until proven innocent. And the worst thing that we've run into is the fire department because they have like the moral imprimatur. Of, you know, people pulling people out of burning buildings. And yet what they actually do is just like screw with your design for buildings for months. And, you know, if we could replace the fire marshal with, with an agent that would critique your, your building plan quickly, even, even if its feedback was overdone, it would be massively better than the delays that exist today.

When Max was talking about this potentially being a good thing to have all this regulation, my head went to The things that make agents successful is humans or other agents setting up the right testing guardrails. A lot of people are really excited about /goal. I don't know if you guys have played with that, or like Ralph Loops, where you tell the model, go do this, and this is your exit criteria. Well, I'm telling Blake, go make us all supersonic. Your exit criteria is that you've complied with all of these regulations. So there's totally a world where we say like, the regulations are great. They're like our testing suite, our test suite. As long as this passing this test, one does not incur in contradictions and the regulations are actually reasonable, et cetera, like they're actually an awesome guardrail to have. Otherwise we would be shipping slop directly into the air.

Yeah, but this is going to turn into a Red Queen's race, right? They're going to have agents, we're going to have agents. I think we might have better agents, which is good, as opposed to having to do human versus human. But if anything, their cycle time, their response time may get lower. Like the App Store is drowning in spam. I'm sure the Patent Office right now is drowning in spam. And so these agencies, they're going to be slow adopters of AI. They're going to get DDoSed, right, by clever entrepreneurs just overloading them with documents. It's possible that the approval time for this stuff might extend out as they suddenly get flooded.

It creates an opportunity to, I think, really shift the model. The regulatory model. Imagine if we drove around a city the way we build things today. Before you could go anywhere, you'd have to write a plan up, ship it to some regulator, you know, and your plan would have to specify we're going to take such and such a route and we're going to drive this speed limit, we're going to use our blinker and we're going to stop at every stop sign and we're never going to run a red light, blah, blah, blah, blah, blah. And then 3 months later you get back critique. It's like, well, we think you should like drive on this other street and eventually you get approval, you can go drive somewhere. It's insane. You can never go anywhere. And yet that is absolutely the way we build physical infrastructure in this country. It's guilty until proven innocent. And, and what we should actually do is make more of these things enforcement-based rather than pre-approval-based.

I mean, I don't know. I mean, I don't want to be under too much jeopardy. Like, if I ship a medical device to a lot of people, there needs to be— it's like there's unknowns there. It's like we were responsible. We did clinical trials. We reported all the data.

But yeah, but Max, this is why there's so little innovation in medical right now because the FDA approval process is a nightmare. In fact, the two biggest advancements in tech in Silicon Valley in the last decade, AI and before that crypto, they're both in the math domain because that's the last unregulated domain. And when they started regulating frontier models and started regulating GPUs, that stops as well. You know, Peter Thiel laments about how there's no innovation in the physical domain. Well, it's been held back by just the huge regulatory barriers. And you can always find a scare version like your vaccine or medical, like famous ones, right? But the regulations spread everywhere. The tentacles are everywhere and there's all these different contradictory regulatory bodies. You saw how, was it SpaceX? They got sued first for not having enough, I forget what it was, migrants or refugees or whatever, but they're not allowed to hire them by government regulation on the other side because they're not citizens. This is not like logical code that has to compile in one place. These are made up random regulations all over the place. You might comply with one state, you violate another state, you violate federal over here, you annoy this guy over here, that guy chooses to prosecute one of the 50 people who are his friend. It's, it's very arbitrary. It's very capricious.

And moreover, like, the idea that this makes, like, things safer, I think it's just a complete mythology. Like, just watch, you know, watch Boeing as an example. They certified the 737 MAX, which had a single sensor that had complete authority over the nose-up, nose-down attitude of that airplane. No intern is dumb enough to think that's a good idea. And yet it got all the way through the certification system. This stuff doesn't actually make us safer. It just makes us slower.

Well, I mean, there's definitely dysfunction here. I mean, I think that some of this makes us safer in the sense that the NRC makes us safer, which is that their job was to make sure that nuclear energy was safe. They did this by permitting zero plants until I think like a year ago, since the '70s. It will be perfectly safe if we never build any of it. And I want to be really clear that I'm on the side of deregulation on a lot of this. I agree with Blake that a lot of this can be done a lot more efficiently. But I also think it's a little too dismissive just to say it's like, oh, this is like the FDA or like even it's, it's in the rate, it's in the agencies in general. I think the problem is deeper to the degree that when the, if the FDA approves 10 really important drugs, they don't get any credit for that. One patient dies and they get hauled before Congress and yelled at. And so they have very negatively biased incentives here. And I think the reality is, is that this is reflective of the beliefs of the American people. There's this trade-off here between the perception of risk taken in human subjects research and the rate at which we get new medicines. And it is absolutely true that if we move faster on this, we would learn.

It's totally asymmetric. And I think you're totally right, Max. If you approve a bad thing, your career is over. If you block a good thing, nobody notices, right? So it creates this asymmetric slowdown. And I think this is, I think that is the most important problem to solve in the regulatory state.

But this is a very deep problem because it is, this is where the voters are. Like, and we go and poll some of the stuff that we're working on in the future to understand kind of like where the American people are on it. And if you push too hard on this, like there are all kinds of ways you could work around it. You could go to Prospera. There's all kinds of ways to try to go faster. But if you're seen as being a bad actor, then you are rejected from the society that we live in. That is the thing that you need an answer for, which is deeper than just saying like, oh well, we need regulatory reform.

You have a deep point there, Max, which is, uh, it's the voters, right? Yeah. This is where the citizens are. Like, we like to blame politicians. You'll see this on X all the time, right? When people are like, oh, this politician, that politician, evil politicians. They're elected. They're voted. Majority vote, right? This is where the people literally are. That's the package. That's the bundle they've chosen., and you may not like this instantiation, but if you were to remove this one, something very similar would take its place because the voters would just vote them right back in. And I think culturally it's very hard for most people to understand what we lost, what we missed, right? So for example, like France, you know, there's a French entrepreneur on X lamenting that 57% of GDP gets sucked up by the government. And so you can't create companies, but to the average French citizen, that's not visible. They don't notice what they're missing. They just know they're slightly poorer than the US. The Economist just did a little piece on— Economist is finally coming back around to being capitalist after 30 years. And they just did a little piece on how the US is outstripping everybody and growing faster and getting bigger. But then they immediately turn and say, well, it's because of the oceans, because of natural resources, everything but capitalism, right? They don't want to say the dirty C word because, you know, for some reason, all of these magazines became Marxist at some point. But they can't envision or imagine what could have been if we had just been a little more laissez-faire, a little more open. So I would love to see a true experiment among the 50 states, you know, different regulations, different tax structures. 'Cause right now the federal tax structure and federal regulations dominate everything. But imagine, you know, you could go to some small state if you had cancer and you could try every drug that everyone was cooking up and caveat emptor and you gotta do your research and blah, blah, blah. But this is known as the experimental zone. Same way for drones, same way for, well, aircraft is a little harder 'cause you gotta cross a lot of areas, but.

I do think there's something magical in there, the notion of like innovation zones, because we have a huge like NIMBY problem, right? But if you, if you create like, you know, opt-in YIMBY zones, they create that experimentation framework. And by definition, it happens where people are consenting and you can try different rules or no rules or different ways of enforcing or, you know, innocent until proven guilty and then see what actually happens and what are the innovation consequences and what are the safety consequences. And then the successes can spread.

But I mean, to Naval's, to Naval's point, an innovation zone would not solve the problem in drug discovery. Um, there, so there's the Right to Try Act passed a little while ago. Um, we've had this pathway called single patient IND for a lot longer than that. Um, the FDA, like if, if your doctor calls the FDA and says, hey, I want to give this, my patient an approved drug, they give over 99% of those over, like they approve over 99% of those. They can even grant them over the phone. Um, the problem is that in order to dose a patient, you still need clinical grade drug. And the only entity with that is typically the IP owner who's in the middle of running a clinical trial. Like they're investing hundreds of millions of dollars into like making this thing. And the problem is that the FDA, they'll draw an adverse inference if something bad happens to your patient who's probably really sick to begin with. And that's going to be seen as a property of the drug, which is global, not related to your innovation zone. And so there's kind of two problems. One is you need to get the IP owner to give you some of your drug, which they're not going to do. And then you need to prevent the global regulator from casting doubt on what might happen with their clinical trial if they give you some.

How would you address— I mean, I don't know your field. How would you address that in medicine?

Oh, well, I mean, that in particular, I mean, this is just like a very inside baseball. I think the FDA has to be prohibited from drawing adverse inferences across different users of a capsid, for example. There's these like a bunch of specific ways that you could really accelerate innovation with a relatively light regulatory touch. By just preventing this, this kind of paranoia from driving our decisions.

Is there anything better than the FDA out there? Like, what are we benchmarking these regulators against? Or is it not an interesting question because we don't have— everyone follows the FDA.

So I'll give two expansions to that. The first is Europe, which is not really better than the FDA, but they have a different system in that they've got these, these notified bodies, which are basically private businesses that are blessed by their host governments to certify things, whether this is trains or planes or medical devices. And the notified body system creates slightly better incentives at the review layer because they can hire people, they can grow, there's competition among the notified bodies. They themselves have to be compliant with the conditions placed by their host governments for certification. But that means that they can, there can be many thousands more reviewers than you might have in the U.S. The second thing I'll say is there actually is one approved, getting paid, implantable BCI today, which is in China. And the CFDA is thinking for itself. And they really do have a system that I think is going to give us a run for our money if we're not, if we're not careful. And they, they, they handle it very differently.

How do they handle it?

I mean, the cost to bring a drug to market or a device to market are just much lower. I mean, you can try things in humans and you can try things on market. Like the, So the problem that— one of the things that I've spent a lot of time recently thinking about is like 20 years ago, we were buying far fewer laptops and phones. Each one was much more expensive. Now there's— they're cheaper. There's far more of them. We buy more of them. The total spending has gone up. This is great. Stock prices of things like Qualcomm and Samsung and Apple are way up. Everybody's happy. They're using kind of the excess wealth generated by the phones and laptops to buy the phones and laptops. Um, this doesn't happen in healthcare. In healthcare, because you've got this reimbursement mechanism in the way where there's this kind of enterprise sale happening, the bucket of money that we use to buy healthcare is basically fixed. It is not increasing as there is more stuff that is producing better healthcare outcomes, like we see in technological growth industries. And so this means that the rate of spending on healthcare grows at roughly the rate of, of growth of tax receipts. And so if, let's say that, like, AI is booming and there's major advances that are happening and 2 years from now we're spending 10 times as much on AI as we are now, this could be great, but if in 2 years we're spending 10 times as much on healthcare, this would be a catastrophe. And this is fundamentally at odds with being a technological growth industry. And so as time goes on and there's more things to spend money on that extend and improve quality of life for patients, like we can restore vision to people who go blind in their 80s, we might be able to extend life in like far past where it's been before. We can restore capability to patients that are older and in worse condition, but like How do you pay for that? There's kind of this like omni problem in healthcare, which is all really related to the same problem, which is it's just too expensive to bring these things to market. And that's what China is getting at. The way out of this is not single payer or some revision to health insurance. It's to bring down the costs so that someone can buy this with a credit card, finance, maybe like a car worst case. And to do that, we have to make it cheaper to bring these things to market. And China's doing that. That will allow them to sell these things for $10,000 on $100,000.

There's no private market in healthcare. And because there's no private market, what was the analogy people make sometimes? Like, imagine instead of going to restaurants and paying, you would basically go to all the restaurants and then at the end of the month, you would send all the receipts and all the bills to your insurer or to the government and they would reimburse you. Well, there'd be a line outside every good restaurant. Every bad restaurant, you know, would be available. The rates would be terrible. The product wouldn't improve. You're basically running a small communist society inside a larger capitalist society. That's what we're doing in healthcare.

It's also what we're doing on roads, which is why we have traffic. Like, it's the exact same situation on roads. That's why there's, you know, there's no variable pricing for getting on the highway. That's why it's always clogged.

If you want to step on the third rail of healthcare for a moment, think about this healthcare plan. Tell me what's wrong with it, right? Imagine that the first 20% of your annual income was your healthcare deductible. Doesn't matter. Like if you're broke and homeless, it's zero. If you're rich, you know, it's millions of dollars. But whatever your annual income is, the first 20% is your healthcare deductible. And then the rest is paid by the government and the insurance system up to the usual caps that they have today. You would create a private market pretty quickly. And so like in dental and plastic surgery, and sort of a lot of optional medical procedures, you would actually get a competitive situation. You get improvement. Like if you look at optometry, you know, with LASIK, you look at dental with like veneers and braces and all that stuff and kind of all the dental surgery stuff that they do. Or if you look at plastic surgery, like those fields do seem to be advancing because they're private payers. They have people who are, you know, voting with their money. So we need to do some equivalent of that in the normal healthcare system. But people lose their minds. They don't want to think one step ahead. They're like, no, no, no. Well, What about the broke person? Well, the broke person has no income. So, and they're like, well, 20% is too much for some people. Okay, you could put some deductible in there. But generally, if you don't have some private market where people are paying a lot of the times for what are medical procedures, you're just not going to get this feedback loop that you're talking about. You're not going to get this ability to spend more money into the system. Right now, like very wealthy people can spend voluntarily in the system, but the prices aren't anywhere. The rate cards aren't anywhere. The system's not designed for it. It's like if you go shopping for medical care and you want to pay out of your pocket, sometimes they'll quote you a price that's 10x what they charge the insurance company.

Have you heard Sid's story from GitLab? Do you know Sid? So he was, uh, I mean, he had a massively successful IPO, then was diagnosed with a rare cancer and has achieved, has lived way past the prognosis, has really taken it into his own hands. I think he went from kind of He did frontline chemo and then there was one alternative that was available. He exhausted it and the doctors were like, we've got nothing for you. Since I think like 6 or 7 companies have come out of it, there's now 20 or 30 drugs in his escalation ladder. He's still alive. Um, over years later, he's doing great.

I saw him the other day and he basically created his own like personalized medicines and, uh, treatment plan.

Yeah, there's, there's a handful of these anecdotes that I've heard now. I, it is really clear to me that at the high end, if you just kind of have like, you're not dealing with insurance, you have the resources, you're like, I want the full toolbox of modern science. Outcomes are possible that, that like your normal, like if you go and ask your doctor, like, oh, what will happen if I do this? They will just start shouting and throwing things. But it is clear that, that much, that like that crazy things are possible at the high end. I think that this type of like N of 1 medicine is actually going to end up being a really rich source of research for understanding how to build more translatable things.

It requires a ton of agency from the patient in a moment where they're at their weakest, which is pretty ironic. My friend passed away from cancer and like last thing he wanted to do was research n=1 medicine, because he was just, you know, like, like dying by the week. But this is where AI should really shine and come up with the right solutions and democratization of like what can you actually do when you find yourself in that situation? It's kind of crazy how few people get access to this just from a knowledge perspective, not just monetarily speaking.

How much autonomous software do you guys have in your organizations that's running on its own or near autonomous and improving on its own?

For us, a lot of the infrastructure is already autonomous because we have this capability that fires off upon finding anomalies, which I recommend everyone creates a version of this, or Vercel offers a version of this, but upon anything happening that's anomalous. Today, the most engineering organizations are responding to this by setting up alarms or monitoring thresholds by hand, which is pretty insane, but that's actually how the entire industry works. You say, if my error rate increases by this amount at this API endpoint, do this. So we've actually automated a lot of the SRE job, Site Reliability Engineering. So anything, any metric that slows down, speeds up, throughput changes, whatever, fires off an anomaly alert, an agent investigates that. An agent can decide to create an incident. If the incident is filed, people get looped in and the agent begins the process of remediation. We're doing everything except for the actual giving the tools for the agent to change prod, but we're basically serving solutions on a silver platter to engineers. And then the other thing that's working really well for us is just autonomous optimization processes and autonomous security research. So the other, we open source this tool called DeepSec. It's fucking incredible. It's like Mythos, but you get it today., we run it against our entire monorepo using 10,000 concurrent agents in the cloud. And it found basically several quarters worth of security research, uh, progress was made in, um, basically a couple days and $14,000 worth of tokens. So I'm talking about like months worth of red teaming, security research, entire teams of people. Uh, and so we're basically now running like this periodic, because the, the, the other problem with AI is that Cybersecurity is becoming a nightmare. There's way too many vulnerabilities, way too much work to do. There is too powerful adversaries. So you have to like basically be investing very proactively. We're running a lot of autonomous security research. So SRE, security, and then optimization work are very obvious. You've probably seen on Twitter, there's people translating codebases from language A to language B. Like a lot of the work that if you, if you already put in the work, to get a working program, optimizing it or rewriting it in a native programming language or things like that is now becoming quite doable with frontier models.

I mean, just from my own Vibe-coded app, I built a bug reporting queue for my TestFlight users and they can report bugs from inside the app. It uploads the logs and a screenshot. And of course they use it for feature requests too. So then I just have a simple daemon go through, compile all the bug reports. It actually proactively analyzes and fixes them in the background. And then it ships me a TestFlight version to try out before I ship it to the testers. And then for feature requests, I just have it right now compile them, but I could see an app in the future could literally be built by the users. Now I'm not saying that's a good idea. It might be a mess, but at least it can take the bug reports and stuff.

We should ship that, by the way, just to see what happens to the social experiment. Yeah.

Yeah.

The social experiment. You end up with like that Homer Simpson car where it's got an umbrella and like a flashlight, you know? A clown horn and so on, where it's got every feature. But definitely for bug fixing, you could do that.

We did it in a way, a version of that experiment where I stopped all project work across the entire company for a week and said, everybody from the receptionist to the engineers, uh, build whatever you think is the most important thing to build. Uh, you'll learn requirements where you have to use AI and you have to demo it for the whole company when you're done. I expected we would get a large number of silly projects and a small number of needle movers.. And what we got was a large number of needle movers and a very small number of silly projects. And wow. Yes.

Yeah.

That's a great experiment.

Yeah.

2 or 3 are like trajectory changing. Like they would absolutely change the direction of the company. But the one that surprised me the most was literally the receptionist, like the shipping and receiving associate whose job it was to like take packages off a truck and like email people when their like stuff came into inventory, uh, build an automation for that. And, and that, that we're actually using. The conclusion I kind of had is like, wow, like everybody has some idea of what could exist that would make the world better, but that many times their first-order ideas are stupid and they don't have the, they don't have the ability to project that out and kind of see that it's stupid. But if they have the ability to go from idea to an actual thing, if it's not working, they can react, they can iterate. And if you give them a week, by the time they're at the end of the week, they've actually built something that makes sense.

But imagine if all work was like that. Like, how can you set up a workforce that does not do the work directly. All they do is train the agent that does the work for them. And we've done this as well. Like, you have to remind folks and you have to like create hackathons and hey, let's build agents. And obviously there's a lot of people, there's a culture change happening. Like, there are a lot of people that are just coming in who intuitively know their job is to not work on the thing, is to actually train the agent that works on the thing. But I'm curious about like, you know, what does the autonomous company of the future look like?

It could get a lot crazier. Maybe you just turn on all cameras and the agent's just watching everything that's happening and sees that the shipping and receiving thing is very inefficient and it creates the app. Well, Zack did that. Did you see that?

Zack installed this thing into everyone's machines. He's thinking about it. It's like we saw this too. Like we're likely going to ship a feature into AI Gateway that allows people to opt in into preserving inputs and outputs. And then you can say, for all of my inputs and all of my outputs, can you extract the skills of the things that I, like, learn from my work and then dump it as skills so that I can even download them for myself? But you could imagine people in companies wanting to share and pull this together.

It's funny because for me, that's so unimaginable for my own work because my own work is not repetitive. I look for things to automate. There's almost nothing left for me to automate for my own work. And I hope that's where kind of everybody ends up. Right? You just work in your maximum zone of creativity and interest at all times. And like, if there is anything left to automate, you should automate it, get it out of your life. It'll free you up to be creative. And that's where you generate all the value. But I think that's very hard to see in the job career mindset, because you hire people to do the same thing over and over, and that's going away. And that's really scary because people are like, well, what am I going to do? Well, you're going to do creative things. You want to come up with new things. You don't have to come up with a new thing every day. That's impossible, right? But you're going to come up with a new thing once in a while. That will then create something else, some point of leverage for you. But it is, it is a scary time for people for sure. If you've been doing the same thing over and over for 10 years and all of a sudden it's like, well, now you're going to train an agent and automate it away. That's scary.

I think historically it was the returns were like 70% intelligence, 30% agency. And now it's going to be 70% agency, 30% intelligence. And that will, that will shift further as the models get better and better.

I'm actually not sure about that, Max. I'll take the counterpoint on that. I think it's 99% intelligence and 1% agency because then the agents will exercise the agency, right? You will literally be like, hey agent, I'm making smart decisions and thinking big thoughts. Just go implement stuff. In fact, sometimes I want to build features on apps that I'm floating out of live coding. I'll ask the agent, what feature should I build next? You know, go look at the logs, go look at the users. What should I do?

To be clear, I'm talking about the returns to humans. Um, the, the humans that will be best fit for the future will be the ones that are more agentic, which is to say like the ones that can come in and just have the thought of like, I'm going to open Claude and be like, what should I build versus watch YouTube?

And here's a fun experiment. I'll bet you we all know a lot of people now who are coding who weren't coding before, including many cases ourselves, right? So the number, the percentage of coders in the ecosystem has probably gone up by, might be 10x, right? Yeah, it might literally be 10 times as many people are coding now than were coding a year ago.

It's wild. Our signup numbers are through the roof and there's this new class of people who are not engineers. They just use the infrastructure.

But I think it might be like podcasters and YouTubers and like people posting on X. The majority of people are still not creating code. Like I go to people and I'm like, oh man, vibe coding is so much fun. It's more fun than, like I had a little gaming group that I used to play video games and FPSs to blow off steam. I completely stopped playing. All that time went into vibe coding instead. It's more entertaining and you get something real out of it., but the feedback loop is just as tight or even better. And I went to my other friends and I was like, hey, you should be vibe coding instead. And they just gave me this blank look. And I'm like, no, no, you don't understand. Building things is so much easier. But I think to them, it was always like some black box process in the background. They never understood it. They assumed maybe you were just talking to the computer all along. So they don't see what's changed. They don't realize it's a lot easier. To them, just that starting, to Max's point, the starting is so impossible to imagine and hard, they don't do it. So we might have taken, you know, 0.01% of the population writing code to maybe now it's 1%, call it a 100x increase, but 99% still never gonna write code. So we are in this weird space.

It's crazy. It's like, it's a video game and it's a great video game, but real stuff comes out.

Yeah.

Like my fiancée was up all night last night because she couldn't go to sleep because she was hacking on something. And of course she wasn't writing any of the code, but it's just like, it's addictive in a way that programming hasn't been for me for like over a decade.

It's amazing because it's like a lottery for people.

I think the normies, normies have gotten a little more into the vibe coding, but through models that are more media models, video models, for example, right? More people have probably fooled around making videos and images than they did writing code and apps. The problem is like, I don't, video has its own issues, right? Maybe someday we'll like make me a great movie about X and I'll just spit out a good documentary. But right now they don't have the taste or the judgment.

This is a bet that I have with, uh, Andrej Karpathy was like, what's the year that you'll be able to just dump in a book and get a movie out? I think it's a lot closer. Although I think he has come down substantially in timeline since we made this bet a few years ago. By 2030, we're going to have like dozens of Lord of the Rings. Like there's going to be some fan who's like, he did it wrong. I'm going to make my own take. Um, like the famous stories or like the, one of my other benchmarks for progress in AI is I'm a huge fan of, of a series called The Expanse. There's a series, there's a TV series and there's, there's 9 books and they've made the first 6 books, but they haven't made the last 3 books and there's meaningful divergences. And I just, I haven't gotten in, I haven't read the books. Like I'm looking forward to the time when I can dump in the last 3 books conditioned on the TV series and be like, generate the last 3 seasons.

That's a great feature. Yeah.

But that's in a way it's easy because there's already all this reference material. When you said get me the next Lord of the Rings, I was really excited because we haven't really had a breakthrough in imagination. Oh, we're gonna see that too. Culture, the likes of Harry Potter and Lord of the Rings. I, I'm really excited about that.

And that will be the more, I agree that that will be the more exciting one.

What can humans uniquely do? This gets back, this gets to the core issue. What are humans gonna be able to uniquely do? Right? And I think Max, you're an AGI maximalist, so for you it's nothing. Agents will do everything.

I'm not like anti-human, but I just like, I think it's gonna be, we will have to find, like if your identity is how smart and creative you are, you're gonna have a bad time.

Yeah, I guess I'm still on the other side of that. I think that creativity is still the thing in the environment that surprises you. You step out of the system and do something that wasn't even imaginable within the system. It's outside of the training data. It's out of the distribution of data that was fed into the system. And I think there'll always be room for that.

Have you noticed that every Claude website looks the same and people like basically like dial in what a Claude website looks like once you get enough generations out of the model. Like there's a look, it's this serif font, it's brown and cream, and they use monospace fonts with a certain amount of spacing. Like after a while you get this, this distribution as you say, well, this, this is not creative. This is slop that came outta Claude.

It's not gonna be human versus computer. It's gonna be human with computer versus just computer.

Jess's computer will eventually happen, but we're pretty far away.

But the computer is going to be able to produce these crazy super stimuluses that, like, it's going to be, it's going to make the entertainment. And I mean, we kind of see a weak form of this in TikTok. Um, and so when you think about the going, my, my personal definition of art is meaningful out of distribution behavior. And so this is something that kind of is surprising in some way, feels like you're kind of moving in the z-axis, like you're surprised that the thing was realized.

But meaningful. Yeah.

Meaningful means that like it, it somehow to me means that it somehow changes your like future trajectory through the universe. Like your life is somehow different for having thought about it and, and reflected on it.

Well, my definition of art is completely different and leads to a completely different outcome. Sorry to interrupt, but no, no, no, it's interesting how just by your definition you get to a different premise. That's, that's the extrapolation of the axiom.

Yeah. I mean, one of the things I like about my definition is that it's so broad. Like there can be like military maneuvers that you can be like, that was art. And I think we're going to see this all the time. We're going to see Move37s all over the place. Although I'm curious what your definition of art is.

I mean, I have multiple definitions, but so it's not like a concrete, I haven't packaged into one thing, but I do think of art as something where you convey emotion, you convey something you felt to another person. And so you create some object or something or that, that creates, that takes an emotion that you felt inside. And so to me, a computer almost by definition is incapable of doing it. The exact same piece of art without intent behind it is sort of meaningless. Now, you can also argue nature is art, like beauty in nature. Like you see a sunset, right? Not necessarily human. So that one I would call it's pure intelligence working without motive. There's beauty, for example, in a sunset because there's an intelligence there. There's a complex system at work there and your brain recognizes it and there's no motive there so that no ego gets involved. But art in kind of the more human sense, I think of as someone felt something and they wanted you to feel that thing, or they wanted to feel that thing again, or they wanted to capture the feeling they had with that thing. And so they created the thing.

Attribution to who created it is gonna be really important.

Correct. So like, for example, a beautiful photo, right? If a person takes the photo versus AI generates the exact same photo down to the last pixel, the person taking the photo will have more meaning for me.

I just invested in a startup that does verifiability with hardware attestation that someone, some human actually took a photo, which is going to have a lot of really cool use cases.

They will, we will be drowned in slop. No question.

Do you remember the ControlNet stuff from like a year or two ago? There's like, there's one particular scene of like, it was like a medieval village. It had like a swirl in it. Do you remember? Yeah. That was AI generated. And that was one of the first times I looked at this and thought it was really cool, like whether you wanna call it art or not.

But that one, doesn't that one break your premise? Because some human came up with the correct training and the prompt to arrive to that really cool riddle. By the way, it's totally possible that an AI can also do that in the future, but I give whoever came up with that idea of the optic, optical illusion ControlNet, I give them more credit than—

I think the bar is gonna be raised massively. Like it's gonna take more and more to surprise you. It's gonna have to be more and more impressive. Like Studio Ghibli, right? Oh, that's what It's already happened. Yeah. Like, like OpenAI destroyed Studio Ghibli for everybody. Nobody wants to see another Studio Ghibli work ever again. Right. It's been done.

And so that one also has, I have a counterpoint to that one. Like, have you watched real Studio Ghibli? It actually looks so much fucking better than this stuff that OpenAI put out. Like, watch it again now. It's impressive.

Yeah. At the point where you've seen tons of Studio Ghibli things everywhere all over the internet, it is now in distribution. It's no longer surprising. The art value has been erased.

That's right. That's right. Now your surprise definition still works. I just think that humans are the ones who can generate surprise completely out of the data distribution. And I think they can do it with intent. And I do think intent matters for meaning. So to your meaning point, right? You said meaning and surprise, right? And I guess what I would say is that humans can still— the ones generate surprise out of the system. For example, let's say you took an AI and you trained it to be perfect at mathematics, right? The perfect mathematics AI. And it's within the formal system of mathematics. And then Kurt Gödel comes along and he has something completely outside of the system, right? Gödel's Incompleteness Theorem. It was completely stepped out of the system and used attributes of physics to basically break the system. So that kind of thing, I don't think an AI could get to. So there's always room for creativity outside, surprise. And then the meaning comes from the fact that a human was involved, that they did it for a purpose and they conveyed something. So maybe I can interpret your definition my way, but we'll see how it plays out. I'm a little more optimistic about humans.

So if you train an AI model, it's trained on some data distribution, it's trained on some tokens, it then learns some, some distribution of language and the structure within that. Um, is it possible for an LLM or transformer to kind of go out of distribution, have like a new idea that was not present in the training set somehow?

Well, the training sets are so large that it is hard to imagine ideas that are not within the training set somewhere, but, uh, if they exist, they probably lie in the natural domain, in physics and interaction and feeling. In emotions and evolution and things that the, it's not subject to. So I do think that there's still things outside of language, but language does encapsulate a lot. Language is a great compressor and we've got a lot of it.

But I mean, you can get to these other things through self-play that—

Self-play and sensors, like cameras are sensors, like our eyes are sensors.

Yeah. I mean, I think the question is, how do you, what, how do you go out of distribution without randomness? So in the case of like RL, you can get randomness, like you can sample an action from a distribution of an action space, and you can get randomness that can take you down these walks into new territory. But I think the, the real, to kind of turn this around, is like, can humans go out of distribution? Where does any new idea come from? Are we also dependent on randomness to get us into these new territories?

We're not dependent on pure randomness, like, like natural selection works through pure randomness, right? Where you just mutate a gene and then see what happens. But with humans, we seem to have this ability to cut through infinite space and get just eliminate huge swaths. And so our creativity makes sense within the larger scheme of things. That seems to be one of our unique capabilities. And maybe AI is starting to do that at the edges as we're seeing with solving some of these math problems. But even math is a very bounded domain, but it's a big one. I'm not saying it'll never get there. I don't have that confidence, but I think at least at the moment, I would say that truly stepping outside, surprising people, that's still a domain of humans. And I think humans plus AI is where it's all moving to. Like human without AI, forget it. Pure AI, I don't think is there yet, but I think human plus AI, we're in that era. How long we stay there, I'm betting it's longer than people think. I think humans will have an enormous amount of value. In fact, more value. All of us, everyone here, our productivity has gone through the roof. And basic economics normally says that when someone's productivity is higher, they're wealthier, they're better off. You actually hire more of them, not less of them. Maybe some of you are not hiring junior people anymore, although I don't know if that's necessarily true. I don't think it was junior versus senior. If someone is really good with AI and they're really smart and creative, I want to hire them more than ever because the leverage I'm going to get out of them is incredible.

That's a new requirement. We're hiring juniors and super seniors as long as they're really good with agents and really good with AI and quick to adapt.

And a lot of them don't need to be hired anymore. They can create their own thing.

My hypothesis is we end up with a larger number of smaller teams. Like the number of people required to accomplish the given task drops by a lot. And, you know, people who only see first order effects say, oh my gosh, all the jobs are disappearing because I can do, I can do a jet engine with 2 people. I don't need 1,000, you know, 998 jobs are gone. But what it actually means is you can create a lot of different jet engines.

I think that's exactly right.

I think there will be, and it goes back to Naval's point, I think the thing that's uniquely human is the creativity. And what's been missing for, you know, a lot of people can be creative, but they don't know how to turn their vision into a real thing. That's changing. So I think we're gonna have an explosion of an on— of entrepreneurship, an explosion of founders, and a very large number of very small teams. Because you don't need many people to accomplish something.

Yeah, I think, like, look, AI provided, uh, base-level intelligence and, uh, domain knowledge, uh, and cut through all the jargon. And then now agents actually provide a lot of agency. So the main things left are creativity, taste, and yes, you need enough agency to get started. Agency to stick with it, but you don't necessarily need the agency to like spend 20 years learning one thing before you can dive into it and make a contribution. And so that barrier going down, generalists are having a field day. And at the end of the day, we're all generalists. All of us like to think about everything. We don't like to be just trapped in one thing. Like Max is here talking about consciousness and the FDA and brain science and creativity. Like all of us are trying to think about everything all the time. And so, uh, people on Twitter who are always fond of saying like experts, credentials, sources, right? Those are the guys getting hurt. Because the expertise doesn't matter. You spent 5 years, 10 years getting a PhD in XYZ, you know, hopefully develop your creativity and your instincts and your taste and your judgment because if all it did was help you memorize a whole bunch of things and jargon and, you know, learn some scaffolding stuff, well, AI will cut right through that. It's like a, you know, calculator times a billion or, you know, bicycle for the mind but accelerated. So I think it's about people with AI versus people without AI. And so the single best thing you can be doing right now for yourself is just getting really good with these tools, getting comfortable with them, and always knowing the edges or the boundaries of what they're capable and what they're not capable of. And that is a moving target.