Google I/O 2026, Karpathy Joins Anthropic, and Cerebras’ $95B IPO | EP #256

If you said 5 years ago, hey, Google's going to 6x its CapEx and the stock will go up, nobody in their right mind would have said that's even possible.

Quadrillions, billions, hundreds of billions, trillions. It gets numbing after a while. There was a lot of conversation that Google was cooked, Google was not going to make it, that their revenue engine was being massively disrupted. And here they are, you know, sort of disrupting the disruptors.

All of these numbers, I think, were inevitable.

Andrej Karpathy joins Anthropic. He was the co-founder of OpenAI. He left in 2017 to run Full Self-Driving for Elon. And he'll start a new initiative focused on using Claude to accelerate Claude's own pre-training research. Cerebras record IPO closes up 68%, market cap $95 billion. Andrew Feldman, the CEO of Cerebras.

It's like a lifetime achievement, like kind of like a Nobel Prize or an Olympic gold medal. Where you carry it for the rest of your life. All right.

I see Andrew Feldman has entered the room. Andrew, a pleasure to have you here.

Thank you for having me on your show. Appreciate it.

Now that's a moonshot, ladies and gentlemen.

Everybody, welcome to another episode of Moonshots. Today we have our extraordinary group of moonshot mates: DB2, our emperor of AI investing, Saleem, professor of all things exponential organizations, and our very own artificial superhuman, Superintelligence, our moonshot mate AWG. I'm Peter Diamandis, your host. Gentlemen, a pleasure to have you here.

And Peter, you're a birthday boy as well. Shall we sing? Happy birthday to you.

Everybody's going to be signing off the pod right here, right now.

We are not singers. Stick to the verbal.

Well, one of us was. Well, my first career was in the New York City Opera Company.

Uh, you know, really, you just seriously impressing all of us. Well, thank you, gentlemen, for your— for your—

we want to hear— I would like to hear an aria sometime.

All right, maybe I'll do an outro.

Okay, uh, uh, for sure.

So we had such a fun surprise birthday party for you.

Oh my God, it was crazy. I have never been so surprised in my life. And honestly, you, Salim, are the most giving individual. So for So Salim was visiting with Lily and his son Milan. We had a birthday dinner for him Saturday night, but his birthday was Sunday. And he walks me along the beach to a surprise birthday party where there are 50 people. I walk in this room and I've never been more surprised in my life. I mean, I literally dropped to my knees in level of surprise.

All right.

But it was your birthday, Salim.

And you were—

It was a perfect decoy. It was a great decoy. It was awesome.

Oh my God.

Very, very special. Very special. And Dave, you sent a great message over.

It was great.

It was wonderful. Yeah.

Thank you all for that. So welcome to Moonshots. Our job here is to get you pumped about the future. No politics, no doomerism, just the science and technology driving us along the singularity. Today we have a special episode, our annual recap of Google's mega event called Google I/O. We'll cover news with Andrej Karpathy joining Anthropic and Elon's defeat in the trial against OpenAI. And finally, we'll be joined by by Andrew Feldman, the CEO of Cerebras, after an epic IPO. So Dave, it was a blast to be with you at Google I/O. Here's an image of us along with Tyler Donahue. What'd you think of it?

Yeah, it's great to be where it all started. You know, the vibe on AI is, you know, it's global, but within this epicenter where everything began, it's like, it's off the charts. And you know, the first hour, just the amount of stuff to talk about in 1 hour compared to a year ago, compared to 2 years ago. I mean, we're taking it all for granted, but it's just crazy, you know, just, just the number of new Google brands, uh, of new AI products is— it's pretty baffling. So we'll go through it all, you know, we've got it all beautifully cut up today so you can analyze every piece of it.

And, uh, and Alex, you were watching online, weren't you?

I was, I was watching in real time, dissecting it all for my newsletter, The Innermost Loop. I thought there were some high points, some low points, some midpoints, if I to say and eager to dive in.

And some probably some no points. And of course, where's Waldo today, Salim? You know, you're a probability function on planet Earth. So where are you today, Salim?

Let our viewers know. I just landed in Brazil and I've got a bunch of meetings and presentations here. So I flew from LA with you, Peter, directly here.

Of course, of course you are. Let it not be said that you're not the traveling dude.

Peter, I'm at the Cerebras headquarters today, too. The vibe here, you know, third biggest tech IPO in history. The vibe here is just— I couldn't resist the opportunity to feel it. You know, right after an IPO, a company, it's a once in a lifetime kind of thing for most people. So the vibe is just, just epic.

Yeah. A record IPO until the next record IPO. Until the next record IPO.

I know it's going to be— it's going to be 10x this year. So you got to savor it while, you know, while you have the record.

It's insane. All right, let's jump into all things Google I/O. I want to kick it off with the opening summary by Sundar. Quite the year. Let's, let's listen to Sundar and then we'll, we'll continue on and dive in.

Two years we were processing 9.7 trillion tokens a month across the surfaces. A huge number. Last year at I/O, that grew to about 480 trillion tokens. Tokens. And fast forward to today, that number has jumped 7 times to 3.2 quadrillion tokens per month. Over 8.5 million of you are now building new apps and experiences with our models monthly. And our model APIs are now processing around 19 billion tokens per minute. We are of course also seeing incredible demand across our products. We now have 13 products with over a billion users each. 5 of those have more than 3 billion users. AI Overviews now has over 2.5 billion monthly users. And AI Mode has been a revelation, our biggest upgrade to Search ever. People love it. In just a year, it's already surpassed 1 billion monthly users. Last year at I/O, the Gemini app had 400 million monthly active users. Today, we have surpassed 900 million, more than doubling in a year. And to date, more than 50 billion images have been generated with our Nano Banana models. In 2022, we were spending $31 billion annually in CapEx. This year, we expect that number to be about 6 times that, approximately $180 to $190 billion. Instead, we can now seamlessly distribute training across multiple sites, scaling across more than 1 million TPUs globally. This gives us the ability to create the largest training cluster in the world. Both chips are more energy efficient, delivering up to 2 times better performance per watt.

Wow.

Quadrillions, billions, hundreds of billions, trillions. It gets numbing after a while.

It does. But you got to— you got to step back and it's like 6x-ing your CapEx. If you said 5 years ago, hey, Google's going to 6x its CapEx and the stock will go up. Nobody, nobody in their right mind would have said that's even possible. And, and there it is. I mean, those TPUs are, I mean, that's the linchpin. And if you know, they're 2x the power efficiency, but they're, they're in the hunt now competing vertically from the, you know, from the transistor all the way through the user experience. And no one else can say that. So yeah, there's just a lot. Also, you know, if you look at the array of logos that have over a billion users, I mean, you have to go back, rewind the, rewind the video, go back and look at that again. It's just a litany of Google things now. Yeah, that's relentless.

The Gemini having 900 million users is pretty incredible because that's pretty close to ChatGPT. Uh, I found that very striking.

It is. Alex, what's your take on these numbers here?

Predictable thoughts?

Easy?

Well, inevitable in some sense. I'm reminded about 20 years ago, I had a conversation with Larry Page during his interregnum when he wasn't CEO. And I was reminded he was asking me for advice on how to get Google interested in spending $100 million to work on AI, if you can believe that. It's unconscionable by today's standards that Google wasn't interested 20 years ago in AI. And now here they are. It's become the central focus of the company, full stack from chips and data centers all the way through applications. Applications, but here we go. All of these numbers I think were inevitable. It's been widely remarked that if Google hadn't leaned in at every layer of the stack to trying to own or at least lead in AI, they would've been toast. The original model of Google based on search and ads would've been cooked. So what do you do? You lean into it. Gemini, 900 million plus users. I think that's perhaps just slightly less remarkable than it might seem given that Gemini is basically being swapped for Assistant. Assistant already had quite a bit of traction, but nonetheless, nice to see that Gemini usage is taking off. It's nice, I think, on balance to have something other than a duopoly between OpenAI and Anthropic, which I think if Google doesn't aggressively lean into consumer and enterprise Gemini adoption, I think that's the default outcome right now.

And Alex, you know, and I do as well, 'cause again, I've known Larry since, 2003, 2004, AI was always his focus. It was— he wanted to build an AI company from the very beginning and had a lot of difficulty for a while.

I mean, it wasn't obvious to everyone else within Google for the first part of its life that AI was where this was all going.

And of course, Eric, the adult supervision in the room, came in and built the, you know, the revenue engine. And of course, Google's only able to do what it can do today because of the massive revenue. The other thing that people don't realize is that Part of Larry's vision early on was also BCI. He wanted to connect the brain to AI.

And space stations. Peter, do you remember that whiteboard that Google used to maintain with their long-term tech tree?

Sure.

And they were going to have a Google space station and, and BCIs and all of these things. That's all happening now. Finally, like 3 decades later, the, the Google whiteboard vision is finally playing out.

Amazing. That whiteboard actually had the tethered satellite on it. You know, they're going to make an elevator into space.

A space elevator.

A space elevator, yeah. And they needed a carbon nanotube wire for things to go up on. They manufactured about a meter of it for a lot of money. I was with Jack Hittery. 20,000-mile cable.

We're finally catching up with the whiteboard.

I was with Jack Hittery at SandboxAQ yesterday talking about what the large quantitative models, the LQMs, are going to be able to do. And one of his objectives are new materials that have the tensile strength to give you a space elevator. So, you know, all of— again, what do we say? What did you say, Alex? We're going to speedrun every science fiction movie ever made in the decade.

Over the next 10 years, like every sci-fi trope everywhere, all at once over the next 10 years. That's the singularity.

Just a big shout out and congrats to Sundar and Sergey and Josh and the team there. I mean, hitting their numbers again, you have to remember that a year and a half ago, there was a lot of conversation that Google was cooked, Google was not going to make it, that their revenue engine was being massively disrupted. And here they are, you know, sort of disrupting the disruptors.

I mean, this is an AI-native operating system company because they're now constantly continuous sensing, execution, adaptation, and they've kind of hitting an inner loop there.

This was the company that birthed the transformer. This was the company that my friend John Smart, uh, I think perhaps insufficiently famously pointed out that if you looked at the number of words or tokens in an average Google query over a period of 15 years— and this is before everything hit its inflection point in 2017 or so with the transformer— but if you looked at the number of words per Google query, it was following an exponential curve that was inevitably going to end up in people having full conversations with AI. So Google had the exponential trajectory of user interaction. They had the transformer, they had the compute. It was just a matter of putting all of the institutional pieces together, and it seems like they're finally coming together.

Yeah, amazing. Moving us along after that epic intro by Sundar, the company Google is launching an entirely new family of AI models called Gemini Omni. It's capable of generating video clips from prompts that include a variety of inputs, including text, photos, videos, and audio. You know, Google says Omni will be the create anything from any input product. So let's take a look at the video here being introduced by Demis. And Demis was a rock star on stage. It was so much fun to see him there.

I'm excited to announce Gemini Omni. Models like Veo, Nana Banana, and Genie are able to create extremely realistic videos, images, and interactive simulations. It's a step change in simulating things like kinetic energy and gravity. Gemini's world knowledge and reasoning really shine in Omni. It can translate complex ideas into highly accurate videos. So for example, you can give it a simple prompt like, "Make a claymation explainer of protein folding," and get this.

Proteins start as chains of amino acids. They fold into patterns like the alpha helix and flat sections called beta sheets, forming a perfect 3-dimensional shape.

Omni gives you a more natural way to edit video with conversational language.

Wow.

What's really cool is you can give it your own videos, for example, this selfie, and change reality in a really fun way.

I hope people are watching this on YouTube because the video clips are extraordinary. Dave, you were going to say?

Yeah, the crowd reaction actually on, you know, Demis had by far the best visuals and the crowd reaction on first, first of all, the crowd is hugely into medicine and science as the use case that everybody cares about. And Demis is the spokesperson for that. But then his visuals on the video stuff were the best too. And you should go watch the original YouTube recording and see the full, the full video of him morphing himself through different places and outfits and everything in real time. It's incredible. And I think, you know, in the worry about AI that's going on globally, we lose the fantasy and the cool factor that, you know, you could look forward to this your whole life, and now you can suddenly play with it. And I really encourage everybody to just get in there and build some stuff and play with it. Put yourself in a movie as a character, change the backgrounds. Once you've experienced that, you really get a sense of the amazing things that are possible. You know, starting now and for the next few years, just new things every week.

It's— Alex, reality is cooked, isn't it?

I think reality is getting enhanced for sure. But I also want to applaud Demis and Google DeepMind for being the only arguably remaining American frontier lab to still be chasing multimodality.

Even though they're in the UK, right?

Well, they're really American. I mean, they may have a lot of personnel in UK, but it's an American frontier lab. They're the only frontier lab still chasing multimodality. So OpenAI cut Sora and arguably deemphasized video. Anthropic has never arguably been chasing multimodality. They've been squarely focused on cogent. That just leaves Google with the, the only credible frontier American video model since video is arguably the hardest modality combined with consumer demand. And then you have all of the Chinese frontier labs. China is taking video as a modality far more seriously seriously, my sense from some of GDM's earliest announcements with multimodality is they have a grand vision of modality scaling. Even though video presents as the most consumer-friendly, the most impressive demo, actually at the backend, they're probably treating biological sequences like DNA or protein sequences as another modality. They probably have dozens of other modalities that they're trying to fold into this omnimodal model, looking for modality scaling in a way that the other American frontier labs just aren't. So, so it's a bet, it's at this point almost an idiosyncratic bet that they're going to get to some form of superintelligence that's distinguishable because it handles all these different modalities, text, audio, video, maybe biological sequence data, maybe other crazier modalities, all in some meta-uniform way that the other labs aren't achieving, but it's a bet nonetheless.

Interesting. Saleem, I mean, I can't imagine a better kind of technology for education and teaching people. I mean, I so wish this existed when I was doing organic chemistry and studying medicine.

I mean, this should clear away so much of the cruft of trying to figure out how to present things and have different ways of showing things, biological models, et cetera. I mean, this could all become real-time full 3D. I mean, it's incredible to see what's going to come from this. Very exciting.

The real-time part of it is huge too, because in a call like this or a podcast like this, you can create real-time graphics and visuals to fit the dialogue just purely with your voice. They can do it at Google. We can't do it because we don't have the token speed to keep up. So we have to wait a minute. So if you said something really cool right now, Slim— hey, Brazil, let me tell you about data center explosions in Brazil— The graphic that backs that up would take a minute to come back. And so you can't do it in real time, but they can. And it's just purely who has access to the compute. Yeah.

It's incredible.

It's gonna be coming where AI is gonna be just creating a soundtrack and a visual track for your life. Yeah. Always present when you want.

Yeah.

Amazing. Let's dive into their new Gemini 3.5 Flash model. So they just launched Gemini 3.5 Flash. It's the new default for Gemini app and AI search mode. And as you're about to hear, compared to 3.1 Pro, it's better across all the benchmarks. And importantly, Google says this new model is significantly faster, in a league of its own. And in terms of intelligence versus output speed, it's better handling agentic tasks, offering improved agentic coding, richer and more interactive graphics. Let's take a look.

And today I'm excited to introduce Gemini 3.5 Flash, our first in a series of models. When compared to 3.1 Pro, Flash is better across the board, almost all benchmarks. It's made huge progress in coding. And look at that extraordinary jump in GDPVal, a benchmark that captures many real-world economically valuable tasks. Second, 3.5 Flash is a very capable model at the frontier and comparable to the best models, but much, much faster, which is why when you look at the intelligence versus output speed, it's in a whole league of its own in the top right quadrant. When looking at output tokens per second, it's 4 times faster than other frontier models, and it's incredible delight to use.

Alex, impressive. What do you think?

Well, remember when I opened saying there were highlights, lowlights, and then to use the colloquialism, mid? I would call Gemini 3.5 Flash solidly mid. If you look at its capabilities, and others have pointed this out as well, just from a raw capability standpoint, not talking about throughput or cost, it doesn't compare favorably with say GPT-5.5 High or xHigh or Pro. Pro. On the other hand, this is a Flash Series model, so it's not, it's not Pro yet. Sundar, sort of infamously at this point, has said Gemini 3.5 Pro, that's coming out in another month. There were groans in the audience at the time. So this isn't intended to be top of range. I think the strategy, if I were to play Kremlinologist here, the strategy is I think, I think Google is sort of solidly tier 1.5 at this point in the race to raw frontier capabilities. 3.5 Flash represents perhaps pushing the optimal frontier in terms of throughput versus performance, that optimal frontier. But I think it's also very telling that Sundar is highlighting throughput versus performance instead of like number of tokens on the x-axis, input tokens versus performance, or rather output tokens on the x-axis versus performance on the y-axis, or, or some other metric. He picked the most flattering possible metric. And if you actually—

and they all Okay.

Everyone picks flattering metrics, but some metrics, some flattering metrics are also more sort of truthful than others in some global sense. And if you look at the metrics that Google's been highlighting, A, their note, it's very telling, 3.5 Flash is being compared primarily with 3.1 Pro, less with Frontier, other models. But secondly, the areas, the benchmarks where it's really excelling are benchmarks where tool use in particular, really aggressive tool use is needed. So if I had to squint at this, I would say the emphasis in Google wasn't necessarily beating the frontier with 3.5 Flash. It was probably, I would say, some combination of throughput maxing and tool use maxing, not pushing the boundaries of the frontier, but solid release nonetheless. It's nice that Google's still in the game.

And Dave, I'm imagining you— we've talked about, you know, the labs pulling their punches. I imagine that, you know, releasing— there'll be, you know, the next version of GPT will come out. And then, of course, you know, Pro will come out right after that.

Well, the scuttlebutt here in Silicon Valley is that it's a two-horse race between OpenAI and Anthropic for the best AI in the world. And the talent is flooding into those two buildings in SF. And nothing in that demo or in the vibe on campus at Google contradicts that. So here you pointed out earlier that, you know, Google— or Alex pointed out that Google is the one remaining horse in the race to the consumer. And this is a very, very fast model that, you know, gives the consumer a much better experience. But the other labs have already pivoted to the enterprise and said, look, we're giving up on that. We're going totally after these massive enterprise budgets. And they, you know, if you try and build something sophisticated with AI, you want the smartest AI that solves the problem. And you're not going to back off to a faster model that's not quite as intelligent. You just can't. And so they're going full bore after self-improvement at the other labs. The other difference from a year ago is, you know, Google's unstoppable war chest, you know, $180 billion in CapEx per year and rising. But the other guys in the interim raised, you know, OpenAI raised $120 billion in cash, and they'll, they'll burn that pretty quickly. And Anthropic is on a similar trajectory now. So the war chests are actually not as different as they were a year ago. So, so yeah, I think that's the only disappointing thing in the whole, the whole show is the best of the best Gemini is not up there with Mythos as far as we know. Now, I will say that Google doesn't, you know, they do soft sell. They don't announce what's coming in 4 months and promote and trumpet it because they don't need to. And so if something really, really big is cooking and coming soon, they didn't roll it out, but they don't need to roll it out. They'll wait until it's, it's proven.

You know, I love the naming nomenclature here. Of course, we've got GPT models, you know, 5.4, 5.5, 5.6, and, and Gemini jumps from 3.1 to 3.5. Uh, it's, it's fascinating. Salim, what do you make of this?

I thought one thing that's clear is you're seeing this kind of bifurcation now between premium cognition and ultra cheap but very fast cognition. And I think that's going to continue. I think Alex makes a great point about throughput. This will allow a lot of throughput, right? And there's this continuous march for marginal intelligence, cost trends towards zero.

Yeah. Here was the next segment that I pulled out. And again, I want to just shout out to Gianluca, who clipped all these beautifully and provided them, you know, in record time for us for the show. Hear, hear. Yeah, a conversation about SynthID and content credentialing, you know, really important, especially as we start to encroach on reality. How do you know if something is or is not AI-generated? It's going to become more and more important than ever before. Let's take a look here at Sundar talking about SynthID.

Since launch, SynthID has now watermarked over 100 billion images and videos along with 60,000 years of audio assets. We are now going a step further and adding content credentials verification across products. This will show you if the origin of the content was AI or a camera, and if it's been edited with generative AI tools. In this example, Gemini can tell this photo was captured with the Pixel camera and then edited with Google Photos. Of course, this only works at scale if more partners decide to watermark their own AI-generated content. NVIDIA signed on to SynthID last year, and today I'm thrilled to announce that OpenAI, Kakao, and ElevenLabs are adopting SynthID too.

I love the fact that we're getting to standards and everybody's, you know, picking the best. Alex, how important is this?

You know, the irony is so many people were hand-wringing over the past few years that we won't know what's real and what isn't. And my response was always, we're going to get cryptographic, eventually cryptographic chains of custody from reality capture to what is ultimately presented to the user in the same sense that when you use a browser, you can maybe see a little lock icon to indicate end-to-end SSL encryption. We're going to get the same thing for reality. And I view SynthID, which basically which, by the way, was also just adopted by OpenAI. It was created by Google, now also adopted by OpenAI in the same breath. I think it's sort of ironic that we're going to get, it seems, end-to-end authentication of realness, proof of reality, if you will, not coming from the camera end, not coming from the reality capture end, coming from the synthetic end, coming from Altair. Vendors that want to claim credit in some sense that they were the ones who generated the reality. And then the cameras, the camera makers and all of the recording device manufacturers, they're going to be downstream and the ones who adopt the same protocol. But either way, we're getting our end-to-end proof of reality one way or another.

I think there's a much broader story. I'd love Saleem to get your thoughts on the bigger, bigger societal implications because a big topic at Stanford last night you know, with Erik Brynjolfsson and his entire team there. The rate that AI is innovating can't be kept up with by Congress. And there's going to be no regulation of any value coming out of Washington. So the industry is starting to self-regulate. And that's the only— you know, only AI can keep up with AI. And so we may look back on this moment as one of the first moves by the self-regulation community where, okay, now we're going to start watermarking images. Hey, everybody in the community, please adopt our standard for watermarking images. And then there'll be something else a week later, something else a week later, something else a week later. And that'll become the way that we govern ourselves in the future, much more so than any law coming out of Washington, purely because the pace can't keep up. Salim, you think that's basically— this is the first move in that direction?

I think that's exactly right. When you have intelligence becoming abundant, then the scarcity goes towards trust, and scarcity creates value. So we may end up at a point where authenticity is more valuable than creativity, and that line between something being created and knowing how it was created, et cetera, is blur— is now merging because of the systems that are being created now. And I think, Dave, the point you make is really, really important. Once you have that trust layer, right, now you can scale. I go back to Jerry Mikulski, my community member, who said scarcity equals abundance minus trust, right? And so if you can solve for trust, you solve for abundance. And one of the biggest challenges today, I thought this was really a big deal because we're moving from the information age to the verification and trust is becoming infrastructure. And I think that's a very powerful, valuable pillar for the world going forward.

So, playing arithmetic, does that mean that abundance equals scarcity plus trust?

It does.

It does. Scarcity plus trust gets you to abundance.

How about that?

How about that?

All right.

Thank you for—

That's like grade 5 math. Even I can do that.

All right, so—

But because the cost of generating content is collapsing, then the value shifts from signal filtering authenticity. We saw this with photography, where the big problem in photography is how do I take the best photograph? Because each click costs you a dollar. And you had a bunch of business models crop up around selling expensive cameras or offering courses in photography. Publishing books on composition. Then we moved to digital photography. The cost of creating a photograph went to zero. And now the big problem everybody has in photography is I have 6 copies of my photographs on 7 different online services and you can't find anything. And the value then comes in that filtering system, right?

And I'll tell you, uh, Salim, you know, we're moving into this intentional world that we design and everyone's like, what will happen next? What will happen next? Well, it's whatever we design. Is what's gonna happen next. But Dario and Demis are two probably dominant architects in the future of how we live. And it's just great to hear both guys go back and forth. But if you do a raw word count from Dario, Dario Amadei, CEO of Anthropic, and you go back, you know, the words were all about transformer architecture, speeds, intelligence, benchmarks, and then they transitioned to UBI, ethics, and now they're talking about the way the world should be governed. Going forward and writing papers on it. And so if you just track the word count, it's also on this exponential change rate. Same with Demis. You know, Demis has to be a little more cautious because technically he's an employee of Google, even though he acts very independently. But those are the two guys just very much determining the future of all humanity right now. And so, you know, this is a— the watermarking is just like move one, act one of the whole future way we live.

And, and look at what's happening, right? Our, our trust in legacy institutions is collapsing. And at the same time, AI is building up the capability and the infrastructure and the foundation for delivering trust. So hopefully if that happens elegantly, we'll have an elegant shift from scarcity to abundance rather than a messy one.

Everybody, you may not know this, but I've got an incredible research team. And every week, myself, my research team study the meta trends that are impacting the world. Topics like computation, sensors, networks, AI, robotics, 3D printing, synthetic biology. And these MetaTrends reports I put out once a week enable you to see the future 10 years ahead of anybody else. If you'd like to get access to the MetaTrends newsletter every week, go to diamandis.com/metatrends. That's diamandis.com/metatrends. The next product that they dove into, and a central part of, uh, Google's plans, is anti-gravity. They released Antigravity 2.0, a standalone desktop app built to orchestrate multiple agents to execute tasks in parallel. Let's take a listen, and then Alex, I'm coming to you for, for your evaluation. Mid-tier, high-tier, no-tier? Let's go on.

At the core is Antigravity 2.0, a new standalone desktop application that delivers fully on that original glimpse of a truly agent-optimized experience.

The new Antigravity is unabashedly agent-first, focusing on the core agent conversations, agent-produced artifacts, and multi-agent orchestration.

Like I said, unabashedly agent-first. As Sundar mentioned, this is the exact experience teams here at Google have been using to drive massive value. Let's take this live and actually show this operating system in action.

Try running Doom right now.

It just doesn't work. Turns out that the OS is currently missing some necessary video and keyboard drivers.

So let's just try and fix it in the new Antigravity. I have a prompt prepared, I'm gonna paste it in. Antigravity ended up doing a whole host of research, ended up writing over 100 lines of code, and then finally built the operating system.

Let's take a peek and see if it works.

Amazing!

So first off, Alex, what is Anti-Gravity 2.0 and what are you thinking about it?

Yeah, so let's remember where Anti-Gravity came from. Do you remember Google's hackquisition of Windsurf during that debacle? So Anti-Gravity is basically Windsurf rebranded from the Windsurf team that was hackwa-hired by Google. And then Anti-Gravity 1.0 versus 2.0, I view— you were asking earlier, Peter, is this high, mid, low? This is sort of mid, if, in my mind. If you look at what Cursor has been doing by contrast, Cursor was much more aggressively leaning from their old interface, which was sort of a reskinned Visual Studio Code, code-centric editor from their 1.0-oriented user interface to their more recent interface, which is agent-first. I view this almost as like a copycat fast follow or slow follow or somewhere in between from the windsurf team within GDM. They're basically following the same metaphor of saying, no, we're no longer about direct code editing access. Now the primary metaphor is orchestrating fleets of code agents that are doing all of the hard work. So I would say Google's almost hamstringing themselves a little bit by announcing this now as part of I/O versus, say, in a more timely fast follow or even lead when Cursor was doing this months ago. I also, I've used Antigravity quite a bit, certainly was using it even more when Google first announced it after the, the Windsurf acquisition. And I would say not super impressive. It was very buggy. I think 2.0, I haven't had a chance to use 2.0 yet, but hopefully it's a good deal stronger. But really, I don't know anyone who's doing their primary development work with Antigravity at this point. The development is happening either with Claude Code or with Codex Or maybe with Cursor. Anti-gravity, don't know anyone who's using it.

You know, everybody's trying to leapfrog.

Except every engineer at Google.

Except for Google, maybe.

So Dave, you were trying to get on—

Actually, even that's not true though. So I mean, it's been publicly reported that within Google DeepMind, they're all desperate to get Claude code access for everything.

Yeah, we talked about that a couple of pods ago. Dave, you were playing with this actually during Google I/O yesterday.

Yeah, Tyler and I both installed it in real time as they were rolling it out, which maybe not the smartest move in the world because there's 16,000 people behind us in the crowd, probably all trying to do the same thing. So that was not a great first experience, but I don't blame Google for that. But this morning it worked fine and installed great. And I completely agree with Alex's assessment. It looks almost identical to the new cursor agent-first Windows. I mean, like, I almost can't tell where I am. Am I in cursor or am I in anti-gravity? Antigravity. What they did do, which is a little more extreme than Cursor, is it completely replaces Antigravity 1.0. You can't even see the code anymore. And you have to go and launch the old thing if you want to actually edit code. So, you know, Cursor didn't go quite that far, but it's really obvious where the puck is going. If you want to build things in the future, you're not even going to look at code. You're going to describe what you want and you're going to debug think at this much higher level of evals and functional comparisons and, you know, I don't like where that button is, move it. And so, I think, you know, in the future, nobody's going to want to go back to autocomplete code editor view. And so, they leapt ahead and said, we're just going to eliminate that entirely. And if you really want to hack, we'll give you a way to get back to it. But we're going where the puck is going and not where it was. But, yeah, it really is exactly catch-up like Alex said.

Everybody's got the same, you know, it's Codex and it's Claude Code and Antigravity, and they're just going to be leapfrogging each other. I'm just curious if there's going to be some new sort of breakout approach to this that's going to materialize. Alex, do you think there's anything in the future?

Well, code is clearly going away as a human endeavor. It's all being abstracted away by code agents that handle all code, and humans aren't maybe be in the near-term future trustworthy enough to even be allowed to write their own code. So I think that's one obvious arrow of time in this space. I think recursive self-improvement is another arrow of time. So not even old generation models are trustworthy enough. Maybe older models are trustworthy enough to rewrite themselves and generate newer, better models, but code's going away. I think that that's the obvious trend here.

Well, I think the— also, Peter, to answer your question on the next paradigm, we've only had this paradigm for a couple months, so. So, you know, let's, let's, let's settle for a little. But no, clearly the next paradigm is exactly the Star Trek holodeck, which Alex has been saying for a while. So right now you're talking to it, it's building things for you, it's incredible, but it's not natively graphical and visual and, you know, you're not moving things with your hands. So if you say, I want to move that button, I want to change this, I want to connect this to my email, you're not actually seeing the button move in real time. It's regenerating and then you see a new rendering and, and in the future it'll be a real-time graphical experience that's interacting in your comfortable physical space, kind of, you know, native human environment. And that's, you know, that next iteration is certainly within this calendar year.

Hmm.

All right. Next up is Gemini Spark in Google's— it's Google's, you know, take on OpenClaw, I think is the most obvious thing to say. It's a new always-on AI agent that can write emails, create study guides, keep an eye out for, you know, financial fees that you're being charged. It's Google's "we have Open Claw at home" moment. It's powered by Gemini 3.5 Flash, and it offers you a 24/7 operation. Let's take a listen to the conversation about Gemini Spark.

Introducing Gemini Spark, taking action on your behalf and under your direction. It runs on dedicated virtual machines on Google Cloud, and it's 24/7. A task right off the bat. This is a pretty straightforward example, but it's so useful. Help me draft an email to the team, compile everything about our recent Gemini Live launches and wins from the last week. And what's amazing here is Spark will go through step by step. Look at all these steps, all the time it saves you going through. And again, work across the various skills and apps that you have. And what's really amazing is it'll break it down and also be able to generate files for you. So the first one here, this is a live RSVP tracker right in Google Sheets. You can see that it shows who's confirmed and who hasn't. What's amazing about this is it'll actually update because it's connected to Gmail. So when L. Thompson row 8 RSVPs, it'll update, which is pretty amazing.

I mean, one of the things I find fascinating is it's the integration across all of the Google products is very powerful, right? There's a point at which there's such a cost for not being inside the Google ecosystem that everybody defaults to it.

Isn't that— that's the most ironic thing I've ever heard because 'Cause, you know, people who are younger don't remember that Google only exists because the FTC stopped Microsoft from killing it. So Microsoft had just killed Netscape, taken total control of the browser, and integrated it with the operating system, and made it impossible to do anything on the internet unless you went through Microsoft. And that triggered the FTC. Mike Hirschland came in. The whole lawsuit stopped Microsoft cold in its tracks, and they paid a $1 fine. I don't know if everyone remembers that. Hilarious outcome. But they had to unbundle, and that opened the door for Google to come into existence. Microsoft pseudo-competed with Bing, but they were prevented from competing aggressively and tying it back to the operating system. So here we are all these years later, and Google is coming out with this series of kind of exact copy of Cursor, exact copy of, exact copy of, exact copy of of, but it's perfectly integrated with these other— you saw on the other slide, what, a dozen Google products that have over a billion users. A billion users out of this world population is a massive installed base. So if you want OpenClaw, yeah, you can be over there, but if you want OpenClaw equivalent that works natively with Google Docs, Gmail, everything else, you know, Android, everything else, your Google Pixel camera, Then you have to use this. And so it's exactly history replaying itself. It's so ironic, 'cause they were so anti-Microsoft back then. The whole "Don't be evil" motto was a direct attack on Microsoft, implying they were the big guys that were evil. So here we are, you know, years later. I'm not saying Google's evil in any way. I'm saying they're tying as their competitive advantage in the exact same way that Microsoft used to.

Yeah. Alex, what do you think about this compared to Open Cloud?

I think it's a lazy copycat product. I, I think it's, it's obviously Google trying to take advantage of the resources that they have. So note, it's hosted in a GCP VM, not necessarily pushed all the way to the edge, although they have aspirations, cite to, uh, to Gemini and Android Halo for that.

But it—

if you're Google and you see OpenClaw and you see Jensen out there saying OpenClaw is the next ChatGPT, really, what's the smallest, what's the minimum viable response that you could take? It's, okay, we're going to host GCP VMs with Gemini Flash that integrate together all of our products that run headlessly. That's the sort of minimum viable strategic response. What I would've liked to see from Google DeepMind here was the maximum response. Show us the art of the possible. Show us what a next generation OpenClaw or Hermes competitor actually looks like. Create the benchmarks, show us next generation capabilities. And they didn't deliver that here.

Well, Alex, I mean, just to be fair, they're delivering on a lot. It's not just one piece, right? It's a lot that is being deployed on Google I/O Day. But having said that, you know, I still love Skippy, which is an OpenClaw on top of my Mac Studio. I love it because it's got a personality versus being sort of a generic, you know, agent that's ever present. I don't know if you can do that with Gemini Spark, but I think the personality side of these are critically important. Saleem, any thoughts for you?

Two thoughts. One is I agree with Alex. They really could have gone for a little bit more bite here. But on the other hand, when you can make agents generally available to the average Google user, the, the, there'll be hundreds of millions more people training up agents. And I think that's generally just good. OpenCLAS still has lots of room to be experimental, power user oriented, very opinionated, doing the weird things like the Nat camera stuff. But I think this is a very solid entry into that world to give people a taste of what an agent world could look like. But it is boring.

It's safe. It's playing it safe. Like, is it a safe entry? Yes, it's a safe entry. Will a lot of people maybe use this to clean up their Gmail inbox? Yeah, probably. But it's not pushing the frontier, which is really what I would have loved to have seen here.

Yeah, I think that's your recurring theme on a lot of these, Alex. Is that true?

I think— I mean, look, as accelerationist, yes, I'd love to see frontier labs pushing the frontier. Here. And to the extent that this is an avatar of Google DeepMind and not just Google corporate, I would love to see more frontier coming out of the frontier.

Dave, you want to close this out?

This— well, two quick—

two quick—

go ahead, Slim. Sorry.

Yeah, two quick thoughts. Like, there's something very powerful happening here because this is giving everybody an operating system, uh, for their lives because of the deep integration with all the other Google stuff. So I think, uh, the, the next productivity jump is going to come from persistence And this will create a massive enabler across the board. And it's going to— I go back to the earlier comment. I just want to double down on that, which is it's going to train a lot of people on how to build agents and run agents. And I think that's going to then enable another class of things to come forth from that.

Yeah, so there's no doubt that this is all fast follower, exactly the way you're characterizing it. On the other hand, Peter, you love your Skippy. I love my agents that I set up too. But when you talk to somebody on the street and you say, hey, have you Have you set up an OpenClaw or a Hermes? Overwhelmingly across the world, people say, no, I haven't done that. And that install and onboarding experience is just too much friction. So I wouldn't underestimate the power of default behavior. Now, over half the world uses Google. And if Google says, okay, Gemini Spark is going to be one click away from a Google search, it's completely integrated, massive fraction of the world is just going to click the button. And then their first experience with a personalized agent will be via that click. I don't think that anything can slow down Google because of their massive distribution advantage. They don't have to push the outer boundary. They can afford to be fast followers. I am equally disappointed, Alex. I'm not saying that. But from a strategy point of view, they don't need those risks. They just need to be as good one day later and integrated with Chrome, integrated with Google Search, integrated with Android, and they will win.

I think Google's magic potion is making it user-friendly, making it easy, making it intuitive. And I think they're going to deliver with Gemini Spark on that particular promise. All right. Here's another part of Google's resurrection and dominance. It's Agentic powering of AI search mode. AI everywhere. And remember the conversation we had, search is dead. Well, search is not dead. It's just been reinvented. Did. Let's take a listen.

I'm excited to announce we're launching a brand new intelligent search box. Before, the search box was a contained space, but now it's totally reimagined with AI. It expands with your curiosity, and as you ask, Search helps you formulate your questions with AI-powered suggestions. This goes beyond autocomplete. It offers nuances that you might not have even thought to ask. Now we're taking an exciting step toward this vision where you'll be able to create and manage multiple AI agents for your many tasks right in Search.

Now, let's say you're apartment hunting.

You can do a total brain dump of what you're looking for with all your criteria, like location and natural light and availability, and your agent will continuously scan the entire web across sites, social, and forums.

Persistent search here, right? So this is your agent, you know, whenever you've asked a question, It is going to persistently be looking for the latest and greatest. Yes, this new apartment just became available. This product just got cheaper. You know, your wife loves this topic and, you know, here's a new product delivered to her. The other side, though, is that autocomplete function. I wonder where it's going to take us, right? You're going in asking or thinking about asking one question. And then, of course, Google can sort of drift you into asking a different question. You didn't intend to actually ask. A lot of interesting perturbations here.

Oh my God. Yeah. Well, think about a vacation plan where you're like, you know, I really think I should go to Barbados and it autocompletes to Bermuda and you're rerouted to a different hotel. The revenue power of that is astounding.

Oh, that's evil.

And really, you know, OpenAI rolled out their first ads and a lot of the companies I know have adopted it, but it's very ham-fisted, you know, it's like, here's some ads on the side. They're obviously ads. Ads, but the Google version of it has to preserve $200 billion of existing 90% margin revenue. So they haven't quite figured out how they're going to surf that, but their power of the user decision-making is like nothing we've ever seen before.

So Google AdWords is now going to— is going to now gently sort of drift you towards a different question that you weren't there asking.

Absolutely.

That would be amazing.

I mean, remember Google Instant as well, which also offered relatively fast suggestions. I don't think Google ended up directly monetizing that, but Google does, to Dave's point, have a long history of steering users toward more profitable queries. So I think that's, that's probably quite likely. What, what I probably underline here is the shape of the rectangle changed after decades How big a deal is that? After so many armchair commentators saying that Google was about to be disrupted by ChatGPT with web search, turns out Google is able to self-disrupt and able to change the shape of that multi-decade-old rectangle. And it changed the shape in the direction of building AI modalities, AI search natively into their search experience, which I think many people were scared wasn't going to happen.

They did it in the end. —my very first venture investment ever was TripAdvisor, back when it was first starting. And the big quandary at TripAdvisor was, how are we going to have completely unbiased, accurate reviews and still get paid by the hotels? We need to make money somehow. How's this going to work? And it turned out that just by sorting the list, you know, the human default behavior is so dominant that they'll go— You mean overwhelming.

Laziness. You mean laziness is so dominant.

Yeah, you know, laziness, but we're buried in decisions now. There's so many things coming at us from so many different directions that we have to be lazy. Only Alex could actually study every single pathway and make an optimal choice. Everyone else, you just have to fall into the default buckets once in a while. 80% of people will click on one of the first 2 or 3 hotels. You can have perfectly accurate reviews and just resort the list so the ones that are paying you are at the top. Then you have your cake and eat it, too. I think that default behavior will hugely benefit at Google because they will steer the users, but they don't have to be super overt. And, you know, they're not going to misguide you into some fraudulent product. They're allergic to that like crazy. But people will still follow the default suggestions from Gemini, and then Google will collect the revenue from whoever is willing to pay.

Saleem, uh, just— I just love the fact that they have the courage to risk disrupting their own business. I think It's such a hard thing organizationally to do, and I gotta give them full props for going after it.

Yeah, and for everybody, you know, everybody listening to this, you know, our goal here is to give you an overview of what Google has just done. It's so dominant in the planet. It does steer a lot of humanity's sort of abilities. So I hope folks are enjoying this summary. Please dive into these. You know, your mindset of curiosity is your single greatest tool. Google. So go and play with these things. You know, when you finish listening to this podcast, go and jump onto Google and play with the new AI search or its capabilities.

Well, one more note, Peter, if I may, just on Google's self-disruption via search. I think there's this misconception out there that the main obstacle to Google self-disrupting their search with so-called modern AI was somehow on the business side or the business risk or the advertising side. I think actually the main obstacle was more technical. That Google engineers for a couple of years there were concerned that there wasn't a cost-effective way or a time-effective way to squeeze generative models into the very narrow and latency-sensitive and cost-sensitive parameters of just powering a search, that the models were too expensive and too slow to yield search results that would be competitive. And this is, I suspect, one of the reasons why you see, going back to, Gemini 3.5 Flash emphasizing throughput so much. Yes. It's reflecting Google's own internal dogfooding needs of having ultra-high throughput models that they could use to power search and some of their own internal applications that maybe OpenAI and Anthropic aren't feeling that demand function as much.

Makes sense. Yeah, makes sense. All right, next subject is, uh, Google is launching a universal cart that users can add products to from YouTube, from Search, from Gemini, from Gmail, Google says this intelligent shopping cart works across a multitude of different merchants from Nike and Target to Walmart to Shopify. So you could literally add a product when you're searching on the Nike site or a Target site and then have it monitored and bought at the same time. Let's take a look. You know, again, this is part of Google's incredible revenue engine. Okay.

Journey. I am excited to announce the Universal Cart.

A truly intelligent shopping cart.

It works across merchants and across services. You'll be able to add things to your cart when you're browsing Search, chatting with Gemini, watching YouTube, or even reading your Gmail. The moment you add a product, your cart goes to work for you in the background.

It finds deals, uh, looks at price drops, gives you insights on the price history, and alerts you when something comes back in stock.

So reinventing the shopping experience. I've got some comments, but to hear from you guys first. Elephant in the room.

Yes, the elephant is Amazon. So I look at every announcement relating to shopping, quote unquote, from Google through the lens of how are they going to compete with Amazon for retail e-shopping and whether it's trying to commodify, create sort of virtual storefronts storefronts, for individual retail vendors, whether it's crawling third-party e-commerce websites and assembling virtual pages, and now universal carts. This is all through the lens of how they're going to compete with Amazon. So I think the elephant in the room here is, is Amazon even going to contemplate going anywhere near complying or adopting Google's standards? My guess is not.

Well, you know, the, the follow-on here, we're going to be seeing in a few moments the reinvention of Google Glass. Where you've got imaging capability. And we're gonna see probably the next invention of shopping where shopping is always on. Wherever you're looking and you see something and your AI agent realizes, oh, I'm focusing on Alex's beautiful orchid in the background, which is, I mean, is that orchid real, Alex? I just need to ask.

I thought, Peter, you said reality was cooked, so you tell me.

Yeah. Oh, okay. But literally when I look at something, my AI agent will say, oh, you're focusing on that. Do you wanna purchase it? Or as you're walking through the day, right? Instead of shopping becoming, becoming sort of a, uh, something you do for an instant of time. It's a continuous function. And Universal Cart is aggregating all the things and then probably at the end of the day saying, hey, do you want to purchase that? Just say yes. We're going to be seeing— this is an early step, but not the full instantiation of reinventing shopping. Salim, you're going to say—

yeah. So today we go from human to website to shopping cart to checkout, right? And tomorrow we're going to go from intent to agent to transaction. Yes. And I think there's a— every CMO in the world in going forward is going to be asking, how do I convince 100 million agents to choose my product? They're going to have to market to the agents. Right. And so I think this is powerful. I mean, look, Google helped create a trillion-dollar company by helping people do search. The— somebody is going to create a trillion-dollar company by helping agents buy. —But are you ever gonna market to an agent?

I mean, my agent knows what I want, knows my genetics, knows my taste from everything else. It may just be buying stuff for me all the time that could be returnable, sort of surprise and delight, something shows up on the front doorstep, "Oh, I thought you'd like this, here it is. If you don't want, I'll have it picked up and returned." And you're not—

We'll definitely go to that. The disruption to e-commerce is not the better shopping, it's getting rid of shopping. Yes, right. There's this ambient experience where, hey, Hey, you may— your shoes look a bit dirty from the camera I looked at from your doorway camera. I'm shipping you new shoes. You know, it'll be that kind of thing.

Dave? I think it's just amazingly shocking the degree to which the big guys don't care anymore about consumer shopping. Because Amazon was nothing but consumer shopping originally. And, you know, built their entire empire on consumer shopping. Shopping, and then added AWS. AWS is so much bigger at Amazon now than the entire Amazon we know, all of the shopping. Google was already competing with that side of Amazon with GCP vs. AWS. They're already in this battle royale over compute and data centers and enterprise use and everything. Google had already tried to compete in retail with Froogle. Remember Froogle? Of course. Completely. Long time ago. Yeah. Frugal.

Right before shopping. Yet another rebrand. Yeah.

So I think that this is another attempt to take Amazon head on on the shopping side. But I think AI is a big game changer. But it doesn't matter too much whether Amazon defends its turf or whether Google encroaches and wins. At the end of the day, the battle on the cloud and the backend is so much bigger. And it's already raging. So this is kind of cool. I think, you know, it's just the next stage of this, you know, trillion-dollar retail battle that, that will rage on for a while.

All right. Let's jump into conversation about Gemini app and NotebookLM. Here's Josh Woodward, who heads Gemini. I think we're going to get Josh on the pod here. We'll talk to him about what he's up to. Let's take a listen.

More than 900 million users are coming to the Gemini app app every month. Just on its own, Notebook LM has now been used to create more than 1.5 billion notebooks, podcasts, slide decks, and more. It's now available in more than 230 countries in over 70 languages. It now opens up immediately and inline, and soon you'll be able to pick a regional dialect that resonates with you. You've got a right good mix of different accents knocking about, like this one from Liverpool. Gemini Omni is coming right into the Gemini app. Let's look how this plays out in the real world. I want you to meet Sashu. She is working on a new song and she wants to create a quick video teaser. So she shares the raw video, she adds some reference visuals to it, Ready for it? Let's take a look at what it looks like. The third update today is about how agents are coming to Gemini. One of our newest out-of-the-box agents called the Daily Brief. It's a personalized digest that's designed to be your first stop every morning. Here's how it works. You can see here that it's synthesizing information from across my inbox my inbox. And with this travel info, I can just take the next step right in line.

So this is an integration story. This is Google integrating across all of its capabilities and making it so magical that you can't afford not to be in the Google ecosystem. Dave, what do you think about this?

Yeah, it's amazing when you got one guy on stage demonstrating, here, we can build an entire operating system in real time. Let's go. I'm consuming a trillion tokens right now building— and nobody Nobody gets it, right? Nobody can relate to building an operating system. Then Josh gets up there and says, "Here's a real human musician. And here's her trying to portray herself to her fan base." The crowd goes crazy. And it just shows you the human aspect of this is so dominant, even in Google, even within the empire. The human aspect of it is so dominant in people's minds. And, you know, it doesn't come through on the videocast when 20,000 people people see Josh present something and they go, "Wow! Oh my God, I can totally—" and the vibe is contagious across the whole crowd. And it just is hard to capture that in a video clip. But it's— yeah, it's just remarkable. It's going to unleash so much creativity and it's such an exciting time. And I wish, you know, I wish that was the only vibe, you know, that everyone could just capture and then hold and bottle that in. But this was just a great moment.

Alex, your take. Why is Google still branding Notebook LM as Notebook LM? It should have been folded directly into Gemini or maybe Google Workspace or something else. Why does this still have an independent brand? I don't understand it.

Well, look at all the other, it's like Spark and Flash and, and Anti-Gravity and like all this is really fragmented all over the map, like divisional kind of branding.

Google has this reputation for better or for worse of launching lots of products and having a culture where product managers get promotions for launching but not maintaining products. I'd really love, under the spirit of more wood behind fewer arrows, to see all of this functionality just unified in a way that gets sustained. Yeah.

Yeah, especially 'cause AI is such a unifying force. You can put one voice and interface on top of all of this mess. Yes. And like, and Google's branding originally was so good. Good.

You know, all other searching, it was so good.

Yeah, colorful and, and humanistic and, and friendly and all of that. So it's right.

Speaking directly to, to Josh, uh, and the Google team, please just unify all of these offerings and maintain them. Don't keep launching 10 different products and product names that we'll forget about a few months from now. Just please unify all of them and maintain them.

You know, the idea of a daily brief, I love it. I, you know, Skippy gives me a daily brief. This is a beautiful integration here. But being able to know what you're doing, when you're doing it, what your intention is, and giving you updates all the time on your flight. You know, there's a new flight that's available now. It's 5 times cheaper or whatever the case might be. And the weather is, is going to be hotter than expected. So make sure to, you know, to pack different. That level of overlay intelligence This is going to be magical, right? It is.

But at the same time, remember Google Reader, Google's RSS reader that Google abandoned despite having a rabid user base, myself included? I know Google really wants to own the newsfeed. That much is obvious. But please just maintain it. OK. Salim. So two thoughts.

One is I've always thought about NotebookLM, per Alex's earlier comments, as this weird thing sitting out there because it incorporates creates presentation and learning and interaction that kind of does— what's the difference between doing this in the Gemini app and doing it in Spark? I mean, people are going to create a lot of confusion around this. The point that Alex made, I just want to double down on, which is when in any big company— we had this at Yahoo as well— you're rewarded for getting something out there. But then you've got a strategic project manager looking across resource allocation about 10 different projects. And so you get a peanut butter problem where you're very thin across all the different projects. You don't iterate very well. One of the few companies that iterates very well is Apple. They will relentlessly iterate on their products and most other companies just go too slow.

On a small number of products.

Yeah. Yeah. And they go very— there's a whole thing written by Brad Garlinghouse called the Peanut Butter Manifesto when I was at Yahoo that mirrors this whole challenge of of how do you navigate this in a limited thing? 'Cause they're not run as a start— as each individual team doing startups with KPIs of their own and targets, et cetera. They're run across in hierarchical structures in many cases, and you suffer a lot from that.

All right, Google's new product called Audio Glasses. We heard earlier about their partnership with XREAL. Here's a partnership with Samsung and a couple of different glass manufacturers. Let's take a look And how is this going to impact our lives? Two videos to show, then we'll discuss them.

The next big milestone for Android XR is intelligent eyewear.

Today, I'm excited to announce that our first audio glasses will arrive this fall. They are designed to give you all-day help with Gemini that is spoken into your ear privately rather than shown on a display. And these glasses let you stay hands-free and heads up up for things like listening to music, taking photos, making calls, or tapping into your phone apps, all without reaching for your pocket.

All right, video number 2 from Samsung. At Samsung, our vision is to enrich people's lives and help shape how we live tomorrow. In close partnership with Google, we're introducing intelligent eyewear that empowers you to connect to the the world with confidence.

Built with Samsung's precise engineering and craftsmanship, we're merging form, function, and helpful intelligence to create something you'll want to wear.

In eyewear, every millimeter counts.

Today, we're thrilled to share a first look at the upcoming styles co-created with our eyewear partners Orbee Parker and Gentle Monster.

All right. The elephant in the room here. It's got forward-looking cameras, but you're not seeing words or images on the screen. You're being spoken to by your AI. Interestingly enough, I think that being present in life has just been cooked as well. I mean, imagine you're walking around and you're just having— you're not talking to your wife, your girlfriend, your kids. You're just having the agent whispering to you all the time. Saleem, what do you make of that? This?

So a bunch of things. I was really disappointed in that there's no visual on the screen. I mean, doing the audio is— it might be in Alex's words, very mid. But we're moving to that point where human-computer interaction becomes continuous and becomes an ambient layer that's just ongoing. And I think that is the bigger story here because that will kind of just continue to play out as we merge with technology. Already we pick up our phones 80,000 times a day. This just continues that in a very kind of unnoticeable way. The form factor is very workable.

Google should have owned smart glasses. Instead, Meta is running away with this space. Apple is also playing catch-up. Where was Google? Google had— I was one of the earliest users of Google Glass. Remember that?

Yes, the glass holes out the front.

Did you get punched? I did not, fortunately, but they had a battery life of like 5 minutes and they self-bricked through operating system updates. It was— I think even Google would recognize it was prematurely released. Google could have kept iterating. To this earlier point about doubling down, Google could have kept iterating on smart glasses from the Google Glass era, and they didn't. They basically abandoned Google Glasses to enterprise and then abandoned them completely, and now this is This is, I think, represents a complete reset except without all of the conveniences that Google Glass had. Meanwhile, Meta was iterating away, spending billions of dollars, sure, but iterating away at smart glasses. And now Meta has the lead in the space, not Apple and not Google. So I would love to see a very competitive smart glass market between Meta, Google, and Apple. But I really would like to see Google XR in particular, Android XR, stepping it up a notch and shipping much more quickly. Right now Meta's running away with the space. Space. Dave, what do you think, pal?

Well, I mean, this is where society is going to have a huge rift because, you know, the, the punching in the face was a very real thing last time Google went down this path. And they are trying to own the consumer and be a, you know, a consumer-friendly brand. But if they roll out a product where half of society is walking around recording everything all day long and the other half is offended by that, then that's going to be a major, major problem. They're stuck. They want to own the space. They have the technology to do it. People are going to want to talk to their Skippy, their agent, all day long. They're not going to want to lose touch with it. This is a great way to stay in touch with your agentic world that's working for you behind the scenes. I'm very eager to use it. I'm also not super eager to get punched in the face. There were 3 commencement addresses this week, including Eric Schmidt's in Arizona. Where as soon as the commencement speaker said AI, the whole crowd went boo. And so, I mean, if you're not aware that that's what's going on, and it's very easy to live in your echo chamber, especially here in Silicon Valley or in Cambridge, but you gotta walk around Mississippi or walk around Nebraska to really understand how big a deal this is gonna be. So I think the glasses are gonna be a huge forcing function in this inevitable conflict.

Really important points. Audio glasses, hello, oxymoron.

Oxymoron. Well, audiovisual oxymoron. Now they go together very well.

But the audio feedback layer, I think it was a, you know, they want to provide a product that actually works consistently. And I think probably the image, you know, the imaging up on AR glasses is still kind of weak. You know, the version of the Meta glasses I've tried is okay, but it's still a far way off off from being able to turn on an ambient AR layer that's convincing and compelling. The brightness isn't there. You gotta look at it in the right, you know, but audio, having your AI being able to say, oh, okay, I know you were shopping for that. Do you like that one? And I'll order it for you right now. Or importantly, I don't need to, you know, when I see Alex approaching me and I've forgotten his name and he's coming at me and the glasses can say, oh, Oh, you know, that's Alex Wiesner-Gross, you know, he's got an IQ of 100,000, uh, you know, that should be enough.

I'll never forget your name, Peter, for the record.

Thank you, I appreciate that. Um, nor I yours, my friend. But I think the audio interface is a smart move to make it clean and compelling and consistent, uh, and, and something that you can interface with on a regular, regular basis. I am concerned learned about this issue of, you know, you guys all have this, right? You're with your family, uh, or with friends, and something pops up on your phone, and you focus your attention to your phone. Um, you know, the loss of presence in life can be really costly, really costly.

Yeah, it really can. And also, I think the, um, you know, the, the cameras are always on. They're very low battery consumption. So you can run the cameras continuously and it's just seeing everything you see and then talking to you about what you're looking at. So that's Alex, that's Celine. But if you really want a display, you know, it can talk to your phone and you can look on, on your phone to see anything it wants to tell you there. So it's all integrated through Bluetooth anyway. So, but I think the consumer would rather have the longer battery life and not have to worry about it dying every hour like Alex was saying. So this is a good, this is a good temporary, you know, stepping stone. Phone. And, you know, like you said, Peter, putting, putting the display in front of your eyes— if you think being not present in the moment is bad with this in your ear, imagine, imagine when it's flashing, like, between you and your wife. Yeah, flashing images on your— it's— this is a good— it's a, it's a good product design.

Yeah. You know, I know that, uh, Alex, we're going to lose you in a moment, but I wanted to have the last two segments with you still with us. Uh, let's jump into Demis's presentation on Gemini for science.

I'm excited to announce Gemini for Science, which brings together the powerful AI tools to help accelerate research. Gemini can already assist in solving complex problems, but our new labs prototypes streamline daily scientific tasks, whether it's staying on top of newly published papers, transforming research goals into usable code, or generating new hypotheses. Another powerful tool for science is simulation. AI simulations are going to be critical for understanding and predicting dynamic systems that are simply too complex to model directly today. Our state-of-the-art WeatherNext models can predict hurricane paths faster and more accurately than traditional systems.

At Isomorphic Labs, we're modeling molecular interactions to massively accelerate the development of new medicines, supported by leading industry partners.

We're now in preclinical stage with multiple projects, including potential treatments for immune disorders and cancer. When we look back at this time, I think we will realize that we were standing in the foothills of the singularity.

Standing in the foothills of the singularity. I love that line.

I wonder where Demis got that line.

That's such a nice line. Yeah, you've said that before, Alex. It's a nice line.

Thanks, Demis. Alex, what's your take on this?

I think it's wonderful.

I'm broadly supportive of what Demis, Sir Demis, excuse me, is doing for DeepMind in science. I think it represents DeepMind at its best when it's challenging what Demis calls root node problems like fusion or like protein folding. I think it's wonderful. I don't think Google as a business has deeply vested interest interest in monetizing this. I think this is more for the public benefit from Google's perspective. But I have portfolio companies, companies that I've founded, that work very closely with Google on issues and technologies relating to this. And I'm broadly super supportive of DeepMind pushing these out to the public. I think it's very important. And they've made so many interesting— I would call them innovations relating to meta-science. How do you produce more science at the algorithmic level and hope to see much more from them in the future on this front. Amazing.

Alex Sisson, thank you for joining us on this segment. I know you need to jump. Uh, love you as always. Thank you for your brilliance. Appreciate you, pal. Thank you, Peter. Thanks, Alex. Thanks, everybody. Welcome to the health section of Moonshots, brought to you by Fountain Life. You know, AI is impacting every aspect of our lives— how we teach our kids, how we do our business— but one of the most important things that AI can deliver to is health. And one of the things I think about when, you know, shooting for 100, 120 is, am I going to have the cognitive health to be able to think clearly and keep my wits about me for the next 50 years? I'm joined here today by Dr. Dawn Musalem, the Chief Medical Officer of Fountain Life and a member of my Fountain Life medical team. Dawn, a pleasure. So, Dawn, talk to me about brain health.

Brain health. You know, you're right. This is the number one concern people coming into Fountain Life have is, "Will I remember the name of my child and the face of my loved one?" 45% of dementia cases are entirely preventable with lifestyle.

And what was really intriguing to me, Peter, is that a quarter of our members had advanced brain age, but over 13 months of us really helping them live healthier lifestyles, eating healthier, moving their body regularly, and optimizing sleep, people overlook that so often, but that sleep optimization is critical for our brain health. What we showed is that we were able to improve the brain age in 46% of those individuals.

That's a powerful number. That's amazing. You know, one of the things I love about Fountain is we're constantly searching the world for the most advanced therapeutics and bringing them to our members. So for me and all of you, I hope that you appreciate the fact that you can become the CEO of your own health. You can make sure that you've got the cognitive clarity, clarity for the next 50 years. Come and check it out, fountainlife.com/peter, to learn more and become the CEO of your health. Now back to the episode. All right, one more story from Google I/O. This one is very personal. Uh, we just announced the $2 million Build with Gemini X Prize. Let's take a listen and then I'll provide some detail about it.

The ultimate platform to make an impact. We are officially launching the Build with Gemini X Prize Hackathon. This global hackathon is going to offer up $2 million in prizes for builders who create apps that solve actual real-world challenges. And the premise is simple. Pick a problem worth solving, build with Gemini, and let's all try to positively impact the lives of a billion people. To build at that scale, you're going to need some serious power.

So a call out to all hackers, all builders out there. Uh, you know, we've launched two X Prizes in the last couple of months, the Future Vision X Prize, asking people to create a film trailer and a film treatment for the movie you'd love to see that shows a hopeful, compelling, abundant vision of the future. Uh, that X Prize is meant to help shape people's view of the future and actually to help shape agents' view of the future so they are positive and supportive and aligned with humans. This is one, you know, we've talked about on this pod all the time that the future is one of entrepreneurship instead of getting a degree to go get a job. Instead, find a passion, find a problem, and build on it. So I want to say thank you to Google for, for funding this. They put up $3 million, $2 million for the prize purse, $1 million for operations. And again, if you want to compete, go to geminixprize.com. You're going to look for a problem that impacts 100,000 people or more. And then you've got basically 90 days to build your product using AI. And again, very famously, you're going to describe what you want the product to do, the market, how you want to market it in English. The agent's going to build your website, your interfaces, and the team that's able to build it, market it, and scale revenue the most in 90 days wins this XPRIZE. Teaching people to fish instead of giving them fish. That's the goal here. Salem, any thoughts? Very exciting.

It reminds me of the— there was a problem, there's a medical problem called lazy eye and there was no cure for it. And then a team built an app which had a gamification system that solved lazy eye. So I think there's all sorts of things as we take obscure problems that hit a reasonable number of people and really go, go after it. Very, very excited. We've seen the incredible outcome when you put open innovation seeds, ecosystems like this throughout the history of XPRIZE. So couldn't be prouder.

Yeah, I'm super, super pumped. And, you know, full disclosure, both Salim and Dave are on my board at XPRIZE Foundation. Dave, you know, I think I'm excited.

I think this is the highest calling for XPRIZE yet. You know, being involved with XPRIZE, I've been incredibly proud of it for a decade now. But now with Google behind it, also OpenAI has a $100 billion charity now and they need to put that money to work as well. And so, you know, all these mega-funded companies are suddenly very, very interested in turning AI toward good, which is not an easy problem. And XPRIZE is really, really good at taking very hard problems and making them actionable. And so unleash that capital in ways that actually benefit humanity. You look at the oil cleanup XPRIZE and the impact that has had, that's had, and the kicking off of all of the space activity that we have via XPRIZE. So this is just the next chapter, but it's the biggest chapter by far.

I love it. And a reminder, everybody watching and listening, if you want to be there at our Moonshot gathering on September 25th, go to moonshots.com. We're going to be awarding both the Gemini XPRIZE as well as the Future Vision XPRIZE on that day. We're going to have the 5 finalists, the 5 creators with the 5 top film trailers, and the 5 finalists with the Gemini XPRIZE who've created the most revenue They'll be there pitching, and if you're in the room, you're going to be helping to vote on who the winner is. So, again, September 25th, go to moonshots.com to join the Moonshot Mates and be there with us at the Moonshot Gathering. All right, I see Andrew Feldman has entered the room. Andrew, a pleasure to have you here.

Thank you for having me on your show. Appreciate it. Of course.

By way of introduction, Andrew is the co-founder and CEO of Cerebras, a pioneering wafer-scale computing dedicated to accelerating AI training and inference Cerebras just raised $5.5 billion. I should say you just raised $5.5 billion, the biggest U.S. IPO since Uber in 2019, uh, you know, up 68% and market cap of $95 billion. Uh, quite the, uh, coming out party, Andrew. It felt good.

You know, you had to work for it too. It's super exciting for the team. We were able to, to a portion of the organization and their families and to share it. I mean, my parents were there and my wife and my stepdaughter, and we made it into a family event. And it was really something special.

Amazing generational wealth for everybody in your organization. Extraordinary. And we'll come to that story in just a moment. A little bit more in AI News before we turn to Cerberus and Chip. A big story. Andrew Kaparthy joins Anthropic. Andrej is an extraordinary individual. You know, he's work in— he's joined the pre-training team at Anthropic and he'll start a new initiative focused on using Claude to accelerate Claude's own pre-training research. You know, his announcement, he said the next few years at the frontier of LLM will be especially formative. And that's where he wants to be. Goparthy was, you know, has a stellar resume in AI. He was the co-founder of OpenAI. He left in 2017 to run Full Self-Driving for Elon at Tesla. He returned to OpenAI in 2023 and '24, and then he founded Eureka Labs. Let's take a quick listen to Andrej on a podcast called No Priors. This is a conversation he had before the announcement. So listen up when he's doing a shout out and a call out to which frontier lab wants to hire me.

And they're working on what's coming down the line. And I think if you're outside of that frontier lab, your judgment fundamentally will start to drift because you're not part of what's coming down the line. Right. And so I feel like my judgment will inevitably start to drift as well. And I won't actually have an understanding of how these systems actually work under the hood. That's an opaque system. I won't have a good understanding of how it's going to develop, et cetera. And so I do think that in that sense, I agree. And something I'm nervous about, I think it's worth basically being in touch with what's actually happening and actually being in the frontier lab.

And if some of the frontier labs would have me come for some amount of time and do really good work for them, and then maybe come in— Guys, he's looking for a job.

This is super exciting.

Then I think that's maybe a good setup because I kind of feel like it's kind of, you know, maybe that's like one way to actually be connected to what's actually happening, but also not feel like you're necessarily fully controlled by those entities. So I think honestly, in my mind, like Noam can probably do extremely good work at OpenAI, but also I think his most impactful work could very well be outside of OpenAI.

Noam, that's a call to be an independent researcher, an honor researcher.

That was Andrej on 5 cups of coffee. His quote that he put out on X, I've joined Anthropocene. Topic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and remain deeply passionate about education.

All right, Dave, your take. Andrew, he was calling to you. That was, that was the moment he was saying, Cerebras, I need your compute, please call me. He does. Everybody does, actually. So, uh, uh, Andrej came out with that auto research repo a couple months ago and I installed it. It's been auto researching on my cloud for a while now, but it desperately wants much, much more compute. And it doesn't need mega models, you know, it can run on lean, highly focused models. But I think he finally realized that every other OpenAI co-founder has either raised billions and billions of dollars, or is at a foundation lab. And he's the only guy who doesn't have access to the big machine. And it's just, he can't just sit there posting on Git anymore. He has to be part of one of these these big machines because we're on the cusp of the singularity and you can't miss it by being outside of the big game.

Salim, any thoughts?

I thought his point was, uh, well taken that he— where he said that if you're not in the foundation, in the labs, you're missing out. Things are moving past you, uh, and so he's trying to be at the cutting edge of knowledge. I just— I think there's— it's also notable he would have had his choice as to where to go, and I think it's very interesting that he joined on Really interesting.

And it's interesting that this was announced on the same day as Google I/O. A little bit of marketing strategy, perhaps. Andrew, what do you take about— what's your take on this?

Look, he is sort of one of the most important and prolific thinkers in the space. Right. And not just, I think, as reflected in what he's built, but reflected in sort of how he's taught the community. And I think that's interesting. I think his point that there are a small number of frontier labs that are sufficiently far ahead that if you're not with them, you're not on the frontier, probably applies to hardware too. Mm-hmm. Right, that if you're not building hardware for, engaged with at a fundamental level, one of the three labs the three most important labs, Google, Anthropic, OpenAI, you are not seeing what they're thinking. And just like your ideas will drift if you're a model maker, your hardware will drift from what they need as well. And I thought that that was really sort of, I applied that to our domain. And I think, I think he has just an extraordinary track record for useful stuff he's built. Extraordinary.

Yeah. And he's also just a super, super ethical guy and known for it. So he's a talent attractor. You know, a lot of people here in the Valley want to work for the most ethical organization that they can navigate to. And he's got that aura just all over, as do you, Andrew, actually, as does Cerebras, where we are right now.

Well, speaking about ethical organizations, let's jump into the conversation of— let's jump into the conversation about OpenAI versus, uh, versus Elon. Okay. All right, how's that for a transition, guys? Yeah, that's a stretch.

Yeah, you had to work for that one.

Uh, I did. I took advantage of, uh, of Dave's, uh, Dave's point. So the jury rules against Elon Musk in the OpenAI lawsuit. So federal jury unanimously rejected Elon's lawsuit in just 2 hours deliberation. Wow. Uh, uh, jurors ruled that Musk waited too long to sue, outside the statute of limitations, uh, for, for, uh, his claims. Elon's legal team, of course, is going to appeal. Um, I don't know what to say other than, uh, I, I would love this story to, uh, sort of not cloud would the entire AI data center conversations that we're having.

Well, very curious, Andrew, you know, OpenAI— you have a very close relationship with OpenAI. Uh, did you have a lot of skin in the game in this outcome? It, it's only going to be good news for Cerebras, but was it relevant?

I think this was a giant distraction, and, and, and billionaires in pissing matches interest me not at all. I, I think These are two of the most important thinkers of our generation. I think what Elon has built is breathtaking. I've met him. We've had dinner together. He's a polymath. He's a brilliant thinker. What Sam has built, sort of one of the fastest growing companies in the history of capitalism. But some of his ideas also in the invention of the SAFE at Y Combinator. These were enormously meaningful in the structure of Silicon Valley. I think everybody loses when they battle. I think, you know, I want Elon building cool shit. I want Sam building cool shit. And I don't want to waste time or read about sort of disagreements. I just want sort of these guys doing what they're the best in the world at, which is building stuff.

Well, well said, Andrew.

Well said. I'm just really angry it got to this point. Surely they could have looked at it and gone, the statute of limitations has expired. You just don't even bother. I mean, why? This is very upsetting. How much angst and bullshit time to Andrew Swan has been spent on this? And they could have been building. How much more could they have done?

Really good point. I mean, the ruling they made was obvious from the beginning if they were ruling based on on the timing. But this is the reason Elon's going to appeal, because he says that's not the point. I looked into this.

I looked into this and the appeal will very, very likely fail because it's a factual-based decision and the courts rarely overturn those.

Well, I think to Andrew's point, though, you know, Demis said earlier in this podcast that we're standing in the foothills of the singularity. Clarity. Like, an appeal is a slow, long— it's going to be irrelevant in the timeline that really matters, I think. Yeah.

All right, let's jump into, uh, part of the innermost loop: chips. And here we are, Andrew, Cerebras record IPO, closes up 68%, market cap $95 billion. Uh, and Dave, you're in the Cerebras offices this morning, aren't you?

I am, I am. I love the energy. I, you know, it's such a I think all of America is driven by this moment. You know, people building toward this event, and Andrew's journey was longer and harder than a lot. And to actually get there, you know, it's so hard. I pushed that button, about, you know, 1% of the size of Cerebras, but I still pushed that same button on the NASDAQ. And I didn't realize till— that picture's been hanging on my wall for years now. And it's like a lifetime achievement, kind of like a Nobel Prize or an Olympic gold medal where you carry it for the rest of your life. And so, so few people get to experience that. So I couldn't resist the opportunity. I was only a few miles away anyway. I couldn't resist the opportunity to come here and just feel the aura of— it's still settling in here, clearly. Andrew, is that the way it feels? Yeah.

And you're always welcome to come on by. Thanks. And enjoy the mojo that we've got going here. We try and create an environment where exceptional people can do extraordinary work.

Andrew, you and I met at the Citibank event, I don't know, 6 months ago. I was there giving the dinner keynote on longevity. I remember. And I wish I had only had a chance to go on your friends and family round before you had this epic, epic release. If you don't mind, tell all of our listeners and viewers here a little bit of the backstory, the founding story behind Cerebrys. What was the moment where you said this needs to exist? And you took a very different route than NVIDIA and other chip manufacturers. We did.

We, the founders had all worked together at my previous startup and AMD bought that in 2012. And by 2015, we'd wandered off a little bit and we started meeting and we saw AI on the horizon. And what we knew was that this new workload would eat through extraordinary amounts of compute. And we made two really big bets. We made a bet that said, like graphics produced the GPU and like mobile compute supported the development of the ARM processor, that this technology, this work would be big enough enough to require dedicated silicon. And the second bet we made was that, uh, the right strategy wasn't to build a derivative of the GPU, that you needed to start with a clean sheet of paper and you needed to do something fundamentally different. And these were enormously contrarian bets at the time, and both proved to be dead right. And from that foundation, we, we continued the, the innovative thinking, and We said, what AI is going to need is memory bandwidth. That's the sort of speed with which you can move data from memory to compute. And the way to innovate on that dimension is to use a different type of memory than everybody else uses, right? We have two types of memory. We have memory that can store a lot that's slow. We call that DRAM or HBM. And we have memory that is fast but can't store very much per square millimeter. And so what we hypothesized was that if we could build a chip the size of a dinner plate, right, a chip sort of 58 times larger than any chip ever built before, we could stuff it to the gills with SRAM, therefore thereby overcoming its weakness weakness in not being able to store very much per square millimeter and benefit from the strength. And that proved to be a very difficult problem to tackle. But when we got it, proved to be right. We are somewhere between 15 and 20 times faster than the GPU on any inference problem. And so, uh, the, the challenge along the way was that nobody ever built a chip big, not once in the 75-year history of the computer industry.

Yeah, they kept on slicing them thinner and thinner and smaller and smaller.

That's right. And that, that even sort of those on the, the, the sort of Mount Rushmore of our industry, people like Gene Amdahl, had failed, crashed and burned. And interestingly, even after we solved this problem, we had people come and visit our labs and then try and build it, and they also And so what it took was years of perseverance and innovation. And all the credit goes to the engineering team, Gary and Sean and Michael and JP and the team we had. We failed for years. And in August of 2019, we announced we'd solved this problem that had been unsolved forever. And we thought everybody would rush to our door. And the world didn't care one bit. Why? Why? The world was utterly indifferent. And over the next— I mean, in the first generation, I think we sold 12 systems. Wow. And in the second generation, we sold 300, 350. And this year, we're selling many, many, many thousands. And so what happened was, uh, we solved this problem them and were way ahead of the market. And it wasn't until 2024, late '24, early '25 that the models got fast enough and the models got smart enough that people wanted to use inference everywhere. And that, that's what happened. And sort of there we were with the, the fastest inference machines on Earth by orders of magnitude, and suddenly people wanted to use AI. And the way we use AI is with inference. And we were just crushed with demand. And then in December of 2025, we signed a deal with OpenAI north of $20 billion over several years, one of the largest deals ever signed in Silicon Valley. In March, we signed a term sheet with AWS for deployment in their data centers, and business has been pretty good since. Okay. Well, congratulations.

Let me just take a second and welcome Alex back. Alex. Hey, Alex. Good to have you back.

Good to be back. Amazing to meet you, Andrew.

How you going, Alex?

Doing well. I wanted to ask you, Andrew, I was talking to Valavan yesterday, your chief product architect, brilliant guy in Toronto, awesome, awesome friend. But I didn't realize that the company had a whole history as a training side company. Inference is now what, 80%, 90% of the market? It. And moving a huge amount of data from SRAM through the processing massively benefits on the inference side. But did you see that coming in the initial design? Because I don't think a lot of the research people even knew that inference would be so dominant.

I think that we got many, many bets wrong. And I think any CEO who looks back over a decade that moved as quickly as ours and, and says they got it all right is, is probably not a guy you want as your— at your birthday party. Um, we got an enormous amount wrong, but one of the things we got right was an understanding that we, we make AI with training and we use it with inference. And if AI is going to be smart and if it's going to be useful, you need to have an inference business. And that bit we saw Absolutely. And the real problem between 2020 and 2024, '25 was that it wasn't smart enough to be useful. And so everybody was focused and all the labs were talking about number parameters and now people don't care. The only question is, does it write good code? Does it give me good answers? Can it do things that I want done? And that's because we've moved into a regime, into a world in in which it's useful, and that's how it's measured. And so we did recognize this. We are really good at training, but right now there's such demand for fast inference, such overwhelming demand that we're allocating a lot of our attention to it.

Alex?

I'm curious, Andrew, SRAM, you mentioned SRAM earlier. The largest models, really the standard models at this point that offer frontier capabilities, and in some cases up to 10 trillion parameters. How do you think about SRAM when— correct me if I'm wrong— the Wafer Scale Engine version 3, I think, has maybe in the tens of gigabytes of SRAM. 40, 50. 40, 50, something like that. But the largest models are in the trillions of parameters. How do you think about the future of SRAM, given that, as you said, you're stuffing it to the models, right?

I think the following: I, I think that, that models that size have to be divided up, whether you're using GPUs or, or TPUs or, or, or us. They have to be cut up and they have to be spread over, over multiple chips, right? Remember, models that size, Alex, they have a very large matrix multiply in the attention head, and that doesn't fit on a GPU. You have to cut it up and you have to go tensor model And you don't have to do that with us. But what you do have to do is spread that over 4 or 6 or 8 chips. And what you do is you divide the model very carefully, and you divide it such that no layer runs over 2 chips. And so what you're moving, it results from one layer to the next. And you can move it because that's a very small vector. That's a results vector. Vector. You can move it over 100 Gig Ethernet, and it is slower. That little hop is slower, but the calculations that take up such a big portion of the time are so much faster that you pay a very small penalty for breaking it up into 4, 8, 16, 20 chips, on the order of 2%. Now, other SRAM solutions that are small, like for example Groq that NVIDIA required. They have to break it up because they have only 800 square millimeters to use. They have to break a big model up over 2,000 or 3,000 chips. And each of those hops hurts their performance. While we have to do a few hops, they have to do thousands. And so there's no way ever to fit everything on any size, right, amount of memory. But there is a very nice and simple way way for us to cleave models to spread it over multiple chips. And yesterday we announced and posted numbers on Kimi K2, which is a trillion-parameter model in the open-source community. We were, of course, an order of magnitude faster than anybody else.

Oh, really? How many chips? How many wafers on that?

I forget, uh, I've been busy, but it was about 1,000 tokens per second, where a really good GPU shop like Fireworks is running at 70. Yeah. Yeah. All right. And so, and they're a really good shop. And, uh, so, uh, you know, 15x, you know, that's pretty good. Yeah.

Peter and I were at Google I/O yesterday and they, they showed a whole rack of TPUs operating together, generating 1,400 tokens per second, writing code. And you see that and you're like, I need that. I, I need that tomorrow.

Well, like the trick there, Dave, sometimes is— and NVIDIA's been, been masterful at this sleight of hand— is not telling you whether they mean tokens per second per user or aggregate throughput. Ouch, right? The GPU is an extraordinarily good machine at generating slow tokens. Yeah, you can generate an NVL72 at 35 tokens per second, which is painfully slow can generate millions of tokens. On the other hand, if you ask it to generate tokens at 200 tokens per second per user, it can support 1 or 2 users. That's a $4 million solution working on 1 user, right? And so it's really important when you sort of dig into these is, are they telling you gross throughput? Is this a lot of customers who are unhappy with their performance? Or is this the, uh, are they able to, to serve that to, to individual customers and how many of them?

Yeah, because Andrew, you said— go ahead. Yeah, you were being complimentary of Elon as an extraordinary builder, entrepreneur, you know, sort of polymath earlier.

One of the best in history, for sure.

I'm curious, he steps up and announces TerraFab, uh, produced 50 times the amount of chips on the on the planet that exists today, outstripping, you know, TSMC and everybody else. What do you think of TerraFab?

I'm super curious. Look, I think Elon has proven himself on multiple dimensions. He's proven himself to be a visionary, right? You know, the number of people said you're an idiot to try and build cars in Fremont right? I mean, we've got the highest labor rates in the country, maybe among the highest in the world. We've got a regulatory regime that is unfriendly to business. The number of people who said you shouldn't build a rocket company, the number of people who didn't understand that he was building a rocket company because he wanted to— satellite company needed— I mean, he has been ahead of everybody for a very long time. Okay, but that's the vision side. That's the vision side. Side. Um, he's also been able to execute on some of them, not all. And that, that's what's cool, is he, he is trying to do things that other people can't do. Now, this particular problem I know a little bit about, and building fabs is very hard. And it is hard in a different way than some of the other problems he's attacked. I'm not saying he can't do it. I'm saying it will always take longer than he It will cost vastly more money. And that's the challenge of building extraordinary things. It is not a 5 or 10-year project in my humble view. I've been wrong before, but I put this at a 15 or 20-year project. Wow. I think it's interesting. It's probably good for the US that we have domestic fabs. But I think that there is a reason why Why? Even with the exact same equipment from ASML, right, Samsung and TSMC aren't at the same node. TSMC is ahead and they're extraordinary. And the amount of received wisdom and learning from the fabs they've built over generations cannot be underestimated. But if anybody can do it, Elon can do it.

But what does that mean for Cerebras? Because, you know, obviously US manufacturing And like you said, you know a lot about this topic. But US manufacturing of chips is critical. I mean, critical for everything, for national defense, for, you know— Global security. —the economy. Everything, yeah. And so if it's going to take 15, 20 years, that's just TeraFab. And the TSMC migration to the US is going very slowly, way behind schedule. So what does that mean just in terms of— well, first of all, supply-demand, just the raw ability to get things made. Made. You must deal with this every day.

These things are hard to build, right? I mean, fabs are pyramids. They are our pyramids. And TSMC is the greatest manufacturing company on Earth. And the challenge is these things take 5 years to build, 6 years to build, and $50 billion, $40 billion from the people who built the last one. Right? Right. And that's true whether it's TSMC or Samsung or any of the great builders here. These are unbelievably difficult to build. And that's why they're— they've been behind schedule in the US. I think they encountered some challenges that were unforeseen. I think we have political challenges in that these things take a long enough time that they cross administration boundaries, right? When your projects are more or can't be done in 4 years and have to cut across multiple administrations, maybe 2 or 3 different administrations, right, over a period of time. When you have local ordinances that get in the way of building, as happened with Samsung's fab in Texas, they redesigned the fab because of a local fire ordinance that made no sense. Um, these are painful problems that our system hasn't found a way to overcome. And so I think that we have to find sort of a way to do better because I think the reshoring of fab and not just the fab, fab gets all the glory, but the packaging business, every bit as important and something we lost entirely when the fabs left.

Yeah, yeah, actually good question. You know, by the time you get something ready to put into one of your data centers or a third-party data center, how many different manufacturing partners has that wafer been through? A fair number.

Yeah. A fair number. It goes to SFP and it goes to someone who, ASE, who deposits RDL on the backside. It's diced, it's cleaned. It comes to us for a step. I mean, it is a long process. I think, you know, when we stopped caring about fabs in the '90s and and IBM sort of left and GlobalFoundries' fabs sort of, we didn't do anything to keep them. We lost this collection of surrounding expertise, right? When a chip comes off a fab, right? It's a dead piece of silicon. The package is how you breathe power and life into it. How you get I/O and data into it and how you get power into it. And that's also an enormously challenging technology. It takes material scientists, it takes manufacturing engineering, process engineers, deposition engineering. And we punted all of that by not caring about this industry. And it's all sitting in Taipei and in Korea. The materials are manufactured in Japan. Kyocera is one of the leaders there. Yeah. And, uh, we got to get it all back, and, uh, we got to make a decades-long commitment to, to this industry.

Well, if you said that the TerraFab is, is 10+ years out, if I look 5 years out, uh, do you think that your, you know, Cerebras is able to manufacture on Intel, Samsung, and TSMC, or are there any other choices?

Or, um, I, I, you know, we, uh, we've committed our our 3-nanometer design to TSMC. So that will take us out a little bit. We do manufacture some components at Samsung and have a great deal of respect for Samsung's fab capabilities. We have never used Intel. You know, Lip-Bu is an extraordinary leader and a longtime advocate for hardware in Silicon Valley. As you know, Dave, there was a period between about 2007, 2006, and 2015 or '16 where every VC firm was filled with somebody from VMware who didn't know anything about hardware, who thought compute was made by flea feces in the cloud, right? That there was some sort of thing in the ether that somehow was generating compute. And we tried to explain for a long time that the way you make more virtual compute is to begin with real compute.

Well, I gotta tell you, you've inspired so many people that are in campuses right now that are eager to be part of your mission to get that back. So the more— I'm going to route as many as I can through this building. Please do.

I mean, guys like Andy Bechtolsheim and Lipu, uh, and a few others were continuing to put money in into hardware, continuing to— uh, Pierre Lamon, uh, continued to, to do it in support us as it was tough going to raise money over that time period. And I think, while I know Lip-Bu, they've got a lot of work to do. And he's done great things so far, but they've got some work to do before we could move to Intel.

Alex or Salim, do you have a question?

I have a quick one. Andrew, you've gone from raw invention, solving fundamental big problems, to going to now production. When you want to scale these things. What's the— can you say how long it takes to create one of those chips? And then over time, as you get better and more efficient in the manufacturing process, what do you hope it shrinks to?

Well, I say that the first one took 4 years and maybe half a billion dollars, somewhere between $400 and $500 million. That's why I take it to dinner when I go with my wife. Like a 10-year-old with his first dirt bike. I mean, it's coming to bed. I mean, it's in the bedroom. It's not outside in the garage. It is being carried around everywhere I go. I got a wafer. I think the inventions cut across lithography, chip architecture, packaging. Cooling. Cooling, power delivery and cooling. Tooling. They included sort of compiler inventions, algorithmic inventions. In fact, some of the hardest problems that we encountered were packaging. And, you know, we solved them 7 or 8 years before others encountered them. So the B-100 was— or the B-200 was 18 months late. And it was late because they had a problem with the COAS. What's that What does that mean? CoWOS is a process step where TSMC uses a 65-nanometer chunk of silicon as a motherboard. And so they put on that, they put on NVIDIA's chips and the memory. And instead of putting it on a green board, right, when you— a traditional motherboard, they put it on a piece of silicon. And the wires are more efficient in silicon. They can be narrower. And so this was a, a big invention, but we knew that there would be a problem with the coefficient of thermal expansion. And we knew that 'cause we'd solved that problem in 2018. And so there they were in 2024, '25, uh, struggling with a problem that we'd solved 7 years earlier. And that's what happens when you do pioneering work is, uh, you encounter problems, you have a chance to solve them long before the rest of the industry. Industry even encounters them and knows they're a problem. And so that is one of the joys. Obviously, in everything we do in engineering, there's a trade-off. The downside is there's some low days, right? There are some days you go home and some of these days stack up. And we had about 18 months where we're spending $8 million a month and we couldn't solve the problem. And when you have board meetings every 6 weeks, oh my God, you come in and you still couldn't solve it, you still can't solve it, and you're $100 million in more in the hole, and then you're $120 million in the hole, then you're $140 million in the hole, and you still can't solve it. This is some, some low days.

And then you have the IPO of the year and it's a high day. That's right.

I, I think, uh, Uh, Peter, you know, such an archetypal story of an entrepreneur.

It's the entrepreneurial journey.

Yeah, it is. I think, Salim, one of the things I've learned along the way— this is my 5th startup— is that, um, this shit will kill you if you can't modulate the highs and the lows. Yeah, it will. And that for, for every entrepreneur, every CEO, and I, I tell them first that this is a pressure test on your soul. And second, the number of times you can get kicked in the gut before lunchtime and have it still be a good day as a CEO of a startup is amazing.

Would you rather be doing anything else?

No, this is all I know how to do. I'm a professional David in the battle with Goliath. This is what I know. I have no interest in doing other things, and I have no interest in working with people who are other than those who want to attack the hardest problems. Problems. Yeah, amazing.

Uh, Alex, please.

Speaking of the hardest problems, and it's almost in the name Cerebras, you have, I think, a 4 trillion transistor budget with your third generation wafer-scale engine. I, I'd love to talk maybe a little bit about what's at the end of the rainbow, projecting out, say, 10 years when you're on your nth generation model. What does the future look like? Does it look like brain uploads running on WSE-8? What's the killer app? Where does this look? What does this look like in 10 years?

So Alex, I think one of the fun things about being an infrastructure builder is you don't have to have those ideas. No, really, I was with the team and many of them are here in the mid-'90s that helped drive down the cost of networking. We built some of the first and fastest Ethernet switches. And we had no idea that WhatsApp would arrive and that it would make possible even for the poorest members of our society to communicate home. And when I grew up in the '70s, the only thing I heard my grandmother say on the phone was, "Put your brother on, it's expensive." It was $4 a minute for my mother to call Australia, where her mother was. They spoke for 6 minutes a week. And the only thing I heard my grandmother say was, I'd say, "Hello, Bubba." She'd say, "Put your brother on, it's expensive." Right? And we put in a company called Yago, along with many others, with Juniper and others, we put a small brick in the wall that made the cost of IP transport so low that somebody else could invent a technology that made it such that every person can talk to their grandparents no matter how poor they are anywhere in the world. And that's something that we didn't know. Those aren't— that's not the problem I set out to solve and our company set out to solve. We set out to solve a problem as an infrastructure builder that we build roads. And what you drive over those roads and how far you take them, that's other people's work. What we're trying to do is allow people to do extraordinary things on our infrastructure. And so when I think about what we're enabling, that's work for Sam, that's work for Ilya, that's work for others. What we're trying to do is make a compute platform on which their ideas can take flight. And what we know is you need faster calculations.

So what I think I heard you say is is that you're very deliberately not having opinions as to the future shape of the workloads that will run on your infrastructure. And you're primarily at this point deferring to the frontier labs to steer the future architecture of workloads.

Today's frontier labs or new frontier labs, right? We are making bets that the world will continue to depend on sparse linear algebra as an underpinning all these calculations.

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

One final question. Orbital data centers, fiction, real, must-have? Are you going to put your chips up there?

No, I think first we have serious advantage in space. In space, some of the most expensive work is the chip-to-chip communication. Right? We've had chips in space for a long time. That's what a satellite is. A satellite looks like a PC motherboard with a big camera stuck on it, a big telescope, right? If you unpack what's in one of these small satellites, every computer hobbyist will say, holy cow, that looks like a server motherboard with a big camera on it. Big telescope stuck to it, and then it's hardened. Communicating in building a cluster is actually much more complicated because you have to do a lot of communication in this work. Moving the data from the land to the, the cluster is a problem that we've solved a long time ago. They will continue to improve it. So being a big chip and having to move things chip less often is a huge advantage. I think this is an exciting domain. I think like many hard problems, the last 10% don't take 10%, they take 90%, right? Self-driving is one of those categories, right? The last 10% we've been sitting at for 8 or 10 years and we're just now sort of getting over the hump of the last 10% because it isn't really 10%. I've got it in the 7 to 10 year category.

Category. So it's interesting, Andrew, just to pull on that very briefly. What I would've expected you to say to that would've been something like, with the wafer-scale engine, you had to design around faults. You had to be incredibly fault tolerant at wafer scale. And in a space environment with lots of ionizing radiation, you also need to be fault tolerant. And that Cerebras, with its experience with fault tolerance at wafer scale, is like the perfect computing platform for highly ionizing environments.

There are, you know, I think we have lots of advantage, Alex, and you put your finger on one of them that you have to try and sort of shield your silicon very differently in space and you will get more flaws. And the, you know, there are single-bit errors, there are hard errors, there are a whole collection of errors that you have to contemplate. And our ability to shut down a core and route around it is an enormous advantage in that environment. I think we've got, as a community, some work to do over the next 4 or 5 years before we have sort of the— the truly hard part of getting them in space, orchestrating the software, getting them to communicate. So I've got it sort of out the better part of a decade before we have sort of production in space. I think it's a very sort of a worthwhile project to pursue, but it's out of the way.

All right, Andrew, we close out these segments with an AMA with our incredible subscriber base and would love to have you join us. So we've chosen 8 questions from our comments, which we all love and read, and we'll be peppering them along. So I'll put them up here. Saleem, I'm going to give you first shot. Andrew, you can look at the others and see, get ready for one of them. We have a second page we'll go to. So Saleem, why don't you pick one of these? I'll go with the first one.

If the world becomes compute constrained, does the Tencent lawyer for everyone thesis still hold? Does AI become a luxury only the rich can afford? And this comes from @georust1. So this is a fairly— if you've been listening to the podcast, there should be a fairly clear path here, right? Because every major technology starts with scarce and very expensive. And then you saw this with computing bandwidth, DNA sequencing, solar energy. They all looked very constrained and very expensive initially. But then the learning curve kicks in, infrastructure kicks in, Jevons paradox kicks in, rights law arrives, competition arrives. And the cost collapse. AI compute is going exactly down the same path. You may have some bottlenecks like chips and power generation, data centers, but those become investment honeypot pots and capital floods towards those bottlenecks, right? But the bigger insight is that AI is not consuming compute. It's helping design chips and optimize infrastructure, what Alex calls the inner loop. It's improving, it's compressing the models, et cetera, for that example. And the system becomes recursive. And as you have intelligence building more intelligence, this is why we get so excited by this future. It's going to drive the cost of, of everything down to that. Maybe it's a $2 lawyer first, maybe it's 50 cents, but over time it's going to get to 10 cents.

And it's down from $1,000 an hour. Yes, that's number one. Number two, I, I think the problem with lawyers and accountants, uh, Uh, is the structure of their business is wrong. Business model for the future is exactly wrong. Selling hours. Yeah, right. What their business is to stand between ordinary people and obscure knowledge, right? That's what your accountant does. You don't want to figure out what the tax rules are with related to depreciation on a property you bought or was gifted to you in 2020, right? Who wants to know that? And so what their business is, is sort of the acquisition of obscure knowledge and the application of that knowledge to particular problems. That's exactly what language models do.

Don't you think, Andrew, that generalizes though? Like, what are you other than a gatekeeper of obscure knowledge regarding the highest— Oh my God. I so disagree with that. What are we all?

We're all specialists. Andrew is like an engineer.

Lawyers are at their best. They're actually, you know, the reason they're called counsel is when they're giving advice, not on legal matters, when they're giving good business counsel, when common sense is challenging in a confusing environment. Those are when they're at their best. I think when you're drafting drafting, all the documents you need for most things have already been drafted. Right? We don't need another lawyer reviewing another NDA.

We don't need— Let's give you next shot. Which question, 2, 3, or 4, you want to choose?

I think number 3 is interesting. Yeah, me too. I think there is a profound misunderstanding about how— So let's read the question.

Why can't money buy Elon or Zuck a lead in AI? Can't they just buy the best talent? And that's from @NovaRift. Great question. No, the answer is no.

And I think, you know, why couldn't Intel build a cell phone processor? At the time they had the best fabs. At the time they had the best computer architects and they destroyed tens of billions of shareholder dollars failing. Same with AMD. It turns out in our industry that money money and the acquisition of talent isn't enough. What is? There's something else. Well, MTP. What?

Massively transformative purpose is incredibly important. Intel could have said yes to Apple, though. No, no, but they could have.

But the truth is, is what led them to believe that they were in a position to say no to it?

Intel was chasing margins. Intel was infatuated with its own profits. They had an ARM division that they sold off. Intel could have sold— why?

That's the thing we're trying to understand, is the innovator's dilemma.

They were fat and happy and lazy.

Maybe, maybe. Or, or maybe that there's something in your DNA that makes luck— big mutations. Well, look, I, I'm sitting here saying all day long, we'll take luck over skill. But I'll also say that all day long, that extremely hardworking people with tremendous grit end up more lucky. And that both of those are true. That is life, is really hardworking people over long periods of time who have integrity and ethics, right? They get lucky more often. And that's— luck is not equally distributed to those who work hard and those who don't. Carlo, go ahead. Go ahead, go ahead.

I think I'll throw in a quick plug here. We're launching this service next week. It's at shapingluck.com because one of the things is that luck is nonlinear. And in a world that's going exponential, you want nonlinear outcomes. So if you're interested in, go join us at a webinar.

What's it called again?

What's the URL?

Shapingluck.com is the URL. Okay, awesome.

Free webinar coming a lot. But I think, Alex, the question isn't, you know, of course they could. Why didn't they? Why did they miss it? Why did AMD miss it? Why did— right, why did, for example, uh, why did, uh, NVIDIA fail for decades at everything that wasn't a GPU? They failed to build an ARM processor that worked. I think it was called Snapdragon. I think they failed at Northbridge, Southbridge And they succeeded beyond anybody's expectations at a GPU. I think that the same question, I think, holds true for the Yankees, right? No, no, no. Why doesn't the team with the biggest money win every year in the NFL? Why? I mean, there is something that we we have, in thinking about organizations and talent, that we don't do a very good job of describing that says there is something that is very hard to buy and that has to be made and that we don't seem to be able to articulate it well and buying the most talent doesn't seem to be sufficient. You have to have a lot of talent. Necessary, but it's clearly not sufficient.

Uh, Alex, let's go to you next for a question. Want to get through our lightning round here? Yes, of course.

I have— suffice it to say, I have some pretty different answers to some of these other questions, but I think I have to answer question number 2, which it looks like might have been a response to a comment that I made in a previous pod episode. So the question is, why wouldn't Sam— I think referring to Sam Altman— cut a deal with Bezos and Blue Origin to become the other counterweight to Elon? And this is from from Scott Ray Broomfield. So I think the answer is that's probably on the table. If I were Sam, I would be exploring a variety of potential heavy launch partnerships to become a counterweight to Elon and SpaceX AI's Dyson Swarm. I think heavy launch is going to become, is already arguably part of a critical element of the stack now. Now for the future of compute. Space is becoming—

yep, as you know, right? New Glenn is more akin to Falcon 9 and doesn't hold a candle, uh, to Starship, which by the way will be making a launch attempt, uh, probably by the time this is out. Good luck to Elon on that launch attempt, super excited. But Starship is coming in at a factor probably 100 times cheaper.

I'm not sure that matters. Oh, go ahead, Andrew.

Andrew, go ahead. No, the—

I, I think Alex, all your points are right, and I, I I think that you underestimate Sam at, at your tremendous cost. I think what Sam has sort of done again and again in our industry is see around corners that other people missed. He was trying to lock down data center capacity in space last year and the year before when all the other foundation labs didn't see it. Really? He was trying to lock down memory. Oh, yeah, for sure. I didn't know that. His ability to look at an exponential and not be afraid of what it says in 2 or 3 years while everybody else is afraid, saying, oh, we're not going to need that much, is extraordinary. And his reach is extraordinary. And so with 100% certainty, I will tell you that he is exploring Zuckerberg, uh, uh, deals with every possible way to get access to compute and data center capacity. And I, I can say that having watched from a distance, I have no inside information, but I've been dazzled by that ability of his. I, I think you underestimate that guy. I mean, you would think it would be enough to build the fastest growing company in the history of capitalism to, to, to get a lot of respect. Right? You think that might be a, a sufficient feather in your cap, but I, I think, I, I think he will certainly, uh, be in conversations to, to get compute, whether it's in space or whether it's under the sea or whether it's on, you know, you using, uh, falling water, using geothermal. Uh, he will be in those conversations and his team will be there. Every single day.

Alex, take one point for me.

It's so cool to have another friend who's on the big, big stage. It's just that inside perspective is awesome.

Alex, one point here is, again, if, you know, Elon's Dyson Swarm is 500,000 satellites to a million satellites, it was like a launch every couple of minutes of a Starship. You don't get that when you Glenn. So that vehicle isn't designed for the frequency of launch that we're talking about here. So could, uh, you know, could Sam put up a mini constellation with Bezos? Sure. Could he put something up to really compete with, uh, with what Elon's proposed? Not without new launch capability. That capability is unique on the planet if it pans out as expected.

Yeah, a few thoughts. There are lots of options. I, yes, I agree with the contention that SpaceX is completely dominating mass-to- orbit, no question about it, including dominating historic mass to orbit. So if I'm, if I'm Sam, I would be exploring probably a multi-channel strategy. A, I'd be exploring a deal with Elon and SpaceX to leverage SpaceX launch for my own Dyson swarm. I would be exploring alternative launch capabilities. And then if you really believe— again, I think this is the elephant in this space room, if you really believe that we're on this singularity-esque exponential, then the fabs don't need to be on the ground. We can build fabs in space, and we may not be addicted to heavy launch 5 to 10 years from now. And if you're Sam and you're playing the long game, then you're looking for ways to build fabs on the Moon and in LEO that don't need the SpaceX near-monopoly.

Amazing. Dave, do you want to take us to question 4?

Andrew, you—

did you have another thought on the NASA space? No, I, I, I think, uh, I think building a fab on, on land is hard enough for me.

Yeah, yeah, sure.

I mean, for $40 or $50 billion in 5 years doing something nobody else— I mean, that, that one or two other companies in the world have ever successfully Nicely done. I mean, I, I, I can't really think hard about building fabs in space.

No, there's a great, great segue to question 4, which is directly related. China is building massive compute capacity. Could they sell tokens to U.S. users— losers— users at very low prices and disrupt providers? It's, it's the same answer you just gave, Andrew. Like, no, China is not building massive compute capacity when looking at tokens per second, you know, the driver of AI, they have— that's why they so desperately want to import from the US. So they're building as quickly as they possibly can, but it's not, you know, 5-nanometer, 4-nanometer, you know, 3-nanometer technology. And so it's all bottlenecked today— ASML machines, fab construction, everything Andrew's been talking about. So if they had the ability, they would love to do that, but they just don't have the compute.

Yeah, the, the dimension in which they've chosen to invest so far is in power infrastructure. Yep. And at that, they're just playing better than us right now. They have upgraded their grid, they have tremendous power infrastructure, and we've made bad decisions there. We are stuck with a grid that's built in the '50s, it's designed not for, uh, what we'd like it for today, and we have trouble politically at the local, at the municipal, at the state, the federal level doing projects like infrastructure. And so what they have done is a tremendous amount of investment there. They are obviously starved of compute, but they're going to try and build on what they have, which is an absurd amount of power infrastructure.

Yeah. I would say when enterprises are going to be doing most of the token purchases, You're not just buying the token, you're buying trust, you're buying governance, you're buying reliability, you're buying compliance, et cetera, et cetera.

I would perhaps just add to this, it's worth noting that in the past week or so, there's been quite a bit of public reporting about how China is operating proxy services that are selling American tokens to Chinese users at incredibly low prices. Yeah. Like, they are. 10x discounts in order to siphon the reasoning phrases for training their own models. And that is quite disruptive, and Anthropic is pursuing that.

Amazing. Andrew, we want to thank you for being on. We close out every episode with user-generated content. This is sort of our outro music. And so let's enjoy this one. It's called "We Are as Gods," a, I guess, comment to my new book. It's from my son, Benny.

Today Andrew is a god, so we have to like bow down to you.

I think being a CEO is sufficient, man. I don't want the responsibility. Congratulations. I don't need any of that at all.

Congratulations on an epic IPO. Amazing. Amazing. All right, let's, uh, let's listen to We Are As Gods by Musad Zamani. All right, enjoy.

Everyone's a founder now. Abundance is the land.

AI is the lever.

The future is forever. The mission is your soul. Take the full control.

Oh, Peter sees the vision and Alex writes the code while David finds the logic and Salim leads the road.

The agency of one, see the power rise, the exponential life beneath the digital skies.

All right, lyrics. That was a good one. Very cool. Uh, again, thank you for joining us. Gentlemen, always a pleasure. I think we could have kept those conversations going for a couple more hours. Easily. Easily.

Yeah. Yeah. Thank you for having me. Really appreciate it. Be well now. Thank you. Thanks, Andrew.

If you made it to the end of this episode, which you obviously did, I consider you a Moonshot Mate. Every week, my Moonshot Mates and I spend a lot of energy and time to really deliver you the news that matters. If you're a subscriber, thank you. If you're not a subscriber, yet, please consider subscribing so you get the news as it comes out. I also want to invite you to join me on my weekly newsletter called Meta Trends. I have a research team. You may not know this, but we spend the entire week looking at the meta trends that are impacting your family, your company, your industry, your nation. And I put this into a 2-minute read every week. If you'd like to get access to the Meta Trends newsletter every week, go to diamandis.com/metatrends. Metatrends. That's diamandis.com/metatrends. Thank you again for joining us today. It's a blast for us to put this together every week.