AI泡沫破裂，Zuck冻结招聘，Newsom的2028年崛起，俄罗斯/乌克兰终局

Sacks told me he's a little broken up about the empty-nest thing. By empty nest, I mean his bulldog, Moose, is coming to live on the ranch.

Has Moose landed?

Moose is landing next week. We're giving him a 3-month trial. If he's on his best behavior, he stays.

He's going to be great. He's so well trained.

What's going on?

Moose is our bulldog.

Yeah. And Moose is loose.

He needs the space. He needs a ranch. He's a very athletic bulldog. This dog is not a couch potato. He's very active, and he just needs the space to run around.

Why don't you just buy a ranch and give him his own ranch?

Why buy a ranch when a bestie has one already?

Yeah, a bestie who loves bulldogs.

No, he's a spectacular specimen of a bulldog. I mean, this is a stunning bulldog.

Yeah, he's a beautiful dog.

Great bulldog.

This is a trial adoption, a Calacanis adoption cycle. Oh my God, this is sad. You know these bulldogs resist training.

Like Jason.

You know, they are the most stoic, loyal—

Overweight, stoic, stubborn.

Overweight? This is a svelte dog. How dare you? How dare you? Moose is a specimen.

In terms of personality, this is founder-market fit. This is owner-pet fit—founder-bulldog fit.

What Moose would do is jump on the couch and just lie there. We'd say, “Moose, no. Go down.” So we'd put him down, and he would jump right back up. We would take him down, and he would jump right back up. He could go on all night doing this. Finally, you just give in and let him sit there.

It's literally JCal's couch.

Now it's his couch.

It's literally J. [Music] We'll let your winners ride. Rainman David, and it said we open source it to the fans and they've just gone crazy with it. Love you.

All right everybody, welcome back to the number one podcast in the world. We got a banger episode for you today, all the topics that you want to hear about in the world. And with us, our amazing crew, our chairman-dictator, Chamath Palihapitiya, is enjoying the final moments of summer. How are you doing there, brother?

I come back in a couple of days.

Are you getting a little melancholy there, brother? A little sad?

I think everybody should have a 2- to 3-week vacation in August. It should be mandatory around the world. The world should stop, and we should all go on vacation.

You proved that to the audience by not showing up for the show for a couple of weeks.

It's the best time to go on vacation.

I agree. I just got back from a little visit down to Mexico City, which was absolutely delightful. Rome is amazing. Have you been there before, Jim?

No, but I've heard it's incredible. I heard the food scene in Mexico City is bananas.

It was bananas. The startup scene there is amazing. It's kind of like if Spain and Williamsburg, Brooklyn, had a baby—that would be Mexico City. It's super hip, super fun. Lots of young people, lots of energy.

Hey, speaking of young people with lots of energy, Friedberg's back with us, in his cell, working throughout August. Friedberg, you gave up your summer once again for the All-In crew and the audience. How do you feel? Are you resentful, or do you feel good about your decisions?

I live for the work, JCal.

You do. How's Ohalo doing? It seems like you're doing a lot of work there. You've got your head down. Just give us a little update on the company.

I appreciate it, besties. There's a lot going on. We're very busy. I actually had a really cool test tube here that we had a customer visit today. I'll go find it. It's the world's first true potato seed, and we are ramping up production. There's a lot going on. It's pretty exciting.

Yeah, sounds thrilling. Sacks, any thoughts there on potatoes and just how thrilling this exercise has been?

What are we talking about?

We're just talking about—Friedberg had a thrilling summer. He's got a new potato in a test tube that he's working on.

Wait, did we invest? Is that in—what's going on with that?

Yeah, you're a shareholder.

I'm in. Okay. All right. Well, okay. I care about you—

No, you did not make an investment in me.

I didn't get one.

Wait, is this a sore subject?

No.

Why?

No, it's a sore subject for me. When this thing becomes worth $10 billion and I didn't get to slide in a little $250K slice—not even a slice, huh, Friedberg?

Look, guys—

A slice for your friends.

They want value-added investors. JCal, what's your value-add? Exactly.

A value-add? How about spreading disinformation on this podcast?

Okay. Yeah. All right.

Well, since I'm not on the cap table, good luck with your mutant potatoes.

Hey, Sacks, you and I did a little interview with a certain senator. Give us an overview.

Yeah, we did a great interview with Senator Eric Schmitt from the state of Missouri, who was previously attorney general and wrote a book about his experience as attorney general. He brought the lawsuit Missouri v. Biden, which exposed the Biden administration's censorship apparatus that they ran through social-media companies. This was the government's side of the Twitter Files, and the discovery he brought, when married with the Twitter Files that Elon released, really paints a picture of how the government was in bed with social networks to basically censor Americans' free-speech rights in favor of their point of view on COVID and many other issues.

Yeah, it was spicy, and it's everywhere where you can find All-In. But we got this big event coming up. The summit is coming, folks. Here we go for the fourth time in just a few weeks: September 7th to the 9th. I think the 8th and 9th are the main days. Man, this lineup is stacked. Alex Karp, Vlad from Robinhood, my guy, Energy Secretary Chris Wright, and I are going to have a great discussion, part 2. Wow. Uber CEO hitting an all-time high. Circle CEO—just an incredible lineup here. And, Friedberg, it looks like we've got some great partners.

First of all, the event is going to be very expensive to put on this year. We actually have, in addition to the great speakers, Diplo headlining one of the main parties, the Diplo. Yeah. Gary Richards and Dylan Francis. And so we have a huge set of blowout parties every night. It's going to be awesome to put all this on. They pay for the event. We want to thank the sponsors. So Salana, big shout-out. They were a sponsor last year and they're stepping up as a big sponsor this year. Big shout-out to our friends at Salana. They're amazing. They support everything we do. They're doing over 100 million transactions a day. It's unbelievable what they're getting done over there. Great team. And they're going to do a full open bar for us Sunday. Juice Bar Coffee, the works, classy team over there. And we don't just have typical sponsorships of events. What we do is we ask our partners, hey, make an activation, make a place for the besties and for the fans to hang. And they've all created these amazing experiences. And there'll be matchas and espressos and craft beers and sushi and all kinds of great stuff. And to our friends at OKX, they're going to be coming back. They're building an onstage. Got a matcha bar coming. All these great things by our partners. Google Cloud's back offering 350K in credits to Google Cloud for startups. What mentions they are over there. Google's crazy. Look at this, man. They're putting out a mini campus at the summit and, yeah, 350K in cloud credits. Athletic Brewing's back. They're doing the scholarships again. We have a couple of scholarship slots left. Please don't apply for the scholarships if you have means to buy a regular ticket. But if you do need a scholarship, please do apply if you're an up-and-comer. Allin.com/summit or allin.com/yada yada yada.

Did you guys see this?

What's going on now?

Apparently, I was in the new South Park episode.

The whole thing was about the sick offensiveness of AI. That's you.

Do I say anything in the show, or am I just in the background?

No. They had a line of people paying tribute to the president, based on the Apple gift that Tim Cook gave to our amazing president. It's a funny episode. You know what I like about the administration? JD Vance actually retweeted all this stuff, and it's good fun. We get to mock our civil servants.

That's the vice president to you.

That's what I said. My guy, JD. What did I say?

Suddenly on a first-name basis.

My guy, JD. Hold on a second. Let me just say hi to JD here and get his feedback. I'm going to invite him to the Diplo party. Hold on.

Your biggest nightmare is here. Me and Chris Wright also talking directly now. I'm directly plugged in.

Oh yes. Chris, right? I'm bringing him a special gift to the summit. You're going to love it. All right. We've got to talk about AI. AI mania hit a bit of a detour this week. Over the last 3 or 4 days, AI stocks were down across the board because of this MIT study that went viral, as well as Sam Altman's comments about a bubble and Zuck instituting a hiring freeze in AI after going on a complete blitzkrieg.

So, let's get into it. Act 1: Monday, Fortune dug up a generative AI study that MIT published last month, I should say. In that study, MIT found that 95% of GenAI pilots are failing to make it to production because of employee resistance, poor-quality output, and resource misallocation, which seems to be the most interesting problem.

According to the study, 70% of GenAI budgets are going toward things like building sales and marketing tools, which have poor ROI. The highest ROI was found in back-office optimization, like automating tasks that cut back spending in various departments. Basically, these pilots aren't working. Chamath, this is what the study found: They evaluated 300 AI implementations and interviewed 150 leaders across 52 companies. You've been grinding it out with your own software company, now called 8090. Does this align with what you're seeing in the field, Chamath?

I think the first wave was just a lot of boards who read the words “AI” somewhere in an article and then went to a board meeting and turned to the CEO and said, “What's your AI strategy?”

Mhm.

Then the CEO turns around and sends that down, and it eventually hits the CTO's desk. I think the first wave is mostly people just spending money because they had large existing budgets, so they were like, “Let's just go and try a bunch of different things.”

I think now we're going through the sorting function of realizing that there's a big difference between probabilistic software and deterministic software. That's probably the biggest reason why you're seeing so many failure modes in sales and marketing. It's very hard to codify sales and marketing into a set of heuristics that never change.

But back-office processes are a great target for AI because, ultimately, you have so many people who have been hired to deal with edge cases. I think that's what people do in most companies: They're in charge of a process, and they're dealing with edge cases. You can get extremely high rates of accuracy if you implement AI correctly in back-office tasks.

I think the real question is what happens to all of this revenue that has been generated. You're seeing companies generating $50 million of ARR in a matter of months and then raising huge rounds. I think what we haven't seen is whether there'll be any sort of logo churn or dollar churn as new companies come in with even cheaper solutions, the foundational models move up the stack and just absorb capability, or things simply don't work and get abandoned.

All of that churn happened in social. I remember when I was in the middle of helping build Facebook, we went through that whole cycle. There were 7,000 or 8,000 social companies, and within 6 years there were 5 of us left. It happened in SaaS when I was investing in SaaS. There were a couple of very early and important successes, like Yammer, which Sacks started and in which I was very lucky to be an investor, but then it took many years for the handful of winners to get really sorted out.

I suspect we're about to go through that same cycle in AI. I think that article basically paints a very accurate picture. There's been a lot of trying and experimentation. We now need to go through a sorting and a cleansing, and then we'll rebuild from first principles.

Not surprising to our Sultan of SaaS, David Sacks. The sales and marketing departments are very promiscuous when it comes to new tools. Getting a great lead and closing a sale are things you can directly connect, so we always see them test stuff out. It doesn't surprise me that we'd see sales and marketing go after this first.

But what do you think about the brittleness of this revenue—the churn, Sacks? Are we going to see a lot of these companies rocket up to 100 and come back down to 50? Is this something you're seeing, or is your firm seeing? I don't know your status at the firm. Maybe you could tell us how that's working out in terms of your intelligence there, but what is Craft seeing in the field?

I think we're seeing a lot of interesting AI applications being developed, but it's still very early days. Over the past week or so, there was a correction in sentiment toward AI, but I think it was a healthy correction. I don't think this was the beginning of a bust cycle or something like that. I still think that we're in a boom. I still think we're in an investment supercycle, but I think there was a healthy dose of skepticism applied to some of the more fantastical claims that have been made about AI.

I think this is why you saw roughly a 10% correction in public AI stocks. There was also that MIT report that said 95% of projects and companies are not making it to production yet. I feel like we're getting in the weeds a little bit here, and what we should be talking about is just where we are in this AI supercycle.

And where do you perceive us to be? We're in the experimentation phase. We're in the pilot phase. But this issue around probabilistic versus deterministic makes it hard to trust this software. Is that what you think the key issue is?

Well, let me tell you why I think this correction is actually healthy. After ChatGPT launched at the end of 2022 and throughout 2023, the dominant narrative in AI was that AGI was just 2 or 3 years away.

Everyone had their own definition of what AGI was, but it was this idea of smarter-than-human superintelligence and magic AI. AI would be able to do everything. As a result, you had both utopian and dystopian narratives proliferate. You started getting this job-loss narrative that, within a few years, 50% of knowledge workers would be out of jobs.

You got this rapid-takeoff narrative that basically the leading AI models would be able to turn their intelligence toward improving themselves, toward recursive self-improvement, and therefore, within a couple of years, the leading models would achieve superintelligence and leave everyone else in the dust. They would then capture all the value of humanity.

Based on that narrative—which was the same underlying narrative that fueled both utopian and dystopian, or doomer, takes on AI—I think you got a huge backlash, which has already been forming. You have 1,000 bills running through state legislatures right now, and you have all this AI safety legislation. You got bills like California's SB 1047, which would have applied a tremendous amount of new regulation to AI. So you saw this policy backlash happen as well.

It was all based on these fantastical and magic views of what AI was going to do in just the next 2 or 3 years. I think the reason why this recent skepticism is healthy is because I think it's rebutting all of that. It's showing that AI is a powerful tool. I definitely think it's a new and important form of computing, and it is going to unlock tremendous value in the economy, but it's going to take us a while to get there.

You can't just tell the AI, “Be a sales rep. Be a customer service rep,” throw it over the wall, and expect that it's going to replace a human. It takes a lot of prompting, iteration, and validation to make the AI work and to make it generate business value.

If we were on a path toward rapid takeoff, what you would see is the leading AI models increasing the distance between the top 1 or 2 models and the rest of the models. Instead, what we're seeing is a clustering of model performance around the same performance band.

Incremental, right?

The progress seems to be a little bit more incremental. It's more evolutionary than revolutionary. I think this really crystallized around the launch of ChatGPT-5, where a lot of people were expecting GPT-5 to be this huge breakthrough.

Sam Altman was teasing this concept by posting photos of the Death Star, with the idea that this model was going to blow everybody else away. But the reviews ended up being very mixed, and then we saw that in the performance evaluations.

It's not that the model didn't represent progress; it just fell short of these lofty expectations that had been created. So, Friedberg, let me get you in on this.

Sorry I've been long-winded here, but let me sum this up: I think what people can now see is that we're not in a loop of recursive self-improvement. We're seeing that there are a handful of great model companies, but the development of this technology is going to be a more normal technology race.

It's not like the leading players all of a sudden achieve AGI very quickly. As a result, because it is a more normal technology race, I think we can apply more normal logic to it from both an investment and a policy standpoint.

A lot of the narratives that were hyped up about imminent doom or imminent utopia, depending on which side you were on, were massively overhyped. This is why I think it's a very healthy thing.

By the way, it's also concerning when models successively don't get meaningfully better, especially because there's a tendency for incremental models to get overfit to benchmarks. That's another concerning thing that I think people were scratching their heads about.

If you look at the last big set of model updates from xAI's Grok, they didn't really meaningfully improve their separation from the last-generation model on the big benchmarks that matter. That's also very concerning because they tend to overfit to begin with.

Yeah. I would just say that another piece of evidence here is that the model companies are still leapfrogging each other to some degree with their latest versions. Again, that should not be possible if one model achieved rapid takeoff. The other thing you're seeing is that it's not like one model is dominating all the other models. You're seeing that models are developing areas of competitive advantage. They're becoming increasingly specialized, and they have different personalities.

If ChatGPT has been accused of being sycophantic, Grok is maybe more based. They have different modes that they excel at. Google, I think, has gotten really good at video, for example. Anthropic is really good at coding. The specialization really belies this idea of one model becoming all-knowing and all-powerful. We're seeing a proliferation now of specialization.

Let me try to get Friedberg in here. Altman appeared to compare the AI craze to the dot-com bubble, according to CNBC. Here's the quote: “Are we in a phase where investors as a whole are overexcited about AI? My opinion is yes. Is AI the most important thing to happen in a very long time? My opinion is also yes.” And then another quote from Sam Altman: “I think we totally screwed up some things with the rollout of GPT-5.”

So he's kind of admitting that, Friedberg. What does this mean in terms of the massive investment OpenAI was talking about—$500 billion? If you were to game out what Sam is saying here, maybe he wants it to cool off. Maybe these investments are too large, given the problem and the progress. What's the game on the field here, Friedberg, in terms of the massive investment that people are claiming this is going to require if we're not seeing the results in corporations?

The big capex investment—most of the capital that's been raised has been raised by the model companies, and most of it has been deployed into chips, right, in data centers, right?

Right? So does that make sense right now, or does that need to cool off as well while these companies prove, “Hey, we can get some things done in corporate America”?

So I think one of the key things that's being understood is that the internet was being connected; there was connectivity and fiber being deployed at a rate that exceeded the economic value creation by the end users on the internet during the dot-com boom. As a result, the dot-com boom became a bubble that burst, but the ultimate promise of the internet was realized many orders of magnitude greater than I think most folks declared and proclaimed at the time. It was just delayed by a few years. I think it's likely to be the same here. Whether this particular capex cycle is going to have that ROIC or not, I think, is a bit TBD.

The first is that there's a realization now that you need to pair humans and human engineering with the generative tools that these generative AI systems are providing. You can get a bunch of code written, but you still need to debug the code. You still need to get it into production. You still need to get it to integrate. This kind of pairing of humans with AI is something that's being learned. What's the right model? It's not that you just turn on generative AI and it runs your business for you. It's: where does it fit in the org? Who are the people that run it? How do they use it? Et cetera.

Similar to that, I think we're seeing a lot of partner models being developed. Rather than have Sora or a video model just generate a movie, it turns out that you can use generative AI to generate the objects and then use an existing rendering engine, like Unity, to actually make the video. Those sorts of pairing models, where you use what's being called a deterministic system or an existing piece of software that integrates with generative AI, can accelerate outcomes.

Filmmakers can now make AI movies by coupling the AI generative tools with an existing, off-the-shelf rendering system. That works a lot better than just telling the AI to make a bunch of video, where all the people look different. You can't control the objects, change the camera angle, change the framing, or change the color, lighting, and all the features that you may otherwise want to control.

Layering AI on top of the existing system seems to be the right answer. I think we're seeing that not just in filmmaking or digital content creation, but in other areas as well, where the AI provides almost like a substrate that can sit on top of existing tools and just be like a marionette and do a better job.

The third key thing that we're seeing that's causing a bit of the observed hiccup is this movement to SLMs. You guys have seen a bunch of these papers from NVIDIA and others that have shown performance improvements by actually moving away from one large model that does it all to much smaller specialized models that work on a specific task in a specific application, or that work together as a network of models.

My personal belief on this is that these models are going to move toward SLMs that we don't understand—how or why they work the way they do—in a network fashion, because today we're designing the SLMs in a deterministic way. We're saying, “This is an SLM that does X, Y, or Z.” But if the models eventually get developed by other models, those SLMs will become really not well understood, but they'll work in a network.

The reason that's important is because this rearchitecting of these systems dramatically reduces the energy cost and the dollar cost per token generated. It brings down the cost and the runtime of these systems. So, to your point, JCal, the capex investment suddenly gets much higher ROIC rather than the challenged ROIC that it's dealing with today, because the systems are so energy- and cost-inefficient today.

If the architectural redesigns work, then these systems become much more efficient—10x, 100x more efficient. I actually think at that point token production will go way up, not way down. Those are 3 big trends that I see underway right now that might be architecting the overall ecosystem and probably causing some of the observed hiccups.

JCal, what do you think? Do you think there's a big slowdown underway?

Yeah, it's following exactly the technological curve that we see. Nick, you can pull it up here, where we've seen this with smartphones and the internet itself. The trough of disillusionment is probably where we're living right now because there were so many expectations. An innovation comes out, people get really excited, massive amounts of money are invested, and then people wonder, “Hey, was this a good investment or not?” Then it just takes incremental improvement over time.

We saw this with Uber, as an example, where people were like, “Oh, it'll never be profitable.” DoorDash will never work. Then they just added a couple of bucks and it scaled globally. The network effects kicked in. We're going to see it again with self-driving.

Self-driving is actually probably just getting through the peak of inflated expectations, where people are starting to realize, “Oh, these edge cases are pretty serious. You probably don't want these things going too fast. You're going to need to be very thoughtful about putting them on roads.” That leads to a really interesting debate we should have at some point on this program: should that be federal or local? Should local states get to decide self-driving rules on their roads, or should the federal government come in and make a blanket one?

Looking specifically at sales, we're seeing in the field that maybe the salespeople aren't being replaced by AI, but the SDRs are becoming 10 times—literally 10 times—more effective at finding the next lead and organizing leads and targets. So it's definitely happening, but it's incremental inside these organizations, and it's just going to take specialization. People have to actually embrace the tools.

That's probably the biggest thing: there's a lot of resistance to these tools already. Once you get people actually embracing them, they start to work. But if people aren't going to embrace them and there's some resistance, it's kind of like what we saw with the internet. Yeah.

Here's a big question that I have. At some point, you get so invested in a way of doing things that when things change, it's almost too hard to change the way your company orients. You see it in every single phase of these big innovation cycles. The question is: is there a point at which these foundation-model companies have invested too much money in LLMs, where so many of their employees only understand that, and all of a sudden there are completely different ways of building these models?

State-space models are one example, but then some new entrant decides to build a completely different kind of representation. They may also then build custom silicon for it. All of those things have yet to happen, and the reason is we're really only a few years into what should be a multidecade innovation cycle.

Those are some really important questions, because then you have this problem of the sunk-cost fallacy: “Well, we spent all this money on this LLM-based approach, and now all of a sudden there's this thing that comes around the corner.” That's the thing that I'm really curious about and waiting for: will there be these small teams that may take advantage of what Friedberg said? They use an LLM as a jumping-off point but then build an entirely different approach and are able to generate entirely different and cheap silicon. That hasn't happened yet. The reason it hasn't happened is not because it doesn't make sense, but because we're so early into it.

Yeah.

And then what do these big guys do? The big guys have spent tens of billions of dollars.

Yeah, on all kinds of stuff: human capital, opex, capex. So it's going to be really interesting.

Well, this study actually addresses the purchasing of AI tools from specialized vendors. Those succeeded 2 out of 3 times. So the statistics in that early study seem to indicate you're right about that, Chamath: if you're doing something very specific—and we've got a couple of companies doing very specific things, like TaxGPT, which is one of the companies we incubated, and they're only going after tax and CPAs and being a copilot for them—those things seem to be succeeding faster than going to a general model and asking tax questions of it.

There's also all this other stuff that people never talk about that are very practical realities of building AI. For example, we've been selling into the enterprise at 80/90. This is our first full year of business. We'll do about $40 million of bookings in the first year.

In the first year.

Yeah, that's crazy.

And it's all back-office-type stuff. But our first customer, where we delivered enormous gains, fired us. The reason we were fired was not because the technology wasn't working; it's because as it goes lower and lower from the board and the CEO into the bowels of the organization, you actually run into huge pushback.

I think there's going to be that cycle that we have to deal with as well, which is that in all of these AI businesses, there are all kinds of things that scare people. There are many people in decision-making roles at many of these companies who will have to sort out where their comfort level is.

It was one of these things where I was like, “What are we supposed to do?” We escalated it, and it got resolved. But my point is that I think this is a cycle that has yet to play itself out in all of the Fortune 2000 and Global 2000. I would not expect that the AI boogeyman that has been created won't have some negative ramifications over the next couple of years as this stuff gets sorted out.

The last piece is the talent war. On Tuesday, The New York Times reported that Meta was looking at downsizing its AI division as part of a larger restructuring. The Wall Street Journal then reported that Meta has a hiring freeze across its AI divisions, which is crazy because just 8 weeks ago, Zuck went crazy. He was trying to buy Ilya's superintelligence startup. You remember, Ilya declined.

So Zuck poached Daniel Gross. Meta acqui-hired the Scale AI team, and then they agreed to invest $14 billion in that. Sam Altman claimed Zuck was making $100 million offers regularly for OpenAI talent. I'm curious what you think of that, Sacks. The talent war seemed to go insane for a couple of months and now is pausing. What's your take on all this? Is this an uncertainty, boom-bust cycle in such a compressed period of time, or is this just—

You were seeing founders turning down multibillion-dollar acquisition offers for startups that hadn't even released a product yet, as if those types of offers grow on trees, and they don't.

These types of $100 million or billion-dollar job offers don't come along very often. You have to be at the sweet spot of a boom cycle, and you need a huge company with tons of money that feels like it's strategically vulnerable and at risk of being left behind. If you get a confluence of those factors, then you can get crazy offers like that.

I think what Meta is doing is probably digesting a little bit. They've now made a bunch of talent acquisitions. They've done some acqui-hires at very expensive prices, and they're probably just consolidating a little bit. Like I said, I don't think this is the bust part of the cycle. I don't think that a bubble has popped or anything like that. I actually think that we're still probably early to the middle of this investment supercycle.

It's just a healthy correction in sentiment here, where people are realizing it's going to be a little bit harder and take more work than just, “The AI is going to figure out how to improve itself, and we get to superintelligence.” That was always a little bit of a fantasy. Mira should have taken the $1 billion offer from Zuck, or Ilya should have taken the $30 billion, Sacks.

I mean, these things never happen. And as you're saying—

I guess it depends how much they have in the bank account. If they're already billionaires from whatever they did before, then maybe it doesn't matter. But if you were just starting out, for example, and didn't have any money in the bank, that's a pretty hard thing to turn down, realistically.

Yeah.

They probably sold a bunch of OpenAI equity at $300 billion and $500 billion.

They're both OpenAI co-founders, so—

They seem like they're free-rolling, so they have no incentive to sell.

Yeah, which is fine. It's just that there are a lot of people who've never been through a bust cycle before. If they think that this is the normal state of the world, they're going to be sorely mistaken.

By the way, you're right, Sacks. It's hard to build, as you know, because you did a billion-dollar company that then exited for more than a billion. It is hard.

Right, where you justify that valuation based on fundamentals.

So right now we're in a part of the cycle where you can justify that valuation based on its strategic value to a multitrillion-dollar-market-cap company. But that only lasts while those companies are in the market for strategic acceleration.

Yeah, if they're behind, if they're stuck—

Then you have to make your company work on its own as an actual business.

And to get to a $30 billion valuation based on fundamentals, that is extraordinarily difficult. That implies multiple billions of dollars of actual revenue.

Does somebody want to underwrite $500 billion for OpenAI common shares? Anybody think that that is a good trade?

I'll make the bull case.

Yeah, please.

Yeah. I think the simplest way to make the bull case is if you look at the CAGR on their MAUs and the conversion from MAUs to DAUs. You essentially take a small, minor percentage of the Facebook or Google terminal ARPU and apply it to some number of MAUs and DAUs 3–4 years from now.

If I had to guess, if you take 400–500 million MAUs growing at—I'm guessing, let's just be conservative—

700 million weekly active users right now.

750. Okay. So then I would probably put that at like 500 million DAUs, to be conservative. It's probably doubling every 2 years. So 500 million goes to 1 billion, and 1 billion goes to 2 billion in 4 years. At 2 billion DAUs, they generate a tenth of Facebook's revenue. Just to be very conservative, you probably get to a $1.5 trillion valuation there.

Right. So you could triple up on that bet.

Yeah. I mean, look, OpenAI has actual revenue because they have subscriptions, and they appear to have the dominant position in the consumer space. It is a replacement for a lot of people for search, which is the most lucrative franchise on the internet, and this is one of their applications. So I actually think that you can make that case pretty comfortably.

But can we go back to a point you were making before, JCal, about how, in the survey, the attempts to just apply some sort of generalized AI model didn't work very well? Like, 95% of the time it didn't succeed in these large enterprises. But if they used a more specific vertical application or vertical model, or an SLM approach, which is a smaller, specialized model, then it showed much greater success.

That makes a lot of sense to me. In order to drive business value, there are a lot of what I would call last-mile problems. LLMs need context, so you have to first connect to all of your enterprise data sources, and you have to prompt them in a very detailed way in order to get to a good answer. Then you have to validate that answer to make sure it's not a hallucination, and then you need to iterate on it.

This idea that you're just going to have one superintelligence that figures all this stuff out is not the way it's playing out in the real world. You're seeing a lot of very specific business problems that have to be solved. But I think that this is ultimately a great thing for the ecosystem because it implies that you're going to get lots of vertical applications and lots of specialized models that capture value in lots of different markets.

That's actually the way that we're going to drive this throughout the economy, as opposed to it just being one foundation model eating all the value. I think this is a very healthy thing for the ecosystem.

Yeah, it makes total sense that the vertical systems would feel more deterministic, Sacks, because they have a tighter problem set and a tighter data set to actually come to an answer. Whereas the probabilistic approach—just asking an LLM to come up with a business plan for you—it just feels like it could be 80% correct or 90% correct, as opposed to 99% correct, which the vertical ones are actually getting very good at getting the correct answer.

Okay, let's talk a little bit.

That last 10% is fundamental, and that's where you get all the last-mile problems. You have to understand the industry in order to solve the problems of how you go from, let's say, 90% accuracy or effectiveness to 99%, which is where the business value is.

Well, and you use “last mile,” Sacks.

I think it's quite appropriate if you look at self-driving, which is the tip of the spear in this argument. Waymo has a bit more deterministic model. They don't have safety drivers. Robotaxi has safety drivers for a good reason. They're using a probabilistic model, and they're getting an intervention every few hundred miles, let's say, which means if you were to use it daily, you're going to have an intervention every week or two. Obviously, they'll iterate and get that down to the same level as Waymo.

But on what timetable, Friedberg? On what timetable do you see a probabilistic model like the one Elon is pursuing with his robotaxi program versus the more deterministic one? When do you think those cross over? Do you have any insight as a scientist, or any guess?

I don't know.

Yeah, that is the $10 trillion question: When will that problem be solved? All right, let's talk a little politics here. According to Polymarket, Gavin Newsom is the early favorite in the 2028 Democratic presidential race. So, who gets the nomination?

Here's some background. Newsom is surging in a couple of polls this week. Here's one from Politico that asked California Democrats and left-leaning independents who their preferred candidate would be in 2028. Newsom came in first with 25% of the vote, followed by Kamala Harris, your favorite, Sacks; Pete Buttigieg; AOC, your favorite, Chamath; and Tim Walz, another one of Sacks' favorites.

The odds have jumped about 9 percentage points on Polymarket, with Newsom at 28%, AOC at 14%, and a bunch of people in the single digits that you could probably guess. So, your thoughts on the early information here? Is any of this even valid, Sacks, in your mind, or is it just too much time between now and then?

Well, I think it's valid in the sense that you're seeing Democrats react positively to what Newsom is doing. What Newsom is doing is mirroring Trump's style in an attempt to again give Democrats what they want, which is that they want a Trump of their own. Despite Democrats feigning objections to Trump's style, this is really what they want: They want someone who they perceive to be a fighter.

That said, I think there are 2 problems with presumptively crowning Gavin as the nominee 3 years in advance. One is that what Newsom is doing here is completely performative. Everyone can see that he's acting and aping Trump's style, or what he thinks is Trump's style, whereas Trump is completely authentic. Trump is always Trump, and everyone understands that. There's something very authentic and likable about that, whereas I think Gavin comes across as being synthetic or plastic, and you don't really have a sense of who the real person is.

I think trying to pretend to be the left's version of Trump only reinforces the problem that he already had. So, one is authenticity. The other is that I think Democrats make a huge mistake if they think that Trump's popularity is based on his style. It's true that Trump has a trillion-dollar personality, and that is very helpful in politics, but ultimately I think the reason why Trump has done so well for a decade is because he's been on the right side of a bunch of core issues in American politics that other people just weren't focused on or weren't talking about.

The cities issue seems to be one that heats up against Gavin Newsom and will become one of the attack vectors, or one of the grand debates. Friedberg, you have California's descent into madness in the major cities, and then obviously Trump is cleaning up D.C. from a very first-principles basis: “Hey, you can't camp here. Period. Full stop.”

What do you think the chances are here, Friedberg, that this new strategy—going on podcasts, having a podcast, being spicy on social media—is going to work for Gavin? Or are you in Sacks' camp that it's a bit performative and maybe fake? And, by the way, Gavin Newsom, can't wait to have you on the pod. Go ahead, Friedberg.

I would say that the current polling is a reflection of the non-Trump-aligned market's response to Trump. It will certainly change over time as the results of the administration's actions bear out, et cetera.

One way to think about this, or forecast this, is: What is the reaction going to be to this administration? I think there's, as I've talked about many, many times, a very high probability that someone who's a little bit more socialist-aligned is going to end up being thrust forward here. I think AOC fits the bill there.

The question is, what's the Republican response going to be to the choice they may have in trying to prop up or frame up one of those 2 candidates? Frankly, there may be some group that would say, “We would rather be up against the socialists because they're going to look like idiots and we're going to win.” So there's going to be weirdly support for a socialist-like candidate.

The other side is the honest moderate question: Who would you rather have as president, a socialist or Gavin Newsom? I do think a lot of moderates would say, if I had to pick between those 2, excluding the Republican candidate, I would want to pick the moderate. I would want to pick Gavin.

Chamath, what issue will determine the midterms and then the 2028 election? Look at your crystal ball there, Chamath, and tell us what you see.

Yeah, midterms might be easier to think about, right? I think the midterms are going to be about the economy, but the economy is going to be a window into a lot of these other issues: the safety and security issue, the immigration issue, the tariff issue, and rates. I think rates are going to be an enormous swing factor here.

I think Powell is really holding on white-knuckled to a position that probably doesn't underlie the data. So, in the short term, it's just going to be more of an economic up-or-down vote for the Trump administration. Many things, I think, are positive and in their control, but there are a few things that are not in their control.

In 2028, I don't know. The one thing I would say about the candidacy, though, is that when the Republican primary process started, the person at the top of the polls at the time was Ron DeSantis. I said, “You do not want to be ahead early. These people never win. They never win.”

They take all the arrows and peak too early.

No, it's not even that. It's just that they never win. So, if I had to apply that rubric here, what I would say is the practical choice that Democratic voters will have is to actually ask whether they want a replay of California on an American 50-state scale.

If you look at it through that lens, I think Gavin Newsom is going to have a very difficult path because of what has transpired over his tenure as governor, and he's going to have to explain away a lot. So, in the absence of real choices, I think he's the clear winner. What is that phrase? In the land of the blind, the one-eyed man is king, sort of a thing.

I think when you actually go into the election cycle starting after 2026, in 2027, honestly, it's going to be up in the air. What I will tell you is that the way the Electoral College is set up, and the voting dynamics for the presidency are set up, a socialist cannot win. But the way that the Democratic primaries are set up, a socialist can absolutely win.

This is the sort of Sacks point: I think for the Republican side, the best outcome is a very radical agenda that isn't really supported by substantive facts and is more supported by vibes. “This is what I feel. I just feel it should be right.” I think those things may win a primary but will just get destroyed in a federal election.

What about you, the world's greatest moderate? What do you think?

It's pretty clear that the socialist movement is happening, and that is going to inspire a lot of young people to come out like they did in New York for Mamdani. You have to think about what the platform is.

I agree the economy will determine the midterms, but I would look ahead at the core issues we've discussed here many times, Friedberg. If I were advising the Democrats, I would smash the Republicans on the issues where they don't speak up about important issues to the American voter and where young people feel they are being ignored.

What would I say if I were them, if I were running their platform in 2028? I would go with real wages that are stagnant, and Trump doesn't talk about that. In fact, he's cutting taxes on the rich. They've got a credible attack vector right there. Nobody's talking about the minimum wage, and Trump doesn't care. J.D. doesn't care about you, the minimum wage, or your wages.

I would go after education being unaffordable, and again, Trump and J.D. don't talk about that. Then I would go after housing. Republicans don't talk about that. They don't talk about housing.

Then I would go right at the immigration issue and say, “Listen, America's for immigration. This country was built off of immigrants, and J.D. and Trump want to pull up the ladder behind everybody and not allow other people to immigrate to this country after they benefited from it.”

That's the perfect attack vector for beating the Republicans in 2028: housing, wages, education, and health care. These are things you never hear Trump talk about. You never hear J.D. talking about these. I think that's the way to do it if they want to beat them.

It's way down at the bottom of the list.

No, they don't talk about them. When's the last time? What's the proposal? Okay, so tell me, Chamath, what is the proposal?

Wages went up significantly in Trump's first term, and then they started stagnating again under Biden. That's the whole purpose of his trade policy: to encourage reshoring of manufacturing in the United States.

Tax bill,

He passed a huge tax bill that locked in all these gains. No tax on tips. You can debate the impact on the broad population, but there are enough examples where actual take-home wages are higher than what they would be in a counterfactual.

Yeah, and by the way, Gavin Newsom has a supermajority in the California legislature. He can run the state however he wants. He also has a very compliant court that will rubber-stamp his decisions, completely different from Trump, who has a very narrow majority and a court that basically tries to stop everything he does.

So Gavin Newsom can run the state. He could be operating the state for the benefit of its citizens instead of holding all these performative press conferences, doing podcasts, and tweeting all day. So where is the program that you're talking about? It doesn't exist.

It's a fair criticism, but if I'm talking about the Democrats as a group, if the Democrats came out and just said, “We're going to add $1 to the federal minimum wage every year for the 8 years we're in office,” that would fly with the American public. If they said, “We're going to build”—that's not a strategy.

You can make that argument. I'm just telling you what the winning argument is. The winning argument is—

Well, no, you're just saying whatever it takes to get elected, but giving stuff away is not [what] adults [do]. Adults make hard decisions.

You mean like giving people no tax on their overtime and no tax on their tips? Those kinds of giveaways, Chamath—

Those are helping the middle class and working class.

I don't disagree, but to the point that giving away free stuff doesn't work, it actually does. It does get people out to vote. People were very motivated when they heard those proposals by Trump. I'm saying they should just take that proposal and 10x it. That's a way to win. I'm not saying it makes sense. I'm not saying I'm endorsing it. I'm just giving you the playbook.

Okay. Well, what I would do to fight your playbook of empty promises is I would actually talk about the record of the candidate.

Exactly.

And I think we're talking about a state like California, which is thoroughly blue and has been run by Democrats now for a long time. Gavin probably will be the nominee because I do think that his performative antics are working with the base. They wanted an anti-Trump, but the way to attack this is to talk about the record here.

So let's just go through some of the numbers. As we all know, we pay the highest taxes in the country for the worst public services. Gavin's spending turned a $75 billion surplus into a $20 billion deficit in just a couple of years. Somehow, none of that money went to fill the reservoirs or fire hydrants, and we had entire neighborhoods burned down as a result. They can't get permits to rebuild.

California has the 4th-largest economy in the world, so that's good. But we have the highest rates of poverty, homelessness, inequality, illiteracy, and wage stagnation. So yes, wage stagnation, JCAL, is the highest in the country. We also have the 3rd-highest unemployment among the 50 states.

We're 2nd only to Hawaii in housing costs. You want to talk about making promises to lower housing costs, but if you look at blue states like California, they're among the worst. We're the worst in energy costs, and we have the highest rate of both retail crime and frivolous lawsuits. That's why businesses are moving out of the state.

You may have seen that Bed Bath & Beyond said that even as they're opening 300 more physical stores, they're not going to return to California. The CEO was mocked by Newsom because he never actually takes any of the criticism to heart. He just sort of deflects it.

So until the Democrats, I think, confront why they really lost—and it's not just because of the style of Trump. They want to attribute it all to Trump's style, but the reality is that it's because of the policies and the record. Until they're willing to confront those things and change their policies, I think they're going to continue losing.

I put the following question into Grok. The reason I asked the question this way is that I think the 3 most credible candidates who have a record to run on, on the Democratic side, are Gavin Newsom, Gretchen Whitmer, and Wes Moore.

I asked Grok to rank Michigan, California, and Maryland—3 Democratic states run by 3 very popular, well-known Democratic governors—on 3 metrics: quality of life, cost of living, and crime, and summarize that into a ranked overall list. A pretty standard, plain-vanilla question.

I can publish this later if you want, but at the end, the final overall rankings were: Maryland 1st, excelling in quality of life and having the lowest estimated crime rates, with a moderately high but manageable cost of living. Michigan was the most affordable, with decent quality of life and moderate crime, making it a balanced choice. California had high quality of life in terms of opportunities and lifestyle, but was severely hampered by high costs and elevated crime rates.

There's all kinds of detail, but I was just giving you the final summary. I think that when you get into a Democratic primary, those are 3 people who will run. Democratic voters will have to choose: What record do I want to put up to vote for the broad American public? Is it California's? Is it Maryland's, or is it Michigan's?

If you just do a quick AI-driven analysis: Wes Moore, then Gretchen Whitmer, then Gavin Newsom, just on the facts.

Yeah, Friedberg, your thoughts on the platform I put out there, or anything else with this 2028 discussion as we wrap up?

No, I think the socialists are going to continue to grow their momentum, and it'll be fun to see what happens.

Okay, here we go. Trump had a busy week. He met with Putin, then Zelenskyy and some of the leaders in Europe. Sacks, this is your time to shine.

We didn't talk about this yet because it happened after we taped, but last Friday, Trump met with Putin in Alaska. There was a lot of hand-wringing around that. It lasted about 3 hours. No ceasefire was reached. That was the stated goal of it.

On Monday, Trump met with Zelenskyy and a bunch of European leaders at the White House—the first in-person meeting since that argument, back in the Oval Office in February. It included the EU president, the NATO secretary general, and the chancellor of Germany. Italy and the UK showed up as well.

Your thoughts on where we're at with Ukraine, and whether we're going to get a ceasefire, a peace negotiation, or a peace settlement, or whether the United States is just going to walk away? It seems like Trump is a little perturbed by the lack of progress here, and he has said, “Maybe it's just up to them to figure it out,” and that he'll be done with the whole thing.

Your thoughts, Sacks?

Well, I think that the Alaska summit was major progress. First of all, I think the president deserves enormous credit for reestablishing diplomacy and communication with the Russians. There literally hasn't been a meeting between the U.S. president and the Russian president since 2021, while this war has been going on and while the war has continued to escalate.

I guess the Biden administration's policy was just to ignore the Russians, roll the dice, and hope that the war didn't escalate out of control. In any event, I think Trump deserves a lot of credit for reestablishing that diplomatic connection and trying to use diplomacy to end the war.

It was shocking to see how much Democrats in the media were openly rooting for him to fail in these efforts to end this war. Effectively, they're rooting for hundreds of thousands more people to die or get maimed.

They were pretty hysterical on the news, right? It was pretty crazy.

Yeah. And by the way, they claim that they care about the Ukrainian people, but the Ukrainian people, by large majorities now, want this war to end, and they're willing to make meaningful concessions to do it.

There was a poll by Gallup, which, Nick, maybe you can throw on the screen, where a year or 2 ago, something like 70% of Ukrainians wanted to keep fighting the war. Now that number is down to 24%. Again, the vast majority are looking to achieve peace.

I think that President Trump and President Putin made some important progress in terms of getting to an eventual peace deal. I think there are a few planks here.

Number 1 was the recognition that we should try to get to a comprehensive peace deal, not a ceasefire. A ceasefire is temporary, and the Russians have already said they wouldn't do it because they understand, from their point of view, that this would just give the Ukrainians time to regroup and rearm and try to reset, but it wouldn't end the war.

The truth of the matter is that when you're fighting a war and losing, you don't get to call a timeout. The Russians have been rejecting this idea of a ceasefire for months. I think that it was wise, if our ultimate goal is to get to a deal, to recognize that a comprehensive peace deal is what we want to get to.

So that's point number 1. Point number 2 is that President Trump acknowledged that Ukraine would not be joining NATO, which was the key source of the friction and the thing that ultimately led to this war. I think definitively taking NATO off the table was a very important precondition to getting to an eventual peace.

The 3rd thing was that the parties—both the Russians and the Americans—agreed that there would have to be territorial concessions of some kind.

And we don't know all the details yet. There was some conversation about land swaps. But the point of the matter is that we're not going back to the prewar status quo ante. The Russians have taken a big chunk of territory. The Ukrainians have not been able to get it back. People are trying to blame Trump for this somehow.

But the real cause of this is the failure of the counteroffensive in the summer of 2023. Remember, we gave the Ukrainians something like $100 billion worth of weapons to retake that territory, and it was an abysmal failure. They didn't even make it past the first Surovikin line. That is the real cause of why the Ukrainians are going to lose territory in this war: they're simply not able to retake it.

I think there was a general recognition coming out of that Alaska summit that there would have to be territorial concessions. And I think in these 3 pillars, you have the basis for a comprehensive peace deal, which is, again: number 1, we're going for a peace deal, not a ceasefire; number 2, no NATO; and number 3, we have to recognize the realities on the ground.

Now, I think the big question mark is that we could get to a peace deal pretty quickly if Zelensky and the Europeans were willing to acknowledge these 3 points. But so far, they have not been willing to acknowledge those 3 points. And I think this is one of the things that, if you listen to the remarks the following Monday, when the Europeans came to the White House, there was no recognition. They were still clamoring for a ceasefire, which, again, the Russians had rejected, as opposed to a comprehensive peace.

They were still clamoring for NATO, and they were still rejecting the idea of any territorial concessions, including even Crimea, which was lost over a decade ago. So, I think that the issue we have right now is that Zelensky and the Ukrainian government—not the people, but the Ukrainian elite—and some of these European countries that are backing him are not willing to make the compromises necessary to achieve a peace deal.

And as President Trump has said, at the end of the day, you need the Russians and the Ukrainians to agree. So, I think that President Trump deserves enormous credit here for reestablishing diplomacy and for making substantial progress toward a comprehensive peace deal. But at the end of the day, unless Zelensky starts being willing to make compromises, I don't know how you're going to get to that deal.

Yeah. Friedberg, Dave, any thoughts on this? Negotiation? Chamath, anything?

No.

Well, I was pretty curious about the historical pattern of this, so I just went back, and here's what I learned. What's interesting to note is that since 1945—not including conflicts in Africa, because they are not really quite part of the Western political global order per se—it's important to ask the question: How have other territorial disputes panned out?

There are disputes that end in stalemates, and when you end up in a stalemate, you basically arrange a DMZ, a demilitarized-zone-type arrangement. There's one in Korea, 1953 to now. There's one in Cyprus, 1974 to right now, and there's one in Kashmir, 1947 to today. And then there's this kind of thing in between Israel and Lebanon.

Second, there are disputes where there's some action taken by one country to take something by force, but then there's no consensus internationally as to who owns a territory. So, Crimea, as Sacks mentioned, 2014 to present, and then Israel and East Jerusalem.

I think the point is that unless President Trump figures out a way to do something that's frankly truly unprecedented, the historical pattern is that it's been nearly impossible for even the most powerful and influential person in the world to broker deals in these kinds of situations, especially when they become protracted.

And I think that's what he's getting to when he reinforces, “These things never would have started or gotten this far had I been president.” That's what he says, in quotes. And I think this is what he's referring to, which is that the longer these things get drawn out, particularly with Russia-Ukraine and Israel-Gaza, you're talking about conflicts and historical issues that have been simmering for hundreds of years, and in some cases thousands of years, all the way back to antiquity.

So, I don't know. I just think that it's an incredibly difficult position, and I just don't know what's going to happen. But I think that for modern progress to happen, it's important for both of these issues to get put to bed as quickly as possible.

J-Cal, what do you think? We saw the media attacking Trump for even daring to hold negotiations with the Russians. Do you think this was a good idea, or do you think that somehow Trump was caving to Putin?

I give Trump a lot of credit. He is incredibly good at foreign policy, and specifically with dictators. If you look at our past presidents, they didn't want to meet with Putin. They didn't want to meet with Kim Jong-un or Kim Jong-il. They thought isolation was the best policy. It's not. I think talking to these folks and trying to find common ground is the best policy.

Even if it's a 5% chance of it actually working out—which is what I think any president dealing with a dictator and a despot like Putin is going to have—I think you have a 5% or 10% chance. And I think Trump and Biden actually had a lot in common that worked: the lend-lease of the weapons and not costing the American taxpayer any money, where a lot of partisan people said, “Oh my God, this is costing Americans so much.”

Actually, Trump has figured out a way to make it profitable. He's selling more weapons to Ukraine, and he got us the mineral rights. It turns out Trump was able to turn a cost center, or a supposed cost center, into a profit center in terms of this specific war.

And if you look at the sanctions, Trump has increased the sanctions. So, Trump had my exact position that I fought for here on this podcast for the past couple of years in this conflict, which was: keep the pressure up on Putin. And Trump took it over the top by going to Putin's main client for his oil now, which is India, and putting a massive tariff on them and threatening them with a huge tariff. That was a bold, audacious way to deal with Putin.

So, I give Trump a ton of credit for trying to make peace here. And Putin has been—hold on, let me just finish—absolutely humiliated. He can't even get to Kyiv, let alone take over Finland or Poland. And this has cost Russia its greatest customer, the EU—Germany. These people are never going to buy his oil again, not after what happened here.

And you can give a lot of credit to Trump because he told them, when they built their oil pipelines, that the Germans were lunatics for doing that and for creating a massive revenue stream for a dictator who is, in my mind, a war criminal. What he did in invading Ukraine makes him a war criminal. What he did in kidnapping these children makes him a war criminal.

I think it's awesome that Trump is doing this intervention, even if it has a 5% or 10% chance of working, and then making the NATO alliance pay their fair share and selling them more weapons. I think that's also brilliant. So, I give Trump a lot of credit here. He's aligned exactly with what I said we should be doing, which is holding dictators who invade other countries accountable for it. And he's added to it the fact that he can connect with people. So, I give him a ton of credit for it, Sacks. I know that's maybe not the answer you want.

Look, I think you use this highly moralistic language and engage in a lot of name-calling.

Which one? Which name is inaccurate, please: dictator or war criminal?

Well, look, you're complimenting Trump for engaging in diplomacy with Putin, but then at the same time, you want to call him a dictator and war criminal, and that's not going to be conducive to actually trying to reach a peace agreement. Now, it's fine.

But I'm not doing the negotiation. If I was doing the negotiation, I wouldn't come in and say, “Hey, dictator, war criminal, let's fight for peace.” I'm giving you my personal belief, personally, not negotiating.

So, you're praising Trump for increasing the pressure and leverage that he has with the Russians, which I think is fine. My question for you is: Would you increase the pressure or leverage with respect to Zelensky?

For example, would you tell Zelensky in no uncertain terms, “Listen, stop demanding to be part of NATO because it's not happening”? In fact, the president has done that.

Yeah, the answer to that question is yes. I would say, listen, it's not realistic to join NATO. You can get some security guarantees. There's sort of talk of this NATO Lite. So, I would agree with that. I would say, listen, we're going to take NATO inclusion off—and I said this on a previous episode—we should just take NATO inclusion off for 20 years.

Okay. What about territorial concessions?

I think that's a decision Ukraine has to make. We should encourage them to make whatever decision is best for them. But if they want to keep fighting this thing, that's going to be up to their people to decide if they want to fight for their land.

Well, the people have said they don't want to keep fighting. The problem you have right now is that 70% of Ukrainians say they don't want to keep fighting. They're ready—

Sounds like they're ready to make concessions.

Yeah. So, it sounds like Zelensky should listen to his people. Yes.

That's his decision to make. I think he has sovereignty, and the people get to make that decision in his country.

Well, but the problem you have is it's not just Zelensky, right? It's his ruling clique. They have canceled elections, and they will remain in power as long as the war continues.

So there's a conflict of interest there, and there's a lot of money flowing into Ukraine because of the war. So what if the ruling elite wants to keep the war going forever while the people want it to stop?

Yeah. They're going to face a revolution right quick. I think they have to take the citizens of Ukraine into account. It's a tough decision for them to make, but it's not our decision to make for them. Is that a reasonable position? It sounds like a reasonable position.

I think it's reasonable.

Yeah, I don't—

Look, at the end of the day, the Ukrainians and the Russians have to agree.

Yes.

I think that, of the major parties, President Trump cares the most about peace. I think he's pushing the hardest for there to be a peace deal.

I know people are a little flabbergasted that I really believe Trump is great at this. I think he's great at this. He is great at negotiating with difficult people, and I think he's going to get it done. I think he's got a 50–50 chance of getting this done, and I think other people didn't even—

Admit it. Just admit it. You like him.

You really like him.

It's not about whether I personally like Trump or not.

It has nothing to do with that. I think I call balls and strikes here. It's all balls and strikes for me.

No, no, no. You're in like with him. You may be in love with him.

I loved it. I'm in love with my guy, Vice President J.D. Vance.

Wait, wait, wait. And you loved it when he said, “Even Jason, you loved it.”

Well, I thought it was hilarious as a troll.

How many people did you send that link to?

Be honest.

I just tweeted it. I didn't send it to anybody. I tweeted it.

Okay, so everybody—a million people.

It was a surreal, funny moment and a great troll.

It's not a troll.

It was a troll on Sacks's part. I think Sacks may have encouraged it.

It's an incredible thing that you got.

Let's keep these great debates going here on the All-In podcast, a platform for important discussions. Another amazing episode, summer months coming to a close, and we'll see you all in Los Angeles for the fourth edition of the All-In Summit.

Back to the office next week. Back to the office.

Back to the office and then back to the grind.

Summer has ended. Summer 2025. So sad. It came way too soon.

Way too soon.

Can't believe it's over.

That's ridiculous. The only thing that's giving me some hope is that ski season is coming.

You know, when your kids are—Sacks, you can relate to this. Actually, Jason, your kids are pretty old, too. The summers go by so quickly, and it's like you just start to see that they are—

I am now 18 months from my oldest leaving the nest.

It's the craziest.

Same here.

Oh my God. It's so—I can't deal with it. My son sits for his ACT in 3 weeks. He's getting his driver's license in a week.

Are you sad about it? Are you feeling melancholy, like leaving-the-nest kind of thing, or are you excited for that new chapter?

I feel like all of my kids deserve to have their own adventure. And so what I tell my son is, “Listen, you're packing your bags, you're building your little toolkit, you're about to leave, and you're just going to go on an adventure.” I'm so happy for him. But, man, yeah—guts.

My oldest is applying to colleges right now.

Oh my God. Oh my God. All right everybody, another amazing episode. All right, love you besties, and we'll see you in Los Angeles in Summit. Love you boys.

See you next week. We'll let your winners ride. Rainman David, and we open sourced it to the fans and they've just gone crazy with it. Love you. [Music]