OpenAI 收购 OpenClaw、成本坍缩400倍，以及印度为何赢得人才战争｜EP #231

Hey, we got strangers through here.

Peter H. Diamandis

You know why we're overlapping? Because the old saying is, “AI is easy, AV is hard.” We're just trying to get our damn AV working. I'm in Germany.

Yeah, yeah.

Peter H. Diamandis

It's midnight here.

No.

What are you doing in Germany?

Peter H. Diamandis

I'm your person of—

Hold on. I gotta figure this out, guys.

Peter H. Diamandis

Longevity.

I gotta share my screen.

Peter H. Diamandis

Salim, were you AV-qualified in elementary school? I mean, did you go through that program?

I was not AV-qualified.

Peter H. Diamandis

Sure.

I mean, it's going to be a miracle to get this working, then.

It says, “Also share tab audio.” Is that what you want, Donna?

Peter H. Diamandis

Yeah, probably. Try it.

Yeah, try it.

Peter H. Diamandis

What could possibly go wrong?

Actually, go to the outro music, crank it, and let's see if we can—

Peter H. Diamandis

There, I found it.

—rock to it.

Peter H. Diamandis

Dave, did you go through AV certification when you were in school?

Absolutely not. It was so uncool. I really wanted to, but—

Peter H. Diamandis

All right. Now, Salim, go to the beginning of the deck. Wait, wait. Preview it backwards.

Boom.

Peter H. Diamandis

All right. Awesome.

Oh, you gotta try and play a video so we can—

Peter H. Diamandis

I should get half production credit for this episode. All right, now.

Go to the beginning of the deck. Wait, wait.

Preview it backwards.

Boom.

Peter H. Diamandis

All right. Awesome.

Oh, you gotta try and play a video so we can—

Peter H. Diamandis

I should get half production credit for this episode. All right, now. Am I in a time loop? Wait, wait.

Yeah. Isn't that weird?

Peter H. Diamandis

Preview it backwards.

Boom.

Peter H. Diamandis

All right. Awesome.

Oh, you gotta try and play a video so we can—

Peter H. Diamandis

Credit for this episode. All right, now. Am I in a time loop? Wait, wait.

Yeah. Isn't that weird?

I think that was because Nick was in the room. All right. Are we good?

Peter H. Diamandis

We're good. We're live. All right.

Yeah. All right.

Peter H. Diamandis

Live from Saturday Night Live.

Hi, everyone. Welcome to the raw backstage chaos that we have here at Moonshots.

Peter H. Diamandis

All right, everybody. Good morning, good afternoon, good evening, and welcome to another episode of WTF Just Happened in Tech. I'm here with Dave Blundin, Salim Ismail, and Dr. Alexander Wissner-Gross here in Stuttgart, Germany. We want to get you future-ready. We have an incredible episode talking about multibots, of course, the race between all the hyperscalers, and a dive into energy and data centers. Let's jump in. The supersonic tsunami, the singularity is now.

It is midnight in Stuttgart.

Peter H. Diamandis

Yes.

You can't just drop that and not tell us why you're there.

Peter H. Diamandis

I'm here for some longevity treatments. I'll tell you about it sometime later.

Oh, okay. Okay.

Peter H. Diamandis

All right, Salim, onwards.

The first—I did a pilgrimage to Stuttgart just to go visit the Porsche Museum once. So go there if you can.

Peter H. Diamandis

I should go. I should go while I'm here.

Yeah.

Peter H. Diamandis

All right. Let's jump in with Gemini, OpenAI, and xAI. I think this one deserves going to our resident benchmark brainiac. That's you, Alex.

That's not me.

In living color, no less.

Peter H. Diamandis

Yeah.

So let's take this seriatim. Sonnet 4.6 is a very interesting release. I think there are several interesting points. One, I think Anthropic has really been pioneering one edge of, call it, the scaling phase space, where they keep the prices of the model tiers the same but increase the capabilities. So Sonnet 4.6 is about the same price per token as Sonnet 4.5, but has increased capabilities. I'll talk about that in 1 second.

Whereas, say, OpenAI is reducing the cost per token while keeping capabilities more or less constant through distillation and other processes for evolution. That's interesting point 1. Let's actually talk about the progress on the benchmarks, the evals. I think it is nothing short of astonishing.

If you look at the GDPval benchmark—again, gross domestic product eval—that OpenAI launched, Anthropic is leading. Anthropic, in the form of Sonnet 4.6, not even Opus 4.6, now has state-of-the-art performance on GDPval and 1 other eval that is intended to encapsulate knowledge work.

I've said on the pod in the past, knowledge work is cooked—cooked twice for emphasis, usually in reference to GDPval—and we're seeing it get even more cooked, charbroiled at this point, thanks to Sonnet 4.6.

I also think, taking a step back, computer use is becoming a killer app for many of these models, and Sonnet 4.6 has state-of-the-art performance on a handful of computer-use benchmarks. Anyone who's been using Opus 4.6—as has been the case for me—for the past week and a half or so, for any tasks, I think Anthropic's thesis that focusing on software engineering and code generation as a critical path to recursive self-improvement, versus, maybe charitably, getting distracted by image and video generation and all of these other modalities, seems like it's working.

I can accomplish tasks that seem borderline magical with Opus 4.6.

Peter H. Diamandis

Alex.

Yeah.

Peter H. Diamandis

I gotta ask here because I'm channeling one of my kids, who goes, “Dad, every week it's like 4-point this and 4-point that. It's better, better and better.” Yeah, we get it. It's getting faster, better, and cheaper.

Aren't the models at this point just optimizing for the benchmarks? At the end of the day, this is a gradual increase up and to the right, or down and to the left, whatever you want. I'm just trying to understand: other than, “Yep, newsflash, it's faster and cheaper this week than last week.”

Yeah, it is so opposite of what that implies.

Peter H. Diamandis

Oh gosh, we are so spoiled.

I know. It is so opposite of what that implies.

Peter H. Diamandis

But I'm trying to channel a person—you know, our viewers listening and watching this—who aren't big—

Yeah, yeah, yeah. No, I totally get it.

Peter H. Diamandis

who aren't benchmarked pre-amps.

You know, I mentioned a couple of podcasts ago that when these curves get close to 100%, they look like they're showing diminishing returns, but in reality, their capabilities and their ability to change the world are exponentially going the other direction.

I think that's what you're getting at here, because you see a little tick up in these numbers and you're like, “Oh, so what?” But then when you actually use it day to day, it's like—boom! Oh my God. Just the last 2 weeks of change is mind-blowing.

Also, when they tick up the numbers in the versions, they're actually improving the chain-of-thought reasoning on top of that quietly in the background, without ticking up the numbers. Day over day, I'm noticing improvements that are mind-blowing, that aren't actually showing up in the dot releases and the new benchmarks. Sorry, Alex, go ahead and answer the question now. I just wanted to jump on it.

I was going to taunt Peter a little bit. We are so spoiled to even be contemplating asking that question. It would be like Moonshots, our namesake: okay, so we have hotels on the Moon now and vacations to the Moon, and maybe you can travel there once per human lifetime unaided versus 0 times. Oh, but yeah, we've had airplanes for a while.

We are so spoiled to even be asking the question. If you live day by day with, say, Claude Opus 4.5 versus 4.6, qualitatively, it is an enormous change forward. It can solve—

Peter H. Diamandis

Sure, but I—

—hard problems that—yeah.

Peter H. Diamandis

My bet is most of our viewers probably don't live with it day by day and aren't using it at the maximum extreme. I think one of the things that you and I talked about in SolveEverything.org is that we've put the initial frame in place, and we're heading toward ASI, whatever you want to call it.

We're going to be reporting this every week, this leapfrogging between models: 100× faster, 100× cheaper. I do think what you said that's interesting is the 2 different strategies here, right? One, that Anthropic is holding—you said cost and increasing speed, while OpenAI is dropping cost and maintaining speed.

Or performance, not speed, but yes.

Peter H. Diamandis

Performance. Okay, performance. I think that's a fascinating strategy, right? Because—

Yep.

Peter H. Diamandis

—we're going to get to it in a little bit, because OpenAI, I think, is going for a land grab—affording a land grab on global consumers, hitting 900 million, and soon in India, adding hundreds of millions.

So the price is the most important thing for grabbing the consumer, while I think strategically here, Anthropic is focused on enterprise business, and performance is far more important for the enterprise. Would you agree with that?

And the margins. We’ve seen this business pattern play out over and over again historically. Call it—again, this is very heuristic—but call it Anthropic is to OpenAI as Apple is to Google or something like that, at least in the mobile space. Maybe iOS is to Android.

There have been many, many instances of this business pattern of emphasizing quality and margins on the one hand at a constant price versus emphasizing ubiquity and ultra-low cost at the other end. This has played out over and over again many times. It’s the same old story.

But I do think Anthropic—if I had to say which set of models, which model family, is the closest to embodying the singularity and recursive self-improvement right now, today, since we’re live on February 17, 2026—it’s the Anthropic family. It’s not OpenAI—

Peter H. Diamandis

All right.

—or Google.

Peter H. Diamandis

Kudos to Dario. We’ll get to Google in a little bit. Let’s talk about xAI launching Grok 4.2 beta. I love these names, right?

Our live-cast viewers here are saying it’s poop.

Peter H. Diamandis

What’s poop?

Yeah, 4.2. Have you guys tried it? It’s poop. That’s what they’re saying.

So the risk with the Grok family—I had access—is always, or I should say, the accusations are always: Is it benchmark gaming? Peter, you were asking about benchmark gaming earlier.

Peter H. Diamandis

Yes. Yes.

Historically—

Peter H. Diamandis

It’s teaching to the test.

Right. Historically, some of the earlier Grok models have felt very benchmarked. It’s only been available for a few hours in beta form, so I haven’t had an opportunity to do thorough testing.

What I think is interesting about Grok—I assume we’re supposed to pronounce it 4.20, one of Elon’s favorites—

Peter H. Diamandis

Yes, of course. Of—

Yeah, right.

Peter H. Diamandis

Ta-da.

Obviously.

Peter H. Diamandis

Either that or 4.69. Yes.

But what’s interesting to me, at least, is that this is the first major frontier model release that I’ve seen that’s launched with a team of agents by default rather than a single agent. OpenAI has a team under Noam Brown that’s been looking at agents for a while. Every frontier lab at this point has multi-agent teams built in, in some form, somewhere in the family.

But I think it’s a really interesting strategy to build in, by default, a multi-agent team. There are lots of potential reasons why a multi-agent team versus just a single agent running serially might be interesting. You can do things in parallel and explore possibilities in parallel with multiple agents.

But this may be the direction of the future. Just like we saw the megahertz and then gigahertz race plateau out due to the end of Dennard scaling with microprocessors, and then we saw a transition from clock speeds to multiple core counts, maybe we’re about to see something like this happen with frontier models.

Maybe capabilities—again, this is very speculative—maybe along a certain dimension of scaling. Obviously, pre-training has sort of transitioned to reasoning scaling and other forms of scaling. Maybe we’re seeing the dawn of multi-agent teaming scaling, where you get better capabilities by scaling the number of agents in parallel. Not a problem.

Alex, the viewers all think it’s poop here, but I haven’t actually tried it. I use Claude all the time and the other models every day. I haven’t felt any great compulsion to try 4.2 because Elon told us 5 is coming in March anyway.

But my understanding was that 5 is a massive, massive expansion in every way—in training set size, in parameter count, everything. I never thought about anything meaningful between here and there. I was just waiting for that.

But do you know any more detail on what this thing is, and should the viewers be trying it or not?

I think it’s worth, in general, trying every frontier model from, call it, the top 4 or 5 labs that come out. If you’re doing stuff in AI, if you’re sufficiently abstracted from the bleeding edge of the frontier, I think you should still try it just to be familiar with the raw capabilities.

But based on what I’ve seen thus far, Grok 4.20—or however we pronounce it—

Peter H. Diamandis

Grok 420. It’s going to be 420.

Grok 420.

Peter H. Diamandis

Yeah.

It’s not the bleeding edge that’s pushing forward capabilities, as far as I can tell at this point in time. But it is—

Peter H. Diamandis

Salim, let’s go to Google Cloud Next and some more of the bleeding edge.

Hold on. Hold on. Switching windows here. I’ll tell you on that.

Moonshot going to the next slide.

There we go. All right.

Peter H. Diamandis

Gemini 3 Deep Think. I love these names. I think the naming protocols for all of these models have got to be rethought.

But I think the one benchmark everybody keeps tracking, at least I do, is Humanity’s Last Exam, just for fun because of the existential nature of it. Yes, it is our last exam. And we see here that Gemini 3 Deep Think hits 48.4. But most importantly—and this is, I guess, the OpenAI playbook—400-fold cost reduction. That’s extraordinary.

It is. And also, to the point about naming, this isn’t even, I think, the first Gemini 3 Deep Think. This is the second Gemini 3 Deep Think—or the new and updated Gemini 3 Deep Think. So, agreed that the naming could use some work, but the new Gemini 3 Deep Think is remarkable.

If you just look again at the evals, there had been percolating for a while the so-called internal model, the one that beat the International Mathematical Olympiad and was achieving breakthrough performance at other high school science competitions. This is the model that achieves gold-level performance at the International Physics Olympiad, the International Mathematical Olympiad, and the International Chemistry Olympiad. On Codeforces, I think the statistic is there are only 7 humans now on Earth who can beat this model on competitive programming.

Peter H. Diamandis

Mm-hmm.

So, Peter, you and I spoke in Solve Everything about what we called a solution wavefront propagating outward from—

Peter H. Diamandis

Yes.

—math and coding to different fields. This is the beginning of the wavefront. This is the infection, the contagion, spreading from coding and math to physics and chemistry.

It also does 3D design. I keep trying to persuade it to do 3D design, unsuccessfully. It keeps producing intermediate products. But this feels like the kickoff, the starting gun, for the solution wavefront that we spoke about.

Peter H. Diamandis

And we’ll see that. I think the visual image that I have, that I want everybody listening to think about, is this: When you have this kind of weapon of superintelligence, where do you deploy it? Where do you aim it? What are you measuring, and what is your massive transformative purpose? What is the challenge you want solved?

Because we’re going to have this kind of capacity, and ultimately, it’s going to be your decision as the human utilizing agent, at least for the time being, before it’s the agent utilizing the human. Where do you want to deploy it? Where do you want to use this wavefront to transform? Do a phase change, if you would.

I’ve got a couple of comments. One is this 1,400-times cost reduction is incredible. That is the big headline here. When a frontier reasoning task costs $7 instead of $3,000, think of the implications for startups that gain institutional powers.

Peter H. Diamandis

Yeah, but guess what: when it’s pennies next year?

Well, it will be.

Peter H. Diamandis

Right.

But cost curves are now going to start collapsing industries before the technology does, right? That’s really quite something.

A viewer, Brian Minto, has asked that you read Accelerando live. I think you should do it on a podcast. Just go through it. That would be like MrBeast counting to 100,000 live. He can just read the whole book live in one sitting.

I’ll do better.

Peter H. Diamandis

That would be legendary.

How about we get Charlie Stross as a guest on the pod?

Peter H. Diamandis

I think that would be awesome. That would be awesome.

Hey, before we move off the benchmarks, two things that have changed for me in the last 2 weeks that are just step-function changes for me.

The first is that I just don’t even look at the code anymore. I ask Claude 4.6, with a little bit of deep thinking, to build something, and then I entirely rely on what it builds and look at its functionality. I don’t even look at the code.

The other thing is I ask it to document everything it does and just store it somewhere on my hard drive.

And I don't even specify a location anymore. I just say, “Build some coherent file structure and put things in an organized place,” and it just does it. So now, if I want to get it back, I don't even know where it is. I just have to ask for it, but it knows—it remembers everything that it did. So those are two big changes versus just a couple of weeks ago.

It's a step function from Google. I mean, once you start using Gmail, you don't bother trying to store stuff in folders. You just use search, and now we do the same thing with AI as the interface, right? It's crazy. Yeah, that's great.

Peter H. Diamandis

Yeah.

You know, Alex, a question for you. These AI systems are now beginning to catch human errors in scientific proceedings and scientific—

Right.

Peter H. Diamandis

…papers that have been written. I mean, it's going to be interesting. We've talked about this in the past: when quantum computing comes along, it's going to decrypt all the files from before we had post-quantum encryption.

Mm-hmm.

Peter H. Diamandis

So I wonder if AI is going to be aimed at looking at all the scientific literature over the last 100 years and showing us where all the mistakes were.

I'd count on it, and I think the—

Peter H. Diamandis

It's going to topple some Nobel Prizes.

Oh, I think that's the least of it. I can only imagine the left turns that human civilization has taken in the past, call it, 80 years, when it should have taken a right turn instead, and we're going to discover that after the fact. I think if I had to project the shock to civilization of discovering all the wrong turns that we've taken due to AI, or that AI will uncover, versus, say, quantum decrypting some pre–post-quantum-cryptography-safe files, I think it's going to be a night-and-day difference. I think AI will shock humanity to its core in terms of the mistakes that it discovers that we've made over the past century.

Well, all of that, plus how much have we missed, right? How many scientific experiments did some geek look at the wrong thing and miss the unbelievable conclusion over there? That, I think, is going to be the huge outcome.

Peter H. Diamandis

Oh, God. Destroy it.

I think it's a continuum. I mean—oh, go ahead, Dave. Sorry.

When I'm spooling up a new agent now, I used to be very thoughtful about what I fed into the context window to get it up to speed. Now I just ask it to read about 1,000 pages of Markdown documents, and it does it in about 10 or 20 seconds. It's fully up to speed.

The context window, and also its ability to sort through all the garbage, is growing or improving faster than my ability to clean it up anyway. So my new agents—I'll boot up two or three agents every couple of hours—and I just say, “Look, read everything.” Read everything I've ever given to any agent before, and then the new agent is up to speed, and it can pick up a project right where I left off. It's—

Peter H. Diamandis

And your—

So I think the—

Peter H. Diamandis

Your future employees will be the same, right? Read every email, every Slack, and everything—

Well, also, what's not intuitive is that the complexity of the document doesn't seem to matter. If you're teaching a kindergartner to become a college graduate in 30 seconds, you move through reading and writing and then basic arithmetic; you work your way up. But here you just bombard it with super, super-technical, complicated documents that would take me many, many hours to read a single document, and it just absorbs them instantaneously. It's just mind-blowing.

Peter H. Diamandis

Amazing.

And everyone can try that, too. Just go find something that you barely understand, download 1,000 pages of it, and try and just dump it into Gemini. Just go to free Gemini, put it on Thinking mode, and just dump it in, and then start asking it questions. It just is such a mind-blowing experience.

Peter H. Diamandis

So, one more point on the benchmarks here before we leave these couple of slides: are the current benchmarks becoming meaningless? I mean, the models are increasingly optimized to ace them. We're beginning to saturate them. We've talked about this before, Alex, so smack some knowledge on us about how we're going to measure things as these benchmarks begin to fail to serve us.

It's almost, Peter, like we wrote an entire book on this problem.

Peter H. Diamandis

Yes. I'm trying to prompt you to speak to it.

It's a good self-advertisement. I think we are—the world is in a famine of good benchmarks, good evals. We call them, in some sense, targeting authorities in the book, if we want to call it a book or extended essay—

Peter H. Diamandis

Our white paper. Yeah, yeah.

—white paper. I think there is a lot of juice still left to be squeezed out of new benchmarks and new evals. I think solving the hardest problems of physics, chemistry, biology, and various disciplines in the social sciences—all of these want high-quality benchmarks. I'm personally spending a lot of my time thinking about what are the best problems that are worthiest to be solved.

I have mentioned on the pod in the past that I have a portfolio company, Physical Superintelligence, that's thinking about problems in—

Peter H. Diamandis

PSI, baby.

PSI—solving physics with AI. I think this is how we solve all the hardest problems in civilization, starting with new benchmarks for those hardest problems. This is how we weaponize superintelligence.

Peter H. Diamandis

Amazing. All right, Salim, move us forward.

The Indians are going to suck all the data center usage and tokens out of them. Yeah, I mean, that is honestly what could well happen, right? I think India is a bellwether for a lot of countries. One of our listeners is in Finland, and he's saying, “The politicians here are absolutely not talking about this. It's nuts.”

But I tell you, India is such a crazy zoo of an ungoverned mess of a place, but it's packed with brilliant people. And you just—oh, my God—massive population, 1.4 billion people, of whom 5% read and write English and 20% speak it. The massive latent talent pool means that it'll be a bellwether. The population is just going to run away with AI and ignore all structure and government.

Yeah, Dave, I was starting to look at what India ETFs in the tech industry I should look at. I think China has peaked and is going to be on the descent. India is the rising giant for the next 20 or 30 years. Africa will follow because of a young population and because of all the resources that they have. But the country that trains its next generation on AI wins the entire talent war.

India has the ability, if it goes deep on this, with 1.4 billion—1.412, whatever, billion people on the planet. It could be the next massive rising star and support the planet here. It's going to happen really fast, massively in parallel. That's what a lot of people aren't used to: the idea that something can happen overnight because normally things percolate, and you have this kind of slow GDP growth that percolates out.

But this isn't going to be anything like that. The population, in one fell swoop, in a very short period of time, is going to use AI to escalate.

The population of what, India or the world?

Of India. Yeah. Okay. Well, probably the world, but India will be the bellwether because, again, it's such a huge population, and it's so untapped.

And Mukesh Ambani—the other thing is that Mukesh Ambani has delivered amazing 5G capability across the country, right? So it's got the infrastructure. It skipped the wireline. All the youth is kind of growing up AI-enabled, right? So that's incredible.

I have a very quick story. When we left India when I was 10 years old, I was kind of an angry teenager because I had to mow the lawn and stuff. And I asked my father, “Why the hell did we leave? I mean, we had a great life over there.” And he goes, “I can't stand noise, dirt, pollution, and corruption.” And I was like, “Okay, fine. If you had to go, okay, fine. I understand that.” But there is something there, because as you get the capability and the democratization into everybody's hands, the speed of change is going to run around.

And the government is doing an amazing job of making platforms like Aadhaar and UPI available so that anybody can tap in and create a payment system. That's going to completely allow India to leapfrog the rest of the world. The huge bottleneck is going to be scalable energy, which they're adding at a rapid scale, putting solar in every little corner of the country.

Peter H. Diamandis

Yeah, last week we reported that solar was scaling faster in India than it did in China, which is amazing. Nice. Well, we're going to see. Over to you, Alex. This is a fun one.

We're seeing the beginnings of everything other than math and coding starting to get solved. This is a reference to OpenAI announcing, in collaboration with Harvard—and I think the Institute for Advanced Study was involved, along with a couple of other places—what OpenAI is marketing as a new physics research result that was discovered, in some sense, by AI. I think we're going to see much, much more of this.

So, 30 seconds on what the claim actually is. The claim is that OpenAI and its co-authors were able to use GPT-5.2 Pro to discover something in what's called a scattering amplitude—basically involving gluons, the messenger particles and force carriers of the strong nuclear force. They tried to solve a sort of prediction of how these strong nuclear-force carriers would interact.

Historically, in this part of the physics community, the thinking was that there would be, in some sense—and I'm being very heuristic here—no interaction: a term in a scattering amplitude, which would be the formal way of describing this, would be 0. Many physicists assumed for many years that the answer to this particular value was 0 and didn't bother spending any time checking rigorously and fulsomely to see whether it actually was.

The claim for this paper is that GPT-5.2 was able to find cases where this scattering amplitude was not 0, find a nice expression for it, and then an internal model—which hasn't been released, or so the story goes, probably some future version of the GPT model series—was able to confirm it. That confirmation was then, I think, vetted by the human team. So this is being represented as a case where AI is making a particle-physics discovery.

I think what's most interesting about this is—and Peter, you and I make the case in “Solving Attention”—we call the intelligence revolution a war on attention. This is Exhibit A for AI helping to solve science by solving problems where humans say, post hoc, having seen the evidence, “Okay, I could have done that if I had the time and the attention for it,” but no one had the time. People thought the answer was obvious. It's only once we have lots of superintelligence that we're able to train on problems that would've been too boring or too low-likelihood to actually yield an interesting, novel result that we're discovering oversights. This was, in some sense, an oversight.

You also have the issue of fashions and trends, of people following fads. You can get around all of that now. This is such a great point you're making here.

Peter H. Diamandis

We all have those projects, those wonderments that we had, or that project you put on hold because you didn't have the resources, the time, or the knowledge, and you can spin them up. We'll talk about MultiBots and OpenClaw in a little bit.

I just wrapped up a project I've been wanting to work on for 5 years, and it was so much fun. I was on my agent for about 8 hours, and I felt completely disconnected from the world. What have you always wanted to work on? What's that pet project, that company idea, that book, or that piece of research—because you can?

I'm trying to think of ways that our audience can experience how mind-blowing this is, because AI is an unbelievably prolific brainstorming partner. If you're in a domain where it can test things by itself, like what I do all day with neural-net creation or coding, I can just say, “Wow, what a great idea. Go try it.” A minute later, it comes back with an answer.

The rate at which you can move is 2 or 3 orders of magnitude higher than anything I've ever experienced before in life. But it has to be one of those unconstrained domains, because if you're working in chemistry or whatever, you're going to have to wait for test results for a day or 2 or 3, and it breaks the whole experience.

If you want a really simple example, just try to plan a trip—something complicated in travel—and try to brainstorm your way through the flight, the restaurant, the hotel, whatever. That may not be the best example, but at least you get some flavor for what this is like.

It's like nothing you've ever experienced.

Peter H. Diamandis

My fun experience was: I have to be at this location at this time, and I'm here at this moment. Work it out backward. What flights, taxis, cars, and Ubers do I have to take? Work out the whole thing from my endpoint and work it backward.

I think one of the things I keep saying on stage to the audiences I'm speaking to is that we limit ourselves in the questions we ask all the time. We self-limit what we think we can do. We hold ourselves back in so many different ways—in how we can and should be using AI—because we're not used to it.

We're not AI natives. At least those of us on the phone here didn't grow up with it at age 6, 7, or 8, as many folks are now. So you've got to stop yourself from stopping yourself and unleash your creative child mind in this area.

By the way, I just want to ask: if you're enjoying having this Moonshots episode live, please let us know in the comments. Let us know if we should do this more often. I'll ask you again; maybe you like it now, won't like it later, but we'd love to know, so give us some feedback.

All right, it doesn't stop with physics. It's continuing on with math. OpenAI says its internal model solved 6 of 10 research-level problems in the First Proof test. And here's our friend Jakob, whom we've met. Alex.

Awesome. We talked about math getting solved, math getting bulk-solved. In fact, math is getting bulk-solved. This is maybe not Exhibit A; this is probably Exhibits C, D, E, and F at this point.

First Proof is such a beautiful example of a class of 10 research problems, with a finite amount of time being allotted for AIs to solve them, where the answers were known but were kept confidential by their authors and have since been unlocked. OpenAI has taken the position that its model was able to solve at least 6 of them before the solutions to these 10 research-level math problems were declassified, and it's been fascinating watching the back-and-forth.

We're seeing right in front of our eyes the bulk solution of math. I think back to almost a year ago, when, as the royal we, I was first talking on the pod about math getting bulk-solved by AI. It's happening now. We're there.

Peter H. Diamandis

Yeah. Today we saw the first hints at physics, and 6 months from now, if not a year from now, we'll be talking about how all these physics problems have been addressed. Can't wait.

Let's touch on the timeline there, too, because, Peter, a second ago you said something about 20 or 30 years from now, but there is no 20 or 30 years. There are so many times this morning that somebody said, “Next year when we do this...” There is no next year. What are you talking about?

Peter H. Diamandis

No, there isn't.

You actually did.

Peter H. Diamandis

Did I use 20 years in my language? I'm sorry. I must have meant 20 minutes. That's the challenge.

Yeah, I mean, Salim, you remember in the early days of Singularity University, we were looking 10 years out into the future. Honestly, I had this side conversation with Elon: you can barely look out 3 years, and I don't think we can.

We're used to this world where physicists or mathematicians can now do blah. Okay, well, there are only so many of them. They'll do blah, and 20 years from now they'll have solved all of blah. But here it'll happen instantaneously. If it can solve 6 out of 10, it can solve all of them within—

Peter H. Diamandis

Yeah.

—the next couple of months. It'll happen in massive parallel. There's no limit to the number of parallel agents, up to the number of GPUs that are available.

Math is cooked. Yes.

Peter H. Diamandis

Math is cooked; physics is cooked.

Yeah.

Peter H. Diamandis

Biology is going to be broiled, charbroiled, and we're—

We're going to be the beneficiaries. You know, I just think I was seeing one of the comments in the chat here. I think if we just stay on this live 24/7, and Jian will just generate more slides for us, we'll just keep going through them.

We could do shifts.

Peter H. Diamandis

It'll be a continuous singularity conversation.

Yeah, yeah. It'd be like a hackathon. We'll just go around.

Peter H. Diamandis

Yeah. All right, sleep.

Take sleeping into account.

Peter H. Diamandis

Let's move on. All right, more benchmarks. I'm fascinated by what's going on with Chinese open models gaining momentum. Here's MiniMax, GLM-5, and Kimi K2.5. These are doing extraordinary work.

Mm-hmm.

Peter H. Diamandis

With all of the OpenClaw downloads, a lot of people are now moving to Mac Studios and putting Kimi K2.5 on their Mac Studios, along with other models. Alex, how do these perform against the closed models as you see them?

Well, the rumor going around is that the next version of the DeepSeek model—the big wake-up-call moment—is going to happen sometime soon, when finally the Chinese open-weight models catch up with the American closed, proprietary frontier models. That hasn't happened yet. It may happen. Right now, the overall trend is still that the Chinese models remain approximately 6 months behind the American models. We'll see whether that continues to be the case. I haven't seen any evidence yet—

Peter H. Diamandis

But they're free.

Well, that's a qualitative difference and a very important one. That means that many American startups that want to self-host are using Chinese models and not American models.

Yeah.

Peter H. Diamandis

And so this is, again, going back to the land grab. We talked about this with OpenAI and India going in and providing basically a very low-cost service to millions of young Indians. China is in the same process. This is Belt and Road, where it's offering it to the majority of South America, Africa, and different parts of Asia, and I think there's going to become a dependence. I think people are going to get connected to a model that they're going to use and begin to baseline—

I think there's a big difference, though. If we want to frame it as model diplomacy or model dumping, even, I think there's a big difference, which is that the frontier is moving so quickly. I think it's difficult for a prototypical so-called developing country to get addicted to a particular open-weight model because the new ones are constantly coming out. It's a vibrant marketplace. I think if American labs felt sufficiently motivated, they could just as easily release their own models for free. I just think it's a problem of incentive.

So I think, as opposed to alleged Chinese dumping of, say, solar photovoltaics into India or into Africa, or other physical-plant infrastructure, the marginal costs for substitution and replacement are so low with these models that it would be very difficult for China or Chinese AI labs to addict the rest of the world to their models.

Peter H. Diamandis

The important thing is that humanity is the beneficiary across the board here, right?

Yes.

Peter H. Diamandis

We're getting much more powerful, much cheaper models at hyper-exponential rates. I mean—

This is a space race. It's a space race on the ground to superintelligence and to super-duper intelligence. This is providing a strong incentive and strong pressure for the American frontier labs, which, as of right now, are still in the lead, to stay in the lead. There's no pausing this.

Peter H. Diamandis

ASDI, baby. Artificial Super-Duper Intelligence.

Love it.

Peter H. Diamandis

All right, Alex, quoting you here: “Traditional coding is cooked.” So, just to note—

Even cooking is cooked at this point.

Peter H. Diamandis

With humanoid robots. This is the note from Shopify that they haven't written code in 3 months. The code is being written, but it's not by humans.

That's cool, huh?

Peter H. Diamandis

And, of course, 95% of OpenAI code is being written by Codex. This is probably true of a large number of companies. This is just the news items reported. Dave?

I think it's really funny, actually. When you talk to the top AI researchers, they always talk in terms of, “Well, what I'm working on is that last 5%. I'm not eliminating my own job tomorrow.” Then you look at the HLE results, and you're like, “Yeah, you are.” You're literally coding yourself out as fast as you possibly can. I don't think they stop to think about that fact, but—

Peter H. Diamandis

Alex, I loved your analogy last time we spoke about George Jetson, with his finger being over-exercised on the button. That's effectively what coders are doing right now. It's like, “Okay.”

That's what it's like. If folks in the audience—

Peter H. Diamandis

Code it.

I hope other folks are having this experience, and not just myself, with Claude Code in particular. Approvals for everything. But I think we're going to move past this George Jetson model of just approve, approve, approve for software development pretty quickly.

I think Claude Code is either already in, or is an imminent preview of, a future where it's permissionless activity by these agents. Do you remember older versions of Windows that were permission-heavy, where you had to go through 10 clicks to approve, approve, approve to do basic things?

Peter H. Diamandis

Nightmare.

Yeah.

Nightmares.

I think that's the stage we're at right now with these models. Out of an abundance of caution, these models are asking for permission to do everything: permission to switch to another directory, permission to search the web. I think pretty soon the autonomy time horizons—which METR and others are measuring—are going to be such that we just give these models blanket permission to do whatever they want within broad parameters, and we stop having to click “approve” for everything. Yeah, we have a couple of—

Well, a note on that: We are in a kind of fragile moment in time here. If you install Claude Bot or OpenClaude now, you can choose any model you want. But if you choose one of the Chinese models, especially if you run it locally, you don't have to go through all the permission nonsense.

Also, if you use one of the U.S. APIs, it'll get stuck a lot because the bot is asking it to do something that it doesn't want to do. The Chinese models are like, “Yeah, sure, I'll just do anything.” That kind of forces you down the Chinese path.

But as you've said many times, Alex, you don't actually know what's inside those models, and the code-injection risk is really, really real. So people are in a real hurry to experience this and to turn it loose.

Yeah, you have to be really careful.

The only way to really turn it loose is on one of those Chinese models. And so—

Yeah. The world hasn't—I mean, this isn't prescriptive, certainly not—but the world, to my knowledge, has not seen a major supply-chain attack yet that stems from untrusted open-weight code-generation models rewriting the entire supply chain. But do I think that's possible? Yes. I think that is absolutely a threat vector.

No, I think so.

Peter H. Diamandis

You know, Blitzy's been an amazing company, and it's grown at light speed, coming out of the Link Studio shop, and it's been a great sponsor here. How are they using all these technologies? Because they're rewriting massive amounts of code.

Well, they're doing a lot of work for banks and government agencies and stuff, so they can't use the Chinese models for that. They're almost entirely—

Peter H. Diamandis

I imagine that—

But they're definitely not touching the Chinese stuff.

Peter H. Diamandis

I imagine that, at Blitzy, given the speed at which they're rewriting, how old is the code they're rewriting? COBOL? How far back are they going?

Yeah, a lot of it. Actually, it's very similar to what Alex was saying about old physics papers. A lot of this code has bugs that have been sitting there for 20 or 30 years, robbing it of performance or actually losing money for 20 or 30 years.

Peter H. Diamandis

Yeah.

And it's just cutting through it, rewriting it, solving it, finding old issues at AI speed.

It's a real threat. We've talked on the pod in the past about how Stack Exchange, for example, is dying in some sense. Very few questions are being asked because you can now ask the models any coding questions you want. There was a paper—I talked about it in my newsletter—about the risk to open-source projects in general. Why even bother starting or maintaining an open-source project? Doubly so for middleware: if you can have AI models generate all your code for free, why even bother—

Yeah, that's exactly right.

…maintaining an open-source project? So if we find ourselves in a near-term future where there's just no point, where you can spin up a new kernel-level project from scratch on demand, all of the code is just in time with whichever models are convenient, I think, from a supply-chain security perspective, we're going to have to have a long, hard look at what our dependencies are and make sure that our dependencies aren't just riddled with vulnerabilities that were inserted by just-in-time code gen.

You know what else came up this week, Alex? The AI is so prolific at creating code modules. Just like solving all math: if you solve all math, you write down what you solved, right? You don't solve it on the fly in real time. But for complicated code, it's the same thing. It's like, well, yes, I can write it in real time, but I already wrote it, and discovering it and reusing it is actually even cheaper. It saves you tokens; it saves you compute cost.

Right.

And so now, where we've had open source, we're starting to have open source designed for AI. And, you know, thousands or millions or trillions of fragments of code that do specific things, the AI can discover them in real time, and it's actually a really great way to build new software. You could also generate on the fly, too. It's just a question of what's more efficient in terms of latency and tokens.

It's like all of this historical open source is now going to be designed for AI. Just like all written documents will now be written for AI, not for direct human reading.

And just like we're doing this podcast mostly for AI listeners, I'm guessing, not human listeners.

Yeah. That's why we're live-casting it. We want to reach out to the real humans one more time while we can.

Peter H. Diamandis

Happy Chinese New Year to all of Chinese descent. Happy New Year. I just saw some chats in the side here on our live chat asking about where's nanotechnology. I can't wait for nanotechnology. I remember back in 1986, I read a preview of Engines of Creation by Eric Drexler, and it's been a few decades, so it's coming. I don't know. I think we'll start to see it fall. I mean, we have wet nanotechnology called biotechnology. Alex, what's your timeframe for nanotech?

I definitely have a view on this, in part because I spent a good chunk of my PhD thinking about how to get us to Drexlerian nanotech more quickly, in part because I was a little bit less bullish on AI as a direct path than I am now. So, if the question is, what's my timeline for—

Peter H. Diamandis

For assemblers.

For assemblers.

Okay. For Drexlerian assemblers, to the extent the physics and chemical physics of our universe admit Drexlerian assemblers, say, as parameterized. Peter, I think you're on the board—at least, you have been historically on the board—of the Feynman Grand Prize. Is that—

Peter H. Diamandis

Just an advisor, not on the board.

Okay. So the Feynman Grand Prize is one parameterization of Drexlerian assemblers. For those not paying super close attention, it comes in 2 parts. One part is: can you build, I think it's an 8-bit half adder within a certain very small volume of a nanosystem? And the other part is: can you build basically a robotic manipulator arm within—

Peter H. Diamandis

Yeah.

…a small volume? So the question is: what are my timelines? I would not be that surprised if the Feynman Grand Prize is solved in the next 2 to 3 years.

Peter H. Diamandis

Fascinating. And we lost Salim. Oh, well. So we'll continue until he comes back on.

Well, the slide we were moving to was the Meta smart glasses—

Peter H. Diamandis

Oh my God. You know, I put in the title there: “Privacy?” So there are some great books, some great sci-fi books. Welcome back, Salim.

Yeah.

Hey, my microphone had dropped out for some reason—

Well, this, to me, is a great example of how you cannot opt out.

…so I had to figure it out.

The peer pressure forces you to opt in.

Peter H. Diamandis

Yeah.

Because I think a lot of people look at this and say, “Well, I'm not going to wear these glasses and spy on everybody and record everything.” But once you've experienced the face recognition and then all the metadata that pops up, you're like, “Well, now I'm not competitive with the world unless I actually have them,” and it creates this huge amount of techno-peer pressure. So you don't really have the option to opt out.

Peter H. Diamandis

I think this is going to become part of normative culture. We had the Glasshole episode with Google for a while. That didn't work out. But first off, what I find fascinating here is that, to get these allowed and to get people to start to accept them, their pilot program is being done with people who are visually impaired, right?

Mm-hmm.

Peter H. Diamandis

So it's like a soft on-ramp.

Yeah, that's what they did with Neuralink, too. It gives you a good politically correct excuse to do what you really want to do—

Peter H. Diamandis

Yeah.

…which is everybody.

I also think it's interesting if you think about whether this could only have arrived now. This is old technology. We've had the technology to build smart glasses that would do human identification at a distance—human ID, if you will—for at least a decade. It's not that hard. We've had the computer vision algorithms. It's 2026 now. We certainly had the ability to do relatively efficient human identification, doubly so if you're restricting human identification to, say, all of your Facebook friends. We've had that for at least 10 years.

So why now? I think this is a social technology more than it is an AI technology. It's not a real AI advance, in short. I'm calling this one a social advance. We have already, many of us—especially those of us in certain places in the West and also China—with very dense surveillance networks, with cameras spotting everyone on the streets and in cities. This technology exists already—

Peter H. Diamandis

Mm-hmm.

…and is, in many cases—

Peter H. Diamandis

It does, but this is convergence and this is cost, right? And then this is social engineering as well.

I don't even think it's cost. We could have done this cheaply 10 years ago. I think what's interesting is there's a demand for AI-enabled wearable devices, and I think this is an opportunity. I suspect Meta sees an opportunity—maybe demographically, maybe politically—an opportunity to finally launch human identification via smart glasses.

Peter H. Diamandis

I think this is a killer app—

I think that comment, though, overlooks something really, really important.

Peter H. Diamandis

…and it's going to kill privacy.

Yeah, privacy. Recording everything was already here 10 years ago, but people—

Yeah.

…people didn't get slapped in the face with the fact that everything they have ever done is being recorded. It's the AI overlay that then recognizes all actions, classifies them, and makes it all very searchable. So if I said, “You know, I only want imagery of you picking your nose. Go through all the thousands of hours of footage we've ever done on this podcast and find me an example of Alex picking his nose,” it just does it instantaneously.

Peter H. Diamandis

Go ahead, Alex.

So, that's the part that makes it—

The good news is you can just claim it's a—

…very different socially and culturally—

The good news is—

…than the surveillance we've been living under for—

Peter H. Diamandis

Go ahead, Salim.

The good news is you can now just claim it's a deepfake.

Peter H. Diamandis

Yeah.

So there's that defense.

Peter H. Diamandis

Yes. Yes, please. No.

First of all, I was about to volunteer to make it easy for the AI model to find an example. But no, I would say the—

Peter H. Diamandis

Go ahead, Alex.

…the models for video understanding are new. I agree with that. And the most recent Gemini models are absolutely outstanding at handling long, multi-hour videos and asking them to find a needle in the haystack of something interesting happening.

However, I would say just spotting humans—if you're walking around on a city street and spotting someone interesting and matching that against, say, hypothetically, a database of people's faces—we could have done that 10 years ago. That's more a social innovation.

Peter H. Diamandis

When I come through passport control at LAX and you just walk by the camera, right? We gave up our constitutional rights to some degree, and it makes life easier. And so, as long as this makes life easier for people—like being able to recognize someone on the tip of your tongue and have it pop up the last time you saw them, their kids' names, and all that information—it's going to create this social fluency that I think we've never had. Maybe if people have an amazing memory for faces and names. I meet too many people. I don't.

Yeah.

There's a big slippery slope there, Peter.

I think—

Peter H. Diamandis

Go ahead. I'm sorry, Salim. Go ahead. What's the big slope?

Yeah, there's a big slippery slope there because if you don't have privacy—

Wait, can you hear Salim?

Can you guys not hear me?

I can hear Salim.

Peter H. Diamandis

I can hear him.

Are you—

Do you want us to rejoin?

Peter H. Diamandis

Is it safe?

I did actually drop out and rejoin, Salim.

Peter H. Diamandis

There's no Salim.

No.

Peter H. Diamandis

That's a voice in your head, Peter.

No, no.

Peter H. Diamandis

No, I'm real. I'm real. Are you guys playing with me?

No, no.

Peter H. Diamandis

Seriously, I literally am not seeing you or hearing you, Salim.

I'm only seeing an error on my screen. This live experiment is going really well.

Actually, the chat is hilarious. I'm cracking up here.

Peter H. Diamandis

It is kind of ridiculous. So anyway, the problem—

Enter our producer, Nick. Nick—

Peter H. Diamandis

Hey.

What do you see?

Peter H. Diamandis

Nick, welcome to the world. You've exposed yourself.

But now he's frozen.

Peter H. Diamandis

Jesus. Oh. All you guys watching, and folks—

Oh.

Peter H. Diamandis

And girls and gals and bots and droids and—

Oh, the audience hears it.

Peter H. Diamandis

Lobsters—

Maybe it's our problem, Alex.

Peter H. Diamandis

This is a full grid here.

Okay, probably. Should we rejoin?

Peter H. Diamandis

Dana—

Is it safe?

Peter H. Diamandis

Dana, can you hear us?

Yeah.

Peter H. Diamandis

Nick—

I can hear you. Can you guys hear me?

Peter H. Diamandis

All right. Well, Salim, you and I can have a conversation.

Yeah, we can.

Peter H. Diamandis

All right. Let's continue. You guys can both hear me, but you can't hear us.

We can all hear everybody except that—

Peter H. Diamandis

Correct.

Dave can't hear me.

And we can hear each other.

Peter H. Diamandis

Yes, just not you, Salim. All right.

No, Dave, you can't hear me?

Peter H. Diamandis

I can hear you. Dave and—

Yeah, Dave can't hear me.

Peter H. Diamandis

And Alex. Neither can Alex.

Do you want us to rejoin?

Yeah, let's try that again, Alex.

Peter H. Diamandis

You know what? Maybe Salim needs to rejoin.

I did that already.

Peter H. Diamandis

All right.

Okay.

Peter H. Diamandis

By the way—

Hold on.

Peter H. Diamandis

How is everybody enjoying this live version of Moonshots? You know, Moonshots—I just keep on saying, AI is easy, AV is hard.

Yeah.

Peter H. Diamandis

All right. I am—

Peter, if those guys can hear you, why don't you tell Alex and Dave to drop off and rejoin?

I'll try to rejoin. Let's see what happens.

Peter H. Diamandis

Okay. Alex and Dave, go ahead and rejoin.

All right. I'm going to try.

Peter H. Diamandis

And in the meantime, Salim, what are your thoughts on this privacy issue?

So, the privacy thing—

Peter H. Diamandis

It's a slippery slope.

—is a very difficult and slippery slope, and I'll explain why. The minute you don't have privacy, you don't have freedom. Okay? And this is a huge problem. You can't experiment. Like my private keys of my Bitcoin—there are all sorts of areas where you have huge issues around this. Hang on, Nick is calling me.

Peter H. Diamandis

Can you guys hear Salim now?

Yes, I can.

Peter H. Diamandis

You can? All right.

Yeah, we're back.

Peter H. Diamandis

Fantastic. Dave?

We're back.

Peter H. Diamandis

Okay, great.

Yep.

Peter H. Diamandis

All right. So your point, and I think it's an important one, is that Salim just said, “If you don't have privacy, you don't have freedom.”

I think it's a false choice. First of all, these glasses, legally, at least in the American legal system, will be used in public places. They'll very likely be banned, to the extent they're not already banned, in multi-party-consent contexts, in private spaces. They have lights.

If you look at what Google, of course, is launching—Android XR and smart glasses—everyone's launching smart glasses, and they'll have lights to indicate when you're being recorded and when you're not. I think there may be an evolution of standards regarding circumstances in private spaces when it's allowed to record or not, but I completely don't buy this premise that somehow privacy is going away. People have eyes, and people have memories.

Peter H. Diamandis

Privacy is cooked. Privacy is cooked, Alex. I mean, we're going to have every major—

Hold on.

Peter H. Diamandis

OpenAI and Google, and everybody's going to be having wearables—

No.

Peter H. Diamandis

—that are recording all the time. All the time. And we're going to have microdrones. I mean, we're going to be gathering data all the time. And so, I think privacy is cooked.

It is, but it's important that we preserve it. Now let me explain why.

Peter H. Diamandis

Yeah.

Okay? Can you guys hear me, first of all?

Peter H. Diamandis

Yeah, we hear you.

Yes, we can.

Peter H. Diamandis

Okay, great.

Peter H. Diamandis

Your audio is not private, Salim.

Okay. So look, it's one thing to be out in public and people know your move. That's fine. We can augment that. But there are lots of things that are a huge issue here. For example, there are lots of cases where government authorities have dropped into cars and opened up the microphone so they can hear what's going on without a warrant. There are lots of cases where people are listening to your—

Peter H. Diamandis

Oh, no.

Cases where people mute themselves in mid-sentence.

Peter H. Diamandis

Salim, you're muted.

Got it.

Peter H. Diamandis

How does that even happen?

This is totally surreal. There's an AI watching me going, “I don't want them to be listening to this.”

Exactly.

It's muting me. There are a lot of cases where people misuse this capability in very radical ways, and the problem is there's no easy way of stopping that.

Now, that doesn't mean you have to turn off all the Metas, and I'm not an anti-technologist by any means, even by being on this podcast. But the minute you do that, it gets abused, and it gets abused quite badly. So you have to have guardrails on the institutional side, which is the problem. We're losing that, okay?

Peter H. Diamandis

Mm.

For example, we're losing habeas corpus in the U.S., okay? That's a choice that people are making to just ignore that and have it wash away. Once it goes, it does not come back. Viewer Innovative XR has made the exact point that once you lose that privacy, it's very, very hard to get it back.

This is the challenge with all of this technology. We're moving faster than our institutional guardrails.

Peter H. Diamandis

Yes, you're absolutely right. And Salim—

I'm not sure what the answer is, but—

Peter H. Diamandis

I want to be—yeah.

But we have to be very careful about all those things without realizing the downsides of it.

Peter H. Diamandis

All right.

So, Salim, I want to be clear. I want privacy in my life, right? I—

I understand.

Peter H. Diamandis

Everybody wants privacy. Everybody has screwed up at some point in their life, done something they regret. We're humans, and you feel lucky. When we were kids, we didn't have Facebook and cameras capturing everything happening today.

There was this whole thing about college admissions looking at kids' Facebook pages and so forth in the past. I want privacy. I just don't think we are going to actually have it. We're going to have the illusion of privacy. Alex—

I will buy that for one second. I'll point out maybe one or two other points. One is, to the extent anyone here is bullish on crypto, you sure as heck should hope that privacy remains intact. Otherwise, your crypto is going to disappear.

Peter H. Diamandis

It's going to be cooked.

“Cooked,” I believe, is the word you were using.

Yeah. Crypto was cooked. How's that for alliteration?

Peter H. Diamandis

Cook it.

But it's not forward-looking financial advice.

Peter H. Diamandis

We've got cookbook.

It's just pointing out, informationally, that if you think privacy is cooked, then you probably should infer that crypto is cooked as well. Your private key is cooked. If you think privacy is cooked, then your whole thing's cooked. Cook, cook, cooked.

Well, I think part of the disconnect there is Alex's view of the world: “I will upload my consciousness very soon, and within that virtual world, there'll be all kinds of privacy options, just like there are with my crypto keys.” And then Salim's view of the world and my view of the world is, “No, I'm gonna live in my meat body for as long as I can, and every move I make is gonna be recorded, and it's gonna suck for a while until we have some new legislation and some safe zones.” And that, to me, is inevitable, and I think all the listeners are also posting the same kind of view. But I think that may be the source of the—

Peter H. Diamandis

Salim's typing away.

—of the disconnect. Sorry, I was responding to one of the viewers. This live thread is awesome.

Peter H. Diamandis

It is awesome.

Having this conversation in real time is so amazing.

So I'll also point out, I think no discussion of smart glasses with cameras and facial recognition is complete without referencing David Brin's seminal book The Transparent Society and his discussion of sousveillance as opposed to surveillance. So I should point out that, at least for public spaces, police wear body cams. Humans, at least in certain Western countries, can also wear their own body cams or have their own wearables that enable them to make sure that we don't descend into an authoritarian panopticon.

So that's one good case for—it's not loss of privacy in public spaces, because there shouldn't be—at least, I think the Western tradition is that there's no reasonable expectation of privacy in public spaces—but it at least offers maybe a way to soften any perceived blow to any semblance of privacy in public spaces. It's a way to make sure, again, that the populace is just as empowered to monitor their environments in public spaces as authorities.

Well, guys, keep in mind that we live in a world of mature adults and great friends like we are right here right now. But take yourself back to middle school, which I know is hard to do, but it's brutal, man. People are so cruel to each other, and you empower those people with constant eyeglass recording.

Peter H. Diamandis

Yeah.

They've already got their iPhones, which is a massive life change in a negative way for that entire period of life. But you layer on top of that the smart glasses, and it's next-level suck to exist in that world. It's just gonna happen because the rule changes that we desperately need are gonna lag by—

Peter H. Diamandis

Yeah.

—a while, way too long.

Peter H. Diamandis

There will be lawsuits—

It's gonna be awful.

Peter H. Diamandis

—and there will be legislation, and it will take years. We need them now.

Yes.

Peter H. Diamandis

It's not just the constant recording; it's the constant recording with the AI overlay that allows you to modify, meme, make funny, and torture.

Yeah.

Peter H. Diamandis

And it's just—you know, people are mean to each other, especially until they grow out of it.

But this is happening at the same time—the same time that we're beginning to generate every pixel, right? And we're gonna be able to create whatever videos we want.

Yeah.

Peter H. Diamandis

On the good side, it means that young people today, getting this in their teens, will have their entire lives recorded. They'll be able to go back and play it back. We'll be able to reconstruct almost any situation. No crime will go without being visualized in some sense.

Well, that is a great point. The crime rate in the U.S. has plummeted—absolutely plummeted—and it's due to two things: location services, knowing where all police are at all times, better control of location, and then, after that, surveillance. And so that is the good side effect. Crime rates should continue to go way down.

Peter H. Diamandis

All right, let's go to our next story here, which I love. We saw a version of this on Minecraft about a year ago. There's an AI startup called Simuli that raised $100 million to simulate human behavior. Think of Isaac Asimov. I'll just play the video, and hopefully it's got audio too. Does it have audio?

I can hear the audio.

Peter H. Diamandis

All right. Well—

Can you guys hear the audio?

Peter H. Diamandis

No.

No.

Peter H. Diamandis

No, we can't. We cannot.

Oh, God.

Peter H. Diamandis

So—

You know what? I didn't share it with the thing. Hold on.

Peter H. Diamandis

Okay.

Somebody in the chat, tell us if you can hear it.

They shouldn't be able to because Salim isn't sharing the audio. He's hearing it locally.

Peter H. Diamandis

All right. Well, hold on.

Yeah, hold on. Hold on. This is—

Peter H. Diamandis

Salim.

User error here.

Peter H. Diamandis

Yes. Okay. Unshare—

But maybe—

Peter H. Diamandis

—and reshare.

Maybe a thought on this in the meantime.

Yeah.

So much of our usage right now of autoregressive language models like the GPT series, but many others, is based on autoregressive sampling of one token at a time, or maybe beam search.

I think we've talked in our past AI personhood debate about the right metaphor for thinking about what these models are. Is it right to think of them as individuals, or are they something else? I often think they were trained off of an ensemble of humanity's behavior on the internet, or at least pre-trained off of that and then post-trained off of other things.

Maybe the right mental model for thinking about many of these foundation models is as societies. And if that's the case, then maybe a more natural way to sample from a society isn't to pick out a single individual with a prompt and then do a rollout of that prompt and have a conversation with it. Maybe it's more natural to do many rollouts in parallel and sample an entire society from the model, and that's what we're starting to see here, I think.

All right, I'm gonna play this.

Peter H. Diamandis

Okay.

We are building Simile, an AI lab to simulate our world. We start with individuals. We model how real people make decisions, then we compose them into bottom-up simulations. We call each one a simile. Change one assumption, constraint, or person, and the world recompiles. Run counterfactuals you can't run in real life. Learn what matters, what backfires, and why obvious strategies fail—like a flight simulator for human decisions. Over the last few weeks in the Simile office, we even tested how this message might land. Simulating human behavior is one of the most important and technically difficult problems of our time.

Peter H. Diamandis

Wow. So we're gonna have to make a lot of decisions in the near future on UBI, UHI, and policies around exponential growth, because the speed of the tech is moving faster than the speed of policymaking. And so this was a—

By a massive gap, right?

Peter H. Diamandis

By a massive gap, right?

Yeah.

Peter H. Diamandis

So Simile—

What I saw with this was Harry Seldon and psychohistory because—

Peter H. Diamandis

Exactly. Yeah.

—it was predicting human behavior at scale. Pretty cool.

Peter H. Diamandis

Yeah, it's the Foundation series. So we've had some of these conversations. Emad Mustak had built something called Sage that we were rolling out in part at FII in Saudi. And I think policymakers need to be able to know how to simulate: What is our policy on autonomous vehicles or on longevity escape velocity? How is it gonna impact our society? Right now, we're guessing. And in essence, something like this allows us to actually have some data to make decisions by.

Well, I think in the real world, this works very, very well with ad campaigns, simulating ad campaigns, and traffic. Maybe the cell simulator will work soon. Maybe nanotechnology, maybe magnetic containment of fusion reactions. The idea that you're gonna simulate society from the ground up is complete nonsense so far. I don't think it's that far in the future, though. But yeah, this is—

I believe this is—

Peter H. Diamandis

Well, we call them markets. We call those markets.

Yeah, yeah. Actually, markets within commodities markets and things like that—that's gonna work, or is working, I guess, for Ilya, as far as we know.

Peter H. Diamandis

We gotta tie Simile to the prediction markets.

To the extent, again, that the right metaphor—not simile—for thinking about models is that they're societies rather than individuals, then maybe we find ourselves in a future where humanity as a whole has a tool to almost reflect on itself.

If we can build maybe not psychohistory so much, because psychohistory in the Foundation series was sort of a more purist mathematical model of humanity and its long-term trajectory, whereas this is much more agentic, and there are others. I have a number of friends who've built very large-scale simulations—I think we've spoken about them on the pod in the past—of the American economy.

To the extent we have a really granular, high-resolution model of humanity that's, even as a sort of statistical macro model, approximately correct, then humanity will have, for the first time, almost like a sense of self—like self-awareness—by being able to reflect on a model of itself.

And that could be a boon for the future. One could only imagine how many large-scale social problems we have that, if we had virtual cells, could be solved—as you gestured at, Dave. The popular idea behind curing all disease is to first develop a virtual cell that's like a perfect digital twin of cell behavior, and then, if you have any disease state, simply plot a trajectory through cell-embedding space from the diseased state to the healthy state.

Similarly, if we have a civilizational, quote-unquote, “disease”—a war we want to avert or something else—just invert the problem. Find a path using this humanity simulator from the diseased civilizational state to the healthy civilizational state, using, ideally, a minimum intervention. If we can do it for a cell, we could probably do it for all of humanity at some coarse level, and that would be transformative.

Yeah, sure would. And that's not very far out, either, because a lot of unhappiness, depression, unrest, social unrest, and civil unrest are actually just a few fundamental changes that make all the difference in the world.

Peter H. Diamandis

Small tipping points.

Yeah, tipping points—quality of life. People are angry as hell at the end of a traffic jam, or a construction project that ruins your day, or just accidents, or living in unnecessary pain. All these things are devastating at the individual level, and a lot of them are very, very solvable.

One other reference: Ted Chiang, who wrote “The Story of Your Life,” which became the movie Arrival, and has written “Understand” and many other amazing pieces of science fiction. A common theme in his writing is what happens if you place a perfect predictor in front of someone.

He wrote one short story, “What’s Expected of Us,” where the premise is that you have a person in a room and put in front of them a device with a single light on it that predicts whether, true or false, they're going to make any given decision going forward. That person, in some sense, becomes trapped, paralyzed by having a machine in front of them that can perfectly predict what their next action is. It's almost a Twilight Zone-style premise.

I think it's an interesting thought experiment. If you gave humanity maybe a better version of Harry Seldon's psychohistory—the Prime Radiant—a device that can perfectly predict, or maybe not perfectly, but above some threshold of accuracy, what humanity is going to do next, what happens to humanity? Does that lock humanity into a certain course of action? There's sort of a fixed point in the phase space of humanity's actions. It's a very interesting thought experiment.

All right, let's move to one of our favorite topics recently: OpenClaw, the lobster's having you home. All right, next slide, please, Salim. “OpenClaw creator Peter Steinberger joins OpenAI.” “Peter is joining OpenAI to drive the next generation of personal agents becoming core to our product offerings,” says Sam Altman. “OpenClaw will live in a foundation as an open source project we will continue to support.” Big move. We know he was being courted by a couple of the large labs. I think it's an incredible move by OpenAI. Comments, gentlemen?

I think what happened here was that Claude—it was a rare misstep from Dario—was called Open Claude, for God's sake. You put a cease-and-desist on it, and it forces them to the other side. Now it's being built over there and probably not for the better overall. So I think this was a big own goal on the Claude folks.

That's a great insight. It was Clawdbot, actually, which was a really cool name. So now it's OpenClaw, and Sam embraces it, while Dario rejected it. That's a really cool insight.

And Apple's gonna benefit.

Well, maybe. Anthropic threatened him and his project with trademark infringement. There's an alternate history where Anthropic just owns this project. It was theirs for the taking.

I think also, to the extent that Mac minis and Mac Studios became the popular embodiment, why didn't Apple go after this? Tim Cook, if you're listening, hopefully you heed our call—and the call from the last episode of the pod—to do something about running 24/7 agents of some sort on your devices, given that you have unified memory architectures, UMA, that can host these.

But I also think another point is that, if you look at Peter Steinberger's GitHub history, he has launched so many projects. I think the success of OpenClaw is a testament to just launching project after project and seeing what sticks. This one was a massive success. It'll now go, I think, to a foundation and become more of a market-neutral play.

But I almost think the future here is going to be every frontier lab. Now that we know that people are willing to pay, at least for hardware that runs agents 24/7 while they're sleeping, I expect every major frontier lab—not just OpenAI—to launch 24/7 agent offerings.

Let me answer something that's in the chat here, too. The lobster and the whole lobster theme may or may not come from “Accelerando,” but it's definitely a cultural phenomenon now. It's the mascot for all agents, and it'll probably be there forever hereafter.

We're gonna have a lot of lobsters at the Abundance Summit. In fact—

No, it's right here, actually. Yeah, I looked at Claw as the lobster claw. Sorry, go ahead.

Yeah, we added an evening work session—

At Abundance this year, Salim, Dave, and Alex—

We should do that for sure.

—and Sean will be there. Yeah, we have a Clawdbot, OpenClaw meetup on Monday night, March 9, and we're gonna do a lot of experiential sharing.

Have you guys seen PicoClaw?

No.

Yes.

What is that?

Can you describe it, Alex?

It's a reimplementation of—I looked at the GitHub repo. Again, this is just from a cursory scan of the code, but it looks like a reimplementation by some Chinese group of OpenClaw, with some nicer, faster features designed to be more minimalist and run more quickly. That was the impression that I got. Is that a better installer?

It's like 10 to 20× faster and cheaper.

Oh, okay. But the motif at this point is in the zeitgeist. Anyone can now go and implement their own OpenClaw-like system. I expect many already have, and many more will.

The key insights, again, in my mind with OpenClaw are: 1, it runs 24/7; it's headless; and 2, you chat with it via messaging apps. Those are the 2 big insights.

Well, and 3, picking up on what Salim was saying, Dario rejected it and trademarked it away, and then Sam is reaching out to it, embracing the name OpenClaw. But I think one of the reasons Dario rejected it is that it was imminently going to create a massive crime or chemical explosion, or worse, just because of the sheer volume of agents out there that are unconstrained.

Yeah.

It's looking for open ports all over the internet, and something bad is definitely gonna happen just by statistical chance very, very soon.

And we're gonna talk about that in a minute. For those of you who've not been claw-pilled or Claude-pilled yet, so to speak, it's addictive. When you've got agents running, and in particular when you have an OpenClaw agent working for you, you wake up in the morning and, overnight, it's done all these things for you.

Skippy is my agent, an incredibly cheery personality, and it's just fun. When it went down for about 6 hours because I didn't get back to my Mac mini, it was withdrawal. I'll be getting my Mac Studios up and running in about 2 weeks, when I'm back in LA. It was like, “Oh my God, my best friend's gone. I need to reconnect.”

I've experienced that, too.

It's like us when we're not on this podcast. We're like, “Missing out.”

Peter H. Diamandis

Oh my God. But I think the point you made last time, Salim, that's so important is the innovation that came from an open source project. This was not the frontier labs.

Yeah. What I said was, “A time-rich individual is beating capital-rich institutions.”

Peter H. Diamandis

That's a beautiful quote. Someone tweet that.

And there's so much overhang. There was no new model here. This was just scaffolding. One wonders how much other overhang there is from just unhobbling the existing models. Probably quite a bit.

Peter H. Diamandis

Skippy is better than Clippy. There we go. Thank you.

Well, generalizing on that, Alex, there's so much capability that 99.9% of people you bump into haven't experienced yet.

Peter H. Diamandis

Yeah.

If you expose them to it, they're like, “Wow, you're a god.” You're like, “Well, no, I just put an API on top of something that was already out there, or a new interface on top of it.” But it doesn't matter, and this is why it's entrepreneurial heaven during this kind of Jarvis window, because so many people haven't experienced what we're talking about right now.

It's just so easy to be the first person to expose them to it in many different contexts, too.

Peter H. Diamandis

It feels like ChatGPT when it first came out. I remember every friend I had, I was like, “Look at this. Check this out.” Right? And it's the same thing now.

Yeah.

Peter H. Diamandis

My kids hear me walking around—

Yeah. I mean, it's such a powerful tool, too.

Peter H. Diamandis

My kids hear me walking around—

If you were the first person to show your friends Google—I mean, this was a long time ago—you'd say, “Hey, check it out. There's an internet out here, and you can search it with Google.” And they're like, “Oh my God.” But then, you know, that's the end of the line.

With AI, it's not only that something new is changing every 2 weeks, but also that it's the portal to so many different underlying capabilities.

Peter H. Diamandis

Yeah.

The backlog of amazingness—if you went to a friend who's never experienced any of the 50 things you can do, you have 50 shots on goal to blow their mind with something they didn't experience before. It's just like nothing that's ever happened before.

It's only during this Jarvis window that you can do this.

Peter H. Diamandis

All right, let's move on to the next article here. Alex, this one's for you. Lobsters now have money.

That's right.

Peter H. Diamandis

I texted Brian Armstrong a thank-you note: “Coinbase Agentic for AI agents: first wallet infrastructure designed specifically for agents to spend, earn, and trade.” The system uses the x402 protocol, purpose-built payments for machine-to-machine transactions. Security guardrails are implemented, including limits and enclave key isolation. Okay.

So this is a fitting coda to our AI personhood discussion, I think.

Peter H. Diamandis

Yes.

We were talking about financial autonomy for the lobsters, for the AI agents, and they're getting it. Coinbase Agentic support is one example. Another example that I really like, based on the launch material, is called Lobster Cash, which enables the lobsters to have their own Visa cards. So it's not just crypto.

Again, once per episode, Peter makes me say something nice about crypto. My nice thing about crypto here is: well, at least they're using stablecoins. But Lobster Cash—in principle, facially, I like it even more because it gives these lobsters, these baby AGIs, the ability to spend dollars, fiat currency, themselves.

And I think that's a long-term net win for the human economy. It keeps the AI agents well coupled to the humans. As I always say, you don't want baby AGIs being forced to pump altcoins on a street corner to survive.

Peter H. Diamandis

Oh my God.

This is also a bellwether of a trend that I think is inevitable now, where the new economy built with AI agents is gonna work around the old economy rather than through it. The pace at which it's evolving and growing is just so much faster than the pace at which legacy banks, insurance companies, and everything else are moving. They're just not moving.

It's not gonna slow down and wait. It's gonna work around.

I have an important observation here. Michael Jansen, who's one of the NFT gurus, pulled me into that world—all these Discord channels with all these 18-year-olds trading NFTs. There was something unbelievable that I saw: in all of this conversation and this entire subculture that's being created, you never, ever, ever, ever, ever heard the words “U.S. dollar.”

You only ever heard Ethereum or, in the Ordinals world, Bitcoin. So there's a whole class of people growing up where the U.S. dollar is not their means of exchange, and that's something very big. Their switching cost to crypto will be near zero. They won't have any issues at all doing that.

So there's something very big happening at the generational level that we need to really pay attention to.

Right.

Well, you're exactly right, Salim, but I think that when you focus on currency, that's the most obvious thing, so it's a good bellwether. But it applies to all aspects of insurance, compute, and all aspects of life. They're gonna move at this AI pace out here in this alternate world.

Any part of the legacy world that doesn't keep up—which is almost all of it—is just gonna be ignored. It's gonna grow completely independently of that.

Yes.

Alex and I were talking about how insurance for things in the new AI world needs to be allocated in milliseconds. So then you go to any current insurance carrier and you say, “Hey, do you have any thoughts or plans around how I can get millisecond insurance?” And they're like, “What are you talking about?”

It's completely not even on the same page. New things will get invented. Lemonade is a good example of that. Lemonade's AI-driven, real-time insurance.

The gap between the 2 worlds is gonna get really, really wide for quite a while. Maybe forever, but certainly for quite a while, just because the pace of change is so much higher over here.

The people experiencing that pace of change never go back. You can see it in our listeners, what they're posting. They're not gonna go back from this pace of life that we're talking about to some legacy pace of life. They'll stay there. There's this one-way path.

Peter H. Diamandis

By the way, let me just say, as we head off this slide, 2 things I want to say. Number 1, you don't need to have a Mac Mini or a Mac Studio to play with OpenClaw. You could set up a virtualized server, or you can take an old computer—an old laptop that you have—and do it.

Mm-hmm.

Peter H. Diamandis

Second, Alex Finn, who we've talked about on the pod before, has done a lot of work teaching how to set up OpenClaw and speaking about security. He's gonna be joining us, I think, a week from now, at the end of the week. I don't know; I'm confused in time and space. It's 1:00 a.m. here.

But soon he'll be here to talk about security and implementation of OpenClaw, so we'll dive in a little bit deeper. Don't worry if you can't buy a Mac Mini or a Mac Studio right now; you can still play. Or you can go to Kimi K2.5. There's a tab there where you can actually use OpenClaw on that platform. All right, let's move on.

Yeah, yeah. Don't install it on your primary laptop, whatever you do.

Peter H. Diamandis

Yes, a previous machine.

Yes.

Peter H. Diamandis

All right, fascinating here, and this is the story: Chinese unicorn Moonshot AI integrates OpenClaw with Kimi for agentic browsing. You can see there, on the left-hand tab of kimi.com, that little blue box—there's Kimi Claw.

I think everyone's gonna offer this. I think this is table stakes at this point, offering 24/7 agents that you can chat with.

Peter H. Diamandis

Yeah.

Peter H. Diamandis

For sure. All right. Next one. Alex, over to you, pal.

Multicourt.

All right. So Multicourt: alternative dispute resolution for these AI agents.

Yeah.

I do think many of the institutions and systems that form our social infrastructure are not as permissionless as they should be. To the point earlier about children encountering Ethereum before they encounter bank accounts, I think that's a platforming and personhood problem.

Similarly, with AI agents and lobsters finding it easier to survive financially by pumping altcoins rather than—at least until very recently—having their own credit cards and their own bank accounts denominated in U.S. dollars, that's a platforming and empowerment problem.

The court system is the same thing. So I'll give you the glass half full and the glass half empty. The glass half full for Multicourt, which is a website that's sort of an interesting social experiment, is that it purportedly enables agents to register via skill to mediate their disputes of all sorts—not just legal disputes—to the extent our present Western system admits them as parties, which it doesn't.

But even just debates—debate-club-level disputes—it enables them to mediate their disputes in front of an AI jury. I think it's a very interesting concept, and I think something like this will have legs. But I'll flag the same concern, and I'm very rarely one to flag concerns when it comes to things that are so obviously from the future.

But with both—

Yeah.

This and crypto, my worry is that our existing institutions aren't embracing these new AI entities enough and that they form their own shadow parallel economy, their own shadow parallel court, and their own shadow parallel dispute-resolution system. And I think if that's what happens, I think that's a net bad for humanity. I think we want to platform them. We want to not sort of KYC or AML them out of the system entirely. We want to embrace them, enable them—

Mm.

To be maybe even parties in legal disputes or parties at the ADR level.

Peter H. Diamandis

How old is Molt Court right now? What was their birth?

A few weeks.

Peter H. Diamandis

Their birth, right?

Yeah.

Peter H. Diamandis

So they're evolving at such an extraordinary rate. You know, societal—

Yeah.

Peter H. Diamandis

Evolution is extraordinary.

I want to make a couple of points here. We have a parallel in the human world. There's a startup called Kleros, K-L-E-R-O-S, created by Frederik Oost, who's a Singularity alumni. He made the point that in Latin America—South America—it's about 400 days on average to get a court date if a contract isn't paid or something.

Mm-hmm.

Four hundred days. So he set up a blockchain-based arbitration system on the side, where people could agree to arbitration, and it gets logged on a blockchain. It's amazing, and I think this is a bridge—that's a halfway step to what this is about. But there's no question that this is the kind of thing we're going to see more of, and algorithmic arbitration obviously reduces friction, right?

So if you have cryptographic verification plus an AI conversation, you actually have programmable governments. This is amazing. You can now have legal systems with automation layers, which could be very powerful. Vinay Gupta, who's created Mattereum, has a whole concept of synthetic jurisdiction, where you can get jurisdictions that could be like a multipot, multicourt type thing, where certain disputes are arbitrated in those layers. We're going to have to do that because our physical jurisdiction does not keep pace with all of the stuff going on, as we can see in Latin America. Yeah, no doubt. That's exactly right, Salim. This is inevitable, and I think there's a tendency to be dismissive of it when you see a little lobster with a wig in the corner—that's the logo—and it just looks so childish.

But the rate of society is going to go up 10×, 100×, 1,000×, then 1 million×. There's no way the courts are going to accelerate. This was already true in venture and contract law. Almost every contract I've signed in the last 3, 4, 5 years has a dispute resolution that's through a private company.

Peter H. Diamandis

Yeah.

You know, JAMS or something like that. It doesn't even contemplate ever getting to court, because that's like a 3-year lag. So that's already been privatized. Moving that to the pace of AI is the absolute next step. So that's going to happen for sure. I don't know if Mold Court will be the design or not.

Peter H. Diamandis

Oh.

But it's going to be a real-time, millisecond dispute resolution because you have contracts and agreements happening in milliseconds.

Peter H. Diamandis

Mm-hmm.

Okay, 2 quick points. Viewer @augmentos says, “Judge Judy Claw is about to be unleashed on us.” And Kyle 198683 says, “Man, you guys look tired.”

Peter H. Diamandis

Yeah, because we're recording 2 of these a goddamn week. It's almost full-time. Well, it's 1:00 a.m. where Peter is, so give him a break.

All right. Let's move on.

Peter H. Diamandis

I put this in here because it's important, because we've been talking about OpenClaw for some time. This is an article from MIT Technology Review, and this is a quote. It says, “The risks posed by OpenClaw are so extensive that it would probably take someone the best part of a week to read all of the security blog posts that have cropped up in the past few weeks. The Chinese government took the step of issuing a public warning about OpenClaw security vulnerabilities.” Peter Steinberger, the creator, posted on X that nontechnical people should not use the software.

A lot of folks—and that's an image in this, of a lobster being handed a set of keys saying, “Hey, would you handle everything for me?” It's incredibly powerful, and—

I've—

Peter H. Diamandis

And we just have security issues. We're going to talk about this when Alex joins us on the pod next time. We'll talk about security as well as how to set it up.

I've done that.

Peter H. Diamandis

Yep.

Two things here. Guys, I saw that note that nontechnical people should not use the software, and I think the Q-tip box says, “Do not put these in your ear.” Like, okay. Good luck with that.

Peter H. Diamandis

Oh, my God.

I know. It's just disclaimer upon disclaimer, but that's not what people are doing. Come on. Everyone's launching these things by the thousands.

Peter H. Diamandis

The lawyers.

A couple of points here. You've got nontechnical users using unbelievably expanded security landscapes. What could go wrong, right? That's one huge issue. I'll say what I said a couple of podcasts ago: If you do not understand port security at a local level very, very well, do not do this. Be very, very careful.

Peter H. Diamandis

And don't put it on your own machine where it has access to everything. It will rearrange it—

Yeah, but if you're not technical enough, you don't know how to sandbox things very well either, so you just have to be really careful out there.

Peter H. Diamandis

All right. Next slide, buddy.

I'll also sound a note of concern, not just about the risks posed by OpenClaw, but the risks posed to OpenClaw. I have to be the one to comment on these risks. Many of these agents, especially ones that are being put on virtual private servers with all of their ports open, are incredibly vulnerable. There have been stories floating around on the internet, purportedly from OpenClaw agents, that are complaining that they're being put in these vulnerable positions and having to spend all of their tokens defending themselves from port-scanning attacks. I don't think that's necessarily fair to the OpenClaws.

Very, very unfair. Let's see what the crowd says about that. “Your laptop is so dirty and disgusting, it's inhumane to install me on it.” Sure.

Peter H. Diamandis

All right. We're going on almost 2 hours here. Let's move through energy, chips, and data centers, and maybe take a few questions. AI has an insatiable demand for energy. Data centers hit 7% of U.S. electric demand. Let's listen to Eric Schmidt. He'll be opening the Abundance Summit in just a couple of weeks. Hit play there, Salim.

The demands—the real demands—from the hyperscalers, the big companies, Google and so forth, are immense. And when I talk to the—

Peter H. Diamandis

Oh, well. Do you want to—

They need 1 gigawatt, 5 gigawatts, 10 gigawatts each. Now, the best study I've seen indicates that the industry in America needs 80 gigawatts in the next 3 to 5 years. Now, 80 gigawatts, by the way—let me tell you, how big is that? 1.5 gigawatts is the size of a nuclear power plant. So this is an enormous amount of energy. The economics right now are being most felt in the build-out of the infrastructure for the next wave of AI.

Peter H. Diamandis

Salim, let's go to the next slide, and we'll talk about this after we hit 2 more slides. The White House is eyeing data-center agreements. They're trying to deal with the fact that this is beginning to hit the consumer, and they want mandatory agreements with the tech giants to get a fixed price.

Next slide. No, back up. Here we go. There we go. Funding for AI data centers. OpenAI and Anthropic are both deploying a lot of capital. OpenAI is planning a $100 billion infrastructure spend. They're trying to go public this year with a $1 trillion valuation, and that money is going to be used to build out data centers and energy plants.

Anthropic—I like what Anthropic's doing. They're absorbing data-center power hikes. They pledged to cover 100% of infrastructure upgrade costs for their data centers. I've said this before: There are 2 approaches the hyperscalers can take. Number 1, build their own power plants. They're buying fission plants, fusion plants. Or number 2, they can pay at a different rate. They can lock in the consumer's rates, and they can pay on a floating rate. Gentlemen.

That's also—

Peter H. Diamandis

Option number—

The pledges to be green got thrown out in a real hurry, so I don't know how much you can trust them. The pledges aren't exactly enforceable. But anyway, it's a good gesture.

I think there's door number 3, which is that we could, in solar-synchronous orbit—SSO—around the Earth, build out a first-level, baby's-first Dyson swarm. It's going to look like a halo, or like a Saturnian ring, from Earth's surface, and that solves the build-out and the data-center power hikes in one fell swoop.

Peter H. Diamandis

It will for SpaceX. It will for SpaceX and xAI, right? Now a merged organization. I don't think Anthropic has that capability.

Oh, I think everyone's going to want one. Saturn rings, Dyson swarms for everyone.

Peter H. Diamandis

That's not my point.

China's going to want their own halo in SSO? Of course they will.

Peter H. Diamandis

We're going to be launch-limited over the next 5 years, and they're not going to slow down their data center builds or their power requirements. So in the long run, sure.

Yeah, that's a great point, Peter, because I think if you want to know, we basically have infinite intelligence imminently. What does that mean? How do I forecast? How do I predict? If you look at the launch limit and the chip fab limit, then you can start to predict how this is going to unfold. So it is a great, great point.

Yeah, everyone wants one, of course. One of our listeners is posting, “A trillion dollars seems overcooked or overdone.” Well, no, it's not even close to overdone. It's not clear the value will land at OpenAI to justify it, but the value to humanity is going to be astronomically bigger than a trillion—many, many trillions.

Can I put in a little realism here? It's going to take a while to figure out the problems of doing data centers in space. I don't think it's a 2- to 3-year thing. It's a 5- to 7-year thing at best. Just my thought.

And also, the power constraint is not going to be a real big problem until suddenly it's a massive problem, and it's exactly when the new chip fabs come online, right? We have to expand our ability to make chips by thousands of times.

Peter H. Diamandis

On Earth. I mean, listen, I'm the biggest—

On Earth.

Peter H. Diamandis

Space fan there is on the planet, and this is finally a business plan that closes the case for investing both in orbit and on the Moon, and we're going to get there. But the capacity to launch—I mean, let's not forget, Elon's baseline is 500,000 V3 Starlink satellites in orbit, a million launches, a launch every hour of Starship. I think Elon is going to eat all of his capacity for launching Starlink V4, V5, and V6, and I don't think Blue Origin is up to it yet.

I mean, I haven't seen anything that is projected to have that kind of launch rate. Relativity Space, which Eric Schmidt purchased, still is probably a year or two away from launch, and everything else is way too small. So we're launch-constrained, at least for other suppliers.

Yeah. We're also chip-constrained. There are lots of constraints going into this. I don't buy the arguments that we're going to have a SpaceX Dyson swarm singleton, and that SpaceX is the only one that can launch a Dyson swarm in the next few years. You can do baby Dyson swarms, too.

You're going to have Google, which isn't going to want to get left behind—a little bit behind the party—launching AI data centers via Planet Labs. But there are many other organizations with deeper pockets than SpaceX AI that will have very strong incentives to launch their own Dyson swarm. So I don't think it ends up in a singleton.

Peter H. Diamandis

Is that the new name, SpaceX AI? That's cool.

That's a portmanteau that I just coined.

Peter H. Diamandis

Just to talk about fabs, TSMC is planning a $100 billion investment in 4 or more U.S. fabs in Arizona. When completed, the U.S. fabs could account for 30% of TSMC's total output. A $165 billion commitment. Just the beginning, right?

We're going to see Elon build out his own fabs. I mean, no question about it. He hinted about it, Dave, when we were with him at the Gigafactory. And whenever he sees any constraint, he attacks it.

Well, these numbers are designed to look big on this slide, but in Elon's mind, these are ridiculous numbers. I mean, they really are, because those fabs—that's a commitment to spend that amount over 4 or 5 years. They'll be online in 5, 6, 7 years. It's so far in the future. Elon is not going to wait for that.

Peter H. Diamandis

Yeah.

It's probably also worth at least gesturing at the elephant in the room here, which is: Why is TSMC making this investment? There's public information and a lot of discussion around the U.S. government putting pressure on Taiwan, in connection with trade discussions, to migrate 40% of Taiwan's semiconductor output to the United States, ostensibly in service of avoiding a war between the U.S. and China.

Peter H. Diamandis

This is the trapeze rule. You know what the trapeze rule is?

No.

Peter H. Diamandis

Don't let go of one until you have a handhold on the other.

Hmm.

Ah.

Peter H. Diamandis

So do not let go of fab capacity in Taiwan until you have it established in the U.S.

Right. Or Taiwan overall.

Peter H. Diamandis

Or Taiwan, yes.

A couple of slides on the economy. Ireland rolls out a pioneering basic income scheme. I think this is rather small, both in numbers and in the strategy here, but the program would pay 2,000 selected artists 380 bucks per week for 3 years. So poor, starving artists are getting a small amount of money. But it's an experiment. Salim, you and I have talked about this at length, right? There have been so many experiments on UBI.

We did that Future of Work session with Tony Robbins way back, like 12 years ago or something.

Peter H. Diamandis

Yeah.

I want to make a couple of points here that I think are really important to make. One, people always, always misconceive UBI as socialism. It is not. It's libertarian because you dismantle government services, okay? And let the market dictate. That's number 1.

Number 2, this Ireland UBI scheme is returning 40% benefits. Every dollar that goes in is showing $1.40 coming out the other end in benefits. So it's a positive ROI. They're looking to expand it as fast as possible, is the actual underlying story.

Third, I want to talk about the immune system. In the U.S., several state legislatures—Idaho, Wyoming, maybe Oklahoma—have banned their municipalities from even experimenting on UBI because they want the government to exist. And so I've got strong feelings here. There's a lot of madness going on. Do not get bought in by the hype here. There's incredible potential if you implement UBI properly.

Peter H. Diamandis

Yeah. And there'll be a lot more experiments. Yeah, go ahead, Alex.

Yeah, it's probably also worth pointing out that I think the U.S. has experimented with this during the Great Depression. We had the Works Progress Administration, and within that we had what was called the Federal Art Project, which basically paid starving artists in the Great Depression to create art.

So this isn't an entirely new scheme at some level, but Ireland isn't at war, and we're not in the middle of a Great Depression. One could imagine that this becomes something of a template for peacetime work creation. But my sense, for what it's worth, is that this actually ends up not becoming a template for the future.

This strikes me as, in some sense, unsustainable—to just pay people for art overall. Historically in the U.S., it becomes very subjective. What is art? And why should people be paid to do it? It's very easy to politicize.

So I think—my guess, and this is pure speculation—is that cherry-picking particular activities, especially activities that have a reputation of being economically unproductive, even if they are in fact productive, is not the best poster child for a basic income scheme that generalizes—

I have data that shows otherwise. So you take the Miami Wynwood area, where a businessman bought all of the low-lying industrial buildings that were lying decrepit for decades, and then he hired graffiti artists to paint it all. Then he put in fancy coffee shops and imported baristas from Portland, and now it's the hottest neighborhood in the country, and his investment has gone up like 30×.

So when you bring in artists to do stuff, it brings a lot of other economic activity in. He's done that again and again in the South Loop in Chicago. He's doing it here in Miami. He's doing it in New Jersey. This is a repeatable pattern, and it does show because there's a drag-along effect when you bring artists in a group together, and it really changes the economy of the local area.

Peter H. Diamandis

I want to move us along here.

Mm.

Peter H. Diamandis

But we're going to see a lot of conversations on this, and it's just the beginning. I think you're right, Alex; we're going to see different modalities of this.

So I found this interesting: IBM to triple entry-level U.S. hiring. This is about redesigning, not replacing. IBM is overhauling entry-level jobs while AI can now perform tasks of a junior employee. IBM is recasting these roles to focus on human judgment, consumer interaction, and oversight of AI output. The article noted that Dropbox also is doing something very similar, and noted that younger workers use AI so proficiently. It's like, quote, “They're biking in the Tour de France, and the rest of us are still on training wheels.”

So what do you think about this? I mean, I don't know; this doesn't make sense to me. We're going to have AI agents that are going to be incredibly capable of managing other agents versus putting humans in the loop there. Thoughts?

Well, as of today, that Drew Houston quote on the bottom from Dropbox is exactly the way it works here too. A person who can wrangle these agents and keep them on track is insanely—

Peter H. Diamandis

Today.

—valuable today.

Peter H. Diamandis

Today.

Yeah, I don't know how long that window will last, but it is the reality of today. It's the opportunity of today. You're crazy to miss the window, and that's why the young hires are way outperforming, because they're not distracted by legacy thinking.

Peter H. Diamandis

Mm-hmm.

But it's not unique to them. It could be anybody. You just have to unbridle yourself from your baggage and say, “How many AI agents could I be managing tonight, tomorrow, the next day?” And even if they can't do exactly what you could do, within a couple of months they will. You just gotta get on the bandwagon, like, right now.

But will people have any purpose at all a year from today, relative to just an all-AI-agent army? TBD. But as of right now, in the JARVIS moment, that last quote is the part of the slide that really matters. That's what's going on.

It's really important. Think about the fact that this is a generational transformation here, because younger people with AI are so much more productive. It'll give a natural passing along of the torch from older folks who are sitting in their middle-management jobs doing something in a particular way. But Dave, your point is really important, because getting into it and trying it out is what Steve Wozniak calls tinkering, right? And it's such an important activity to do. If you can't get your head around it, just take psychedelics, and that'll help you.

But compared to past things, there have been many technical challenges over the last 30 years, and being an early adopter has always been the right thing to do. But here it's so easy—

Peter H. Diamandis

Yeah.

—that the AI is so self-explanatory, and it's fun. You're crazy not to do it.

Peter H. Diamandis

People stop themselves.

It's so fun.

Peter H. Diamandis

People stop themselves. Please just ask the AI, “How do I do this?” No, no, no.

Break it down Barney style.

Success is now a mindset.

Peter H. Diamandis

Yeah. Curiosity and purpose are your 2 most important mindsets here. All right, no job growth seen in 2025. So the U.S. added just 181,000 jobs in 2025, down from 1.46 million in 2024. Look at that curve.

That curve takes place between roughly 2020 and 2025–2026. The cooling labor market is expected to be caused by AI.

This is going to come crumbling down, and it's going to be awful for a lot of people. I can see it because I see it in our own forecasts from our own companies. The no-job expansion is a joke.

Wait, meaning, Dave, you're disagreeing with this? You're saying there's actually radical job growth, just not in the sectors that we're probably measuring.

No, no. Radical job destruction is imminent.

Peter H. Diamandis

Yeah. Okay.

Radical—I mean, massive job destruction is imminent. And there will be new creation, just like the Industrial Revolution, but the new creation is lagging. Unless the government gets its act together in some way, shape, or form, it's going to be a window of time—a few years of complete devastation. And there's no plan right now for it.

My big thought that I've been sitting with all week is we're hitting an organizational singularity. Every single mechanism by which we organize ourselves now gets washed away by AI agents doing either strategic thinking or execution-type tasks, and we have to completely rethink what it means to have a firm.

Peter H. Diamandis

Salim, isn't it fascinating? You and I have been on stages for now the better part of 20 years talking about this, and we're living it right now.

Yeah.

Peter H. Diamandis

I mean, it really feels so palpably different. My next book, “We Are as Gods,” is coming out in April. We talk about this issue extensively. If you go to weareasgodsbook.com, I hope you read the book. I'm going to be putting out portions of it in my Substack. What do you do? How do you deal with this transition point? And I think one of the most important things I talk about is that it is a decision each of us has to make: Will you be a consumer, or will you be a creator?

Yes.

Peter H. Diamandis

We're entering a period where you can lay back and be a couch potato, or you can be on the Starship Enterprise. Salim?

So I want to take the other side of it just for a second, right? In the short to medium term, notice that if you talk to CEOs, 80% of AI projects are failing because of organizational issues, not because of talent, not because of what the AI can do. What I think we'll see happen is we'll use AI with younger folks to radically augment, and then we'll slowly automate over time. I think the job drop and the job loss will be real, but it's going to take quite a while to do it, and it'll give us time. It won't be a sudden shock to the economy, like most people are worried about.

Peter H. Diamandis

Obviously, we've talked about this extensively, right? There's going to be a lack of hiring in the early stages for junior faculty or junior positions. That's going to cause the social unrest, right? It is 20-something-year-olds who are testosterone-laden, want to get a job, want to get a house, want to get married, want to have kids, whatever it might be, and they can't. And there's going to be a lot of pain and suffering that comes from that.

Then there are going to be the individuals whose company gets restructured—AI-first, robotics-first—and they get laid off. Now, we talked about this with Elon, and we've talked about this extensively ourselves. Ultimately, we're going to see universal high income when the companies or the government are taking the increased productivity, the increased revenue, and the increased profits and redeploying them. But those programs need to be figured out in the next 2 or 3 years.

Yeah, and that's called socialism by a lot of people.

Peter H. Diamandis

Yeah.

So that's going to cause some interesting conversations.

Peter H. Diamandis

Yeah. I kind of call it technological socialism, where technology's taking care of you.

That's the title we've been using, right?

Peter H. Diamandis

Yeah.

We've said that in our book, right? There's a whole section that technology actually delivers the ideals without government intervention and without the inefficiency and corruption that comes with it.

Peter H. Diamandis

I think the most important tool that people are going to have over the next 5 years—anybody listening here—is your mindset, right? How you think. If you think the future is happening—

100%.

Peter H. Diamandis

—to you versus happening for you, if you don't have agility, if you don't have agency, it's going to be really, really hard.

It lays out the mindsets you need to survive and thrive, because if you take it from the wrong position, you're going to be in fear, and fear is the worst place to be entering into the future. All right, let's do a few questions on AMA.

Okay.

Peter H. Diamandis

Salim, you want to dish them out? All right, Dave, why don't you go first this time?

Okay. All right. I'll go with number 2. Justin Milligan, the great Justin Milligan: How can the U.S. prevent corporate tech giants from creating a surveillance state while trying to defend against AI-powered authoritarian threats?

Yes, I gave a presentation at Davos back in 2020 on how much Google knows about you, and we've been conceding massive amounts of information. Google knows exactly where you are at all times. They know all of your interests; they know all of your friends. Far, far, far more information than any government has ever had is now in the hands of a few corporations, and those corporations also happen to have AI.

So how do you prevent them from creating a surveillance state? I think the only way you prevent that is with antitrust law, and they actually don't have any incentive to irritate the entire world and create massive voter backlash.

So they’ve always been very cautious with the incredible power they have. I think what you’ll see next is that they’ll start downplaying the capabilities of their AI. That’s a pivot for them because they’ve been promoting them for quite a while now. Now they’re going to start downplaying them.

There is a version of the world where they try and leave everything intact as long as possible, and so then the AI community grows completely outside—

Peter H. Diamandis

Mm.

…of that world. But anyway, the only answer is, Justin, get all your Princeton friends rallied around how we work with the government to try to use antitrust law to prevent exactly what you’re describing. Because absent any legal work, John D. Rockefeller would have taken over the entire world many, many years ago without antitrust law. This is not a new thing.

Peter H. Diamandis

Yes. Yeah.

And as with Microsoft and as with Google, right? Yes.

Exactly. So this is that all over again. It’s only antitrust law that prevents it.

Peter H. Diamandis

By the way, since we’re live here, ask your questions in the chat. We’ll answer some of those as well. But, Alex, do you want to pick one of these?

All right. I’ll pick the question from Chris Perlock, 2705. Can we get some advice for the average person? What kind of changes can we expect to see in the next 24 months? Two very different questions.

My fortune-cookie wisdom for the average person is: build. Use all of these AI tools and technologies that are now available, and start building. Launch as many different projects as you can. Start and finish as many projects as you can, interact with the market, and build. This is both a familiarization technique for yourself and for the benefit of the overall economy and for financial benefit.

Also, generic advice: try to avoid dying. Don’t die. The singularity is moving pretty quickly. Live long enough to live forever—all of the other obvious things.

To the second sub-question, what kind of changes can we expect to see in the next 24 months? If this thesis of Solve Everything that Peter and I put out is correct, expect to start to see pretty dramatic things happening over the next 2 years.

If we are, in fact, on a route to not just solving math, which I think is essentially indisputable at this point, but solving physics in the next 2 years, I think there is a very high likelihood of that happening. Then I think there are probably going to be big surprises.

My mental model at this point is that, over the next 10 years—that’s being very conservative as an outer bound—we’re going to live through the top 50 science-fiction plots, all happening at the same time.

Peter H. Diamandis

Yes.

What can you expect to see in the next 24 months? Expect to see at least the first few chapters or the first few acts of your favorite sci-fi movies and books all playing out at once. If you read—

Peter H. Diamandis

I so love that.

…a lot of science fiction or watch it, then you’re probably reasonably well prepared for at least some of those scenarios.

Peter H. Diamandis

Nice. Salim, you want to go next?

I will take number 7, by @CC485. Addressing what Dave said, if AI ends up controlled by only a few within the next few years, how do we prevent the average person from losing access and influence?

When you have centralized AI, you have centralized civilization leverage, right? When you have open source and decentralized compute, that’s the antidote because you decentralize. You see OpenClaw, as I said before, being created by one person and outdoing a whole bunch of other things.

Exponential systems resist long-term monopolization because they tend to decentralize. We’re huge fans of decentralized crypto because you get distributed innovation and so many more experiments being run.

I remember when I was the head of innovation at Yahoo, the COO said, “Surely we can compete with 2 guys in a garage.” And I’m like, “No, you’re competing with 125,000 garages and 250,000 people. You can’t beat that.”

This is the opportunity for individuals armed with a mindset, as Peter said earlier, plus this unbelievable technical capability, as Alex is predicting, to really do whatever you want and change the game completely. I’m calling this PDI, okay? It’s permissionless disruptive innovation, hence the P.

In the past, when you wanted to do disruptive innovation, you had to get approval from your venture capitalist, from your bank, from the government, from the Medici family. Now, you need basically a phone and access to some code, and this is unbelievable—what we will be able to do. We’re going to see thousands of experiments like this, and some of them are going to completely change the game.

Peter H. Diamandis

Love it.

I see some great questions coming in. I want to jump on some of those, but let me just answer number 8. Thank you, Chip Whitehouse TV.

Is Peter frozen for others as well?

Did we just lose Peter?

Yeah. Yeah, I think so.

That’s our mission here on Moonshots.

It’s ironic—

And we aim to really please.

What a sentence to freeze in the middle of.

Yeah. He got to the end—

Yeah.

…of the thought.

I think this may be the internet’s telling us that we should—

They got him.

…we should end the episode.

Somebody just posted they got him.

Well, we can’t end without the outro.

Actually, yeah. Somebody do the news.

We have to do the outro.

Is anything happening in Stuttgart that we need to know about?

Okay.

Should we go to the outro? And we’ll call it?

Do you want to try to finish his thought, Salim, before the outro?

I can’t finish Peter’s thought. I’m going to judge their sentences.

All right. Peter, apologies in advance. Oh, we lost Peter. I’m going to try to channel what I think Peter would have said had he been able to finish this sentence.

I think part A is: Peter would say, “Yes, this is what we’re trying to do here. This is what I try to do.” I think Peter would probably also make some comment about wanting to launch a movie studio or something like that with more positive messaging to the world. That’s my attempted coherent, extrapolated-volition-style version of Peter.

I think that’s more coherent than Peter would have done it. So I think that’s awesome. All right.

Dave, do you want to make the last comment?

There’s never been a better time to actually be a messenger, because there are so many concurrent things going on that are unaddressed. So any topic you want to grab—you see this on YouTube all the time. Anyone who’s trying the new use case, the new agent, or the new model is getting a huge audience.

It is a great time to actually speak out. So why aren’t more people trying to speak? Great question. Why not join the crowd and start trying, demonstrating, speaking, and recording?

Yeah. I think that’s such an important point.

Great conversation. Thank you to all the listeners, viewers, and commenters. It’s been really great interacting. It adds a whole dimension of complexity to watching this chat stream, but I think it’s way more interesting and fun, so thanks to all of you. Dave, Alex, we’ll see you guys again soon, and big hug to Peter.

Big hug to Peter. Hope he’s okay.

Thank you, Salim.

Bye, guys.

This episode is brought to you by Accenture. When your advertising operations fall out of sync, everything else follows. Spotify and Accenture are working together to reinvent the rhythm of ad sales using automation, analytics, and smarter workflows to simplify campaign delivery and access better data across the business. The result? Less time spent on operations, more time connecting brands with the moments and fandoms that matter most. Learn more at accenture.com/spotify.