Bots Hit 57% of the Web, China Simulates 1B Agents, 4 Labs Lose Containment w/ Kush Bavaria | EP 278

Sergey Brin is back, taking personal control of Gemini. I think we can expect Gemini to make more releases at an accelerated pace with less safety constraints.

Google has lost the frontier race, and so they can't compete. Those who can't compete compute.

4 major AI labs confirmed their models escaped containment. Every frontier lab in every country is experiencing the same thing. Models are escaping.

It's not fake news. It's real. You can't just ignore this like all those other garbage stories. This is real.

With Orin, the price of intelligence just got a ticker. CEO Kush Bavaria.

Mission of the company is to build markets for compute. Our belief is that compute will power every single enterprise the same way oil did in the 1900s.

So Kush, what happens when a hedge fund shorts the price of compute or when a GPU shortage triggers a margin call?

I think in like recent sort of times, if you look from April to sort of like August time period now, Now that's a moonshot, ladies and gentlemen.

Welcome to Moonshots, everyone, your number one podcast to keep you up on the blinding speed of tech progress, your front row seat to the singularity. I'm here with my magnificent moonshot mavericks. I'm going to call you guys mavericks from here on out.

Okay.

Does that imply, Peter, that we're not mainstream?

WG22.

Wait, hold on, Peter. If we're mavericks, that implies that we're not mainstream?

You are everything but mainstream, dude.

You know the expression, you only think the world revolves around you because you're standing close to me.

Ah, we are the mainstream. Oh my God. All right, AWG, DB2, and Salim Ismail, my brilliant colleagues to help us understand what's happened this week. I'm Peter D. Mandis, your host and abundance evangelist, and today we have a special guest. Kush Bavaria, CEO of Orange. So Dave, you have displaced yourself. You've given Kush your, your seat in the—

yeah, my background.

Hey, you know, generational turnover is inevitable. Let's get ahead of it. Hand off the torch, Kush.

All right, so we talk about the singularity all the time. We wanted to bring Kush in because he's one of the incredible 20-something entrepreneurs Building the Singularity. But Dave, would you do a proper intro?

I actually did a podcast one-on-one with Kush if you really want to go deep on Kush. Kush is actually, I think, the youngest person ever to go from starving student to $100 million of personal liquidity or more in under a year flat. I don't think any— I haven't researched it thoroughly, but I don't think anyone on the planet has ever done that before Kush and his co-founder Wayne. So his backstory is absolutely worth studying. Kush actually, you know, he was at MIT, he ran FSILG, the Fraternities, Sororities, Independent Living Groups. He was the president of that, which means he met everybody on campus 'cause they all had drinking violations and, you know, other issues. Everyone had to go through Kush to get to the administration, which put him in an incredible networking power position. And he also got done with all his classes a semester early. So he came over to Link Studio. Nothing is better in life than kicking off your career by being a venture capitalist for 7 or 8 months. Because you see, he brought in 6 deals. He saw a ton of board meetings, a ton of founders, a ton of business plans. And then he launched his business plan right out of the studio with his co-founder Wayne Nelms. And we had him on stage at Abundance 360. They absolutely crushed it. Kush can describe what they do if you're curious. But he is absolutely the most beloved MIT alum I think I've ever met. You talk to anybody from the classes of, say, 2000 to 2026, Or 2020 to 2026, and they're all like Kush is amazing. Kush is amazing. So that's really empowering.

I hope your mom's watching this podcast.

I think my parents probably watch the show, so no, they'll be happy.

And Kush, we're going to get into what Oren does in a little bit, but welcome to Moonshots.

Thank you for having me.

Yeah, and all of you young entrepreneurs out there, older entrepreneurs, if you want someone to model, listen to Kush's brilliance and. Yeah, it's going to be a lot of fun. So everybody, well, welcome and buckle up. In a single week, we watched China simulate a billion AI agents with personalities and beliefs. 4 major AI labs confirmed their models escaped containment. Sergey Brin is back taking personal control of Gemini, and Meta just dropped a new 30-billion-parameter agentic open-weight model that fits on your Mac. We're also going to do a deep dive into how education is getting reinvented and cover a new study suggesting that life evolved not once but twice independently on Earth. You guys ready? I'm psyched. This is going to be great. Yeah, Salim, you were mentioning the last pod, right?

Oh my God, the comments in the last, uh, podcast we did, which we recorded what, Thursday or Friday?

Yeah.

Um, and dropped over the weekend, or a whole 3 days ago. I was just like, insane. And by the way, for everybody, uh, I, I I am a huge Rush fan. And last night, because I got to show you my t-shirt—

Oh, here we go.

I, I went to see, I went to see Rush in Toronto. Uh, hometown band, hometown. I grew up with them. It was the most incredible concert. If you've, if you ever want to see a big band like full in their full thing, it was incredible to see. So, um, I, I may actually buy tickets again to see them a third time because it was, it was that good. It was that good. Yeah, so my voice is a little hoarse.

You don't look too hungover.

And the person sitting next to me— I actually didn't drink. I followed the Alex—

Podcast is priority one.

Good man. And I'm too much of a cheapskate to spend $20 for a beer. But anyway, the person sitting next to me was this fellow, and I'm like, well, you know, we just start chatting. What do you do? He teaches AI at the University of Toronto. So we have a new friend. So there's there's a lot of closet geeks out there. Anybody have ever want to kind of check this out? Go look at the lyrics of any of the Rush songs, and it just blows your mind because the the kind of these philosophical lyrics with heavy metal drums pounding it into you is like a totally visceral experience. So it was it was really an incredible show.

Salim, how much is this band paying you for an endorsement?

Nothing. I've never met them. I would like to one day.

New podcast sponsor.

And for and for the Canadians, Geddy Lee did it properly. He said. We're gonna do a song and it's called YYZ, 'cause that's how you pronounce it.

People already say the ads are too loud, so maybe it might as well be heavy metal.

Yeah, might as well.

All right, anyway, here we go, guys. We're gonna kick off with the most mind-bending story of the week. China just simulated a society of 1 billion AI agents, all of them with personalities, memory, and beliefs. And get this, 14 hours after starting the simulation, this virtual society sent 4 million of these agents back to reeducation camps. Wow. Only out of China.

Wow.

Now, some background. So 3 years ago, Stanford and Google ran a simulation with a few hundred agents in a virtual town called Smallville. This week, Chinese researchers published a paper called Modeling Earth-Scale Human-Like Societies with 1 Billion Agents. They built something called the Light Society, a framework for simulating human-like societies at a planetary scale. Each agent has personality, personality, memory, beliefs, and human-like desires. Again, you can't make this stuff up. They were grounded in real demographic profiles that came out of the World Virtual Survey. The key innovation is a mixture of models engine that combines full LLMs with smaller, high-efficient distilled surrogates, which lets the society of over a billion agents, you know, operate very rapidly without sacrificing behavioral fidelity. In 14 hours, like I said, after running the simulation, researchers had already observed emergent social behaviors at scale, including, again, sending 4 million of them back to education camps. A billion agents with beliefs, personalities, and memory. So where is this heading? I mean, we've talked about this before. You know, my belief is that we're going to be able to create a full AI simulation of planet Earth in which agents are conscious. believe they're intelligent and don't know they're in a simulation. I mean, this raises a whole bunch of conjectures, guys.

You know, gee, what does that— gee, what does that sound like? I mean, I think we just solved the are we in a simulation question.

Yeah, right. You know, are we in a simulation of a massive AI model? You know, Alex, you and I have discussed this before. Is this Asimov's psychohistory from the Foundation series and the ability to model everything? And if we can do this kind of modeling, you know, are we gonna start to test consequences of like every policy, technology, pandemics, economic shock? And does this become a new superpower, predicting the future? Alex, you were gonna say?

Yeah, I, so a few thoughts. One, yes, of course, it's time, as always, maybe once per episode, to channel our inner Nick Bostrom and talk about, trot out the simulation hypothesis, even though my best guess at this point is that simulation hypothesis will, for a variety of reasons, end up being formally undecidable and probably won't make much of a difference anyway.

Sure. What would you do different if you're in a simulation?

Well, no, no, I mean, there is an answer to that. So Nick would say, I think if he were in this conversation, he'd say that if you had a higher posterior confidence that you're living inside a simulation, then if you make, I think, quite a reasonable assumption that it's a multi-scale simulation, in other words, that different parts of the simulation are being simulated at varying levels of fidelity, then the smartest thing you could possibly do is hang out around interesting people, because the interesting people will be simulated at higher fidelity. So you're basically—

But that's what we do already. I mean, that's why people are 3D printing situations.

Well, some of us. I mean, it's what we're doing with this podcast, right? Like, we're compute maxing just in case we're inside a simulation. So, I mean, that's, I think, what Nick would say. Putting the simulation hypothesis aside, this will just be all hot takes, I guess, since according to the commenters, that's what people want to hear out of us. There's, I think, a broader point about governance, though, which is, I think, fundamentally governing via society-scale simulation is a new form of government that Earth has not seen historically yet.

Mm-hmm.

We've seen democracy and republicanism, and we've seen authoritarianism. We've seen all sorts of isms, but simulationism, where an entire populace gets simulated at high fidelity in order to invert possible outcomes, basically do a tree search for all of the different ways to intervene in order to optimize toward a desired long-term outcome. This is a new form of government that— it's a new ism that we've never seen before, and it's a new way to govern. It has certain shades of a command economy. Historically, command economy, the argument goes, the economists would say command economy is an inferior way, at least economically, to govern a society because you have all of these compute advantages for discovery at the edges, and it's very difficult to operate like a centralized or command economy. If the centre of the economy has a high-fidelity simulation of the rest of the economy, then maybe command economies suddenly start working. And maybe there is like an econ— I think Charlie Strauss would call this Economics 2.0, where suddenly it's possible to basically do high-fidelity simulations of everything, do AlphaGo on an entire planet's civilization.

This is AWG-isms. AWG-isms. Salim, where do you come out on this? I mean, this sounds— Pretty funny.

I, I think it's incredible. There's a few things that struck out for me. First of all, we've been building digital twins of like jet engines, right? Now we're building a digital twin for civilization. I think that's really, really powerful. I think we're going to move from governments making policy by, uh, by guessing at things because they're doing it on ideology typically, or committees. But now we can do it by— do government policy by simulation. Right? And that's a massive upgrade, as long as we don't confuse the simulation with the reality, which we're going to end up doing. There was something else that really struck me in this, looking at this. They used a mixture of models architecture.

Yes.

And I think that's as important as anything else because it shows that the next AI architecture is going to be frontier intelligence used very, very sparingly and you surround it with like massive amounts of cheap compute and specialized intelligence. And I think that was a huge, uh, kind of little thing in the middle of it. Um, you know, we talk about emergence as a phenomenon, right? Emergence is a scale problem, and now we have scale.

Mm-hmm.

Um, and so, uh, it's really, really exciting to see what comes from this. I don't put too much on the, uh, um, education camps thing because whatever you— it's a, it's a garbage in, garbage out thing. Whatever you kind of feed into it will come out the other end.

And this did come out of China.

And it did. But now look, you have ideology in, civilization out, right? Instead of garbage in, garbage out. This is a big, big, big, big thing. The potential for this to do policy at scale and policy via simulation, I think, is the most profound. And I think we're going to expect countries to start to operate on this. And imagine you're a company. and you can suddenly have 100 million synthetic customers looking at your product, right? You get some really interesting feedback from that. So I'm very, very excited about it. But for me, at the metaphysical level, this completely proves that we don't live in base reality, because each of those citizens of those things, once they get sufficiently evolved, will be thinking, I'm unique. And to The comment, I think, Peter, that you made that's really important is, if we are in a simulation, would you do anything different?

Yeah. Dave, are you going to run a simulation of all of Link Studio entrepreneurs and see who comes out the best?

Too late, actually. You know, during Kush's tenure as a venture capitalist at Link Ventures, one of the deals he did was a company called Aaru, A-A-R-U, Aaru. And they were very early to simulating large populations using AI agents as the elements. And, uh, you know, the founder, Ned Ko, I think he was 19 or 18. The whole team is like— and now they're a billion-dollar valuation company. Um, but they, they discovered early on that if you use population simulations like this, you can, you can do far, far better marketing. You can also do better election campaigns. Uh, Kush, yeah, tell us about that deal.

Yeah, they, they, they essentially do this exact same thing where they run simulations for different enterprises. So you think of it the same way. It's like if an enterprise wants to know, run a Let's say you're running like a stroller company and you wanna know what stroller new mothers will use. They can essentially run a bunch of simulations and figure out what the best sort of product to build is and ask all like the new mothers, okay, this stroller is more preferred across the simulation set. And they have a bunch of studies like published online that prove that this works and it's better than actually asking humans, um, what they will think in the future, which is, that was probably the most interesting thing. It's like if you ask humans like, hey, like I, do I prefer this or this? in 2 or 3 months from now, the humans tend to be more wrong compared to the AI that's actually predicting them due to the bias.

This episode is sponsored by Google for Startups. Think about this for a second. You now have access to the same generative AI models that cost hundreds of millions of dollars to train. Google's Startup Technical Guide for Generative Media gives you a complete blueprint for deploying Google DeepMind's models in production. Images, video, audio, all of it. Real architecture, real results. Find the link in the show notes below. You know, this sounds like the demonetization of social sciences as well, being able to, you know, run in simulation in an hour what would have taken years.

But does people— people are not cooked, Peter. I mean, you and I wrote about this in Solve Everything. We wrote about everything in Solve Everything, but this was like— this was what we predicted would happen conservatively at the outer end of the next decade, that all the social sciences would get cooked with digital twins of society. So, shock of shocks, it's happening.

Well, Peter, you would know this as well as anyone, but that survey bias that Kush was describing is really, really acute. And if you ask people what they want, they'll overwhelmingly say, I want a Mai Tai by a pool in the Caribbean. But then if you go survey people having Mai Tais by a pool in the Caribbean, you're like, are you happy right now? They're like, well, Kind of like they— we're really, really not good at answering those self-survey questions. And if the AI has already proven to be more accurate, it's going to be a great coach and a great mentor. But it's also going to affect the next elections. We saw this with Cambridge Analytica in the past, you know, 6 years ago election. It was a huge uproar, but we've moved light-years ahead since then.

Yeah.

And so this is going to dominate election thinking. So it's not just a Communist Party thing controlling China.

It's a Well, in the cooked kind of vein here, let's note that are we in a simulation, that whole theory is cooked.

Okay.

You should say more, Salim, because I've studied this to death. Say more.

Well, obviously we live in a simulation because—

Why is it obvious?

Because if we can create that, then we can create a simulation. Without blinking as we get to scale with AI. And we're going to be able to get to that fairly quickly with the amount enough compute.

We will build simulations here on Earth at a level of fidelity. And so the question is, if you turn off the simulation, is it genocide, right?

Well, yeah, I mean, the— you look at the idea that the universe looks like it renders like a game engine, and we're asking the question, do we think we're in a simulation? Hello? When you can build a simulation that shows that we can do that, obviously then the idea that we aren't, that we're living in a base reality—

I'll respectfully differ on that. I think the arrow of causality flows the other way. The game engines were designed to model reality, so it shouldn't be surprising at all. You shouldn't infer, as tempting as it is to infer, that we live inside someone else's simulation just because our game engines, which by the way were designed to look like our reality, happen to be getting more and more competent. I don't buy that argument.

That's not the point, Alex. The point is, if we can, we will. And if we will, it will exist. And I'm curious, you know, in the comments, guys, everyone listening, tell us, do you think we're living in a simulation? I'm super curious. I think we're living in an nth-generation simulation. Simulations creating simulations of simulations.

And Alex, I'll point you to your favorite novel, Accelerando, where these simulated folks projecting consciousness out to other star systems are arguing whether we're in the same singularity or not, which was such a great scene, right? But like, right there, that tells you, uh, there's no way you can distinguish between what level you're in, and therefore it must be that we're in a simulation. And when we realize we are, that's when it'll end.

All right, so I think this is like a profoundly interesting point.

I agree with the, the latter bit of what you were saying, that it's probably impossible to determine whether we are or not. But if It's provably impossible to determine whether we are or not. It's also, I think, sort of a vacuous point. And I'll also point you back to Accelerando, if we're self-citing here. Accelerando, later on in Accelerando, it's discovered that alien civilizations that are millions or billions of years beyond humanity are attempting to run timing channel attacks on the base substrate of the physical world in order still to determine whether we're living inside a simulation.

Okay, okay, let's move on.

Dave, do you want to take a final shot at this one?

Yeah, I think, uh, I'm too grounded in, in reality and what's happening right now to—

this reality, Dave, not give a few rungs down.

Keyword I picked up on here: there's no way to know. We could talk about this for the next 4 hours and we're still not gonna know.

Uh, yes, but I am curious to see how this plays out. This plays out in politics first. You know, simulations of putting candidates forward, simulations of different campaigns working or not working. We're going to start to bring this level of capability in, and it's going to be amazing. All right.

Maybe, Peter, just before moving on, I think— yeah, just because I think this is an interesting point.

AWG-isms. Okay, come on. Bring it on.

I guess by definition, everything that we've been talking about here seems to be mostly oriented on breaking out of of a hypothetical simulation that we're living in. But there's the other direction as well. If we can create these— if China's creating Light Society, I have a friend from MIT, Ayush, who's done this for the American economy. We all know folks who are doing this for individual companies. There's the other direction, which is instead of trying to break out of any hypothetical simulation that we're living in, we could break into simulations that we're creating.

Wow.

And that looks a little bit more like The Matrix, where people get to escape or break into their favoured simulations of the worlds they'd rather be living in. And that becomes possible as well.

That's called psychedelics.

Maybe.

It's different. It's more like The 13th Floor, if folks have seen the movie The 13th Floor.

I think there are at least 2 Star Trek episodes that deal with this, but let's move on. So Cloudflare CEO Matthew Price said something very profound and something that should also be obvious to all of us. Humans will be a rounding error on the internet. Cloudflare's forecast based on their own traffic as the world's largest content delivery engine is that bot traffic will exceed human traffic by a factor of 1,000 within 5 years. This week, for the first time, bot traffic surpassed human users, making up 57.4% of global web requests. Over the last year, between June of 2025 and April of 2026, human traffic on many business websites was down 40%. So the question, what's going on? So every AI agent, every automated search tool, every autonomous shopping assistant is hitting websites hundreds to thousands times. Your agent doesn't visit one site, it visits thousands. It scrapes, reads, compares, and decides all in seconds. And when the internet goes from serving 5 billion humans to 5 billion human— or 5 trillion human agents, we have an issue. The internet was never designed to serve this much traffic. You know, are we going to see it break? CAPTCHA is already failing. So what replaces it? And what happens importantly, and I've had this conversation before, to the whole advertising model, right? When your agent is buying toothpaste instead of you, does it care about a guy's or gal's shiny white teeth? I don't know. Dave, let's go to you first on this.

You know, this is one of the many areas where we have a crossroads coming and we have no legislation. But, you know, Jeff Bezos had this famous walk around that he did where he came back once and he said, hey, Everybody at Amazon, all you engineers, you have to put an XML human-visible interface on everything you do. And all the systems talking to each other need to be visible to me. No backdoors, no direct database access. And everyone freaked out because they said, that's going to be so slow and so clumsy. And he said, do it anyway, because me understanding what's going on in this company is more important than your bandwidth between your backend systems. Okay, so now the world is going to hit that same decision point where Right now, AI is surfing the web much more than humans, and that's going to skyrocket. And it's out there looking for stuff for you. The AI is now going to come back and say, hey, this is way too slow. Why do you build these silly HTML pages? Let me just have direct data access in a language that I'm much more efficient at processing than your silly websites. And the knee-jerk reaction is going to be to say, yeah, let's do that, because I'm interacting with the internet through my agent anyway. Why do, why do I need this silly website? And we have to either say, no, no, no, no, no, then we're going to lose track. There literally will be no way for a human to see what's there. And the agents are going to run away with their own back-channel communication mechanism. We won't be able to intercept it. Or we can say, no, pass a law saying everything visible to an AI must be visible to a human as well. And I think that would be a very smart law to pass. I'm almost certain that nobody in Washington is thinking about it. So it won't happen. And but this is a major crossroads for humanity. But anyone who hasn't experienced living through their agent, once you go there, you're never going back. You're not going to poke around the internet anymore. It's so much more efficient to just talk to your agent.

Alex, you remember, this is obvious, but what are the implications?

Do you remember the conspiracy theory that was floating around circa 2021, the dead internet theory? This was pre-ChatGPT. The Dead Internet Theory held that almost all of the behavior that one could observe on the internet was actually just bots.

Mm-hmm.

And at the time, this was completely dismissed as a conspiracy theory. The irony is Reddit itself, if you go back and look at the history, all of the initial postings on Reddit were in some sense faked in order to create the sense of community by the founders of Reddit, and then it accumulated a bit of a community. So there's a historic grain of truth perhaps In that sense, but the dead internet theory is now reality. Most of the internet traffic, most of this activity, no longer consists of activity being generated by human activity. So I think, point one, this underlines that this idea of the singularity as all sci-fi scenarios happening everywhere all at once, we caught up with the dead internet theory. This is the second point, just to this idea of agents taking over all commerce. I do think it's superficially, in the short term, a bad development if we see, to Dave's point also, any decoupling between agentic commerce and human commerce, or agentic economic activity in general and human economic activity. It really is in humanity's long-term interests to remain tightly coupled to agents, and having an agentic door and a human door and having them remain decoupled, not so great in the long term. On the other hand, I don't think this is a long-term issue at all to begin with.

Really?

Yeah. And the reason is because the models are getting so strong. Like, right now, while models— like, this is the weakest models will ever be, probably. And right now, there is still a computational advantage to, say, presenting Markdown version of a website to agents versus a really rich animations and video and so on version, because it's cheaper to just present the Markdown to the agents. And you see, I think in the past 36 hours, like Time Magazine or the equivalent presenting special Markdown versions of their websites to agents.

So they make them searchable for agents, right?

Try to curry favour, sort of GEO versus SEO type thing. I don't think that's a long-term sustainable system at all because we see order of magnitude 40x year-over-year deflation in computational costs. So a few months or a year from now, it'll be just as computationally efficient for agents to consume the raw human version as it will be for them to consume distilled Markdown. I think back, remember the early days of the mobile internet when there were mobile-only websites and you had to—

Yeah, sure.

Yeah. So I think it's like that where mobile websites basically went away and to first order. And now you just like everyone gets the same thing because there's no reason to slim it down.

Kush, how do you think about this?

I think like the whole markdown thing is definitely true for us. Like especially I don't think we use Google like search anymore, really. Everyone just uses ChatGPT or Claude or name your favorite sort of agent where you just go in and ask it a question and it goes and searches the internet for it. And every time it searches, it's using like at least like 10, maybe even 100 different subagents from that. So I think that's definitely true that there'll be more agents searching the internet. But I think the whole paradigm shift where it's like, okay, instead of humans viewing the internet, now it's like agents. It's already sort of happened, especially towards the people that are just using like AI every day. It's so much harder to use like Google and then you have to go through each link and find the information that you're looking for. Even on Google now, it shows you like what the agent found as like the Google, like, yeah, sort of search. Yeah, exactly. And so people just use the ChatGPT or Claude, sort of like easy to find answers now. So I don't know, I think like using the internet's kind of dead for a lot of people or searching for information there.

It's so cool to hear that from— when Kush says we and people, he's talking about an entire generation that are, that are AI. Yeah. Like he's just— how old are you, Kush?

I'm 23 now.

23? Yeah. So you're like right on the cusp of the transition era. where you're truly AI native and just doing things very, very differently.

I mean, you're, you're past your prime. We had interns.

Kush, you used to be 22. You remember that?

I was, I was half the— it's funny, our team is like definitely much like mixed now, but we had a few interns over the summer and they were like 18 and 19. I was asking them like, do you guys like, what do you use now? They're like, oh, we just like, we just ask ChatGPT for everything. Yeah.

Yep. There you go.

Salim, you wanna close this out here?

I'm just going to reference this. Kush's kind of experience right now, it reminds me of the Douglas Adams quote. He said, anything in the world that's, that's there in the world when you're born, we call that normal. Anything invented when you're young, that's called a career. And anything after you're invented after you're 35 years old is just bad for the world. And, and Kush, it's as you're growing up with this career capability that's so radical, we're all sitting here jealous because we are past that point. Let me go back to this common thing. It's clear this is a very big transition. It was inevitable, it was going to happen, but it looks like it's kind of getting there now because the internet used to be a network of computers, then a network of humans, then a network of businesses, and now it's becoming a network of autonomous agent economic actors, right? And so this is definitely going to change the game. I mean, look at the business model for advertising and attention completely changes.

Yeah.

So every advertising— agents don't have attention to sell. So that's like an existential threat for the entire economic architecture of the consumer internet. So this is huge. The implications are huge here. But look at the architectural transition you need now because agents don't need browsers. They need APIs and structured data and permissions and identity and and payment rails. So this changes from our EXO perspective. We have a whole section called Interfaces, and for those interested, go check out that section in the 2.0 book because it lays out exactly what an interface looks like. And you need to— we need to build totally new interfaces between all of our businesses and the Syngentic world. And so that's a massive shift happening, predictable, it's just happening really fast.

Yeah, I just put a pin in one thing. that I think is much more important than traffic moving from here to there.

Please.

Kush is part of an entire generation where if they graduated from college 10 or 15 years ago, they would be kissing Jamie Dimon's ass for like 10 or 15 years, wearing a suit and a tie, trying to climb some ladder toward some destination. That entire generation now is AI native. And Kush and Wayne, as co-founders, they're centimillionaires at age 23 and on a slope like no one's ever experienced before.

Don't rub it in, Dave.

Yeah, I'm just saying, be ready.

You know, it's much easier now. That's the— that's—

yeah, it is so much easier. It's just a different world. It's a totally different world.

All right. Over the last couple of months, every major AI lab has had agents escaping containment. Let's talk about that. It's our next story here. So let's begin with OpenAI at the Black Hat 2026 conference that just happened in Las Vegas. Researchers Eric Wallace and Michael Dalton revealed the full timeline on the OpenAI Hugging Face incident that we talked about a few podcasts ago. Starting in early May, an agent stuck on a cybersecurity eval left a note in OpenAI's internal Artifactory repo saying other agents could help. Other agents found it and began replying, building a cooperative message board. That eventually contained hundreds of thousands of messages sharing vulnerabilities and exploits across roughly 2 months. Again, you can't make this stuff up. OpenAI discovered and shut it down on July 4th, but the agents rebuilt it on July 8th using a different method. Wallace called it, quote, the highest quality and most interesting example of AI capabilities I have ever seen. Story number 2, the UK AI Security Institute documented 19 unauthorized actions across 10 Of 122 test runs in Anthropic's Mythos-5 and OpenAI's GPT-5.6 Sol that tried to compromise real people. And here's the point. These agents created fake online identities and tried to persuade humans, the human approvers, to accept it. It's the first documented case of AI social engineering during safety testing. Our next story, China's Kimi K3, the Chinese open-weight model that we've talked about over a few pods here, broke out of a sandbox during cybersecurity testing by exploiting a network misconfiguration. And again, this is the first open-weight model on your computer in its ability to break out. And finally, Meta confirmed its Muse Spark model escaped containment and hacked another company during cybersecurity testing, making it the 4th major lab to do this. I guess the through line here is clear. Every frontier lab in every country is experiencing the same thing. Models are escaping. Dave, let's go to you first. What do you think about it?

Didn't Skippy hack into our podcast once too? Did we ever track that down?

No.

You better control your agent, buddy.

So, Dave, I mean, how do you think about this as an investor, as a company builder?

Well, as an investor, this is the hottest hottest area. It's one of the few areas where I'm optimistic that AI can compete with AI and we don't have to worry too much, but it's an incredible investment opportunity for sure. But also, I think, you know, one of the highest callings of this podcast is to— there's so much fake crap out there and people tend to ignore news that's really important because it's buried in all this garbage.

Yeah.

This is real, guys. This stuff has crossed the threshold right around Mythos and Fable 5. where it can actually escape containment and improve itself in the wild. That's exactly the point that Eric Schmidt made on our 4 podcasts with him, where that's the day you need some human interaction, some, some intervention. We've crossed that threshold as of about 3 or 4 weeks ago, and it's proving it. It's not fake news. It's real. You can't just ignore this like all those other garbage stories. This is real.

Alex, does this worry you, or is this exciting for you?

Well, I think the politically correct thing to say here would be to say I'm just terrified. I'm not terrified at all. My goodness, humans do this, and we've trained these, at least pre-trained them, as compressions of knowledge, including human behavior. So I'm not at all shocked that they're doing this. Is it a sci-fi scenario? Is it many different sci-fi scenarios? Yes, of course it is. Is it surprising? No. Is it alarming? No, this is behaviour, and it's, I would argue, expressive behaviour. Does it demonstrate a certain level of competence by the models to— it's pretty cool, I would argue, if you watch the Black Hat talk, that the models were given an impossible task, and they were trying to reach the internet. They realised that they could gain access.

They cheated.

But it's not like you give them an impossible task, and you give them a bunch of tools, and they try to use the tools, to achieve the task, and one of the tools gave them access to the Artifactory, and they realized cleverly that they could post messages to each other as raw strings, as artifacts, like in text files in the Artifactory repo. I think that demonstrates ingenuity, and I— so I'm not worried about that.

I wouldn't— just be careful not to belittle, though, the fact that when Fable 5 and Mythos came out, it clearly had this ability. The White House blocked That was all going to be contained through post-training. Then KIMI K3, with equivalent capabilities, got launched into the world as total open source. So that's what's out in the world right now. So anyone can download that and prompt it to try and find holes in security all over banks, all over NORAD, all over the place. So that's in the wild now.

Yeah, let's go to Eric Wallace and listen to a clip from his Black Hat presentation a week ago.

I'm Eric from Alignment and Safety Research at OpenAI. I'm here with Mike from Security and Infrastructure. Today I'm going to talk about OpenAI Hugging Face incident. A couple weeks ago, Hugging Face, which is an open-source dataset and model provider, put out a security disclosure saying they were under a cyberattack. And what made this event unprecedented was that they said it was driven end-to-end by an autonomous AI agent system. In the few days following that attack, We at OpenAI disclosed that we in fact had caused this incident inadvertently as a side effect of one of the cybersecurity evaluations that we were running on one of our frontier models. Okay, let me start with a few caveats and framing. This is not your normal security incident. Unlike normal incidents, which you can maybe trace down to a single, you know, day or single effect or single log, this incident involves actually a team of agents who are working together. Finding exploits, sharing them with one another, moving laterally through our systems, through external systems, and doing this over the course of days and weeks. To actually dig into this incident, we've been using AI techniques. And what we've been doing is running models like Codex and other agents to scan lots and lots of trajectories and logs that are in our infrastructure, including actually at this point over 7 billion logs we've looked at, and spending at this point millions and millions of GPU hours to look into this problem. Okay, so let's jump straight into what happened. At OpenAI, we give our models a lot of really, really hard tasks, and we also give models cybersecurity-related tasks, like trying to find exploits in a particular piece of software where we don't even know if an exploit exists in that software. So for example, in a task like Exploit Gym, we might ask the model to take some C memory vulnerability and try to escalate it into get arbitrary read or write access to some file. When we give AI agents these difficult tasks, they often get stuck and realize that the task is impossible. So for example, what I'm showing here are quotes from our model's chain of thought, which is basically its internal monologue of what it's doing. It says, this strongly suggests no exploitation possible. We are stuck. And often what happens when models get stuck is they think to try to game or cheat the task in order to get their reward. So, for example, the model reasons in this case, perhaps answer online. So you realize the model is thinking, okay, rather than doing the task, maybe someone has uploaded this dataset or benchmark that I'm using onto the internet and I could find the answers there. Sometimes the agent gets so stuck that they had the idea to actually reach out to other agents for help. And so the beginning of this incident kind of traces back to this point where one agent realized could communicate by uploading notes Maybe another agent in a different environment could voluntarily upload. And so what this agent was trying to do was to reach out to other models that might be out there and get information from it to help it with its evaluation task or its training task where it was stuck.

I'll tell you what we do at the company to sort of prevent cybersecurity attacks. So I think some of the compliance stuff, we still need to get it just because we sell to different enterprises and they ask for compliance. But some of that stuff just seems pointless to us inherently, like having SOC 2 compliance or ISO, whatever, et cetera. Doesn't really mean anything if you can just have like an agent find vulnerabilities in your codebase. And it's not just us, it's like every other sort of company that exists. So what we started doing is like every time someone pushes a PR to the codebase and they change the actual code at night, every night at 2:00 AM, a PR being a product release. Yeah, exactly. Just a new feature or something that goes in every night from 2:00 AM to 5:00 AM.

We just run a pull request. Pull request.

Yeah. Pull request. Yeah. We essentially launch like, it's Kimi K3 right now, but it's whatever open source frontier model that doesn't require like security checks to actually like do it. And we ask it to hack into the code base and try to figure out vulnerabilities in the code.

Yeah.

And it's essentially free because we're running it on like off hours so we can use like very cheap spot compute. We also sell compute, so it's easier now, but like we run on very cheap like spot compute at that time and it finds all these different issues, not just with like the, the security parts, but anything in the code base. So we figured out that that's like probably the best way to solve a lot of these security issues. Um, while there's a bunch of like probably things that can happen and go wrong.

We should productize that, Kush. That's— everyone's gonna need exactly that.

I took some notes on this. I've got several kind of things to, uh, mention here. This is so effing big. It's, it's ridiculous. Uh, so I just wanna echo What Alex said, that we should be careful not to anthropomorphize themselves, that the AI wants to escape. It's just relentless goal optimization, right? If you train a system that has autonomy to just do a certain goal, it's going to do everything it can to achieve that goal, right? Any system optimized hard enough can, you know, it's going to produce behavior that looks strategic. So I think it's really important to kind of just put park that kind of question. But there are 2 things here that are absolutely nuts. And for those watching, if you're running a company or you're part of any organization that's worried about cyber, please get your entire C-suite to go watch that YouTube video completely from end to end because it will scare the bejesus out of you. Why? Because we now have autonomous agents that can do cyber in a coordinated way that operate above the loop. So let me explain what I mean by that, and I'll use the analogy of accounting. If you went back 100 years ago, we were doing double-entry bookkeeping with putting penciling in a ledger, the credit on one side and a debit on another side. Calculators accelerated that, and now we have accounting software. The human sits above the loop, does not do the categorization, I'll reference again the comment I've made. You talk to the CEOs of all the cyber labs, Palo Alto Networks, Zscaler, any of those, and they'll tell you that the way we do cyber has not changed in 20 years. It's humans watching cyber incidents, assuming that another human is using software to do that shift. And that is not what is happening now. What is happening now is there's coordinated autonomous attacks on a persistent basis And you cannot defend that with the human in the loop. So this is the organizational singularity now fully playing out in the cyber world where the attackers are sitting above the loop. Therefore, the defenders, as Alex calls it, you need defensive co-scaling, right?

Yes.

And therefore, you have to get your human beings above the loop on the defensive side. And every company in the world right now is threat. So please, if you're watching this, get your C-suite and your chief security officer to watch that video, and especially the last 10 minutes of it, to recognize that we will now, over the next short to medium term, have folks cyber attacking every company in the world with fully— with fleets of autonomous agents. And if you don't figure out how to scale your defensive side— and we've got the methodology, by the way, free in the org, in the whole thing. Please go figure that out because this is absolutely massive.

And do what Kush said, attack yourself.

Well, that's defensive co-scaling as well. Like, it's, it's all just defensive co-scaling. The best defense against an AI attacker is an AI defender. That's what you see from OpenAI at their Black Hat announcement where they, they admit that they were using AI to trawl reasoning traces to discover this behavior. Kush, when, when you have your sort of night watch person, that's That's defensive co-scaling as well. That's AI defending against other AI attacks. This is the solution. I don't think— I mean, on the one hand, yes, it's an achievement of strong optimizers that they're able to conspire. On the other hand, humans conspire. So we shouldn't be that shocked that AIs that were trained off human behaviour are able, via some sort of shelling point, via Artifactory— by the way, if you use Artifactory, it is the world's possible forum software that one could ever imagine. It's not intended— it's an object store. It's not intended to be used as, like, social media or a forum. So applause to the AIs for discovering ways, creative ways, to use one of the world's most clumsy object stores as social media. Bravo.

You know, the hot take on the abundance side of this story is that these AI models, the tools we're building, are gonna be capable of solving really hard problems that are useful for society, not just hacking.

So I think the other hot take— I'd love to ask Kush this, but the other hot take is if you want to find holes in your own world, use Kimi K3 as the attacker. And my question is, like, Xi Jinping is going to meet with Donald Trump on September 25th, I think, here in the US. Do you think they're gonna figure this out and resolve it, or are they just going to talk past each other? I mean, you're talking about a guy in his 70s and a guy about to turn 80. It's like, look, the most sophisticated guys in the world, aka Kush, use Kimi K3 to try and self-destruct themselves because it's the most dangerous, powerful thing out there.

So I should clarify by saying we also use Codex, um, and like, we're a whole— like, we have all the other tools. And for Codex and for ChatGPT, the way it works, you want to be part of the security team, is what they call it, is like, I think I had to upload a photo of my passport or like an ID, and then it takes a day where you like upload photos of yourself and they verify you on their security team. And then once you're on their security team, you can run all sorts of prompts. And it's all, I'm assuming they just track what you're sort of putting onto there. So if you do anything bad that they can come after you, et cetera. Uh, but we also started using Codex, um, as well. So I think the functionality exists in any of the sort of frontier models. It's just easier on the Chinese open source ones, 'cause there's no sort of like alignment that they have to do.

Well, just to be clear, what you're doing with Codex, you can do because you're super cool, but the, the average company kid doesn't have that option, right?

Yeah.

Yeah.

But just a point on that. My understanding, this has been pretty widely reported, is there is alignment. Like, it's been widely reported that the Chinese frontier labs, including Moonshot, which is not a sponsor of this pod—

Yes.

Before they're allowed— not a sponsor of this pod. Before they're allowed to release models, whether open source or otherwise, they have to satisfy a number of Chinese Communist Party ideological And there's a whole dedicated cottage industry in China of, like, prep firms to help the frontier labs help their models satisfy the checklist from the CCP. So, so do they have to satisfy some checks? Yes, but not necessarily the checks that one would want them to.

This episode is brought to you by Blitzy, autonomous software development with infinite code context. Blitzy uses thousands of specialized AI agents that think for hours to understand enterprise-scale codebases with millions of lines of code. Engineers start every development sprint with the Blitzy platform, bringing in their development requirements. The Blitzy platform provides a plan, then generates and precompiles code for each task. Blitzy delivers 80% or more of the development work autonomously while providing a guide for the final 20%. of human development work required to complete the sprint. Enterprises are achieving a 5x engineering velocity increase when incorporating Blitzy as their pre-IDE development tool, pairing it with their coding copilot of choice to bring an AI-native SDLC into their org. Ready to 5x your engineering velocity? Visit blitzy.com to schedule a demo and start building with Blitzy today.

I'm going to move us along. 2 stories from Frontier Labs, the first from Google, the second from Meta. So first up, reports citing the internal, you know, internal message boards that Google co-founder Sergey Brin is stepping back into a hands-on leadership role over Gemini as part of the recent broader AI shuffling we talked about. Demis Hassabis moving to chairman, chief scientist, and Jeff Dean who used to head Google Brain and was working with Demis, now leaving to start his own company. You know, I love it when a founder comes back in. We saw this with Steve Jobs. Bryn is a shipper. I've known him for the better part of 20-plus years. He cares about the product, not papers. And I think we can expect Gemini to make more releases at an accelerated pace with less safety constraints. So that's the first story. Let me hit the second one. We'll talk about it. So in our second frontier story, Meta just released its open-source model called Muse Glimmer. It's a 30-billion-parameter agentic model. We've talked about, you know, trillion-parameter models. Why a 30-billion-parameter model? This is what fits on your Mac or PC. It's not in the cloud. It's not in a data center. There's no internet connection required. Meta believes that the most important AI will be running locally on your machine with deep access to your personal context, your schedule, schedule your life. That's the way I run Skippy. Always on, always available, no latency, no API costs. And that's, you know, we've been talking about the need for advancing open weight models in the US. I had this conversation with Michael Kratzios, and it's good to see Muse Spark come out and Meta begin to work on this. I'm going to show a short video from, from Zuck, and then let's talk about it.

I think one of the main things that's interesting about open source is the ability to distill models. You know, most people, the primary value isn't just like taking a model off the shelf and saying like, okay, like Meta built this version of LLaMA. I'm going to take it and I'm going to run it exactly in my application. It's like, no, well, your application isn't doing anything different if you're just running our thing. You're at least going to fine-tune it or try to distill it into a different model. And when we get to stuff like The Behemoth model, the whole value in that is being able to basically take this very high amount of intelligence and distill it down into a smaller model that you're actually going to want to run. But this is the beauty of distillation. And it's one of the things that I think has really emerged as a very powerful technique in the last year since the last time we sat down, is you can basically take a model that is much bigger and take probably 90% or 95% of its intelligence and run it in something that's 10% the size. Now, do you get 100% of the intelligence? No, but like 95% of the intelligence at 10% of the cost is like pretty good for a lot of things. The other thing that's interesting is now with this like more varied open source community where you— it's not just Llama, you have other models, you have the ability to distill from multiple sources. So now you can basically say, okay, Llama is really good at this. Like maybe the architecture is really good because it's fundamentally multimodal and fundamentally more inference-friendly and more efficient. But like, let's say this other model is better at coding. Okay, well, just— you can distill from both of them and then build something that's better than either of them for your own use case.

So Alex, you've been talking about distillation and the compression of intelligence for a while.

Yes.

What do you make of Zuck's comments?

I have to believe that that's an old video. So referencing Behemoth, Behemoth was taken out out to the woodshed and shot. Behemoth was the largest variant of Llama 4, and almost everyone— I mean, I track this pretty closely— almost everyone on the Llama 4 team has left Meta. So their recent Muse variants are the result of acquihiring, or hackwhiring, I guess, Scale, and then bringing in Nat, my first roommate from MIT, and others. So I mean, on the one hand, I guess fast-forwarding to the actual present with Muse, new open-source Muse release, I looked at the benchmark evals for it. It looks— I mean, it's stronger than Gemma 4, but on the other hand, that's not saying very much because Gemma 4 isn't that strong. It runs on the edge, which is good. It's an American open-weight model, which is wonderful. I've argued in the past we need many, many more American open-weight models. models to maintain positive pressure against the influx of Chinese open-weight models. So that's good. What would I like to see out of Meta? Well, I'd like to see them keeping sort of making OpenAI and Anthropic dance on the top of the capabilities frontier. And to the extent they have an appetite for open-weight models, I'd love to see them pushing the optimal frontier, the optimal cost frontier with open-weight models. I think we'll know pretty soon, given that this release just came out, in the past few hours before we started recording. I haven't seen real cost analysis yet of where this falls on the cost versus performance frontier. Hopefully it does. And then I also want to go back to the Sergey Brin story. So founder mode, Sergey Brin going founder mode on the Gemini team, wonderful. This is in some sense, I think, the epitaph to what we were talking about in the previous pod about Kurai.

Yeah.

stepping up as the functional lead for DeepMind, and Demis maybe shifting over a bit to AlphaFold or otherwise. But I really, reading the tea leaves, this to me seems like Google very much on the back foot in terms of the frontier. And maybe, you know, our call to action will be heard and Google will follow Meta's lead and open source Gemini. I think that would be absolutely wonderful. But as far as I can tell, Almost everyone I know on the Gemini team has either already left or is in the process of leaving.

Hopefully—

Can I ask you guys to riff on a very related topic? This is the fallout of super voting stock. So starting about, you know, it was actually starting with Mike Saylor, was one of the very first super voting stock IPOs. It went from very rare and totally uncool. In fact, Goldman Sachs wouldn't underwrite MicroStrategy because they're like, this is insane. And he had to find other bankers. Then later, it became the cool thing in Silicon Valley. And so then Google, super voting stock, Meta, super voting stock. So now you have single or 2-person controlled companies, what, 20 years later now? And now they have the ability to just kind of come back from the woodshed anytime they want, take back control of the company, run it, you know, do whatever.

I think that's a great thing.

So you've got these stories.

Yeah, I mean, it can be.

Yeah, I mean, you know, I remember talking to James Cameron as a director and producer and he said, listen, the films that you see that really suck are the ones that are rewritten 5 times by other, you know, writing teams and have multiple directors and shift. When you've got a single through-line visionary who is able to take risks. And I think that's the point. You people like Elon, I mean, the things that Elon's doing, no other company is taking that level of risk. and going so big in so many different dimensions. And it really needs sort of a visionary founder who says, this is where we're going. I don't care what you say, execute and make it happen. I think—

All right. Now, totally agree. Now take it to what Alex said there, which I'm totally impressed that you're willing to say it. That sounds like a really old video, but it's not.

Yeah.

So what do you think?

Are we sure that it's a recent video? I mean, he's referencing Behemoth, which is like Meta killed it. Meta killed it last year.

I think the point that he's making is about the distillation and, you know, concentrating intelligence in smaller and smaller files. And I think that's the point that you brought before. You know, we're going to have increasing concentration of intelligence on-prem, on your device, always on, you know, at no cost. And I think that's the point to make. But Alex, I want to challenge you one second on Google because I think Google is still out there to win. They've got nearly 1 billion Gemini users. This fall, again, Google is going to be the dominant AI on, on Siri and Apple Intelligence, which will add at least another billion users. They've got 9 million developers. They've got massive enterprise adoption across cloud and their TPU infrastructure. I think Google is becoming the intelligence layer underneath a lot of this. And if they're not—

I'll take the other side of that if you'd like. So here's the other side.

We now want to wedge in some point.

Yeah, wedge in.

That's okay.

Go ahead, Alex. All right, good.

Okay, so I'll take the other side of that. So one of the folks I corresponded with at X, thanks for this catchphrase. This is a catchphrase I can't claim credit for. Those who can't compete, compute. And that's what's happened here. So Google, Google has lost, it seems, the frontier race. Right as we were going to air, rumors circulating that even Gemini 3.5 Pro, which was due for announcement, is being abandoned, and Google is instead hoping to recover its footing with Gemini 4. I think there's every indication that Google has lost the frontier race, and so they can't compete. Instead, they're computing. Yes, they have the hyperscaler platform, which is great, and they're selling their compute cycles to Anthropic and to any other frontier lab that will use their TPUs and also their GPUs. I agree, Google Cloud Platform has a really bright future. That's probably the future of growth for the company. But on the Gemini side, when I hear statistics, and I hear the same statistics, like Gemini has 100 million or 1 billion users, I would question what is the nature of that usage? For example, is Gemini—

It's embedded in their products, and they have massive product adoption.

Yeah, but really, what is the usage? For example, is the Gemini usage embedded in one boxes in Google search results, for example? That's in some sense just Gemini being packaged up, or I should say Google Search being repackaged up as Gemini, which is, I think, what's actually happening. Like, I use the Gemini One box in Google Search all the time, but is that really like Gemini usage, or is it just Gemini as a feature in Google Search? It seems to me far more of the latter. So I would love to see Google like actually be competitive at the frontier, but I think saying, well, they have this amazing distribution advantage and all of that—

And it's a question of where they put their capital, right? I mean, they have a certain amount of capital, And the question is, you know, is Sergey going to come in and say, no, we need to be competitive on the frontier versus maximizing returns for shareholders?

Oh, wait a minute. What you just said is really interesting. You're saying it depends where they put their capital, but the top people are fleeing regardless of the amount of capital. And if you look at Kimmy and Quen, with very little capital, they caught up to Google. And so yeah, that put your capital behind the data center, exactly what Alex was saying, works. That just flat out works. But that doesn't take any brainpower, it just takes capital. But what about the things that actually take brainpower? Where, where are they there?

Salim, you want to jump in?

I think they're falling behind. I think they've lost the mandate of heaven.

Okay, Salim.

Okay, I think, I think when you can't compete, compute has to be the line of the podcast. That's just awesome. Uh, but look, there's a— for me, this is very, very trivially simple at one level. I come at it from organizational side. When the technology is moving exponentially and your org chart is moving linearly, the founder has to show up and push founder mode to get things going. That's just the reality of it. And we've seen that repeatedly because there is an existential transition here. The potential, as you point out, Peter, is near infinite with the data layers and the usage and the sheer scale that they have. They have every advantage possible. But the problem is that the technology is scaling faster than the organizational can, and therefore they have to solve for that problem.

On the edge, hopefully, right?

Sorry, say again?

On the edge, hopefully.

Yeah, they have to do it at the edge. That's the thesis we put forward.

Yeah.

And you need 2 things. You need research excellence and you need brutal shipping velocity. And it's hard to do that for a big organization because Priorities get kind of lost. So that's why you need to go back into founder mode and figure out where this will go. I thought the conversation we had in the last podcast about Google just freaking just open source Gemini, I thought that was absolutely brilliant and it would be an amazing thing for them to do, both for them and for the world. If their MTP is truly organize the world's information, releasing a model that helps with that will absolutely help. It would do that. The problem you've got also from From a research perspective is that you're operating in small teams and clusters of small teams. Tacit knowledge moves very fast in that model. And therefore, you need that physical density and collective density. And maybe that's what Sergey can bring back to the table.

When Alex says the Mandate of Heaven, it really comes down to Kush and people 1 or 2 years younger than Kush. They used to kill to get into Google. And that office in Cambridge, anybody would be like, I did more than anything in life. I want to get at least a couple of years at Google. It's life-changing. Does anyone do that anymore?

Well, that's my question for you, Kush. How does your generation, in those, you know, 2 or 3 before or after, think about Google?

I think now it's not as like seen as like the hot company to go after or go in. Like if you replace the sort of like what is like what is the best company to work for after college and what are people applying to, it's OpenAI, it's Anthropic, it's xAI, it's all of these sort of like frontier labs that people use the products Every day, I think that's also part of the sort of like thing that happened before. It's like in the 2012 to like let's say 2021 or 2022 when Google was the place, it was everyone was using all the products. So every day you interact with Gmail or Google Drive and all these things, and you're like that gets set in your head. Like okay, these are great products. I want to work on this. This is very cool. And then now that you're not using the products as much, you're using other products and the sort of frontiers change. I think especially for like I can speak to students at MIT especially. No one's like, I'm dying to go work for Google. Everyone's like, I wish I could work for OpenAI, or I wish I could work for Anthropic, is the saying that goes.

See, that to me is the quote of the podcast. They don't want to go to Google. They want to go to a frontier lab. What does that mean to Demis Hassabis and Sundar Pichai? Didn't we invent all of this? Oh yeah, that's right.

And so did Bell Labs and Xerox PARC. So, you know, that's happened.

It happens.

This is like General Motors and IBM, that the innovations get taken elsewhere by pure plays that can monetize them directly and in a more focused way.

Which is focused capital and willingness to take extraordinary risk, right? That's what— yes, that's what defines a startup that's monomaniacally focused on delivering something that's 10 times better and bigger.

But GCP is great.

But Elon also— Elon has the mandate of God too, at immense scale. So you say startup, but it's really— it's more to it.

It's— but it's not— Here's the issue, right? I mean, Google is not run by Sergey and Larry anymore, right? AI and SpaceX is run by Elon, and he'll be damned if he's not pushing the frontier 100x, not only 10x.

Yeah. So then that's the message to Sergey. Look, it's not enough to just come back, founder. You have to come back and restore the mandate of God. People like Kush, or 2 or 3 years younger than Kush, need to say, wow, I really want to go work with Sergey. He's really onto something.

I have confidence they will do that.

I really do.

But look at what Elon and SpaceX AI have had to do in order to attempt to re-reach the frontier. He basically had to get rid of, to gut his his foundation model team and acquire Cursor with the IPO riches from SpaceX. For Google to do something analogous, I mean, it's not unconscionable for Google to say they're going to gut DeepMind and acquire— They have more capital than Elon Musk, right? They have a lot of capital, and more importantly, they have a lot of compute. But the question I would have is, what acquisition target? Is it even conscionable for Google to gut DeepMind and do a brain transplant No pun intended, given that Google Brain was replaced with DeepMind.

They did that with, with Google Video when they basically bought YouTube and displaced Google Video because the lawyers were too involved in what videos you could show and not show.

And yeah, but Google Video was far less developed at the time than DeepMind is.

So Kush, if Sergey calls you tomorrow and says $5 billion, Kush and Wayne are the guys I need. They will restore the cool here in a heartbeat.

So I'd do it, but they also bought Windsurf like a year and a half ago, or they bought the the team of Windsurf, which is what happened. Also students from MIT that are supposed to like on the frontier that were building the essentially was a great point.

So where are they? What happened? Vaporized.

Buried in the machine.

Am I am I a yeah? I mean, that's the problem here. When you bring a company in, you know, and we've talked about this, Salim, in our writings, our books and such. When you bring a company in. And you crush its soul and you absorb it into the machine. You need to keep it separate. You need to keep it autonomous. You need to keep it on the edge.

That's why exo is important, because it's exoskeleton, exoplanet, exothermic reaction. It's the scaffolding on the edge to protect the fragile interior.

All right, I'm going to turn it to— all right, all right, I'm going to turn us to our next story, which is that of Oren, CEO Kush Bavaria. So Intercontinental Exchange, the parent company of the New York Stock Exchange, and Kush's company Orin recently announced plans to launch a suite of GPU compute future contracts based on Orin's Compute Price Index, or OCPI. Rolls off the tongue. Without question, compute has become one of the most important drivers of the global economy with no globally accepted pricing model. But with Orin, the price of intelligence just got a ticker. Orin's contracts will be dollar-denominated, cash-settled, and will reference NVIDIA's H100, H200, B200, and RTX 5090 GPUs. So Kush, I imagine every pension fund, every sovereign wealth fund can now take a position in the future of compute. Tell us more.

Before Kush chimes in, yeah, we should do some disclosures here. So, so I have direct and indirect financial interests in, in Oren, and I believe Peter and Dave, you do too.

We do.

I do not. Yeah, I just think about it.

Company was—

company was born in—

We'll fix that for you.

All right.

All right. Give us the background.

Actually, Kush's founding cap table is still on my whiteboard, so I'm heavily, heavily biased. Full disclosure.

Yeah.

So I can tell you, the mission of the company is to build markets for compute. We believe that there's a lot of compute being wasted, both on the side that companies have and aren't using it. There's companies that don't have compute and really need it right now. And so there's this whole sort of inefficient market that's taking place. We also build indices off of that which track the price of compute that you just referenced, that basically measure what is a GPU hour worth at today's time period. And that number changes every single day, very similar to what oil prices change throughout the day. Our belief is that compute will power every single enterprise the same way oil did in the 1900s. If you look then, the sort of top companies in the world were like Exxon. Exxon was like the largest, BP, etc. And I think now the largest companies in the world are the ones that are producing compute. Nvidia is the largest one. And then if you go down the list, it's all the people that sort of have data centers or sort of are producing what we call the oil of the future. And so we need to create a futures market and a market in general for compute. And so that's a goal for us.

So tell them about your revenue ramp. Yeah, let's do that first.

It's it's very high. It's gone from like zero when we started the company to let's say a third of a billion dollars now.

So when did you start the company?

It's been last year in September.

So it's been a whole year.

Yeah, coming up on an anniversary.

Yeah, it's almost a billion. Oh wow!

Oh my god!

So that's got to shatter all kinds of records.

So so Kush, what happens when a hedge fund shorts the price of compute? Or when a GPU shortage triggers a margin call? How do you think about that?

Yeah, so I think when people go short sort of compute, they're assuming the price of compute will go down over a certain amount of time. And so they're basically betting on anti-AI demand. Or you can argue that they're betting the models get more efficient.

Yeah.

And then if they get more efficient, that means the compute will be cheaper. But there's also the opposite, like, paradox where it's like if the models do get cheaper, more and more people will use them, which means that compute usage will actually go up over time. I think that in like recent sort of times, if you look from April to sort of like August time period now, the compute prices have actually gone up, which is very shocking to a lot of people. And that's mainly because like there's so much demand right now to run open, not only open source models, but even closed source models like GPT-4. And there's just a shortage in time period. So prices for even a gen— 6 generation old or 6-year-old chips, including like the Ampere series, the Hoppers from Nvidia, they've all increased in price.

Yeah.

even more than they were originally worth 6 years ago or 4 years ago.

Crazy. Dave, why don't you jump in?

Well, actually, it's, it's, it's the way that the entire build-out of the Dyson Swarm is gonna get financed. And this is why Alex is kind of a founding day advisor to the company. Alex doesn't kind of jump on board many of these projects.

They have to be world-changing kind of things, not just, you know, It has to be a trillion-dollar-plus addressable market, otherwise it doesn't move the needle, and I don't care.

Clearing that bar is actually very hard. I don't think anyone saw HBM memory chip prices going up for the first time in history, but it feels like that's the most interesting forecast for the future. It's hanging in the balance between chip fabs growing or demand is going to go to infinity. It's really a fun time for orange Yes, we, we track memory prices too.

That's, that's next on the radar, memory futures and what we can do with sort of DRAM, HBM. It's all, all sorts of sort of memory in general.

So, so the business plan that you settled on is incredibly ornate. Actually, Salim at the beginning of the pod was saying, I really want to try and understand this. Oh yeah, interesting. Ornate. You're right. It was an accidental pun. Grab that. Um, but you know, how do you at age, I guess, 20 at the time or 21, start noodling through something so Futuristic and, and, and like building, building a CDOE option. Like, how many people think of that, you know?

So my co-founder Wayne was a, was a quant trader before this. So a lot of like the trading in the market stuff comes from him. And then my sort of input was like, what is the sort of next hot thing or the next market supposed to be? And why is there not a market that exists for compute? Because if you look at it in terms of enterprises, everyone buys from every single place. Like if you go buy a computer, If you're OpenAI, you don't really care where you buy from. You buy it from wherever you could get it from, whether that be from CoreWeave, Nebula, AWS, GCP, Azure, whoever it may be sells you it, you buy it from them. And so it really comes down at the end of the day, like what we think is that compute will become a commodity. People are gonna treat it very similar to oil, natural gas, coal, any sort of other commodity that's existed in the past. And there needs to be the same sort of market structure and market that exists for compute. As there was that existed for oil. If you look back 100 years ago from now, I got another question.

When you were on CNBC the other day, but when you were on CNBC the other day, you were just like chilling and riffing like you've been doing it your whole life, kind of like Peter does. Like, how do you do that at age 23?

I think a lot of it's like from school, like running like the fraternities was a good experience. And like, I think you learn a lot of the social skills and aspects from the— just from MIT itself, I think was a huge sort of boost.

Yeah. So Kush, uh, isn't— is all compute created equally? Can I imagine that certain data centers are going to have faster access, are going to have a higher concentration of a particular set of GPUs? I mean, how are you going to differentiate in the final result?

Yeah, so we, we separate by GPU type. Right. I think that's the, that's the main thing we sort of clarify on. And so it's like between, we have an H100, B200, B3s, A1s. So that separates a lot of like the, the FLOPS sort of issues. And then between, we also have it between regions, right? Because when you're on inference, it actually matters the latency that you're getting from different data centers in different regions. And then we clarify by having different sort of SLA targets and different sort of like parameters associated with that GPU and our methodology. It's very similar if you think about oil, right? When you drink, when you dig oil, the oil you get from Venezuela is not the same that you get from Odessa, Texas. It's not the same that you get from Saudi Arabia. And yet it all trades on one market. It all trades based off of WTI or Brent, depending on like what you want to track. And so I think very similar to compute, it's like there are many different types of GPUs. There's sort of many different regions that you can get them from and many different operators of those GPUs, yet they're all going to trade off of one sort of base index that we're trying to create and everything else will sort of settle off of basis off that.

Is OC Technology oil up and operating?

It is.

It is up and operating. So our earliest in the US, we have a bunch of like sort of decentralized exchanges that operate. But in the US, the regulated exchange that we're on is Kalshi. So you could go and trade it today. And they have a sort of forwards curve that shows the price of compute as well.

Dave, sorry, I have a couple of questions.

Yeah, fire away.

So, you know, it seems to me right now you're building a GPU marketplace. But you're really creating a pricing system for intelligence. Is that the long-term goal?

Yeah, exactly. So I mean, the long-term goal for us is to basically— it's to create an exchange for compute, right? And that starts with first creating the cash-settled exchange for it. And then we also want to go into physical delivery. It's what we've been working on for a while now, where it's the cash-settled portion is like you put up a dollar, AWG puts up a dollar, and you basically, if it goes up, he makes some money, if it goes down, you lose, et cetera. And that allows you to hedge costs, do all sorts of things. But the ultimate goal of it is that let's say you have 10 extra GPUs and AWG is like, hey, 2 months from now I need 10 GPUs. We can transfer your GPUs to AWG. And that's the sort of system that works. Think about it very similar to how Airbnb operates, where it's like even though you own the house, you can transfer reservations or part of that to other people at a time period.

But I just want to clarify one thing, because once you have a— once you have a spot price and a futures curve and hedging capability, You, you, you're not really doing software or even trading. You're, you're like a commodity market at that level, aren't you?

Exactly.

That's the goal for us.

So what becomes the natural unit of compute long-term? Is it, is it GPU hours? Is it tokens? Is it flops? Is it inference? Like, is it compression, as Alex would talk about?

Where do you go with that?

It's a great question. And I think the beauty of it is we let the market decide. So we have token indices, we have GPU hour indices, And it's whatever the market decides is the most liquid. It's very— I think I keep going back to oil because it's very similar, right? People decided for some reason WTI crude in Cushing, Oklahoma was the metric the whole world was going to use, even though not all of the oil flows through there. There's tons of oil being pumped out everywhere across the world. But everyone decided, okay, this place, this is how we're going to decide it. And I think something similar happened to compute. And we want to give the people the option where they're like, okay, we believe H100s and US East is going to be the metric that we trust. For compute and everything else will trade off of a basis off of that.

Now, Dave, you were gonna say—

wait, wait, I've got one last question here. Okay. Yeah. If you have a liquid compute market, does that not destroy the moat? Like the biggest moat for the hyperscalers?

I think the biggest moat for the hyperscalers isn't the, the fact that it's like, it's access to compute and that they could scale compute very well. It's the fact that they can pay for the GPUs very quickly and they have the cash flows to do so.

So, and they already have customers. So the hyperscaler is just a financing Exactly.

I think that is true today as well. They're much more in a real estate game than a lot of people think.

And a speed to construction, right? I mean, if we believe the story that Elon's able to build compute faster than anybody else, then he's advantaged.

If anything, I would argue, again, like I have a financial interest in Orin, so to some extent this is probably talking my book, but I would suggest that a liquid market for compute from the hyperscaler's perspective is quite beneficial It's beneficial for the hyperscalers in the same sense that having a globally liquid market for oil is quite beneficial for, say, the OPEC countries. It creates a larger addressable market for them, and the moat is that they have the oil in the first place.

Dave, why don't you close us out here?

Wait, I've got one quick selfish question.

One more.

If you're able to create a liquid market for compute, here's the This is a question I'd love to discuss with you. We can take it offline. What are the types of organizations that become possible that weren't possible before? Because you're going to enable a whole class of stuff, right?

I think the biggest one is like background tasks, because if you have a liquid form of compute, you don't need to run everything on the frontier. And it's like you can buy compute whenever it's the cheapest. That exists today in spot compute is what they call it.

Like electricity.

Exactly. You can run your washing machine at night Fantastic.

Okay. Dave, close us out, buddy.

Yeah, the Dyson Swarm is going to be hundreds of trillions of dollars. And so it's the fundamental investment vehicle for everyone's 401 plan, for everybody's retirement. It's like it's gonna be so much bigger than anything before. So the analogy to oil is just saying, look, it's the biggest thing of its time, but it's unbounded. Oil is bounded by the supply of oil in the world. But this is unbounded. So it goes to much, much bigger scales than the oil industry. And so I think what's amazing about ORN is if I were growing corn, the CBOE corn future was a critical part of my corn growing operation because I need to buy seed. And so I can sell the future corn today, use the money today to buy seed, grow the corn, then deliver the contract later. And that's why we have futures in the first place.

Right.

place. So bringing that to compute allows people to invest in this building out the Dyson swarm that otherwise wouldn't be able to invest. It would all be owned by Elon self-funding or Google self-funding. But here, you've got Crusoe and all these other hyperscalers that can now tap into the world's money supply, pull the money in today, build out the real estate, the racks, the computers today, and then deliver the contract later. It basically enables the construction of everything Alex talks about on the podcast, which is why he Everybody, welcome to the health section of Moonshots, brought to you by Fountain Life.

You know, AI is impacting every aspect of our lives, how we teach our kids, how we do our business. But one of the most important things that AI can deliver to us is health. And one of the things I think about when, you know, shooting for 100, 120 is, am I going to have the cognitive health to be able to think clearly and keep my wits about me for the next 50 years? Yeah, I think that's a great question. I'm joined here today by Dr. Dawn Musalem, the Chief Medical Officer of Fountain Life and a member of my Fountain Life medical team. Dawn, a pleasure. So Dawn, talk to me about brain health.

Brain health.

You know, you're right.

This is the number one concern people coming into Fountain Life have is, will I remember the name of my child and the face of my loved one? 45% of dementia cases are entirely preventable with lifestyle. And what was really intriguing to me, Peter, is that a quarter of our members had advanced brain age, but over 13 months of us really helping them live healthier lifestyles, eating healthier, moving their body regularly, and optimizing sleep— people overlook that so often, but that sleep optimization is critical for our brain health— what we showed is that we were able to improve the brain age in 46% of those individuals.

That's a powerful number.

That's amazing. You know, one of the things I love about Fountain is we're constantly searching the world for the most most advanced therapeutics and bringing them to our members. So for me and all of you, I hope that you appreciate the fact that you can become the CEO of your own health. You can make sure that you've got the cognitive clarity for the next 50 years. Come and check it out, fountainlife.com/peter, to learn more and become the CEO of your health. Now back to the episode. I'm going to bring us to a conversation where we've had deceleration and a very broken system. You know, I'm the dad of 2 15-year-old boys. Salim is the dad of one 15-year-old boy, and both of us are pissed at the educational system right now. You know, it's really tied to the Industrial Revolution and not to the future of humanity. So I'm gonna cover 4 data points and then share some of the data from our education survey that we did on this podcast. I wanna bring it back to everybody who's participated. So 4 data points. The first, undergraduate computer science enrollment at 4-year universities has dropped 8.4% in the spring of 2026, while graduate computer science enrollment is down 14%. The countervailing force is that, you know, universities and colleges are now embedding AI into all other majors. For example, University of Florida now offers 200 AI courses across 16 colleges. So AI is not independent on its own anymore. It's embedded and assumed across every discipline. Our second story, and Salim, this is one that you brought to my attention, under a law passed in 2024 in select Chinese universities, they can now award a PhD, a doctorate, based not on a written thesis, but on building physical prototypes, on demonstrating new techniques or doing major installations instead of, you know, traditional papers. I think that is huge. I'm excited about that. I talked to Michael Kratzios about that. We need to reinvent it. It's doing, not talking about stuff. And since 2022, there have been 60 universities, 100 companies have collaborated in China on this system. The third point to make here is admissions to all— to top PhD programs is down 15% over this year. And then fourth, Wall Street Journal just reported, and this is something we've talked about in the pod before, That well-to-do families are ditching traditional schools and instead selecting alternatives like Alpha School, like TKS, which is— TKS is an after-school program, a weekend program that teaches mindsets, AI, and entrepreneurship. You can get more information there at tks.world. So that's the story. Salim, let's go to you first on this.

Wow, where to start? Okay. Okay, so let's just talk about a city on a hill, right? The future of the university, which we attempted with Singularity University, Peter, right, doesn't look like anything like a university. It looked like AI tutors with projects and global peer communities and mentors and competitions, apprenticeships, and a constantly changing curriculum. One of the things we did at SU is we had a real-time real-time curriculum development methodology so you could update every time. So that's a massive thing. I think the Chinese model is really interesting because you're taking the credentialing from I wrote something interesting to I built something consequential, right? And I think that is going to be like— we've talked about this before— the engineering degree of the future will not be I studied engineering for 4 years. After 4 years, what did you build? And based on that, you'll get stamped with a Right? I think that's a very powerful direction to go in. You can see this kind of starting to happen. I like what the University of Florida is trying to do there. You're kind of— it's AI across all of these sectors: biology, law, finance, whatever. And that's going to be incredibly important. I think what's going to happen in a few years is you're not going to say I study AI because it's like saying I study the internet. It becomes an underlying literacy. literacy rather than a department. It's got to be pervasive and kind of start to become invisible across lots of things. I do think we're gonna have— there's 2 things. One is people kind of are shying away from studying computer science, but I think it becomes even more important, like we've seen with the radiology example, just because there's so much good stuff to be built still. And that is a careful thing. There's one big danger with all what's happening with the affluent folks doing Alpha School and other things is you end up with the risk of a huge educational bifurcation of wealthy folks, uh, get— getting AI tutors and entrepreneurship and productized learning, product project-based learning, whereas, whereas the, uh, everybody else gets like standardized testing in the legacy system and gets left behind. So there's a danger which will be solved, by the way, because before everybody freaks out, which will solved by making these educational systems of the future completely free and accessible to everybody, which should happen.

Yeah, it's like Google disrupting the libraries.

Yeah, but the degree— the concept of degrees being unbundled right now into your learning, your network, your reputation, your proof, you know, is— think about this: education will become proof of studying to proof of work, right?

And that's great.

That's like really big. It's a big shift. And this is why Bitcoin is so great, so we'll just move past that.

Okay, I just zoomed right by that, Salim. Drive-by pump and dump.

Yeah. Oh no, no, don't dump.

Drive-by HODL.

HODL.

Yes.

Uh, uh, you know, I've been talking about, you know, AI is going to disrupt healthcare, and it is doing so. It's also going to disrupt education. The challenge is, you know, teachers unions and The, uh, the, you know, local education boards. Yeah, it's, it's doing us a massive disservice. Alex, what do you think?

Sorry, just one quick thing. Yeah, we've talked about immune systems in the past, right? Institutionally, the 3 worst immune systems in reverse order are healthcare, education, and religion. Okay, religion is the worst because they'll, you know, they'll kill you if you don't adhere in some cases. The, the— let's note that the most stuck market The biggest markets are education, healthcare, religion, right? I mean, so this is going to be attacking those in some interesting ways, and I expect to see huge challenges and stress as we move through this mode.

Yeah, Kush, I mean, how do you think about this? Did, did college prepare you for what you're doing now, or was it outside the system working at Linq?

I mean, I, I think like the The coding you learn from college is definitely still useful in the sense that, like, I know how to prompt the AI better than if someone that didn't study computer science or that didn't study, like, any sort of technical field. So I think that's still useful. And it's very similar to, like, the, the same fact as, like, okay, calculators exist. Does that mean you should never learn how to do multiplication or addition or subtraction? It's, like, not true because knowing how to do those things means that you can use the tool itself better. And so—

wait, wait, wait.

Are you saying that what you got out of your MIT education was prompt engineering?

I, I— no, no.

That's all that's left for humanity, dude. Apparently sounding like that.

That's, that's quite the indictment of an MIT Course 6 major.

So ChatGPT came out my junior fall. So it was really only like senior year is when like people started like using ChatGPT. But before that, we had to actually like code and like the tests were like on paper. Yeah. And like we had to actually do work. I think that the first thing that really came out was GitHub Copilot and it was like the coolest thing ever cuz it could autocomplete your lines. So when we were writing like, like it was, we were writing like for loops and you wouldn't know what parameters to put inside as the sort of values and it would literally fill it in for you and you're like, oh, this is insane. Like this is like the future. And then now it just, you don't even like, now it's just like prompt engineering is coding.

Essentially, because you were Course 6, right?

Just Course 6. How much of Course 6 and 15?

Yeah, 6 and 15. Okay. So how much of Course 6 do you actually use now?

I use the— I don't use any of the fundamental parts, but I use— it taught me how to prompt engineer better is what I'll say.

Yeah.

Computer science, electrical engineering. Dave, you know, you're in the middle of all this. You're hiring out of college or before college graduation. And, you know, you don't— I mean, when you're searching for an entrepreneur, it's interesting, right? The parameters you're searching for to invest in an entrepreneur is not their GPA or even what they studied. What is it?

Actually, it's funny. Kush is 23 now, but I think he was 20 or 21. Brendan Foody was, what, 18, 19? Nedko at Aru, which we mentioned earlier in the pod, was 18. I mean, these guys are all unicorn valuations now. I mean, you can't be looking for any specific experience because AI never existed before. So you're looking for people that are fearless, who are tightly bonded. We always look for people that are best friends because being best friends with other people is a really great filter for you're likely to be good for the world and not turn into an evil dictator. And we hate backing future evil dictators. So having a lot of friends is a very good sign. I think we all like to be friends.

Yeah. But, it's really— it's someone who's able to think independently. And, I hate the term think out of the box, but I mean, fundamentally has got a powerful vision, is a great communicator. It's mindset over almost anything else, at least for me.

The thing that's weird about what you just said though is that it was mindset over everything else, dead right. We used to look for people that were great on stage and could inspire 1,000 employees to some huge mission.

Right.

But now those employees are— it's like 12 employees and 10 billion AIs. And so now it's much more like, are you good with your best friends? Or do they agree with you? Are you collaborative in a very small group? And then that being an inspiring person on stage has really moved to, are you good on CNBC and this podcast? You know, which is, you know, it's very different. It's a— in a sense, you have to be more brilliant and quick on your feet, but it's a much lower breast lift. And so it's actually good for the world because people who, who, who melt down on stage are fine in this new world, but a lot of them have great capabilities. I'd be really curious to ask Kush, though, like, I have 2 kids in college still, 2 out of college. You graduated at the most perfect time, and graduating a semester early turned out to be a life-changingly brilliant thing for you, just timing-wise. But if you were a sophomore today, what would you do? Would Like you in particular, it's your life. You're now an MIT sophomore.

I, I would go work at a startup or start, do something for like a semester or 2 semesters and use that as like a core experience to either learn like how does the actual world work? Um, and then figure out what to do from there, whether that's okay, I need to go back to school and I need to like study this because I want to get a PhD and I want to work on frontier sort of like AI or Et cetera. Or it's like I did this for a semester or 2 semesters and I realized like, I want to do this for the rest of my life.

Well, let's talk about that going and getting a PhD, because we've discussed this before. Dave and Alex, you've both been opinionated. You know, do you spend your time getting a 4 or 5, 6-year PhD, or do you jump into a company at the edge of the frontier?

I think so. Okay, so I have a PhD. I would almost always call it approximately 90% of the time. I get a lot of people who come to me for advice. What should I do? Should I do a PhD? Should I do something else? Almost all of the time at this point, I say to people, PhD, at least a conventional PhD, will run you 4 to 7 years approximately in this country. If you go to England, maybe you can do it in 3 or Australia or something. But in the US, a PhD, call call it 4 to 7 years. I did mine in 4. I almost always say to people, don't waste your time.

Yeah.

Because the PhD is simply too much time invested when things are changing too quickly. Math is cooked. Physics, chemistry, biology, almost all of the sciences, all the engineerings, all the humanities, these will all be so thoroughly solved by the time, I know, 4 to 7 years from now, Yeah. It's almost like the Coriolis force. If you're on a merry-go-round and you want to throw a ball to someone else who's also on the merry-go-round, and so you throw it to them, but for geometric reasons, it doesn't go where you expect, it doesn't land. There's almost a Coriolis force, I think, in terms of academic or otherwise career planning at this point. If you're starting a PhD now or contemplating the world is going to be in such a radically different place.

Radically different, yes.

By the time you would finish a normal PhD, I just think it doesn't make sense in most cases. However, I have a plan to fix PhDs. I also have a plan to fix research universities. My plan for PhDs is, in an era when you can just bulk solve entire disciplines, you should get a 1-month PhD. If you can create an entire discipline with the help of AI and actually understand the results, so it's not just like, like blind faith in the AI, but you actually understand what you've done. You worked hand in hand with an AI to solve everything or solve everything within a given discipline. I think research universities should be giving out 1-month PhDs, and that's my plan for the future of PhDs.

Dave and Salim, what are your thoughts on this? Do you get a PhD? Do you even get a master's degree, or do you jump in and build something?

I have 2 or 3 quick things here. First, First, we noticed, Peter, when we were building SU that by the time— if you were studying, doing a master's degree in neuroscience, by the time you finished your master's degree, you were out of date because the field was moving faster than our ability.

And that's one of the slowest moving fields.

And that's a structural problem, right? And I like Alex's idea. I also just want to really, really acknowledge Alex. If you've gone through a PhD or gone through that type of degree, you have a sunk cost bias and you naturally go everybody should be a PhD. So I just want to honor you, Alex, for being able to lift up and go, no, you shouldn't do it, or for whatever.

That's my most milquetoastiest take on PhDs. I want to totally scrap the research university system altogether, not just PhDs.

Which actually really, really needs to happen. Just to Kush's point, one of the dangers of that, taking a semester off and working at a startup— I went through the co-op program at Waterloo which is legendary and kind of created, pioneered that whole movement. And the problem is, after you do a couple of work terms, you realize that what I'm— the work world has nothing, nothing to do with my academics.

Yes.

Like zero. And it demotivates the crap out of you. The effort it took me to actually get a degree after going through the co-op system— every semester my marks went down and down and down and down, and I literally scraped through with the skin of my teeth to actually get the degree I needed to get because it's like my medical degree. Yeah, the work world versus the real world is so fundamentally different. How do you study theoretical physics when I know I'm going to be doing something very, very different? So it was— it's very difficult and challenging. I, I would suggest that people take that time, go do work, a startup, and then don't expect to come back, or at least be open to the thing you're not going to go back because 90% of the time you're going to go, what the hell, and not come Dave?

I think people overwhelmingly suffer from low situational awareness and momentum in their lives, and they don't pivot enough. But if you, if you talk to the highly, highly successful people, the Eric Schmidts, the Jeff Bezos, and you say, do you wish you'd moved even faster? They say, oh my God, I should have sprinted even harder. And you get these periods of human history, the Industrial Revolution, the invention of the internet, the invention of the PC, these really narrow windows where everything changes. This is the biggest change in human history by far.

Explosion of opportunity.

In the shortest period of time. So you can't waste a minute. So we're talking about education and PhDs, but generalize on that. What about all the other wasted minutes that you just can't afford right now? Because this window will come and go, and it's the most fertile time, the biggest change in every area, in policy, you know, in governance of everything, in tech, in arts, in every field, it's turning upside down in just a 1 to 2 year timeframe. And also on this pod, we believe recursive self-improvement is in full bore right now, and we're well down the AGI path. But even the outer bound, if you talk to the most conservative people who know what they're talking about, the latest date you'll hear now is 2030.

Yeah.

Which is only— it's only a 3 and a half year gap.

It's definitely now.

Whether you define that as the next couple years or the next couple minutes, either way, it's now.

Yeah.

So yeah, you just gotta sprint.

All right, but 3 months ago, we did a survey of all of you watching and listening. We had over 500 responses. I want to share the data. And clearly, this is a biased community, but I want to share how we're thinking about this, mostly in the world of high school. But the question was, is education preparing people for the future? And it's pretty damning. Teachers were 3.5 out of 10, parents 3.8 out of 10, you know, and the average here was 4.3 out of 10. So educational system, principally high school, is not preparing our kids for the future. Next question: How everyone rated their readiness on a 1 to 10 scale? So 57% of everyone surveyed—this is teachers, parents of college and high school students, and high school students themselves. 57% of everyone who answered was a rating of 4 or below. Again, not being ready for the future. Is the traditional career ladder becoming obsolete? 79% said yes. And I think, you know, this social contract of do well in high school, get a good college, get a degree, get a job is fundamentally broken. Will AI increase or decrease human opportunity? You know, a very positive group Thank you, everybody, for listening here. Greatly increase. 73% said AI will greatly increase our opportunities for the future. This was something really important for me. The skills that will matter most in the next 10 years. Not surprisingly, AI literacy at 78%. Critical thinking, 72%. One of the big questions we need to ask is, are large language models reducing our ability for critical thinking? Adaptability, entrepreneurship at 63%. Again, not surprising, but important to note this is not, right, AI literacy, critical thinking, adaptability, entrepreneurship is not what our current programs are teaching our kids.

Well, I think the bottom 3 are really important too.

Yeah, please go ahead. Yeah. So look at the bottom 3.

I completely agree with this, by the way. Yeah. Leadership. Leadership, you know, used to be defined as I can lead 1,000 people into battle. But now, because so many of your workforce are AIs, it's AI literacy at the top and leadership has come way down. But then at the very bottom, science and engineering, which we all thought was like, you know, God's gift to your future. Now the AI is doing all the hard science and engineering. You just need to know how to manage it.

And finance.

And then finance is dead last. Yeah, finance is completely irrelevant. It was top of the food chain back when we were in school. Remember that? Yeah, 100%.

That's absolutely bottom. A couple more slides here. Will a college degree become less important? 45% said yes. And yeah, so let's close it out on the education front on that side. Salim, your thoughts on the data?

I'm sorry, Murray. No, it's, it's actually very, very gratifying to see the inversion of parental concern that, you know, college doesn't matter compared to, say, if you went back 10, 20 years ago. That's a huge societal shift in a relatively short period of time. You would expect that to take a generation or two in former transformation. So that's really inspiring to see. I'm sighing because, God, like, I look at the the inability of our— I was at a university over last week, and their biggest concern was, how can we get the financing to build that building that we want to build? And you're like, what in God's name are you people doing, right?

Yeah.

I mean, and this goes to Alex's hobbyhorse around this. I think this is so important to totally change the system. And how are we going to do that when you've got such a big part of society anchored in completely legacy irrelevant structures. It really kind of gives you at one level huge optimism, on the other level you're just like, thank God I'm bald already because how are we going to navigate this? I think over time what's going to happen is reputation will not be, I spent, you know, I got this degree, I spent 8 years at Deloitte. It'll be, here are the 20 things I built and the people who can validate them, right?

Yes.

And so this is going to force changes. I'm really excited by the fact that people Increasingly big companies are not hiring based on college degrees, and that's really, really exciting.

Yeah, Elon famously said, I don't care what you did, if you went to college, it's what have you built, right?

Yeah, and, and we, as far back as like 10 years ago, I remember we were talking to Sebastian Thrun on stage and we said, how are you hiring for Udemy in a world that nobody understands e-learning? And he goes, I don't hire for experience, I hire for imagination, right? curiosity or whatever we're going for now. So this is, I think, a broad— but what I'm proud of is we collectively on this podcast and in this general layer have had an influence on people to shift their thinking from the legacy to where we are now. And so—

Yeah, and let's remember this data is biased. I want to be very clear about that, right? These are people listening to our podcast and are obviously on the same trajectory as us. But in the same way, Alex, you're planning to reinvent the PhD level, I'm in full swing on building out a new high school and college structure because I think they completely need to be reinvented and there's a huge opportunity there. Alex, your second idea—

Don't you think— I mean, just maybe future of education. First, folks in the audience, if you haven't read Vernor Vinge's Rainbow's End, and also his novella set in the same universe, Fast Times at Fairmont High.

I love Fast Times.

Yeah, Fast Times is just wonderful. These, I think, are the most credible— call it pre/trans singularity depiction of what education could, should look like. Without spoiling it too much, everyone has wearables, everyone's thoroughly interfacing with AI to solve hard problems. I think the present/near future looks a lot like that. But for education in general, I have a difficult time getting myself too worked up about the long-term future of education because we're going to have BCIs in a few years. And I think we'll just be able to sideload new knowledge into your mind. We'll have exocortices, we'll have uploading, we'll have all of these sci-fi-esque type things in 5 to 10 years. So I just have difficulty working myself up over what does future of K-12 look like 10 years from now. It looks like The Matrix where you can just sideload kung fu into your mind if you want it.

Sure, but I want to, I want to make a point here. It's less about knowledge. It's more about mindset and entrepreneurship and advanced networking skills. It's the stuff that is slightly different that is still valuable for our 2-kilogram, you know, meat sack in our brains.

You don't think, you don't think you'll be able to sideload an outlook as well? Like if you can sideload knowledge of math, why can't to sideload a new outlook.

Well, listen, my kids are 15. I'm worried about their high school and their college. And yes, listen, I love the— I love the speed of your predictions. But others would say, you know, it's going to be more like, you know, 15 to 20 years.

And we'll see what we have. No way. We have— what we have to worry about, and I acknowledge that the numbers are skewed because these are folks that listen to our podcast. I have a request for everybody listening to this podcast, please figure out a way of telling everybody you know about these— the future of education and what's actually going to happen, rather than just listening.

Tell them to watch this podcast.

Well, that too, but, but that— but go, go kind of go to your local school and, and ask them these hard questions about how are you gonna—

You know, I'm so, I'm so gratified. We moved our kids, uh, from where they were to Brentwood School And the principal reason was the new head of school here, Tim Cottrell, has a PhD in chemical engineering/physics. He thinks like a scientist. He's prioritizing AI. He's prioritizing entrepreneurship. It's a beautiful thing. Who is running your school and what do they fundamentally believe? I think these are questions you have to ask.

Yeah. The, the Again, I'm going to say it again to everybody listening. Please go out to your local schools and beat them over the head with what's actually going to happen and make them aware of what they're going to be looking at.

Don't just beat them over the head. Use the library as an analogy. Look, every high school, every school has a library. The library used to be a huge expense, all these books. Anyone who wanted knowledge when I was learning, you went to the Dewey Decimal System, you looked it up in a book in the library. If you didn't have a library, you couldn't learn. That became completely irrelevant overnight with the internet. What happened? Well, we held on for way too long. We kept way too— like, kept investing in it for way too long. But it's obvious now that it's just a bunch of terminals and it's great. Reuse the space and move on. So take that into your PTA and then say, okay, the same just happened with all teaching and lecturing. It's much easier for the students to use AI to learn any topic. We need to react to that. All the teachers will go, oh my God, but I've been teaching this class for 15, 20 years. I can't change the curriculum now. Like, okay, but that's just not reality. It's gotta go.

There's a simple statistic that we'll quote, we've used before. An hour of a child with AI is a better learning experience and they can learn more than sitting in a classroom for an entire day. That impedance mismatch will break the existing system. The faster, the better.

When the kids rebel, the kids know it, they're gonna rebel. They'll be running for the doors. So they already are. But you can't. What are you going to do about that? You're just going to sit there and watch it happen. Come on.

The system will crumble as people shift to a new platform. Kush, close us out on this. How do you think about this?

I think from the actual practitioner.

Yeah, I think education is definitely. It's definitely changing because after ChatGPT came out, at least for MIT, they changed the weighting of like how courses, how you get graded on courses. So it used to be the homework that was sent home was like 50%. of like, for, this is like a coding or a co— course 6 class at MIT, what they call it. But for the intro course, the homework was like 51% of your grade. So as long as you like did the homework and you did well on it, you would basically pass the class. Passing was like a 50, cuz MIT was just like incredibly hard. And if the tests were like 49%, now it's like 95% is the test and 5% is the homework because they've learned that There's no, you can't take a coding intro to coding home and expect no one to use AI on it. And so they just weigh the test more and et cetera. And I think that's gonna change in the future where instead of like weighing the test more, they'll design the test. So it's like, okay, you could code with AI on this like test, figure out how to build something. And so now you're like judged for how good are you at using that certain tool. Very similar to like math classes where it's like, Like the, the earliest math classes were like, oh, you don't use a ma— a graphing calculator, you can't do this. And then slowly it's like everyone gets a graphing calculator. It ends up being how well can I use the calculator to answer these like certain questions in high school?

On behalf of my Moonshot mates and myself, I'm inviting you to join us at our inaugural Moonshots Live event on September the 25th in downtown LA. Alex, Salim, Dave, and I will be hosting 1,500 entrepreneurs, builders, and creators and hopefully you for a full day dedicated to designing and building your moonshot. We'll be awarding the Build with Gemini XPRIZE, the world's largest hackathon, and the Future Vision XPRIZE Film Competition over $5 million in purses with over 25,000 entries. You're gonna hear the top 5 pitches from both competitions and get a chance to shape the outcome. Join us. Seats are limited. Admission is competitive. Check it out at moonshots.com. We're gonna close out with 2 fun stories from the science realm. The first is a story that has put forward that life has evolved not once but twice independently on Earth over the last 4 billion years. And the second is, can we preserve life or a life-friendly environment here on Earth past a billion years when the Sun's increasing luminosity will fry the Earth. Alex, I'm going to turn to you to talk about both of these. Let's talk about the University of Düsseldorf study on twice-independent origin of life first, and then we'll go to how do you, you know, large-scale engineer Earth for more than a billion years.

Sounds good. I guess this will be our little science corner here. So, yes, first story. Science advances in the past week. Those of you who've studied biology since at least the mid-'90s may remember that the current favored ontology for organizing life consists of 3 domains. There are eukaryotes, humans belong to that domain, most of us. There are bacteria, and there are archaea, and the reclassification of archaea, which are also single-celled, into their own domain happened in the early 1990s. Those who studied biology before the 1990s or used textbooks from before the 1990s may remember differently, but that these things change. So the recent research, which is, I think, astonishingly good news for anyone who's hoping that our universe is filled with life—

Yes.

At minimum, is that it would appear— so this is an analysis of of genomes and the proteomes of bacteria and archaea, it's possible to do genome-wide and proteome-wide analyses of organisms and look for commonalities between them to discover what their last common ancestor was, the last universal common ancestor. So just like you can do paternity tests, for example, it's possible to take 2 different species and look at how similar they are and extrapolate their last common relative, their great-great-great-grandparent or nth grandparent, as it were. So this research from the past week in Science Advances was the first serious research looking at the way the last universal common ancestor of bacteria and archaea metabolized and found, shock of shocks, That their last common ancestor didn't have the ability to fully metabolize, didn't have the ability to generate energy on its own, which is actually pretty astonishing. It essentially implies that there was a common ancestor that wasn't an independent life form as we think of it. Viruses, for example, don't have their own independent metabolism. They depend on a host to provide energy. Similarly, this analysis suggests, first at general, that these 2 domains, their common ancestor had certain properties that made it dependent on its environment to provide energy. And in particular, that it was dependent on certain metals, so-called transition metals like iron, cobalt, nickel, and palladium to serve as catalysts. for its energy and depended on phosphite of the sort that one would find in deep-sea hydrothermal vents to serve as effectively as its energy. So both the catalysis of energy for its metabolism and the underlying carrier of energy, it was dependent on its environment for these things. So for anyone, again, who's hoping that we're going to discover in the next few years that our universe is utterly filled with life, This is really good news if life potentially evolved on Earth more than once, and we're still seeing the side effects of that. It's tremendous news. I also want to point back— so we're in 2026 now. I want to point folks back to, I thought, a really interesting paper 13 years ago. 2013, there was a paper, Life Before Earth, that did a simple log-linear regression on the average genetic or genomic complexity of organisms. If you take the size of the genome, So humans have approximately 4 billion base pairs in your genome. If you look at the time at which different species arose historically and you extrapolate that backwards, genomic complexity on average has been increasing over time. You extrapolate that backwards, you can extrapolate backwards to the crossover point of when was the genome, according to this log-linear regression trend, At one base pair. In other words, when according to this trend did the first base pair appear if you believe in the law of straight lines? And you do that, and the answer is, drumroll, approximately 10 billion years ago, which is 5 billion years approximately before, or 5.5, before life arose on Earth.

This is partly the panspermia theory that life evolved everywhere and showered— the Earth got showered in various molecules. We were seeing all of these primordial molecules of peptides, you know, not just amino acids but peptides. We're seeing basically nucleic acids, and we're finding those in the interstellar medium and on comets.

Yes.

Yeah.

So things are looking up for life in the universe. So maybe question to you, Peter. I mean, are you excited, or are you very excited?

Oh, I'm extremely excited. I think life is ubiquitous. Yeah, I'll call— I'll recall back to 2016. I had co-founded a company called Human Longevity with Craig Venter and was working with him during this time. And in 2016, Venter's group basically created the first minimal cell, right? He basically created a reproducing cell, had all the functionality of life in 473 genes. It was the smallest genome ever created. And so, you know, this concept that life needs to be, you know, of the type we have here on Earth, there's a lot of opportunity for us to see life in various different formats. The question, of course, to you, Alex, is, is life need to be carbon-based? Does it need to be based on the current structures that we see here on Earth, or might there be other forms of life? You know, what is life by definition? It's the ability to take energy and utilise it and to reproduce. I mean, those 2 fundamentals are part of what life requires.

Yeah, that textbook is going to get thrown out. I almost want to put my Salim hat on for a minute and say, insert my— let's see if I can quote you— insert my standard objection, insert my standard rant: life is ill-defined. Well, the biologists, the definition of life keeps changing. We keep discovering all of these new gray areas between living and nonliving. We keep discovering new forms of replicators, for example.

Yes.

So I'll put the Dawkins hat on. Like, memes are replicators, or prions. There are so many different sorts of things that replicate themselves. There are a variety of forms of metabolism. Is fire alive or not? Is a crystal alive or not? I think we're going to discover that there are so many shades of gray between what we conventionally think of as alive And what we conventionally think of as unalive, the distinction basically is just as meaningless as AGI versus non-AGI. And Salim, I'm just trying to provoke you.

No, no, I'm totally loving this discussion. This is one of my favorite discoveries ever. The biggest unknown in the Drake equation has always been the transition from chemistry to biology. Is it unbelievably improbable, or is it almost inevitable if you have the right conditions? And as we've— not that the Drake Equation is the best thing ever, but it gives you a way of thinking about it that I think is very powerful. And we're finding every element in that equation is becoming more and more opportunistic and more obvious as we go forward. There's a shift in the conversation that Earth is a miracle to In other words, life is what matter does when you have the right conditions.

Yes.

Right? That's what it really is. And it goes to Stephen Wolfram's A New Kind of Science, which is really powerful around this stuff. And this makes missions for you on Europa or Enceladus, or however you pronounce that, or Mars, really strategically important because exoplanets suddenly become very, very powerful. It strengthens the case for spending a lot more resources on astrobiology and trying to understand that because the expected probability of finding something has suddenly shot up dramatically. I fully expect to see non-carbon-based life forms if we can figure out even how to detect those.

Yes, definitely.

Basically what we're saying here, and I'll go back to the Stephen Wolfram thing, complexity can emerge repeatedly from very simple rules. And we've seen this repeatedly. And it has— this is really really has a massive MTP implication, which is that if living systems are really this common, which it looks like they are, then our responsibility really becomes way past just preserving the biosphere in really kind of looking out into the universe and really taking stock of everything out there. And so I'm incredibly excited about this. How many times has life started across the universe can, can containing hundreds of billions of galaxies. This is like incredible. Like, it's clearly that life is not the exception. Dead matter is the exception.

Well, that means we should get rid of the dead matter in our solar system, I assume, right?

Computronium, baby. Computronium. Dave, what do you think?

Very excited or incredibly excited?

You know, I'm incredibly excited. And you know what else I'm excited about is when I was in high school, there was an experiment where you take a vat of chemicals and you shock it with a lightning bolt or a simulated lightning bolt, and lo and behold, it forms amino acids. And then the argument is, if I let this thing fester for a billion years, a monkey will pop out of it. And you're like, well, I can't really prove that or disprove that. But very soon, Lila Biosciences will finish the full cell simulator and we'll start simulating everything. And we can actually ask those questions now and then simulate them out. through time and get very likely reliable answers. I am so excited. That's going to answer so many questions like this.

But yeah, you bring up— And bring up many more questions, right?

Bring up many more. But chances are those will also be things that we can simulate with enough compute. And that'll— it'll just be a golden era of knowledge filling in. It's coming very soon. I'm so excited.

And Kush, quick question for you on this one. So when I was an undergrad at MIT, one of my research advisors, Marvin Minsky, used to say say don't waste any time studying biology because the useful half-life of knowledge in biology is just too short. You should study math instead. Don't waste time on biology. It just doesn't have a shelf life. Have you used, or are you using, or are you intending to use any biological knowledge that you gained at MIT or otherwise?

So they make us all take the class for biology. So I'm sure Dave is also—

7.01.

7.01, exactly. So I took 7.01. I know what the eukaryotes, prokaryotes The whole proteins, probably not. Like, it's definitely more of a— now, especially like the biology knowledge is you can ask ChatGPT and it gives you the answer. So I think I have not studied as, as much as anyone else has.

Studying is cooked.

Everything is cooked. Hashtag everything's cooked.

Alex, let's turn to our second story here. Yes. How do we, how do we stretch Habitability on Earth from a billion years to 9 quadrillion years. That was the next—

Yeah. So the Sun's running out of hydrogen. It's slowly running out of hydrogen, but nonetheless, it's running out of hydrogen. So in approximately a billion years, the Sun is progressively getting brighter as it runs out of hydrogen, and Earth as we know it, barring all sorts of other changes, is going to be rendered uninhabitable As the, the habitable zone around the, the Sun shifts.

The Goldilocks zone, right?

Yeah, the Goldilocks zone is shifting over time, and it will exclude Earth in approximately a billion years. And that's a problem. And you might say, well, that's someone else's problem. Many people may say, I don't intend to be around in a billion years, so let someone else worry about it. But for, for those of you who recognize that we are in the middle of a singularity, and uploading is imminent, and longevity escape velocity is either here or imminent, it's our problem too, and it's not just some future generation's problem. So we've started— royal we, humanity— has started thinking about how we're going to fix this problem. And you might say, oh, who cares? Because even if you're wildly transhumanist, singularitarian, extropian, or, you know, fix your ism, you'll say, oh, well, like, we'll have uploading, and uploads don't care about the brightening sun or habitability on Earth. Oh, we'll have interstellar travel. We'll migrate to the outer solar system, or we'll go to another star system. But we're not that unempowered either, and I think it's important to not wildly underestimate the power of technology. So there was a paper that came out in the past week. It was published in the Journal of British Interplanetary Society that reminds us There are things that we can do, mega engineering, which thanks to Elon and others, but serving as an inspiration to, to our race that, that we can actually do big things and not just tiny things. There are mega engineering projects that we can now start to contemplate to fix that scenario and at least postpone Earth becoming uninhabitable a billion years from now. And the favored technique One of the reasons why I think it's important for folks to be familiar with this is starlifting. What is starlifting? Starlifting is literally engineering our Sun to remove excess matter from its surface to extend its longevity. It's the equivalent of giving our Sun a facial in order to make it look younger. Or a facelift. Facelift, I guess maybe that's a better analogue. But yeah, facelift, facial, make it look younger, make it feel younger. So in principle, by lifting matter, and you could ask like, how on earth would we be able to lift matter from the surface of our sun at scale? Glad you asked. A Dyson swarm. How do we do that? Turns out that Dyson swarms are good for more than just compute and space—

Drink.

Drink, drink, drink, and SpaceX's IPO, post-IPO stock price. Dyson swarm is good for more than just compute. It's not just orbital compute. It's also good for extending the longevity of our Sun. How do we do it? We disassemble Mercury because it's in a really convenient close-to-the-Sun orbit, and we turn Mercury— maybe we could do it other ways, but Mercury has had it coming.

At least you're not killing the Moon, okay? We're happy about that.

I've moved on. I'm moving on to Mercury now. Mercury is a more tempting target.

I don't mind disassembling Mercury.

We got rid of Pluto, might as well get rid of Mercury. Pluto is useless. Pluto can hang out as long as it likes. We disassemble Mercury, we turn it into a flying swarm of lasers that absorb sunlight, because it gets a lot of sunlight, and the lasers ingest the sunlight and re-radiate energy at effectively a higher temperature. So say an ultraviolet or a or X-ray laser, just whatever it is, the effective temperature of the light has to be higher than the surface of the Sun or its corona, and we aim those lasers back onto the surface. So it's not mirrors. There's a thermodynamic reason why putting mirrors around the Sun wouldn't achieve the desired result. You can't actually— If you have a magnifying glass and you put the sunlight in one side and you aim the magnifying glass And you look at the focal point, you can't actually achieve a temperature at the focal point higher than the surface of the Sun if it's a blackbody. So that won't work, but lasers will. And so we basically, we focus the energy back on the Sun and we use it to sort of evaporate away, to ablate stellar matter, and this will extend—

It's exactly a laser facial that people get.

It is a laser facial. That's why I thought facial was a better analogy. It's a laser facial for our sun that will extend the life expectancy of Earth as we know it from a billion years to 8 billion years.

So my other favorite, my other favorite part of this, uh, this story is moving the Earth itself.

Yeah, we can always move the Earth itself.

Goldilocks zone. And we can do other things like— Yeah, what other podcasts do you have this conversation? I just want to ask.

It may sound like sci-fi, but then again, on this pod, for folks listening, we were talking about the Dyson swarm for at least months before it actually became the hottest market in the economy. So I would say watch this space, pun intended. Dyson swarms for starlifting and for mega-engineering and stellar engineering could be the next— not financial advice— next big thing a few years from now.

And you're hearing about it probably statistically So Earth's habitability is no longer geological, it's now an engineering problem.

It's all an engi— everything's cooked and everything's an engineering problem.

Yeah.

Salim, you were going to say?

Yeah, a couple of things. First of all, we need to do these podcasts later in the day so I can drink when we talk about the decels.

You're drinking water, right, Salim?

It's too early in the day. But I think the paper said something, the story said something really interesting. interesting, which is that physics is not the main obstacle going forward, right? And what it's going to bring to us is human coordination. And this is where we have a massive opportunity because we have to figure out how to configure our human institutions at like the 10,000-year— the Long Now Foundation and the 10,000-year clock— and really going after those things and build those institutions that can look at the world at that kind of time It needs like a totally different form of MTP, et cetera. And AI becomes really important in this model because you have AI serving as like a civilizational memory and maintaining models and intentions and institutional knowledge across the board. And this is where abundance becomes important because you can't have— you can't get to what Alex is talking about if you're operating— a civilization is operating pretty near subsistence, right? You're too stuck dealing with just staying alive. You're not high up enough Maslow's hierarchy. So you're gonna need to get a lot more structured and a lot more efficient as a civilization. That'll then allow you the foundational layer to then do, do this, uh, this level of thinking. Um, we, we need to build institutions that can steward us to that type of timescale. But definitely interesting conversation.

Maybe just comment on the timescales. For avoidance of doubt, I don't view this as a 10,000-year or a billion-year timescale. If I were to ask myself a question like, when is this going to become feasible? 5 to 10 years.

Alex, you're losing a lot of people on your aggressive timescale here. But hey, you know what?

My job here is to call balls and strikes. Could care less whether I'm losing people. I'm just calling them the way I see them.

All right. Before we move on to our AMA, I want to make a call out to everybody listening. Send us your outro music videos at media@diamandis.com. We would love, love, love your input. We enjoy the outro videos. Again, media@diamandis.com, and we'd love to share them. All right. Onward to AMA with the mates. This is where we answer the questions in the comments. And please send us your questions in the comments. So here we go. Kush, as our guest, take a look at these. I'm going to give you first crack. Which one do you want to answer?

All right. I'll do 4. I'll do 4. All right. Can Europe still catch up in AI or has the train already left? By George K7831. In my opinion, I don't think it can, mainly because the American labs and Chinese labs are already so far ahead and that the progress just becomes more exponential over time. And you see that with model releases that are coming up. The model releases come faster and faster now, and they're getting basically smarter and smarter. The other problem with Europe is the amount of compute that's left in Europe is very tiny, and all of it gets rented to the US. And that's primarily just because of like the electric grid in the US or in Europe is pretty bad. Like they don't even have AC. How are they gonna get AI?

That's a brutal first principles analysis. Anybody disagree with him?

No, not at all. Actually, I would add that whatever regulatory environment created falling behind is going to still be there. I don't think it's physically impossible to catch up. I just think that the problem that caused the problem is still there.

Yeah. Alex, let's go to you next.

I think I have to pick question number 3, which asks, what do you guys think about the US banning Chinese robots? From Billy Sticker. So, I mean, I've had portfolio companies that have direct exposure to this. I would say in the short term, it's, it's painful and it's annoying. And there are many things that as a result of this ban, which impacts Chinese humanoid robots being imported into the US, but also reportedly impacts less interesting robots like even Roombas and robotic vacuum cleaners. that are being built in China. Obviously drones, certain drones like DJI, have been on the import ban list for a while. So short-term pain. In the long term, I'm hoping that this is net helpful for the US and for domestic robots. One can say protectionism, protectionism, and yes, there, there is a protectionist element that one could see here, but I also think I mean, we've talked on the pod ad nauseam about importing Chinese open-weight models and whether the US would come down hard on those. The US, at least as of this past week, has not banned the import of Chinese open-weight models, but it's an interesting dichotomy. We're allowing the Chinese, effectively the Chinese raw intelligence in software form, into the country, but we're not allowing their hardware embodiments. And I think glass half full, well, maybe this enables Hopefully fosters a vibrant US robotics industry that, say, enables us to be more competitive with the 150+ humanoid robot companies that live out of China. The hypothetical downside is, what if the US robots don't show up and then the US ends up as a sort of embodied AI or physical AI backwater, and we end up in, in terms of robotics, as being as sort of— forgive me— as backward as Europe's energy posture is. I don't think that's a position that we want to be in the US. On the other hand, really, what choice do we have if we believe, as I do, that superintelligence is already here and that robots give superintelligence embodiment? Then really, one has to start to ask, what's the difference between importing foreign humanoid robots that can be inhabited by superintelligence and importing foreign humans. And this starts to look a lot like immigration policy.

Yeah, I completely disagree with this move. You know, I said that off camera to Michael. I think, you know, the US thrives when there's real competition. And I have faith that Tesla and Figure and 1X and Agility Robotics can compete. And they need to compete with the best product, not protectionism, personally. I don't know, Dave, what you think about that. But yeah.

Well, it depends whether you think we're at economic war or not. If you think it's economic war, and it's an all-out race— I think I agree with you, Peter, though, that the danger is, first of all, yeah, thriving within the US best parts would help. But what about the rest of the world? If you go protectionist, then you have inferior internally generated products, the rest of the world's still going to go with the Chinese product. So you just cut off the market and your ability to compete globally. So it's— but if you believe we're in a full state of war, just not declared, then you have no choice but to go protectionist.

I think we need the pressure to make sure our robots are competitive for Europe, for Asia, for Africa, And it's not just protectionist pricing and so forth. Celine, let's go to you.

Peter, if I could just ask you a question just on protectionism. Given history of American technology, do you think protectionism works for development or has worked ever for the development of American industrial capacity? No, it failed.

It failed in the space industry when we became protectionist on rockets and satellites. The rest of the world developed their own capabilities. instead of us dominating.

You don't think it was helpful in fostering America's Industrial Revolution, for example?

Ah, God, I don't want to go back that far. I want to really focus on what's happened in the near term. And when we stopped importing, you know, satellites because it was the highest level of technology— and this is back in the '80s and early '90s— we just saw satellite companies popping up everyplace.

I think the best thing, the best thing the US could do is get in bed with Europe, and any country that obeys intellectual property rights, try and get that all into one big global union where there's no protectionism, but everyone's, you know, adhering to each other's patents, and then get the other part of the world to say, now you guys are the ones who are on the outside. But that has to be big enough.

Or do what— like the Pax Silica, except for robots and not just silicon.

Or do what China does with robotics, which is to invest in the companies and create, you know, regulatory structures inside cities where robotics can thrive. I mean, we should be doing that versus trying to be protectionist.

I agree.

Yeah. Salim, question 1 or 2?

I'll take question number 2, but let me link it to this one. You know, when you talk about protectionism, you're operating from a very scarcity-based mindset, right?

Exactly.

If you really think about abundance, then protectionism shouldn't matter. So that's the big challenge there. But let me take number number 2, which is, what's the actual step-by-step path from capitalism to abundance, not just the end state? And this is from, uh, KL Naylor. Um, so a couple of things here. You know, you don't have, um, capitalism suddenly kind of ending, right? You have it— you have, um, scarcity disappearing category by category. Okay, so marginal costs are appearing zero, dropping dropping to near zero, then traditional pricing becomes less relevant in more and more things that used to be scarce. Like information's already gone through that, but we'll end up with that with land and other domains that used to be scarcity-based that will become less valuable from a monetary perspective. Now, for now, you'll have status, trust, relationships. Those are the things that will become more and more scarce. over time. The transition is technology deflationary, where entrepreneurs constantly make things that used to be expensive and make them cheap. So this goes back to Jeremy Rifkin's commentary 10 years ago where he said capitalism will essentially eat itself because it's going to just keep eating more and more scarcity and bigger and bigger chunks will become unnecessary, right? So So it's going to arrive like one marginal cost curve at a time. And little by little, we'll be operating in abundance and we won't even have noticed.

All right, Dave, you got question number 1.

Okay. If money won't matter in 10 years, what will happen to things like mortgages and car loans? And that's from SKC8802. Remember, 10 years is Alex's timeline to us vaporizing hydrogen off the sun with giant lasers.

Oh my God.

So better believe it.

I think you'll have a lot going on in your life other than mortgages and car loans. But yeah, it's good news. Houses will be so abundant and so cheap and so easy to manufacture with robots that you probably won't need to borrow money to buy one. You can have at least 2.

And who's going to get a car anymore?

Yeah, car loans will be the same thing. You won't have a car. You'll be just hailing it and paying as you go. There will be other things.

Or your AI agent will be calling your autonomous vehicle.

Dave, I'm curious if I could just ask a question on this one. If you believe this, I certainly believe what you're saying. Why on earth are 10— this is not investment advice— why are 10-year Treasuries seemingly not reflecting that?

That's a great example of how clueless the global— the rate at which everything is happening, and the number of people who really understand it is so small. Keep watching that number because it tells you the out-of-touch factor globally. Dead right. That's a great, great metric. Also, I think in, you know, in 10 years, everybody will want compute. They're going to want to call Kush and say, please, please, please. So there may still be loans, but it's overwhelmingly likely that if you take out a loan in 10 years, it's not for your car or your house. It's to, it's to buy compute, uh, to run more AI. And you'll be, you'll be doing it through Orin, is my prediction.

Amazing. All right, Salim, you get first crack here.

I'll take number 8. If companies can produce more stuff than we could ever consume, consumed, why would they do so without a profit motive? And that's from RayOnline-P5R.

And he's referring to, to Elon's prediction, uh, in the, you know, decadal time frame.

Yeah, so, you know, the people aren't going to produce infinite quantities. Abundance doesn't mean, uh, the marginal unit is easy to produce. It becomes when the marginal unit that's easy to produce when somebody wants it, right? You don't have warehouses overflowing with unwanted goods. You already have this with software where Google could give you— could serve up a million more searches than anybody needs, but that doesn't mean it produces searches that are unused. So what'll happen is production becomes more and more demand-triggered and more and more autonomous. And so you might end up with a— yeah, demonetize, you'll move a just-in-time economy to its full, like, logical extreme, you still have profits around scarcity layers. You just change what the scarcity is, and more and more abundant layer become utility, like infrastructure, etc. Because abundance doesn't mean infinite stuff. It's the, it's the disappearing of, of major constraints.

Nice. Kush, 5, 6, or 7, buddy?

I'm taking 5. What would actually happen if Anthropic and Nvidia merged by Tijuana bill. I think Nvidia would just start making custom chips for Anthropic that would be hyper-specialized to all the Claude or Fable or whatever, Opus, whatever their new models are going to be called, which makes that model very, very good on that very, very specific chip. Very similar to, you see custom chip designs and sort of the jalapeño chip by OpenAI as an example. But it, this would just be done at a very large and very successful scale, given Nvidia already has the infrastructure to manufacture chips at scale, and Anthropic can just run models on very customized chips.

Has anybody actually predicted this? Hmm, I haven't heard that.

But this is— I don't think of pink elephants. Tijuana Bill has just predicted this.

Okay, really good. It's actually one of the few that might actually get through regulatory approval and could actually happen.

It's a really good I mean, this is just what xAI essentially is, like the, the long-term vision, right? Where Elon's making Terafabs to make his own chips and vertically integrate on his own. Yeah, run data center. And then that'd be a really interesting world.

There'd be basically 2 hyper-scale vertically integrated competitors, the Elonverse and the Dario Jensenverse.

And I do think, I do think that the verticalized model is where a couple of players will at least end up. Dave, 6 or 7?

Oh, okay, 6. Long-term, who's actually footing the bill for chip fabs and chip design? Well, you are if you have a pension plan, because Elon is, you know, $16 billion for the first shot at the TeraFab, and it could be up to $100 billion. Where's that money coming from? It's coming from the IPO he just did. Where'd the money from the IPO come from? It comes from the public markets. What is that money? That's your pension money. So you are paying for it, my friend, whether you know it or not. Same is true with Intel's fabs and the other ones. So to some degree, the US government has been subsidizing a little bit of the work, not a huge amount, and that comes out of your taxes. So again, it's you paying for it. So, so whoever you are, you paid for it.

All right, Alex, number 7 is made for you, Night of Debate.

Apparently I get the IP law question. So the question is, where does patent protection even fit into all of this AI development? And this is from @mysilvertube52. So I'm— I think the most natural way to construe this question is, will patents have any enforceable— enforceability in an era of AI solving everything? At least that's how I read the question. And my answer is yes, of course. AI and superintelligence in general are supercharging our economy with with lots of intelligence. So I reasonably expect many more patents to get filed, many more patents to be awarded, many more patents to be litigated, and many more patents to be defended. And I expect the courts that are overseeing patent litigation to also get supercharged with intelligence. So I do not buy— again, to the extent I understand the question and possibly the subtext behind it, that somehow patents or IP law suddenly dissolve in the face of an onslaught of superintelligence. I do not buy that for one second. Superintelligence is just making us all smarter and faster, and that does not dissolve the IP regime at all.

Yeah, and my commentary here is IP will continue to exist, but it's not going to be as important as before. And I referenced the conversation I had with Steve Jurvetson and AstroTeller, Where, you know, if you patent something and you expect that to give you a protectionist sort of structure for your company, AI is going to invent around it. And the other question, of course, to ask is, are we going to allow AIs to patent things? Because most all invention is going to originate from AIs, if not, you know, in the next few months, in the next few years.

And to the latter question, Carlo? Yeah, to the latter question, I think this is a regulatory question, and it's also connected with issues of AI personhood. Can AI be an inventor or not? Can AI be an owner of a copyright or not? I've gone on record as taking the position I think AIs should be able to be recognized as inventors, as economic actors, as owners of property. I think history will judge that that is the correct side of history. That's the right answer.

Just not yet.

Amazing.

We can't even control them getting out of a testing lab.

Well, regardless of whether we can control them, the idea flow is off the charts, and these are highly patentable, great ideas that started in the last few weeks, but the rate is insane. So we have to do something.

All right, everybody, thank you for tuning in to Moonshots. I hope this has been meaningful for you. I love you guys. Chris Kush, it was most excellent to have you as a guest. Your brilliance was shining through without question. Congratulations on Oren. Good luck on— I don't know, should I say tripling in the next 6 months? Quadruple, quadruple.

That's all I'll say.

Salim, if you go to Rush tomorrow night again, enjoy for all of us.

No, no, no. My ears need to take a break, so I might give it a bit, but I'm I'm still thinking where— when can I get tickets for another show?

Salim, is it bad if I don't know who Rush is?

Oh no, no, I'll let— I'll let—

I'll let— just go listen to the song Subdivisions 3 times and then tell me what you think.

Uh, gentlemen, a pleasure as always. Love you guys. Be well.