Elon’s Anthropic Deal, The Next AI Monopoly?, “FDA for AI” Panic, Trading the AI Boom

How do I sound?

Brad Gerstner

You sound perfect.

Great. Do I look?

Brad Gerstner

Yeah, you sound great. Better than you look.

You’ve got that face made for radio.

You don’t look as tired as you have in recent weeks.

That’s true. Yeah.

Yeah.

Somebody was slagging me for the bags under my eyes. This audience is brutal.

They’re brutal. They’re brutal.

It’s a good thing I’m rich.

You guys see those Spencer Pratt ads?

Wow.

It’s one of the best political ads I’ve ever seen.

Oh, there are 3 or 4 of them.

There are multiples.

Yeah.

Whoever that social media team is, they’re on fire.

If you get a good social media team and a good ad production team, I think it’s next-gen, because these things go crazy. Spencer Pratt, if he wins this election—which I think he’s going to in Los Angeles—the reason is what Brad said. Those ads are incredible.

Brad Gerstner

Well, he’s also quite a good debater. Did you see clips from this debate?

Absolutely. He’s so funny. He’s so chill.

Yeah, well, he’s up against Karen Bass, who’s the mayor and is basically extremely left-wing. Then there’s someone who’s a city councilwoman who’s even further to the left of Karen Bass. I mean, she’s often in Fidel Castro territory.

She’s an Indian Fidel Castro.

Raman. She was basically criticizing the mayor for the homelessness problem. Then Pratt pointed out that this councilwoman is actually in charge of all these homeless programs already.

He eviscerated her. He basically made the key point, which is, “Look, the problem here is not a lack of housing. It’s an addiction issue and a mental illness issue.” He said, “Look, if she went to the street, she’d get stabbed in the neck.”

Yeah, which is pretty accurate if you’ve been to Skid Row. You would not want to walk through there. It was like the Spider-Man photo. We’ve got the clip. Play the clip, Nick.

Oh, God.

This clip is brutal.

We played a different clip, but this one went super viral. It reminded me of Trump a little bit.

Let’s see.

Nithya Raman

I’m not sure how to respond to that vision of Los Angeles. This is a MAGA Republican’s idea of what Los Angeles looks like. This is really not—

For those of you listening, he put his hands up and wiggled his head like, “Oh my God.”

Hey, hey, Sacks. You know, if Spencer Pratt wins mayor, the Retirement Protection and Savings Act is going to pass. It’s going to pass with big numbers. This is the referendum that is effectively going to knock out the wealth tax. Can you imagine if California effectively passes a constitutional amendment protecting retirement savings and personal assets and banning the wealth tax, and Pratt gets elected? The message that would send to the country—that’s a very non-consensus view that I’m becoming increasingly optimistic about.

Well, from your lips to God’s ears. But until that message actually is sent, I think I’m going to be—

In Texas.

In Texas.

This is also in the face of—I don’t know, just a follow-up story here—Mondami doing an attack video on Ken Griffin’s house. We talked about it on the pod a couple weeks ago. He literally stood in front of his house and pointed at it, and this is in the face of a CEO getting shot for ideological reasons and Sam Altman’s house being targeted. This is a really dangerous thing for Mamdani to do.

Ken Griffin came out today, or yesterday, in an interview and said, “Hey, listen, I’m out. We’re going to be putting our efforts into Florida.” This is the same thing that happened to Chicago. He basically said, “I really felt offended, and I was nervous about this because of my personal safety.” He called him out. Mondami came out with a mealy-mouthed response that didn’t even apologize for what he did; he just doubled down on it, essentially.

I don’t know if you guys saw it or not, but who cares? New York is becoming a flyover city.

It’s an interesting way to put it. I don’t disagree.

All right, first story. Elon just leased all of Colossus 1.

New data center?

Brad Gerstner

He did?

What?

Yes. Shocking to Dario and Anthropic. Chamath, on this week’s pod, go ahead and give yourself a pat on the back. You said Elon and Dario should do a deal tomorrow. It didn’t happen the next day; it happened 5 days later, so you came close, Chamath, but no cigar.

Because of Anthropic’s obvious compute constraints, Anthropic just added over 220,000 NVIDIA GPUs and over 300 megawatts of energy. The deal is already having an impact, as we’ve discussed here. Claude users have been experiencing rate limits. Claude has now doubled the Claude Code rate limits, removed peak usage caps for paid users, and increased API volumes for Opus models.

xAI is now trading their models at Colossus 2, so they have more than enough compute. Elon made a great bet on compute and built up those data centers really fast, and that is now paying off.

We had the Cursor deal we talked about last week. Let’s talk about the emergence of Elon Web Services, or EWS. He is now in the hyperscaler business, competing against Google Cloud, Amazon Web Services, and Azure. I don’t know if you had inside information or just a brilliant epiphany, but take us behind the call. What do you think about the deal itself?

I think the deal is fantastic. I’ll say 3 quick things.

The first is, as I mentioned a couple weeks ago, Anthropic and OpenAI’s revenue performance has nothing to do with demand—zero. It has entirely to do with the supply constraints that exist in data centers, and specifically in power. If they had infinite power, I think their revenues would probably be even more parabolic. All the breathlessness about either exceeding or underperforming a forecast, in my opinion, means nothing. I think the 5-year view for those 2 companies is quite robust. What they really need is more compute and more power.

The second thing is that, while they need that, we have a very big problem: We unfortunately have very poor leadership at the head of most of these AI firms. I think they are coming off as untrustworthy or too self-interested. The political reaction is now starting to turn negative, and the community reaction is negative. You have about 9 gigawatts that are supposed to come online this year. Almost 50% of it is now being protested. More than likely, if history holds, most of that will get turned off, so they will get even more supply-constrained.

So that’s the setup. What’s the opportunity? I think for Elon, if you look inside how people try to nitpick the SpaceX valuation case—or, let’s not even do that; let’s be more generous—when people try to paint the bear case or red-team the valuation, the biggest element is the on-the-come value around the orbital data centers.

By actually landing a bunch of terrestrial capacity, I think you start to blunt that. You can now start to say that even if the orbital data centers get delayed by a few months or a few quarters, and even if the technological de-risking of them takes longer, he now has a structural core business that will effectively subsidize his ability to train Grok. I think that’s a really important and underreported theme.

You have all this infrastructure. He somehow saw the tea leaves before most people, built to a level of scale, and secured power before most people. It has now become the critical asset. He’s kind of king-making, and I think that’s a really interesting valuation reinforcement as SpaceX goes through testing the waters in the roadshow.

Brad, your take.

Brad Gerstner

Yeah, no, I think it’s well said. First, we know that there’s nobody better on planet Earth than Elon at converting electrons to tokens. It’s a critically important evolution to the story.

Our friend Shawn Maguire sent out a tweet that summed it up well. He said SpaceX has this 5-layer cake: launch, connectivity, compute and hyperscaler, space data centers, and then applications and models, and then other bets.

The question on the roadshow has been, “But xAI doesn’t have the revenue trajectory of OpenAI and Anthropic, and yet there are huge commitments.” Now we see the ace card that Elon’s playing. He said he was building AWS all along—or EWS all along.

I estimate that this is going to generate an incremental $45 billion of revenue this year, on top of what I’ve seen analysts estimate in the mid-$20 billions. That’s a material amount of incremental revenue to offset the cost of the investments that he’s made here. That will subsidize, to Chamath’s point, all that he’s investing to build the next generation of growth.

Remember, too, that he has 3 facilities.

Colossus, Macrohard, and Macroharder. There are 1.2 gigawatts in Macrohard and Macroharder, in Blackwell. So he's given the one that's kind of less connected—H100s, which are great for inference—to Anthropic. He's monetizing it in a big way. It's terrific for Anthropic, and it solves what I think was the biggest question in the valuation story: what if he spends ahead of xAI's revenue?

It takes the pressure, Chamath, off xAI delivering immediate revenue. Now he becomes an immediate competitor in hyperscale. I don't think this is the last announcement. I think he's going to make a lot more moves in this direction. I think it will be a material part of their story and their revenue projections as they come together.

And finally, I would just say, again, everybody has talked about how we don't have enough power, how we don't have enough compute, and how the revenues would not show up this year. But the chaos that is American capitalism somehow finds a way, okay? There's tremendous demand for Anthropic, and we find a way.

Brad Gerstner

I was so happy to see the détente and the kind exchange between the team at Anthropic and Elon, because we need all of this in order to produce American frontier models and stay at the frontier. And finally, I would just say, Chamath, you referenced these activists that are protesting and delaying these data centers in these localities. One thing I want to dispel is this myth: these are not organic, hyperlocal protests by people in a community who aren't being spurred on. These are highly organized activists moving across the country to stir up trouble in the exact same way they did to stop all fission reactors from being built 30 years ago in America.

Now we have no nuclear reactors being built. China's got 100 of them. Who was funding those activists? I think we need to really look into who's funding the activists now. I'm not saying that there aren't any concerns, but there's misinformation about water and electricity bills. Electricity bills are going up in the places that are not building data centers—New York and California—because they haven't built any supply on the grid. In Texas, where you're building the most data centers in the country, electricity costs are going down. So I think that's a boogeyman that we've got to take on.

Well, look, the deal is highly complementary, as Chamath and Brad pointed out. SpaceX has a profitable—I think very profitable—space and telecommunications Starlink business, the satellite business, but the xAI business had huge losses. The reason is pretty straightforward: you need these superlarge training clusters, but they cost a lot of money. Until you have a model that's capable of competing at the frontier, you're not making any revenue.

That problem was compounded by the fact that right now all the revenue is in enterprise, which is to say coding. We know that xAI just did that deal with Cursor to try and catch up, but they don't have a coding product yet. So they're not participating in the revenue, but they're participating in all of the cost. This deal fixes that problem. Elon is now able to have a frontier model company, but he's also able to avoid these massive unpaid-for CapEx commitments because he's able to lease that capacity. So I think it solves a major problem for them and their balance sheet.

And then you have to say that for Anthropic, this is a really great thing because they were compute-constrained. To build on that point, let me be the first to congratulate Dario on winning the AI race.

Let's be honest, Sacks. You have been on this podcast and been moderately critical of that company and Dario himself for being a little P-doomer 110. On your X account, you've been even a little spicier. So now that there's peace in the Middle East of the AI business, what's your take here?

My take is, look, let's honestly and accurately assess where the state of this AI market is right now and Anthropic's place within it. For the last 3 years, Anthropic has been growing at a rate of 10x a year. I think going into this year, probably the conventional wisdom was that there'd be no way to sustain that kind of rate of growth at this level of scale.

What happened in the first 4 months of the year? First, we find out that from January 1 to March 31, they grew from roughly $10 billion of ARR to $30 billion. So it tripled. And then in April, if anything, the rate of increase seemed to accelerate. They went from $30 billion to $44 billion of ARR. Nobody in Silicon Valley has ever seen anything like it. Forget about the rest of the country. All we do in Silicon Valley is deal with exponentials, and still people have never seen that kind of growth at that level of scale.

The only thing holding them back in the future was compute. Now they've made this deal. They've made other deals as well to get that compute. I think it's pretty much a foregone conclusion that they will hit that forecast of 10x this year, exiting the year at, call it, roughly $100 billion of ARR. And now the only question is whether they hit $1 trillion in 2027. And we can debate whether that's—

On board.

No, we can't. We can debate whether that's true or not. But look, if they do that, I think they'll easily be the most valuable tech company in history. In fact, they might even be more valuable than the rest of the Mag 7 put together.

Just to give people some basis for comparison here, the biggest tech companies—you know, Apple—

Nvidia.

Nvidia, Google. I think they kind of do around $400 billion to $500 billion a year right now of revenue. I mean, I guess Nvidia's a little bit of a different category, but you look at Google—

The three—

The hyperscalers, yeah. I mean, Google is doing what? $120 billion a quarter? Something like that. $100 billion a quarter.

Correct.

But growing at what? 20% year over year? Not 100%, and certainly not 1,000%. So the fact that Anthropic could be on track—in fact, let me correct that—

Them going to the Mag 8? It'll be a Mag 8.

I'm saying something else, which is that unless something about their current trajectory changes, Anthropic will be the most powerful monopoly ever created in human history.

Go.

Again, it will be $1 trillion of ARR, growing at some exponential—

Interesting.

Dario calls it AGI. I call it the biggest monopoly in human history.

Interesting to hear that word “monopoly,” Sacks. Very interesting placement. Chamath, go ahead, and then I'll move on.

Apple in 2025 was $420 billion. Microsoft was $300 billion. Alphabet was $390 billion. Amazon was $700 billion. Nvidia was $190 billion. Meta was $185 billion. Tesla was $110 billion. Total: about $2.3 trillion to $2.35 trillion. So if Sacks is right and Anthropic can tack on $1 trillion, it won't be the Mag 7; it'll be the Mag 1.

Brad Gerstner

Just to put it in perspective, though, Dario and Dwarkesh said he thought the combined AI revenue of the market leaders would be about $1 trillion in 2029. I love what you're saying, Sacks. I think there is unlimited TAM. We may be over our skis a little bit in terms of the forecast. If you back your way from compute, they're expecting to have 5 gigawatts by the end of this year and 10 gigawatts by the end of next year. It's hard to get to those numbers for a single company, but I do believe that the trajectory they're on—I totally agree with you—is an exponential curve that not many people believed in 4 months ago.

Right. So then the question is, okay, I think we all agree they're on an exponential curve and that the TAM is big enough to support that. Just one data point on TAM: my understanding of the total market size just on coding is $1 trillion, meaning that roughly $1 trillion a year is spent on software developers and all things related to the creation of software. Now, I'm not saying that they eat that entire market, but I can easily see the market for software doubling.

Well, hold on. It's doubling from $1 trillion to $2 trillion, given that coding tools basically 10x or 100x the value of that market and the ability to generate code. So I think we all agree that the TAM here is large enough to support $1 trillion of revenue.

Brad, I think you bring up a couple of really important constraints. First, there may not be enough compute, and there's not enough energy. I'd say the second big one is: what's the competitive reaction going to be?

At the beginning of this year, all these frontier labs were playing around with a lot of different things. Anthropic was the porcupine; they believed in one thing. All these other companies were acting like the fox who thinks they're good at a lot of different things. They were doing Nano Banana, they were doing Sora, they were doing image generation, and they were doing fantasy-character chatbots. In hindsight, they were doing a lot of things that appeared to be a waste of time.

The whole market appears now to be coding and the things that we built on coding tokens, like co-work, like agents. So there is going to be a competitive response here where all the other guys realize, “Oh, wait a second. We were misfocused.” They're going to get focused. I just don't know how much share they're going to be able to take.

It does look like OpenAI has already made the pivot. We hear very good things about Codex now based on GPT-5.5. GPT-5.5 is based on a new base model called Spud. I think they're very optimistic about continuing improvements. Their rate of growth appears to be accelerating now because of GPT-5.5.

So, look, there's reason to believe that OpenAI can take some share here. I'm sure that Google won't be asleep at the wheel. They're very, very good at coding.

They've got a really good team. Elon just tied up with Cursor, so there is going to be more competition. But still, what you have to say—and I think all of us know this from Silicon Valley—is you always want to be the company in the lead, on that trajectory where all you have to do is maintain inertia, whereas the other people have to change something in order to put themselves back in the race. So this is when I say, somewhat facetiously, “Congratulations, Dario, on winning the AI race.” I don't mean that he's won it, but he is winning it right now.

Well, here's the brilliance of what Elon's doing. If you look at the existing business, which is Starlink and basically the launch services at SpaceX, it's an incredible business—$20 billion this year, I think, is the estimate. But if you look at the footprint of Amazon Web Services, Azure, and GCP, you're looking at $300 billion in revenue and a combined market cap of $5 trillion—or $4 trillion if these were independent companies.

If you look at Elon's core competency at Tesla, it's building factories. And if you look at the footprint of these factories, they're huge. What are data centers? They're basically big, giant factories. Then, if you look at energy, what else is Elon extremely good at? Battery deployment. He's also got solar deployment from the often-criticized SolarCity acquisition he did years ago.

You put this all together, and if this is $5 billion, as I think you referenced, Brad—if it's $5 billion in incremental Elon Web Services business—and he's a neocloud, what could he build on planet Earth? What could he build inside of Teslas in terms of extra compute? What could he build inside the Powerwall? What if the Powerwalls had his new fabs in them and you built a distributed system from home to home?

The Powerwall has compute in it, the cars have compute in them, and, of course, the ultimate manifestation of this, where nobody can complain, is you go right out into space. That's what he's going to do. The sneaky small part of this announcement from Elon, Anthropic, and Dario was that they're also interested in space.

So look for the race to go from factories and data centers to homes, with compute in the Powerwall. It's already online, right? Starlink also gives him the ability to do distributed compute in people's homes. Again, you could be paying people to put Powerwalls with computers in them. That's going to be the next shoe to drop, I believe.

Did you guys see the deal that was announced yesterday between Pulte Homes, which is a huge builder—

Yes. They're doing it as well.

—and SPAN?

Yes.

Nick, just throw this up here. It's super cool. What's happening is that these guys are putting mini data centers with NVIDIA GPU clusters beside every home and then allowing people to actually run those things. That's just incredible. I thought that was so cool.

It's a great pivot. What this company did originally, Chamath, was make smart Power Panels. When you flip your breakers, all those breakers are in an app. I looked at it for my house, but I guess they pivoted to add this. I think Base Power Brad urine industrial unit they're going to do the same thing.

Yep, Zach Dell's doing that. One of the things I would say in response, Jason, to what you just said about Elon is this: This is why the SpaceX IPO is going to trade at 40 to 50 times revenue. Next year, if they do $40 to $50 billion and this thing goes out at $2 trillion, they're going to trade at a really high revenue multiple compared to the Mag 5, which are trading at around 25 times earnings.

There's only 1 person on the planet who has a future pipeline of innovation and the largest TAM in the world because he's playing in all these different spaces that can command that multiple, and it's Elon. It's deserved, and it's great for the country.

Tesla has that same Elon variable in it as well. People value his companies at, I would say, 2 times the market, 3 times the market, 4 times the market because of the future pipeline. You guys devalue Apple because they don't have somebody like Elon or Steve Jobs there who is giving them the future.

I don't think it's devalued. I think—

Or properly valued if you don't have an Elon.

I think that's exactly what it is.

We talked about this last week, but explain why you think it's different.

All of these companies are actually very fairly valued, and then Elon-world gets a premium.

Totally.

And that premium is because of what you guys said. That I agree with. The big message that I take away from this, which the markets and retail are telling you, is: You guys have stopped innovating. There's a lot of incrementalism, and we as a society aren't benefiting broadly the way that you told us we would be.

So maybe this is the best way for them to get this message, which is to whack their valuation. And by the way, I'll just say it again, when Tesla and SpaceX merge and we have all things Elon in Elon Corp, okay, which will happen probably by at end of the year, maybe it'll happen the middle of next year. It's going to then break everybody's brains again because you'll have this one asset, as you guys said, that will trade at a valuation premium. And some people will say it's unexplainable. And I think it's logically explainable, which is everybody else has stopped innovating. People know how to draw more blood from the stone, how to target better ads. That does nothing for society anymore.

Yeah. That's it. Literally.

In fact, it does the opposite.

There is no good left. That was literally the exact point I was making when you cut me off. If you look, Tim Cook's greatest innovation—Tim Cook's greatest innovation, before you cut me off—was Apple TV. Not even the hardware product. It was just spending money and making a Netflix knockoff. There's been no other product—

Hold on. Let me finish again before you interrupt me making my point.

That? You don't like being interrupted?

Oh, no. Oh, yeah. Well, okay. Okay. Interesting.

Pot meet kettle. If you look at their track record, and I think this is why we had a change there, they have not done anything innovative. In fact, the things they were doing that were innovative in AI or self-driving cars, they shut down. They won't take any swings at the bat. So they are getting penalized in their valuation.

They're just not getting a premium. They're not getting penalized.

I think they're getting penalized.

On what metric? They're trading at incredible valuations. Just look at them.

Oh, no. I mean, if you compare the 2 valuations, I think they're being penalized. Anyway, let's—anybody else want to get in on this before we move on to the next one? Yeah.

There is no world in which Google, Meta, Apple, and Amazon could be viewed as being penalized in valuation. There is very clearly a world where Elon gets a massive premium because he's innovating.

You're saying the same things. You're saying the same things.

We're saying the same thing. It's not the same thing.

We're I think we're debating semantics here. I'm not letting you off the hook, Saxby poo. When Sacks is very deliberate in how he speaks, they said he's the captain of the debate club in his 20,000 word article this week and that he's a master debater. He's a masturbator. And you slipped it in. Are you saying that the FTC or whoever should be going in and looking at Anthropic? Oh, Brad's book is getting attacked headwinds. You said they're a monopoly or they're heading to monopoly tactics Sacks is are that what you're saying?

Well, look, I mean we know that tech markets have a history of consolidating down and turning into either monopolies or duopolies. And if you just look at the revenue right now, there's only two companies making substantial revenue on AI. It's Anthropic and Open AI. We know that Open AI is growing at three to four X, which is incredible at the level of scale they're at. Anthropic though we said is growing at an exponential 10 X a year. And if they just do that for 18 more months, they'll be by far the most valuable company in human history and they'll have unprecedented control over the most important technology of our time. So, I don't know what you call that, but it is something to think about. And I guess I do have a thought experiment for you guys, which is I just want you to think for a second about the case of John D. Rockefeller, who I think is known as probably the most successful greatest most ruthless monopolist in American history. But he wasn't very good at PR. He was terrible at PR. Everyone sort of recognized how ruthless he is. We see movies like There Will Be Blood, which is basically about him. In any event, imagine if John D. Rockefeller was way better at public relations. And instead of calling his company Standard Oil, he called it Safe Oil. Okay, let's just play this thought experiment.

Clean beautiful coal. Yes.

Safe oil. He called it safe oil because as we know kerosene is dangerous. Their first big product was kerosene. And kerosene can light your house or it can burn it down. And in the wrong hands, it can torch a city or you can use it to make a bomb. So, John D let's say should have called for the creation of a new government agency to regulate the safety of his product. And they could have done rigorous testing, licensing, common sense regulation. There would have been a very intense debate over safety standards. You know, what should the proper wick thickness be? And should we allow all those dangerous independent refiners, right? And I think people would have gotten so wrapped up in this debate over what constituted safe oil or safe kerosene that they would have missed what was really going on, which is that Rockefeller was building the richest, most powerful monopoly of all time. In fact, people might even have called Rockefeller an effective altruist because of course he was so concerned about the safety of his product.

That's it. Shout-out to David Sacks and the writers. Great, great, great writers.

Newman? Newman wrote this? No, I wrote it.

Well, an Emmy Award for Best Writing in a Dramatic Monologue goes to Newman. Wow, Sacks.

No, that's my writing.

You landed it. Very good, Sacks.

Brad Gerstner

I thought after the Elon-Anthropic détente, where Elon complimented Anthropic, and David started off with a bit of a compliment, that maybe we were past this. First, it's ridiculous to think of this as a monopoly. We're talking about annual run-rate revenues, David, but on a GAAP basis, they're doing about the same revenue as OpenAI in the month of March. We're way ahead of ourselves.

By the way, 5 months ago, everybody thought OpenAI was going to run away with this. Google's revenues are very substantial in AI. And, by the way, Google, Amazon, and others are producing $100 billion of free cash flow to justify their incremental investment.

At the same time, you have these 2 startups that are still fledgling and fragile in the scheme of things. You, of all people, should know we've got the best competition in AI on the planet, which is why we're at the frontier and kicking the tail of everybody else on the planet. I just want to see these companies compete. I want to see D.C. stay out of the way. The last thing I want to be doing is seeing people talk about this and throwing roadblocks in the way of competition.

Translate Brad for you: “Don’t fuck with my paper” is what he’s saying. He’s got bets on these. So, Sacks, Washington, D.C., don’t fuck with Brad’s paper. Sacks, do you want to get into the regulation stuff right now as a segue, or—

Let me respond to Brad and also translate what I’m saying satirically, okay?

Satirically.

First of all, nobody wants to see these companies compete vigorously more than me. That was the whole premise of the action plan that we worked on last year: We want to bring out the best in everyone. This is how America is going to win the AI race. We have 5 major labs vigorously competing, and as long as that competition is taking place, I think that’s a good thing. It doesn’t mean we can’t have guardrails and the rest of it, but basically, competition should be our North Star.

All of that being said, what I am pointing out—and I think it’s historically true—is that people in Washington have woken up to monopolies on the late side, not early, right? Once a company has won 80% of the market, that’s when they wake up and say, “Oh, we have a monopoly here.” I’m not saying that they have a monopoly yet, but if the trajectory continues for just 18 more months, then I think it will be in this unprecedentedly powerful position.

I don’t think people should be distracted from that fact by this rhetoric around safety, because someone like Rockefeller could have used it, too. Let me just make 1 last point on this: If you actually look at what a lot of the safetyist policies are calling for, they’re basically calling for a form of regulatory capture. They’re calling for things that would create a stronger moat around this monopoly or duopoly that’s in the process of being created, and it would get in the way of competition.

Again, I think that people might not have such a terrible view of all of this safety rhetoric if they understood that what was being created here is the biggest monopoly in human history. We should just be a little bit more skeptical about some of these altruistic claims.

Brad Gerstner

I can’t believe that David is talking about monopolies when we haven’t even left the starting gate of AI. I think this is—

There are only 2 companies with revenue.

Brad Gerstner

The thing I want is D.C. trying to preemptively get in the game of picking winners and losers at the starting line of AI. That would be a disaster.

Brad, did you just put another soapbox on top of the soapbox you were standing on?

Look, Brad, like I said, my North Star is competition. As long as there’s competition going on, I support it. Hold on. We know that monopolists want to stop competition, and they use regulatory capture to do it. Furthermore, they do things like ban their competitors from using their product.

What conceivable reason did Anthropic have for banning OpenClaw from using its models? That is anticompetitive, is it not?

Brad Gerstner

I would double-click on it. I might not file, but I would double-click.

Chamath, the girls are fighting. Let’s keep moving through the docket. We’re going to be here all day with these 2.

All right, the White House allegedly, possibly, is considering, according to reports, an FDA for AI. That would vet—you heard that correctly, folks—new models for safety. The thing we’ve been talking about not doing here. The thing David Sacks has spent the last year on. The White House is considering it.

The New York Times reported that Trump is considering an executive order to create an “AI working group.” This group would include tech executives and government officials who would “examine potential oversight procedures,” including “a review process for new AI models.” Oy.

The catalyst was, wait for it, Anthropic’s model, which reportedly scared, spooked, and made people really nervous at the White House. “The White House wants to avoid any political repercussions if a devastating AI-enabled cyberattack were to occur. They want to see why AI…” according to The New York Times.

Kevin Hassett, the director of the National Economic Council, confirmed the report on Fox Business. Here’s your 15-second clip.

Kevin Hassett

We’re studying possibly an executive order to give a clear roadmap to everybody about how this is going to go and how future AIs that also potentially create vulnerabilities should go through a process so that they’re released into the wild after they’ve been proven safe, just like an FDA drug.

Additionally, friend of the pod Scott Bessent had something to say.

Scott Bessent

What we saw in the past month was a step change in the power of 1 large language model, but we’re going to see it from the other AI companies. What we are determined to do is work with our AI companies to allow them to continue to innovate, but our charge as the U.S. government is maintaining safety. There’s a very important calculus here between innovation and safety, and at the U.S. government, we’re going to make sure that things stay safe.

There you go. Kevin Hassett and Bessent. Slightly different positions here, Brad. What do you think?

Brad Gerstner

Actually, I don’t think they’re slightly different positions, but I would agree that Kevin bringing up the FDA kind of muddied the waters. I talked to Kevin last night after that clip ran, and I asked him, “Do you think the FDA is the right analog here?” And he said, “You know, I was only bringing it up to say that we want them to show us the models so that we can coordinate them.”

Obviously, our job is to make sure that the government is prepared, that we harden our systems, and that our intelligence agencies are up to speed. But he does not think—and I can’t find anybody on the right who believes—that we’re going to move to an approval regime, right?

The approval regime, this idea that you’re going to have to share every model with an FDA in Washington, and they’re going to have to preapprove the model, is a disaster. Sacks has been effectively fighting against this over the course of the last year. It would lead to 3 bad things.

Number 1, we do not want to put Washington in the position of picking winners and losers when it comes to these models. We’re winning. We’re on the winning horse in America. We’re out in front of the rest of the world. There’s no reason to change horses and regimes at this point, and we don’t want to burden this with more democracy.

At the same time, obviously, I call these pre-AGI or AGI models—Mistral, Spud, etc.. I see a lot of coordination going on between the industry and government. I think we can do an even better job of evolving that framework so that everybody in government is on the same page.

We need to build more capacity in government to quickly be able to do the cyber review on these models. Right now, it takes too long when the coordination does occur. We need to have a finite amount of time in which they give the government feedback. But the last thing that we want is an FDA of models sitting in Washington.

Brad Gerstner

Kevin understands that. Scott Bessent understands that. So, I expect that we will continue down the path that we’ve been on.

Chamath, obviously, I think we all agree we don’t need an FDA for AI, but there are things that reasonable people would want to have guardrails around AI. I’m sure you would agree. It shouldn’t be a total free-for-all.

So, what’s your take on this? Is it just somebody who gave a bad analogy here, or maybe some people were weaseling their way into the White House to try to shift things when Sacks was back at home or something? What’s going on here? That’s what people say: The last person to talk to Trump kind of has his ear, and things can bend a certain way.

I don’t think it’s that. I think there’s a pretty profound vibe shift with respect to tech oligarchs, Silicon Valley, and particularly AI. That vibe shift has already happened on Main Street, and I think that’s starting to seep into Washington. I think regulations are coming. I think they’ll be worse under a Democratic regime, but I think some form of oversight is going to exist under a Republican regime.

The question that I think is worth asking is why. If you listen to everybody’s tone, it’s all around the negatives of AI. So, I think we suffer from 2 things. Number 1 is that we have horrible messaging. Nobody spends the time and the money to articulate the positive upside case so that there’s broad-based support.

Number 2 is the idea that there are going to be, as Sacks said earlier, a few winners and many, many, many potential losers. I think that’s really disconcerting to everybody. The response from the tech community, again, should be the leadership of the tech world coming together and actually reinvesting in America writ large. They’re not doing that on enough of a scale to blunt this.

So, what you’re seeing is the buildup of antibodies. Is it avoidable? Yes. Are we doing a good job of avoiding it? Absolutely not. We’re doing a horrible job. I’d give the community—the tech leaders—a D-minus, trending to an F.

So, I think the question, Jason, isn’t regulation or no regulation. It’s why did we get here? I think we got here because the other version—the glass-half-full version, the demonstrated investment, the broad-based uplifting of American society—hasn’t happened. And if it has, it’s been very poorly communicated.

The response is, “Hey, hold on. We’re going to give 3 guys trillion-dollar net worths, and we’re going to allow them to control the keys?” That’s why this is happening.

Exactly correct. And it’s very easy, Sacks, to imagine all the bad things that can happen. Our minds are constructed to do that. We’re vigilant. We look out for the tiger or the tornado to keep ourselves safe. Humans have a bias toward safety, and they’re going to think about deepfakes. They’re going to think about robotics. They’re going to think about self-driving cars taking people’s jobs. They’re going to think about all the dark things that could happen—bioweapons, etc. We don’t have anybody out there really talking about all the positives that could happen.

What’s your take on the palace intrigue we all have here? What’s going on in the palace, in the 47th administration, around this debate? Who’s leading Trump down the path of regulation and creating this AI FDA? We know you’re part of the camp that wants to keep this train moving, not overregulated, and not have regulatory capture. Who are the people trying to slow this down?

Well, look, I think there are several things going on here. The first one is that there’s a lot of fake news. This whole idea of an FDA for AI—I don’t think any senior official supports it, just like Brad was saying. I spoke to Demis Hassabis as well. That’s not where his head is at.

So, I don’t think anybody in the administration is saying they want an FDA for AI. Certainly, I don’t think that’s the way the president thinks about these issues. He’s the most pro-innovation president we’ve ever had. The White House chief of staff, Susie Wiles, just put out a statement last night that I think pretty much shoots this down. So, I think there’s a big fake-news component.

Remember, it was not really the White House who was saying it; it was The New York Times. And really, I think Andrew Ross Sorkin—I’m not criticizing him, but he’s a commentator—was the one who said this first. Somehow, that spin or gloss took on a life of its own, and I think Silicon Valley reacted accordingly. There’s a very visceral negative reaction here because we know how damaging that would be to innovation. But look, I think the good news is that that was fake news.

Second, I think there’s another thing going on, which is a straw-manning of what the Trump administration did on AI in its first year. In the same way that they want to spin this FDA for AI, they’re also trying to spin what we did as this completely laissez-faire attitude where there’d be no regulations whatsoever and no guardrails. It’s a way of criticizing what we did. They’re trying to portray it as unsafe.

In fact, if you look at March 20, the White House released a national AI regulatory framework that I worked on, in which we put out a 4-page bulleted list of legislation that we would support if Congress wants to pass it. So, we have not been against every conceivable regulation or every conceivable law. We just believe that there should be specific solutions to specific problems, as opposed to a giant power grab by Washington that would squash innovation.

So, I think that’s point number 2. Point number 3 is that there is a legitimate thing happening here with—let’s call it Mythos or cyber. We know that it’s not just Mythos. OpenAI now has a model that’s just as cyber capable as Mythos, and within 3 to 6 months, all the major frontier labs, including Chinese models, will have cyber capabilities.

In response to that, we do need there to be a hardening of systems, and we do need there to be a scanning of codebases to find these vulnerabilities and patch them before the hackers do it, because the hackers will have these capabilities in a matter of months. That’s a certainty. The same capabilities you use for cyber defense can also be used for cyber offense. It’s the same toolset. The open-source models will have these capabilities anyway.

They have it to a certain extent, let’s be honest. They have 80% of it, 90% of it.

It’s simply the case that AI will be good at cyber, and so we do need a response to that. My view on what that response should be is, first of all, we should want the government and the private sector to work cooperatively. I think they are. We have a giant cybersecurity industry in the United States whose sole job it is to protect systems and protect against breaches.

We have the best companies in the world at doing that. We have CrowdStrike, we have Palo Alto Networks, and we talked about that before. We have the best defense.

Right, exactly. And so, what we should be doing, I think, is getting these tools, Methos and then the OpenAI model, and others like it—into the hands of our cybersecurity industry.

And by the way, not just the public companies like Palo Alto Networks and CrowdStrike. Those are certainly 2 of the most noteworthy, but there are also some incredibly strong startups on the way up.

A long tail of hundreds of companies that are doing—

Yes. AI-powered pen testing and all the rest of it. We need to get these tools into their hands as quickly as possible because they’re a force multiplier. For all the companies out there that aren’t that good at cybersecurity, or maybe they’ve got IT departments, they can use these companies as vendors. So, I think there is a role for us to play.

Do you think that the models should have a KYC wrapper going forward?

KYC, for the audience, is “know your customer.”

Yeah. So, what it would mean is that before you can use Methos, you have to identify yourself so that we can try to know that you’re not a state-sponsored actor or a bad guy.

I think that’s the type of thing that we should be thinking about. So, first of all, I want to say that both Anthropic and OpenAI acted responsibly here. No one was trying to release these super-powerful models. So, in a way, all the people who are saying that we need pre-release approvals for models, they’re trying to solve a problem that didn’t exist. Anthropic wasn’t—

Which is the ideal situation.

—trying to release this. They all understood the power, and they were all acting responsibly.

They understood the ramifications because they would have been sued. So, there is self-policing going on here, which is the ultimate way to do this.

Yes. But to your point, Chamath, I think that before giving your API for a super-powerful model to a company or an actor when you don’t know who they are, you should identify them. Some basic KYC makes sense. They should know who they’re giving these tools to.

My view on the Mythos preview and whatever the equivalent is of what OpenAI is doing is that we very rapidly need to get these tools into the hands of more good guys. You need to know who those good guys are. So, yeah, KYC is a predicate for that, right? You’ve got to know who they are.

Just to be clear, we’d all agree that if you did have identity for those frontier models—which they’re probably doing anyway right now—and you logged what people were doing with them to look for security breaches, that wouldn’t necessarily happen when you released it to the public because of privacy issues.

To our partner at Polymarket—man, I've got to get in here. Do I have inside information here, being the world's greatest moderator on this podcast, or can I collect this money, Chamath?

I would not. Do not place a bet, J Cal.

Don't place a bet.

Brad Gerstner

Don't place a bet. But to Chamath's point, I think we're workshopping this in real time.

We are. We should say—

I think that for the preview period, we should definitely have KYC. Maybe—

Logging? What about logging?

Well, once you're past the preview period and it's in general release, I'm not sure if the KYC matters as much because so many people are going to have it. But during the preview period, there should be KYC.

Brad Gerstner

Let me just say one thing. All the labs are already tracking API use, okay? Anything suspicious is being flagged because there are major anti-distillation efforts going on by all the labs. There's a ton of coordination going on with the government. There's way more happening, I think, in terms of our API and API use, and anything suspicious is being flagged and shared with the government.

The idea that we have no idea who's doing it, I think, is not the case. In fact, in some cases, we may want to allow people to use it so that we can see exactly the types of things that they are extracting. I would say we're already down that path, but better coordination may, in fact, be called for.

Yeah. Just one last point in this whole thing: I want to build on my point that prerelease approvals are solving a problem that didn't really exist because, again, Anthropic and OpenAI weren't trying to release these models yet. There is a substantial faction of, let's say, AI ideologues or doomers who are basically employing the classic “never let a crisis go to waste” strategy.

We do have this cyber issue that is real. Everyone needs to harden their systems now, over the next 3 to 6 months. That is a real issue, but it is a problem that we will solve over the next 6 months. We have to. What they're trying to do is use that issue to create a permanent new infrastructure in Washington.

Again, I don't think that's the administration's intention. That's not the administration's agenda. But you saw a lot of people on social media, a lot of the think tanks, and even Bernie Sanders weighed in. He said, “For the first time, I like something that—”

This is great. The administration understands the 1% of the 1% tax, and everybody understands that this is out of control. AI is going to take the jobs. They're going to take my summer home. It's going to be terrible.

So, there are people who have this agenda. Bernie Sanders just wants to stop the progress. I mean, he's—

Of course, I do.

He wants to ban data centers. He's put out a bunch of things. He basically has bought into the whole doomer narrative. That's why he likes the FDA idea: It would put the kibosh on innovation.

Already. Let's go back to paper and pen. It was a better society, Sacks.

Jason, what do you think?

I think there are 2 really interesting things I want to build on here. The first is your point, Chamath, around how we turn around the sort of bad vibes around AI. I think we have to have 2 strategies here.

One is giving. What you've been working on, Brad, with your project—we should see more people giving. There's no reason why Nvidia, SpaceX when they go public, Anthropic when it goes public, and OpenAI if and when it goes public, or if it stays as a nonprofit, couldn't give a portion of the IPO to every American citizen.

So, IPOK—IPO for Kids—they all take, whatever it is, 5% or 1%, whatever they choose, and put it into the Invest America accounts. We should see some major giving from the people who are becoming trillionaires, hundred-billionaires, whatever it happens to be. There's no reason not to.

But those people haven't been doing that. We had the Giving Pledge, which was a little bit of virtue signaling, and it wasn't real. It was just, at the end of your life, you promised to give away half your money. Let's have something real. Let's have something where people say, “I'm going to give away 1% of my stock over the next 20 years of my life. Every year, 1% will go into Invest America,” whatever it is. It won't cost anybody anything. You can't spend this money, whether it's Bezos or whoever.

Second, in terms of giving back, we have not talked about how massive this could be for health, extending people's lives, and reducing suffering. We need to work on that. That's where contributions to basic science could come in, and obviously education and lowering the cost of education.

If you look at what Americans at the bottom have, you were talking about the cup half empty. There are really 2 or 3 things they feel anxiety about. One of them is income, and the second is health care. On the margins, it's housing and their kids' education and the cost of those things.

We should take a deep look at this, and I know this is very unpopular among capitalists, including myself. We should really look at the minimum wage and study what happened in New Zealand, Sweden, Switzerland, and Australia when they raised it. What actually happened when they raised it? There was a lot of hand-wringing about it, but when they slowly raised it, what they found was that those consumers don't save money; they spend it. They're always behind the eight ball in terms of their spending.

We should opt in to trying to raise the minimum wage company by company by company and give people at the end of the spectrum the understanding that, “Hey, year over year, whether it's Amazon or Target, restaurants, et cetera, we're all collectively going to add a little bit to that minimum wage and try to lift the bottom third of society.”

That's the stuff we're not talking about. We don't talk about it here on this podcast. We don't talk about universal health care. We don't talk about the minimum wage. But that's what capitalists should be talking about.

If we did that—if we increased the minimum wage, and I'm not a socialist, I'm a capitalist who thinks this is good for capitalism—if we increased the minimum wage just modestly each year and opted into doing that, and figured out a way to give universal health care, companies wouldn't have to deal with universal health care and we would have customers. We're a customer-driven economy.

Sixty, 70%, 80% of what happens in this country is driven by the consumer. We need consumer spending. It's great for companies if we had more people being able to buy Netflix or order on Amazon.

Anyway, that's my TED Talk. Thanks for coming. How did we get from AI to the minimum wage? I'm still a little bit confused. I don't know. No—the black eye we have in this country, with polarization of wealth and people scared of losing their jobs, we should look at why they're scared, David.

I've talked to you privately, and you said to me privately—you can strike this if you want—that you wouldn't necessarily be against figuring out a way to do universal health care if there was a way to do it. You want to see every human have health care, yes?

Brad Gerstner

Sure. The issue is not the desirability of it; it's the cost. I haven't studied that issue, so I don't know. I just know what P. J. O'Rourke once said, which is, “If you think health care is expensive now, just wait until you make it free.”

You take away all the incentives, and you have an even bigger problem.

Well, what do you think of the minimum wage? Yeah, go ahead.

Well, let me just—can we just get back to AI? Listen, you guys are right about the unpopularity of AI. We've all seen those polls, but I want to put up this additional poll that came out about the salience of this issue, which is: How important do people think it is? AI ranked 29 out of 39.

Although AI is not very popular, it is certainly not top of mind for voters. It's not in the top 10 issues. It's not in the top 20 issues. What is top of mind for voters? Number 1, cost of living. Number 2, the economy.

We know that AI is deflationary. It helps with the cost of living, and it's creating an economic boom right now. It's 75% of GDP growth in Q1. By the way, that economic growth is not just limited to startups in Silicon Valley. We're seeing a construction boom. We're seeing a blue-collar boom. We're seeing 25% to 30% wage increases for construction workers, and so on down the line.

And Brad, if you look at that chart, there's health care in there, too.

So, look, my point is that AI may not be popular, but the effects of it actually are popular if the media would honestly report what was happening, which is that AI is creating an economic boom right now.

Brad Gerstner

It couldn't be better said, David. Bernie Sanders is calling for a moratorium and to shut down all data centers. We'd have negative GDP growth this year. The stock market would be down 15% to 20%. Unemployment would be on the rise.

There is a consequence to the government controlling the economy through command and control. In 1929, we had 4% unemployment. 3 years later, unemployment was 23% because the government got involved in regulating everything and shutting down what was working.

That is the greatest threat we have here. AI is delivering huge net benefits today in terms of the unemployment rate, economic growth, and productivity growth.

We need to tell the story. But to Chamath's earlier point, we also need to deliver net benefits. Jason, thanks for the shout-out. Yes, every American having an investment account that compounds with the upside of AI—we're going to do that. We're going to deliver that, and that's going to be massive.

But I also think ideas like, if we're going to put a data center in Abilene, Texas, let's make electricity in Abilene free for the households in Abilene, Texas, right? There are ideas that can deliver net benefits. We've got to deliver those. I think optimism will be on the march. I think we're in the trough right now.

What's your take on the minimum wage, and how do you think about it as a capitalist, as an innovator, and in the face of AI, which could have a dramatic impact on these issues?

Brad Gerstner

I haven't spent a ton of time thinking about those, except that as a society, we're $38 trillion in debt. We haven't been able to afford to deliver those things. I generally think the market works out those issues better than the government trying to control things top-down. The government gets more and more involved in health care, and the only thing that happens is it gets more expensive.

We've all seen the charts of the most expensive categories where we've had inflation: education, health care, et cetera. It's where the government's involved. I actually think if you just let the markets work, we're entering into an age of abundance. A lot of these problems are going to be solved. People are going to have a lot of AI coaches in health care and education, et cetera. Let the market work. Government, stay at bay, keep things safe. We're on a good march.

I think this proves my point perfectly. If you talk to any founder, they're not thinking about housing, they're not thinking about higher education costs, they're not thinking about the minimum wage, and they're not thinking about health care all that often. Some do, though—there's some innovation there. It's because it's so regulated that entrepreneurs and VCs are just like, "That's kryptonite. The government has poisoned the well. We can't participate in that."

That's the roadblock, and that's where Americans are suffering. It would be great if founders actually put their minds to it, and the government has got to get rid of all that regulation and let us cook in those specific verticals.

All right, the market is in hyperdrive. Hyperscaler revenue has made the markets move up. We hit on this briefly, but we didn't have you here. Fifth, B2B cloud computing is on a tear. I referenced it earlier, but AWS is now on a $150 billion run rate, Azure at $108 billion, and GCP—Google Cloud—at $80 billion.

There's a little bit of fun with the numbers there, Brad. Azure and Microsoft include some of their software products in there, and Google Cloud includes things like Google Office or Google Suite in there. But the growth numbers are tremendous. AWS, which is the more pure play of the 3, is growing 28% on a very big number. Azure is at 39%, and Google Cloud is stunning everybody with 63% growth.

It is incredible what the ARR numbers are. Google Cloud added $10 billion, AWS $10 billion, and Azure $9.5 billion—so basically $30 billion collectively. Jamin Ball, who works for you, put out some data on Twitter.

Brad, the markets are at all-time highs. The Magnificent 7 are cooking. Uber is blowing out growth, Disney is blowing out growth, and the consumer seems absurdly strong based on those 2 bellwethers. Tech seems extremely strong based on cloud computing. What's your take on the overall market and overall economy? Obviously, inflation is up a bit, and people are hand-wringing about the never-ending war and the cost of oil.

Brad Gerstner

Let's just telescope way out. The level of criticism directed at this administration—tariffs were going to cause hyperinflation, we were going to destroy GDP, and conflicts in Venezuela and Iran were going to do the same. We've heard all of the negative stories, but what's happening? Accelerating GDP. A 10-year that's sitting at 4.3%, inflation totally under control. AI, AI, AI. Compute, compute, compute. We're leading the world. It's contributing massively to GDP growth in the country.

We see the S&P only up 8% this year, right? So we're not in bubble territory here. Meta's trading at 17 times fully taxed GAAP earnings, Nvidia at 19 times, Microsoft at 20 times, and Google at 24 times. Then, for the memory stocks that everybody's excited about, we have 25% of our portfolio in SK Hynix at 5 times fully taxed GAAP earnings, Samsung at 6 times, and Micron at 7 times. This is not the stuff that bubbles are made of.

David referenced it earlier: We started the year with OpenAI and Anthropic doing a combined $30 billion in revenue. Now, combined, 4 months later, they're doing $80 billion in revenue. The policies of this administration on the economy are working. They're working in spades. Our gap with the rest of the world in AI is growing.

From my perspective, we've been all-in on the market. I talked about it earlier in the year. We're heavily tilted.

You made that switch to go all-in on the market because you were bearish?

Brad Gerstner

I'd say toward the end of last year, the market had run up a lot. We had a lot of these questions. Listen, entering this year, there was a huge question hanging over the market: Would the AI revenue show up?

If the Anthropic revenues hadn't shown up and we didn't see this reacceleration out of the hyperscalers, the market would be down 10% to 15% because people would say there's no ROI on all of this investment in infrastructure. Exactly when I saw the numbers start showing up in December and into January, we went from medium to large in terms of our exposures, and 80% or more of our exposures have been in compute, AI, memory, et cetera.

This is why it's great to operate in the private market and the public market, because you can see things in the private markets that inform the public markets. But the question remains, Brad: How much better would the economy have been doing, as much credit as you're giving to the administration, if they hadn't started a $100 billion war that we did not need to go into, according to all reports, and if we hadn't done a bunch of tariffs that wound up being unconstitutional according to the Supreme Court, which Trump himself put in?

We would have been further ahead. That's my take on it. We would be ripping even more if we didn't have those silly diversions.

Brad Gerstner

It's hard to imagine, okay? Just to set up again here, it's hard to imagine a more Goldilocks situation for the United States of America. We have reset the table geopolitically. The discount rate globally is actually coming down, not going up, evidenced by markets at all-time highs and the bond market in control.

Look at the private markets. We have multiple trillion-dollar companies that have been created in the private markets that are now coming public. SpaceX coming public is going to be a multitrillion-dollar company. OpenAI, Anthropic—at some point, you just have to acknowledge the USA is winning.

Of course, there are always things that we could be doing better, but there's not a country in the world that wouldn't trade all of its fortune for the United States' fortune today.

I'm 100% in agreement. American exceptionalism, as embodied by the great companies in America—SpaceX, Google, and all the ones we've been talking about here—that is the story, and I give infinite credit to this administration for being business-friendly.

I do think they've made 2 critical mistakes. I think the tariffs were poorly executed, and I think we shouldn't have gone to this war. We should find a quick resolution to it, which the administration seems to be desperately trying to do.

Sacks, your take on the economy?

Well, look, we have an AI boom going on right now, and I think that's thanks to President Trump's policies. Remember, the first week he was in office, he rescinded the Biden policies on chips and models. And what were those policies? It was the approval regime that we're talking about.

Models would have to go to Washington to get approved if they were trained with some number of FLOPs, and then every sale of a GPU worldwide would have to be licensed from Washington unless it fit into some narrow exemptions. So the whole approach of the Biden administration—

Exactly.

—that President Trump inherited was everything approved in Washington. He rescinded that. He declared that we had to win the AI race, and he unleashed our companies to do that.

Now, one other really important thing is energy. Remember, it was this president, going back a decade, who said, "Drill, baby, drill." He said, "We have to unleash American energy. That's the basis for the American economy. It's also the basis for AI."

He also has said that he wanted to allow our AI companies to become energy companies so they could bring their own power to these data centers. That way, they're not drawing off the grid, they're not competing with consumers for electricity, and they're generating their own power.

It's thanks to this president that we have seen this blue-collar construction boom right now powering all of this infrastructure. What would the alternative have been? We know—I mean, Bernie Sanders has said it. It would have been a ban on data centers.

Yeah.

Right now, Biden has banned data centers. That would be a much worse choice.

Chamath, I'll give you the last word here as we wrap on the economy generally.

I think the markets are going to keep going up for a while. And then at some point, they're going to go down.

Okay, I wrote it down. Chamath, you said markets are up, and then eventually they're going to come down.

Those are the 2 things: up and down. Let me put a U here and then a D here.

I think you're doing an impression of the taking-notes emoji. Is that what you're doing?

Yes. Okay, I'll shut down.

Good. Chamath, in all seriousness, what makes you bullish, let's say, in the next 6 to 12 months? What are the catalysts, and what do you think the headwinds are as well? Let's take the short- to mid-term—6 months to 2 years.

I think that in the short term, the people that make the new thing need to get valued and need to demonstrate value. Who are the people making the new thing? It's the NVIDIAs, the memory makers, the Anthropics, the SpaceXs, and the OpenAIs. But eventually, it all comes home to roost. You can't just make things for a market that doesn't then have a measurable benefit itself.

To be very clear and blunt, there is literally not a scintilla of evidence that AI has helped lift the operating margins of the S&P 500. There's all kinds of bluster. There's going to be an important fork in the road, probably 2 or 3 years from now. One path will be that OpEx shrinks, hence margins increase, and the other path is revenues grow and margins expand, and OpEx stays flat—or maybe it even goes up.

Those 2 things are very important differences, because in the former, you're talking about a shrinking workforce and shrinking OpEx as a percentage of operating margin and revenue. In the latter, you're actually growing through it. The answer to that question, I think, is critical to how the markets will respond and how society will respond.

So, I think we have 500 days where you just have to be net long. But I think it's literally 500 days from now. You're going to have to have an important reckoning moment. The people that are paying for all these tokens need to see an actual benefit.

Yeah, that's reasonable. That's a really interesting point. Let me connect a couple of dots here between something Chamath said and what Brad said, which is that Brad said at the beginning of the year, we went into this year with this massive CapEx, this massive investment in infrastructure, but people weren't sure that the ROI was going to be there in terms of model revenue. And that was true, and then the model revenue has proven out.

Now what Chamath is saying is that we're going to be at another fork in the road soon in terms of whether there's going to be ROI on all those tokens that are being sold and generating the revenue for those model companies. I agree with you that that is not proven out yet, but I'm optimistic that it is going to be proven out. Otherwise, you wouldn't be seeing—

Look, I—

Enterprises continue to buy. Hold on, let me just finish my point. You wouldn't be seeing enterprises continue their month-over-month spend on coding tokens if they didn't feel like the ROI was going to be there.

But you make a good point, which is: What is the impact on the economy going to be when all of this new software—this bespoke software that's being created through all these coding tokens that are being bought—is going to power a wave of productivity like I think we've never seen before? So, I think what you're seeing is the ROI trickling down from infrastructure to model to application to end user. And I think it's going to create an economic boom.

I got it. I'm with you, Sacks. This is déjà vu all over again. We watched this happen with the PC revolution, the internet revolution, the cloud revolution, and the mobile revolution. We had all this hand-wringing: Will this ever pay off? Should I build an app? Should I build a website? Should I not? Should I move to the cloud? Should I keep it on-prem? All of these questions, over and over and over again. And then they went from question marks to exclamation points.

I can tell you, inside my firm, we were using agents, and then we started building code. I've got 3 people on the team who are making all the interfaces and products that a 22-person investment firm should not be making internally. They should be using SaaS software. And they are shipping product day in and day out. The ROI is fait accompli, Brad. It is fait accompli. I think this has been decided. I think it's been decided.

Brad Gerstner

It has not been decided at all. It has not been decided at all.

You have 80/90. You're working with the big enterprises. I invest in 100 startups a year. I work with the small ones. It is fait accompli with startups. They are building software. They're shipping. They're getting massive value from these tokens. And they're getting so much value that they don't have to add half the number of employees that they would with the same amount of capital. They're getting further with less money.

It is working in startup land. I don't know what's happening at 80/90. You would have a better picture, obviously, of the enterprise. Tell us what you see there.

Brad Gerstner

Our business is doing well, but what I'm trying to get across to you guys is that you can't will profits to go up. Ultimately, what happens is—I'm just going to take a company randomly—Anheuser-Busch. They have to eventually sell more beer. Take Nike. They ultimately have to sell more shoes. Take a medical device company. They have to sell more artificial hips and knees.

The point I'm trying to get across is that right now there's an enormous amount of very constructive and creative experimentation. But I think it's also true that a lot of that has not yet proven value. I don't think that means it's going to stop. All I'm trying to say is, until a company can trace very directly, "I spent X and I made Y," where Y is now greater than X and it's lifted my margins, that is the thing that causes the flywheel to spin faster.

And right now we've started the first part of that equation. We've spent the X, and we have not seen the Y. You would see it in global GDP. You haven't. You would see it in global productivity. You haven't. You would see it in the global profit margins of the S&P 500. We haven't. It doesn't mean it's not coming.

Brad, do you want to pick up on this? I'm definitely taking the other side of it because I'm seeing, with a lot of these companies, a massive lowering of costs. Their ads are getting more effective. At the same time, they've stopped hiring. They're not adding positions in a lot of cases.

Take the Nike example. A lot of the imagery they used to make for their app is now being made without having to hire photographers and do that work. I know this example because we have a startup that does this specifically for brands like Nike. They've seen a massive drop in costs. We have one that helped DoorDash with its food pictures. All those pictures used to have to be taken by photographers. Now it's all done by AI.

There's been a massive reduction in cost, and they're using ads and ad creative now that are a double-digit percentage more effective while costing half as much. So, I definitely think we're seeing it in the earnings, but is that true? Are you seeing it in the earnings of these companies yet?

Brad Gerstner

Yeah, so 2 data points. Number 1, we just saw Azure grow 39% in the quarter. We saw Google Cloud grow 63% in the quarter. Headcount growth for those companies over the last 3 years—the Magnificent 5 combined—is about 3%.

If you look at the S&P 500 writ large, in Q1 of 2024 operating margins were about 11.8%. That was up from 11% in 2023. This year, they're 13%. So, we've had a 200-basis-point improvement in the operating margins of the S&P 500, which is massive.

Do you think that's AI across those businesses?

Brad Gerstner

Yeah—

So, but any amount of margin is not AI. We saw the same financial engineering that got these earnings to rise in the last decade.

Brad Gerstner

I think that's the question: Is this margin expansion durable? The consensus forecast and estimate is that margins are going to continue to expand over the course of the next 2 years. You and I both know that back in '22 and '23, we went from the age of excess to the age of fitness, right?

A lot of these companies were able to shed people with the excuse of AI, just because they had become too excessive during the period of COVID. So, I think it's a legitimate question whether or not that's all from AI, but I will tell you, anecdotally, it maps for me. I'm hearing, like Jason and David, that a lot of these companies are really growing their top lines at an accelerating rate without expanding head count nearly at the same pace.

Okay, Sacks, I'll give you the final word while we're at it.

Brad was talking about how we got all these operating-efficiency improvements. The unemployment rate stayed at historical lows during that time. Economists consider full employment to be 4% to 5%, and we've stayed at the low 4%—roughly 4.2%—during this time. So, you're able to get these efficiency improvements while unemployment is still extremely low.

Moreover, there was just a big article saying that the unemployment rate for young college graduates has dropped. There was this whole narrative recently that recent college graduates were going to have the hardest time finding jobs because there was going to be no work left for entry-level jobs because of AI. In fact, it has gotten easier for recent college graduates to find work.

Brad Gerstner

Maybe that's because they're AI natives. Maybe that's because they know how to use AI better. In any event, we're just not seeing any evidence yet of these theoretical downsides of AI around job loss and unemployment. We are starting to see big productivity gains.

Brad Gerstner

Yeah. This is going to be a circular discussion, but there's a lot of conflicting evidence. The last piece of conflicting evidence, obviously, is the labor participation rate, because if you're not even opting into participating, then you don't get counted as unemployed. That's been, I think, a big challenge. The labor participation rate was 61.9% in March. Back before COVID, it was 63.3%.

Yeah, and college graduates are hearing different stories. Certain degrees are getting jobs, and other ones aren't getting jobs. It's too early to tell, I think, is probably what we all agree on. And it's a mixed bag.

Brad Gerstner

No, no, I don't agree with that.

No. Look, whenever I have data to refute one of your narratives, you always say it's too soon to tell.

Brad Gerstner

No, no, no.

Headline

Wall Street Journal. Nick, put it on the screen: “College graduates are finally catching a break in this job market.”

JCal, you should be happy about this.

Listen, I'm happy anybody gets a job. But what you do is then you say we don't trust the numbers, and we should get rid of the Fed, and we should get rid of the numbers.

We should get rid of the Fed?

No, that was Chamath. You chose Chamath. Let's get rid of the Fed because we don't like the numbers.

What did I say?

You said, “Abolish the Fed.” Abolish the Fed. What is the Fed here for? All right, listen, enough. We're getting into Trump Derangement Syndrome or Trump Bend-the-Knee Syndrome. It's the end of the show. We had a great show, everybody. We had some laughs. We all learned. We workshopped some stuff. Let's leave it where it is. Great job, President Trump.

Brad Gerstner

I want to congratulate all of our innovators, and I want to congratulate Elon and Dario D. Rockefeller on their recent deal.

Oh, here we go. Shots fired. Come on the program anytime, Dario. Hey, you know Dario well, Brad. Get him on the program. Next week, I want him on the program. Have him come on.

Brad Gerstner

I'm going to ask—

Ask him for me. Will you ask him for me?

Brad Gerstner

Well, sure, I'll ask him. The fact of the matter is, I thank our lucky stars that we have Elon, that we have Anthropic, that we have OpenAI, that we have Google, and that we have Amazon, all innovating in this country. I know we like to poke fun at some of these companies for some of the things they do, but I am happy that they are American companies and that they're innovating here. So, congratulations on your monopoly, Dario. Absolutely.

We just covered on this podcast how beneficial some of these economic trends are. You never get it from the media, and they are trying to derail us from the policies that have been so successful. I totally agree with everything Brad said. We need to stay the course. We're on the winning horse. We just had the Derby last week.

There it is. The winning horse. Stay on the horse. America for the win. There it is, Senator Brad Gerstner. I think if you're going to run, you've got to get rid of the red glasses. We've got to get rid of the corduroy shorts in there, but I think you've got a serious shot, Senator.

Brad Gerstner

I like secretary better. Secretary.

Secretary. Who's the Secretary of the Treasury? Brad Gerstner. Secretary of State, David Sacks. Secretary of cashmere and wine, Chamath Palihapitiya.

There's so much fake news out there. Look, I totally agree with everything Brad said. I poke fun at some of these companies for some of the things they do, but I am happy that they are American companies and that they're innovating here. So, congratulations on your monopoly, Dario. Absolutely. And look, we just covered on this podcast how beneficial some of these economic trends are. You never get it from the media, and they are trying to derail us from the policies that have been so successful.

Yes, but they did some great, inspiring coverage of micro-looting, so get your micro-looting on. Congratulations, New York Times.