Trump AI Speech & Action Plan, DC Summit Recap, Hot GDP Print, Trade Deals, Altman Warns No Privacy

How much founder mode did you do? Are you saying that I popped an ALP? I need an ALP right now. Hold on. You don't need anything right now. Are you chewing it? What are you doing? No. You put this nicotine pouch, you upper deck it, releases it, and then you become a god. Is that the app that Tucker sent you? Yeah. Tucker and I are going to do a crossover. Wait, did you work out a side hustle here? I haven't presented it to the group for a vote yet. You're pre— Wait a second. Are you being paid for this plug right now? Yes. I'm just saying if you use the promo code JCL. Wait a second. Promo code J 15. Okay, he broke up, which is good. Is he on drugs? Is he taking drugs? He's on drugs. No, I'm not on drugs. And he's doing a deal with— This is like a PSA for not taking this stuff. You're so out of control. Did you take two of them? What are you doing? I thought this stuff relaxes you. What the hell is going on? Your internet's on the fritz, too. I fixed it. I fixed it. I fixed it. That was Putin. Putin's got my internet. Putin's got my— Oh my god. What flavor are you eating or using? Oh, today's chilled mint. Today's chilled mint. You don't seem very chill. You seem— This is the first one. Agitated and angry. No, I'm trying to get us back to that original all-in energy where we laughed and we had fun and we enjoyed each other's company. No, but Jakeal, seriously, do you have a side deal going on with Alp right now? No, I don't have a deal yet. I don't know if I have a deal. There's no deal. I'm texting Tuck right now just to cut you. [Music] Let your winners ride. [Music] We open sourced it to the fans and they've just gone crazy with it. Love you guys. Queen of—

All right, everybody. Welcome back to what Jensen from Nvidia has confirmed is the number-one podcast in the world. Yes, the All-In Podcast is here. We had an amazing time in D.C. last week, and we'll get into that.

Friedberg, you crushed it on all those incredible speakers last week. You had 10 days to pull off that event, and you did it. Chamath and I just parachuted into D.C. last week for the AI summit. Sacks was busy working with POTUS to get all those executive orders done.

Take us behind the scenes, Friedberg. You had Lisa from AMD, Lutnick—I liked him—and Bessent—I liked him. We had to say no to a lot of tech company CEOs who found out about the event and wanted to speak onstage.

There was a big cutoff that we had to make around ensuring that we got our message across. If you watched the content, we talked briefly about it at the beginning, but the focus was really on trying to dispel the negative AI narrative and the myth that AI is just here to destroy jobs.

There's this big economic boom happening, both with respect to new industries that are emerging—which is why we showcased Hadrian and others—and the infrastructure needed to support the AI race, with data centers, chips, mining, and energy. We highlighted each of those 4 industries.

Then the Cabinet people found out about it and wanted to get involved, so we were unfortunately squeezing people on and offstage. It's kind of crazy to tell the secretary of the Treasury that he has to get off the stage because he's passed his 20-minute allocation. We had to line everything up so that the president could get his Secret Service detail to clear the stage and get set up in time. That's why we were rushing everyone.

But what a week. What a rush. It was awesome. Thanks to David Sacks for the leadership and for pulling it all together and bringing those folks to the table. And Sacks, congratulations on getting your executive order signed and your action plan published. It was pretty cool and pretty awesome to meet the president, meet all those Cabinet members, and have all of this come together because of the work you've been doing. How does it feel? Sacks, how are you doing in the afterglow there? I can see that you're in the afterglow. You sent me a picture of the 4 besties with our incredible 47th president. How are you feeling right now?

Are you going to put that on the screen?

I may have it. I don't know if that's allowed. Are we allowed to put that on the screen? I don't know what the protocol is.

Yeah, I think we can. I mean, I haven't gotten my picture. I did notice that, unfortunately, when they took the picture of the 4 of us with the president, somehow I got cropped out by accident. I think maybe they weren't using a wide lens.

Wait, Jason, what was it like for you to meet the president? Just for the audience, we all stood in line, took a photo with the president backstage, and then took a photo with the 4 of us. But Jason, when you had your moment with the president, what did you say? Did you ask him about immigration?

I have your photo with the president.

Oh, it's on my phone.

Did you bring up solar panels with him? What was your big moment all about?

I didn't know we were taking a picture. It was sprung on me. My brother Josh, who runs security for us, said, “They need you in the back to take a picture with the president,” and I said, “Yeah, I'm good. I have to prepare for some—”

Oh, you were going to pass?

Well, I thought he was joking with me. I said, “Yeah, I'm good. I'm good.” So he said, “No, no, I'm serious. They're taking pictures with the president.” I was like, “We are?”

I ran back, and they put us in line. Then I thought I was getting punked because they kept repeating to me, “Okay, Jason, you're last. You're last.” I know you guys like to put in a joke or two, so I just got in line last.

Obviously, it's a big deal to take a picture with the president, so I didn't want to use that time inappropriately or anything. I just said, “It's a pleasure to meet you.”

Just say it already. You like him. Just say it. Let's get it over with. Get to the end. You like him. You tried not to. You're all Mr. Big Shot, Mr. Big Talk, and then you got in front of him and you liked him. Just say it.

What I will say is—

Jesus Christ.

Like him or dislike him—

What a joke. You're such a goon.

I had a great time. I had a great time.

Predictable. You're a predictable goon.

You don't even know what “goon” is. Stop rizzing. Stop aura farming. You don't know what gooning is. I had a great time meeting him. It was a great event. Obviously, he's trying to get his rizz up to impress his kids. But it was great, and I didn't know what to do in the picture, so he did.

We can move on. What do you think of his speech, Jason? After the president gave you a shout-out, how did you feel?

I don't know about love. He said, “Even Jason.”

How many times have you listened to that clip over and over? How many times?

How many times? How many people have you shared that with? How many? Play the clip. Play the clip.

Donald Trump

Say hello and thank Chamath and his wonderful wife, Nat. Thank you very much for being here. Thank you very much. It was great seeing you again. Great couple. David Friedberg and even, as we know, Jason Calacanis. Thank you, Jason.

I say “even.” I appreciate that.

Yeah, he's a good person.

I mean, he's a good person.

He called you a good person.

He called you a good person. So here we are.

I mean, it's obviously surreal for all of us, I think, to be this close to the administration and then for Sacks to be part of it. You have to give a lot of credit to this administration for the velocity they're going, what they're accomplishing—even if you disagree with certain items on the margins—and their ability to engage with leaders doing important work. If we compare that to Biden and Kamala, they weren't even letting people come to the White House.

I love this administration. I love the administration. I like Trump. This is a Cabinet of CEOs. Let me just say this: I'm not in love with Trump. I'm in like with Trump. That's where I'm at. I'm not in love with Trump; I'm in like with Trump. But what better team has ever been put together? It is a Cabinet of CEOs. It is a Cabinet of managers. It is a Cabinet of people who know how to get things done. Every time I go there, I'm impressed by this Cabinet. I pull my hair out when I meet.

You admit that you're pro-Trump, finally, Friedberg. You've been splitting it and dancing around the issue. Are you 100% in support of Trump? You want to sit here and put me on the spot?

I support my president. I support the president.

Okay, so you voted for him and you love Trump. You voted for him and you love Trump.

I love what he's doing, and you voted for him. I have issues with the spending, and that hasn't been resolved. So, like I said before, my full-throated endorsement will come around when DOGE actions are taken seriously and/or the White House puts pressure on Congress to take action on spending in the budget.

Okay, great. Here we are, folks. What is everybody's favorite moment—favorite other than Trump being absolutely amazing and giving a great speech?

He's hilarious. Whatever. We'll put that aside because it's hard to compete with the president of the United States. Sacks, did you have a couple of favorite moments? Give us a couple of favorite moments.

First of all, I think we should talk about the substance of the speech, because I think this was the first speech that President Trump has given on AI since the AI boom began. He's spoken about it before, but this was a full-length policy speech, and he declared that the United States was in an AI race. It's a global competition.

I think the language that he used was reminiscent of how President John F. Kennedy declared that America was in a space race. In a similar way, President Trump declared that we have to win the AI race. I think you can argue that the AI race is more important than the space race. It's going to reshape the global economy. It's going to determine who the superpowers are of the 21st century.

President Trump was really clear that we had to win it and that he was going to support a strategy for winning it. Then he laid out what some of those key pillars are. Number one was innovation. We have to get the red tape out of the way and let our geniuses cook, and he was clearly very supportive of a lot of the CEOs and entrepreneurs in the crowd.

Number two is infrastructure. He touted the hundreds of billions of dollars of investments in energy and power generation, grid upgrades, and data centers that he's supporting. Then he also supported AI exports. He said that we have to make America's tech stack the global standard.

I think those were really important messages. On top of that, I think there were also some parts of the speech that may have gotten less attention but are also important, where he said that it's not only important that we win; it's important how we win. He sort of mentioned three non-negotiables here.

Number one was that American workers have to be at the center of the prosperity that we create. Number two is that the AI models that the government procures and buys must be free of ideological bias. So, no woke AI. He also signed an executive order to prohibit woke AI in the federal government. We could talk about that in a second. That probably was my favorite moment.

That was your favorite moment?

That was my favorite moment.

The red-meat moment. I thought that was—

That was the red meat.

That was the red meat for the base.

Yeah. The third thing is that he did say that we do want to prevent our technologies from being misused or stolen by malicious actors. We are going to monitor for emerging and unforeseen risks, so we're not going to disregard the risks.

He had this really good line in the speech about how, even though AI is a daunting technology because it's so powerful, and like any revolutionary technology, it can be used for bad as well as good, the daunting nature of it is all the more reason why we have to do it in the United States. The United States has to be the pioneer and the leader because we don't want the power of that technology being developed in other parts of the world. At least, other parts of the world are going to have it, but we want to be the ones on the cutting edge who are defining it and leading it.

Fantastic.

I think it was a really important speech. I think this idea of an AI race that is similar to the space race is going to be the dominant frame on AI policy for years to come.

Well, it's pretty clear that this presidency, this term, is going to be earmarked, I think, by 4 key initiatives: AI, crypto, immigration, and tariffs. I think that feels like what they're locking into as what's important for the next 3 and a half years. I think you would agree with that.

It's just great that you're spearheading and helping the president with 2 of those 4. The velocity, to me, is what's super impressive. Is there any way you could take us behind the scenes of how this stuff is getting done so quickly? It feels like there's some operational cadence here that we didn't see in his first term. We certainly didn't see it in the Biden term, but there's a cadence here that's different. Startup speed. How is that?

Well, yeah, we call it working at tech speed. I just think that the president's constantly working. He's so energetic. He basically works 2 full workdays. I think it's well known that he doesn't need a lot of sleep, and he continues to work late into the night. I just think his energy propels everything forward.

I also think that there's a very cohesive team at the White House under the chief of staff, Susie Wiles. I think that's very important. I think she runs a tight ship. Then you've got the deputy chiefs of staff under her, and I think most of these people have been working together for a long time. It's a team that works well together, and it just feels very coherent and cohesive to me.

So, I think it's a very effective team.

It does feel like that. The pace is great. It means you're going to get more shots on goal, and you'll be able to try more things and get more accomplished, just like we see in startups.

Chamath, outside of the president's talk, we'll go around the horn here. Top 2 or 3 moments from the discussions. Just a lightning round here, rapid fire. What do you got? Top 2 or 3 moments for you, Chamath, in the discussions that were enlightening, inspiring, or notable to you?

I came out of it very motivated. I think that the combination of the speech, the executive orders, and the clarity of the Big Beautiful Bill now give those of us that are in these markets a ton of runway to go and execute. Those things, reinforced by the various members of the Cabinet, were very important. That was one.

The second thing was the market commentary from both Lisa Su and Jensen Huang. I thought it was really valuable. The third was Chris Wright and Doug Burgum talking about energy. I tweeted this yesterday, but we are sort of back to basics, almost, in a sense, where in the absence of power, I think AI is not going to be the thing that we think it can be.

That's going to create an enormous amount of appetite by the federal government to do deals and get players on the field. That's, to me, very exciting. I came away really risk-on, I guess, is the best way to say it.

I love it. Friedberg, you have 2 or 3 moments outside of the president's speech. Obviously, that's the pinnacle there, so let's just go below the pinnacle. What were the other 2 or 3 moments for you that were salient, inspiring, or notable?

I thought Jensen did a great job. I don't know what you guys thought, but he is very compelling and has an incredible vision and view on where AI is taking us, where it's headed, and what the challenges are. I really appreciated him taking the time to come and join us. At the last minute, he rearranged his schedule to come out for it, and it was great.

By the way, on the point on energy, which I still think is the biggest unsolved issue right now in America, besides the federal deficit and the debt problem, Chris Wright agreed to rearrange his schedule to come and join us at the All-In Summit in September.

Oh, great.

To continue the conversation. We didn't get enough time to talk about it, so we are going to hear more from Chris, particularly with a particular focus—which is what I wanted to spend time on and didn't get a chance to last week—on nuclear. Where are we? Because he actually is very passionate, like he said at the event. It's where he's spending most of his time right now, and I think it's very good to hear the deep dive on where we are in the cycle of trying to accelerate nuclear-energy deployment in the United States.

Same question to you, Sacks. You got 2 or 3 moments that stood out?

Let's just talk about the executive orders for a second, because I think it's pretty cool that the president of the United States signed 3 executive orders at the All-In Summit that we just hosted. That was pretty amazing.

One of them was to promote AI exports because we want the American tech stack to become the global standard. The second was around AI infrastructure, to make permitting easier so that we can help solve those energy problems you're talking about, Friedberg. The third one was on preventing woke AI in the federal government.

That, to me, is probably my personal favorite, because we spent a couple of years on the show talking about how, when we're talking about woke, you're really talking about censorship, right? We were talking about censoring people's views based on ideological bias and ideological dogmas. We saw what was happening in social media before Elon bought X, which helped bring things back.

But we were on a track, I think, before President Trump's election, to repeat that whole social media censorship apparatus in the form of AI bias or AI censorship. We saw this with the whole Black George Washington, and where some AI models were saying it was worse to misgender someone than to have a global thermonuclear war.

Yeah. And this wasn't an accident, because if you go back to the Biden executive order on AI, there was something like 20 pages of language in there encouraging DEI values to be infused into AI models.

So, again, we were on track to repeat all the social media censorship, all the trust-and-safety stuff, in this new world of AI. But it would have been even more insidious, because at least when someone gets censored, you kind of find out about it.

It's explicit.

It's explicit. But with AI, it would have been worse because you wouldn't have even known. It would just be there, rewriting history in real time to serve a current political agenda.

It would have been brainwashing our kids.

Oh, and people trust these AIs more than they should. These things are making a prediction of the next word coming. This is not God-given truth here.

Friedberg, you wanted to interject about this one because, I'll be honest, Sacks, I'm surprised you're saying this was the most important one to you. I like that you clarified it because it was the one that was mocked, or people were like, “What? Why is this important?” I think you made a good case for why it's important. Friedberg, your response.

Yeah, but, Sacks, this is not about broadly making AI non-ideological. Private companies should still have the right, through freedom of speech, freedom of expression, or freedom to operate, to make AI that does whatever they want it to do.

What the executive order was was that the federal government would not procure ideologically biased AI. Is that correct?

Yes, exactly. No, we're aware—

Just to make sure that the federal government is not trying to instruct private companies how to operate, it's simply saying, “If you want to sell to us, these are the rules of the road.”

Yes, that's true. We were very careful about the First Amendment issues. You're right that if a private company wants to put out a biased AI product, we're not going to tell people they can't use it.

And it could work. It could be successful. People might like it.

Yeah. We're just saying that the federal government is not going to spend taxpayer money buying AI models that have compromised their accuracy and quality because they're beholden to some ideological agenda.

Which is similar to the approach with the universities, right? You could have a biased university. We're just not going to fund it. We're not participating. I think it's quite reasonable in that way.

Yeah, and I would just say that we were a lot more careful about this than the Biden administration was when they required that DEI be inserted into all these models. They didn't distinguish between public and private money or government procurement versus private models. They were trying to suffuse DEI into everything.

What we're looking for here is just neutrality, right? We're looking for a lack of ideological bias. The first step was to get rid of that Biden executive order, which the president did in his first week in office. This goes a little bit further, and it's a little bit of a shot across the bow of these Silicon Valley companies saying, “Look, you need to play it straight. You need to be ideologically unbiased.”

As the default, when you sell to the government, you can't insert your values at the expense of accuracy. At the end of the day, accuracy and truth-seeking are the standard, right? You can measure. That's the goal.

So, we don't want quality, accuracy, and truth-seeking to be sacrificed because of this. Are you still seeing that? When you say these Silicon Valley companies, is this still a widespread concern or widespread deployment, from your point of view? Are you still seeing a lot of the models being trained on ideological systems that are preferential to one group and not to another?

I think it was a much bigger concern 6 months ago, and I think there's been such a huge vibe shift since President Trump's election and taking office that the woke stuff is sort of going away on its own. I think that's the trajectory we're headed toward.

But you still think it's important enough to make sure that there's any error?

Yeah. It's like, look, this is making sure this thing doesn't come back from the dead. I think there's been a huge vibe shift since President Trump's election, and woke has definitely fallen out of favor and seems to be going away on its own.

But we could still get Orwellian outcomes with AI. I do think it's very important to keep underscoring that what AI models should be focused on is truth and accuracy. We don't want ideological agendas to sacrifice that. Even though this is a less salient issue now than 6 months ago, precisely because of the vibe shift, I still think it's important to underscore this point: We don't want AI taking an Orwellian direction.

Would you go so far as to limit free speech and make AI non-ideologically biased? Would you make that law if you could? Again, the decision is about the federal government procuring versus what these private companies can choose to reflect as their values in their systems.

No, he just—you already answered that he would not.

Yeah. No, look, we understand the difference between public procurement and private speech, and again, in a way that the Biden administration did not, because they were saying that all AI models had to adhere to a specific ideology—to the DEI stuff.

Yes, it was an ideology they wanted embedded in it. You're saying, “Don't put an ideology in.” But just to be clear here, I want to make one point: This is the default.

Anybody who wants to could, when they start their prompt or set up their preferred language model, say, “I'm an atheist. Here's what I believe. Please speak to me with this in mind.” Or, “I'm a Catholic. I'm a Protestant.” Whatever you want: “Here's my belief system. Please never reference these 3 subject matters in this way.” This is the default. I think it's a great thing to—

I think that's a great example, JCal. I do think we'll end up seeing religious AI. I think we'll see AI that's tuned to people's religious ideologies.

One of the startups we did was doing a learning app, and they were struggling. They just made a prayer app, and their prayer app went parabolic. Now they're just printing money. There is definitely a huge market here.

Sacks, what were your highlights?

It was great to be included in everything, so I appreciate that we had it. No, I'm being dead serious.

Your invitation finally didn't get lost in the mail.

No, but here's the thing. I think this could have been a non-All-In thing. It could have just been—you could have done it and invited whoever you wanted to. I like that it was under the All-In umbrella, that we didn't censor anything, and that we went right at hard topics.

I'm a moderate. I know people want to make me into a stupid lib, but I am an independent moderate. There were moments in time when we had great debate, too. This wasn't just a love letter to the administration.

One of the great moments was J.D. Vance. It was just great that he wanted to come chop it up and hang with the besties. He came out and went right at me. He was like, “Hey, you treated me like a [bleep] at the thing.” We had a big debate, and he went right at me. Then I was like, “Okay, it's on. Want to talk about stuff?” And he said, “Yeah, let's get into it.”

That's what I love about J.D. J.D. seems like the politician of the future to me. I know this is the Trump administration, but—

So, you like him?

No. I'm in like with Trump. I'm in love with J.D. because he's young, he's opinionated, and he likes to mix it up. He's on Twitter all day long. He engages people on Twitter. He engages people in other groups. I'll leave it at that.

We had a really honest discussion about immigration, and we got back to the high-skilled immigration question. That's the third rail for MAGA and for the country right now: immigration recruiting.

You mean you brought it up right off the bat?

No, no. He said he wanted your hobby horse. You brought it up—your hobby horse. I wanted to continue the debate, and I said, “Okay, let's continue the debate.” So here we go.

He was super spicy, and he made a great, super-spicy point that I want to point out here on Amplify. If companies are going to be laying people off—and there was an incredible chart in the Financial Times that showed male college graduates versus non-college-graduate males—there was usually a huge gap in unemployment between those two. In other words, if you had a college degree, you had a much better chance than the non-college male.

Now those two things have flipped, or they're neck and neck. If you have a college degree, you have no advantage as a man coming out in this 20-to-27-year-old range. This is men. Women are actually doing better. There are more women in college than men.

He's very attuned to this, and he said he's got big concerns right now. This is again why I love J.D.: He's very tuned into the fact that people are asking for more H-1B visas, and that typically is to save money and is supposed to be for very skilled people.

But why is Microsoft laying off 9,000 people and then asking for more H-1B visas? This is a really honest, truth-seeking question. It's hard for this administration to talk about this issue because you've got Stephen Miller, Bannon, whatever—people all the way on one side who want to deport 20–30 million people, Tucker, and then you have other people who are more moderate.

I thought that was a really great moment in time for America and for us as a podcast to challenge and have a really important discussion. He made some great points there.

Number 2, we had a great debate, I think, about energy.

Disagree.

You disagree. Okay.

I disagree with that because I think you challenged him with things that were not facts and not true.

And I'm happy to debate that with you offline. I think he was caught off guard, but I think it was pretty rough and inappropriate.

100%. If you think it's inappropriate, that's fine. I—

Jason's favorite moments were all the ones where he got into debates.

That's what you're describing.

No, no. Where there were debates. You got into it with the vice president. You got into it with the secretary of energy. Those were your favorite moments when you got to—

Okay, great. Fine. I like when there's a little conflict, a little debate about an important issue. When I polled the audience, which was 90% Republican, GOP, MAGA, et cetera, people said that was a great moment.

I really liked that debate because he kept saying, “Unreliable energy,” or whatever. I said, “Are you talking about solar?” I think there was a little misinformation.

Reliable. No, it's not.

You put it with a battery. Right now, Texas is 30% wind energy on some days. I can tell you—I live in the great state of Texas.

Texas is roughly 5% solar, just so you know.

What's that?

Texas is roughly 5% solar, right? Wind puts it up to 25% to 30% on the top days. My point about that is, it's cheaper to put in a solar and battery farm than a new coal plant. It is 100%. We can pull up the stats.

I said coal.

It is twice the cost to do solar than it is to do natural gas. It takes 4,000 acres, whereas gas takes 20 acres. The big advocacy with these guys is to use natural gas, to use methane. He was saying “clean coal, clean coal, clean coal.” He said it 50 times.

These methane plants are half the cost of solar. They can be stood up in less than 2 years to generate a gigawatt. Instead of being 4,000 acres of solar, you can get it done for, call it, 20 acres. Now, talk a little bit about pollution. That's a big part of why they're doing this. A big part of methane is that it's actually cleaner than coal, which is why they're using it—cleaner than oil.

Cleaner than oil. There are 2 ways of getting energy. Now, science guy, now do—

I'm trying to give you the facts about why it's cheaper and faster, which is what he was making an argument for, right? It's not about solar. Yes, you're right: It has a lower carbon footprint when you're running it. But at the end of the day, what these guys are focused on—and a big challenge for America—is how we scale energy production in the states.

In scaling energy production, I personally think we need to fix the regulatory roadblocks in nuclear. Chris Wright's been very vocal on this.

We all agree on that.

But the fact is, this natural-gas supply that we have in the United States, and the fact that we can deploy natural-gas energy production very quickly, is what makes it such a reliable source right now.

If the U.S. wants to have a chance at scaling from 1 terawatt to 2 terawatts faster than projected today, that's the reason. It's not about solar being bad. Solar is bad—that's not the argument. It's just, dude, we have to get moving fast, and we have to have reliable energy.

In our debates, I just want to point out that when there are bad-faith moments, I think it's a bad-faith moment when I say coal versus solar and then you say, “No, you're wrong. It's solar versus natural gas.” That's what he was doing. This is what politicians do.

Here at All-In, we like to do fact-based, truth-first stuff, not biased stuff. Solar—you're comparing solar and how fast it is versus how fast it is to go to natural gas. Of course, it's faster to go to natural gas if we have those available. Let's put that aside. It's an important debate. The fact that you and I are debating it is important.

I also thought Lisa Su from AMD was fantastic. I haven't heard from her. By the way, I want to point out that when I got back to the conference—I left for a time to go back to the White House and then came back—the first thing everyone said to me when I got back was, “Did you see JCal being a jerk to Chris Wright?” Everyone was like, “Oh, yeah, a jerk.”

He's a civil servant. He has to answer hard questions. You didn't talk to him in a way that you would talk to someone. Basically, everyone thought you were a jerk to Chris, right? And you were kind of a jerk to J.D. What are your favorite moments from the conference? You're reminiscing about that.

You were an asshole to me.

Anyway, the point is, one thing you're going to get here at All-In is that everyone—two out of 3 people—was saying this. You almost derailed the whole thing.

Nobody derailed. You're a civil servant, Mr. Sacks. You're a civil servant. You're all civil servants.

I've been putting up with you for 5 years on this podcast.

The hard questions.

It was perfect training for government service, beyond the podcast being interrupted by you for 5 years.

Yes.

That's why I'm so ready.

You work for us, all of you. And you're all going to take hard questions. And you're all going to take hard questions on September 7th, 8th, and 9th when we have the All-In Summit in Los Angeles. By the way, one thing I'll say is Chris Wright's chief of staff came out to me afterward. I said, “I'm sorry. I heard JCal was a jerk to Secretary Wright.” And he's like, “Oh, no. Chris loved it. He loves mixing it up.” Okay, of course. And he's coming to the All-In Summit on September 8th. So, can't wait to debate him more. Can't wait to mix it up more. So he likes mixing it up. Kudos to him.

And so did J.D. Vance, the vice president, with you.

Stop calling him J.D., by the way. Listen, I just want to say Vice President J.D. Vance and I have been directly communicating. We had a—

No, you haven't.

David Sacks, your worst nightmare.

Oh, my God. Your worst nightmare. The nation is ruined. We let JCal into Washington, and now look what's happening.

Yeah. And listen, I want to level-set with everybody: I am going to ask whatever question I want to whatever guest we have, and nobody's stopping me. The only way you're going to stop me is by writing me a huge check to buy me out of this podcast and replacing me with some mid—

Or if the Secret Service keeps you offstage, which might be an option.

Or if the Secret Service keeps us offstage. But the truth is, this is one of the great things about this administration, Sacks: They love to mix it up. They like great debate.

You know who didn't like great debate and ran from it? Kamala Ding-Dong. She wouldn't even come on this podcast. You know who doesn't like debate? TikTok. We get that with Bernie Sanders and Biden, who didn't even know what a podcast is.

Tim Walz doesn't own any equity. He literally doesn't own 1 share of any company. He doesn't own his home. Tim Walz is on there giving a hard time about the Trump savings accounts.

I don't even know if that's a hot take, though, which you loved. You thought that was going to win the election.

I thought he might be able to speak to the middle of America, and then I find out, when they do the deep oppo research, that the guy doesn't own 1 stock.

The guy doesn't own his home. He's financially illiterate, and we're making him employed by the government. He's been employed by the government his whole life.

I mean, have you—there it is.

That's what Jason thought would win the election.

You're never going to live that down. I remember when you tweeted you thought that was it. You thought it was the masterstroke.

The masterstroke that was going to win them the election.

Hey, listen. Nostradamus does not bat 1,000. No, even Nostradamus cannot bat 1,000. But it did come out, by the way, that Nancy Pelosi wanted to do the speed-run primary. I don't know if you saw that. Just not to rehash too much stuff.

Sacks, I want to say there was 1 point of difference, if you want to get into it, around the content part of it. This is something the press was having a field day with and really keyed on: respecting IP, respecting copyright. What's the feedback been so far on that? It was a pretty spicy part of President Trump's speech.

Well, I think what the president said was just very pragmatic. He said we had to have a common-sense approach toward intellectual property. He said if you have to make a deal with every single article on the internet, every single website, every single book, every piece of IP, in order to train an AI model, it wasn't feasible.

He said, “Look, I appreciate the work that went into people creating these works, but you're not going to be able to negotiate a deal for every single one of them. And if we require our AI models to do that and China doesn't—and they won't, they're just training on everything, whether it's pirated or not—then we're going to lose the AI race.”

So I think he took the side of a fair-use definition. I don't know if he used the term “fair use,” but effectively he was taking the side of a reasonable fair use.

What did you think of that part, David Friedberg? Do you have any thoughts, Chamath, on that part?

I think he's absolutely right. I've said this before: If something's on the internet, if something's in the open domain, I strongly disagree with the idea that AI getting trained is the same as AI replicating copyrighted material.

If AI outputs text, audio, or video that contains copyrighted material, it is 100% in violation of copyright.

And he said that, by the way.

Yes. If the AI is learning, it is understanding patterns, reasoning, and concepts by reading copyrighted material, just like humans do. A writer or author reads a bunch of fiction, learns good techniques, concepts, and theory from reading all those books, and then goes and writes his or her own book. They are not violating copyrighted material.

In the same way, Friedberg, what if it’s all the New York Times on the open internet?

100%. You’re 100% correct. That should be paid for or licensed. I’m talking about the open internet. I’m talking about open material. I’m talking about stuff that’s in the public domain.

Common Crawl.

There’s a thing called Common Crawl. If somebody stole 100 books, let’s say, and put them on their website, and it was a pirated Russian website with 1,000 books on it, and you accidentally crawled it, you would be obligated to take that out. I think we all agree.

Correct.

Okay, correct.

Because that’s what a lot of the lawsuits are about. I think we’re reaching something. I just want to say, this is such an important point, especially to me as a content creator and somebody who has spent his career in this.

I’ve been thinking about the endgame. I’m here in Park City, and I was just giving a keynote. I wanted to show you something I made, Sacks, because I think we have to get to the endgame here. In my talk, I talked a little bit about how we can get through this fight and then maybe get to a solution.

I had my team mock up the New York Times website here, with ChatGPT doing a deal with them. Here you see you’re on the New York Times website, and you ask it a question powered by GPT. You might ask it, “Give me the earliest mentions of Putin,” if you were a fan of Putin or something. You log in with your ChatGPT credentials, and it could be Grok or Gemini. It would then go through that and give you your Putin references.

Obviously, this would be an exclusive to ChatGPT. It would be one of those things where they get an exclusive. On the Disney+ channel, imagine you could make yourself into a Jedi Knight. You could upload your photo—kids might really get into this—and it makes you into a Jedi Knight. You talked about this, Friedberg, a couple of times: the future of narrative storytelling. There’s Darth Calacanis.

That looks to me like you’re infringing on their trademark.

What’s that?

Are you infringing on their copyright?

This is fair use. This is a perfect example of fair use for editorial.

You’re also infringing on some Ozempic. That’s absolutely infringing.

Trust me, I am definitely infringing on some Ozempic here, guys. I’m past that. I’m on to peptides now, man. I’m on the Wolverine protocol. I started doing peptides. I mean, what could go wrong?

Don’t take a podcaster’s advice. Please don’t take a podcaster’s advice on your health care. Rule number one. Take Chamath’s advice because he’s got 6% body fat, which I think contributes to much of your pomp and circumstance around your privates. I think it has to do with the lack of fat, but I’m going to leave it at that.

First of all, it’s 11.5%. That’s right before I go on summer vacation. Then it ends up at 12% or 13%.

Did you go get that gelato? What was that place we went to that we loved? Lulu’s? I’ve gone there every day so far.

Did you do 2 or 1? Be honest. 2 or 1?

Bro, I’ve been doing—

No, no. Per session. Do you do 2 or 1? Be honest.

Per session, 2. I start with the medium, and then I finish with a small.

Yeah, exactly. This stuff is so good.

I’ve never tasted any gelato like this. It’s incredible.

I mean, it’s unbelievable. We have to license it for the United States and the All-In brand. We have to license it from them.

It’s really incredible. But, Chamath, generally speaking—and anybody who wants to have at it, Friedberg, Sacks—what do we think about the endgame here? There are some major lawsuits here. They’re going to get settled in the next year or 2. What do we think about the future I’ve shown here today?

I think what Sacks just highlighted is exactly right. Look, we have to have a common-sense approach here, or we’re going to lose the AI race. One of the key determinants of AI quality is the amount of data that you have.

It’s very simple, right? There are a few building blocks: energy, chips, data, and algorithms. If you lose on any one of those dimensions, then you’re in trouble.

You just can’t have a situation where China can train on the entire internet and AI models are hamstrung by needing to contract—

Negotiate contracts with every single website.

But right now, Elon owns X, right? He owns Twitter—for now, X. Does Sam Altman have the right to use X in his corpus?

It’s public.

No, it’s not. It is not a public endpoint.

I honestly don’t know the answer to that. There are some edge cases here. We’re going to have to come up with—

It’s not about whether it’s behind a paywall or not. It’s whether these APIs exist and whether you’re actually contractually allowed to use them or not.

The terms of service. Correct.

The terms of service are published on every website—what the terms of service are with respect to the content. I think it would be okay to let people opt out. We already have this with Common Crawl. You can put it in the footer of the website, put it in robots.txt, and opt out of Common Crawl.

Common Crawl is a nonprofit organization that basically archives the entire web every few months.

Funded by Gil Elbaz—

Formerly of Google.

Great fan of the pod. He shows up to our summits. Great guy.

All of OpenAI was built off Common Crawl originally. They’re very clear, by the way: They say you have to clear copyrights. You don’t get to just use Common Crawl.

Can I go out on a limb? I don’t know if you guys saw this Amazon deal with the New York Times for $20 million a year. Did you see that today?

No, I didn’t see it today. Explain it, please.

I think the New York Times licensed all of its content to Amazon, including The Athletic and a bunch of other things, for training. $20 million a year.

Okay, here we go. I read that, and I thought this is the peak of these deals. These deals will only go down in dollar value from here. It actually brought me to this point: Is it even realistic to believe that patents and copyrights actually exist in 5 years?

I went through this exercise: If a computer studies the periodic table and also understands the laws of physics, biology, and chemistry, and then independently derives a material that is otherwise patented, what will happen? Separately, if 2 competing AIs invent a new material from scratch, how will the international courts deal with this?

If you take all of these examples to the limit, the idea that there are enforceable copyrights, I think, is a very fragile assumption. I’m thinking more that we have to spend some time understanding the landscape of a world that doesn’t have copyrights and patent protections. Instead, what is the surface area in which you compete? What is a trade secret? What does that mean in a world of AI? I think it’s quite an interesting thing to think about.

Patents are a totally different piece. I think that’s a fascinating string to pull on.

I will take the other side of the bet. If we want to make a Polymarket on this, I will guarantee that this will be the beginning of the deals, and the deals will go up from here. I’ll tell you why.

The reason the New York Times made that deal is to make it apparent that what OpenAI has done has damaged their business, because now they have a customer. Their customer is Jeff Bezos at Amazon and Andy Jassy. Now they can show damages, and those damages could give them an injunction against OpenAI.

OpenAI has to take it out of its crawl, its corpus, and that’s going to be really expensive for them. It’s doable, but it’s going to be expensive. Let’s think on a societal basis about what we want as a society. Do we want a society in which journalists, writers, artists, musicians, filmmakers, actors, and podcasters cannot make a living, or do we want a world in which they can?

As technologists, we typically think, “If we can crawl it, it’s ours.”

What I can tell you as an artist is, if I make it, it’s mine, and you need my permission because it’s my art. I think the industry will do better if they respect artists, because now the New York Times can hire more fact-checkers.

But can I just ask you a question?

Yeah, go ahead. Sure.

Why do you have to connect the 2 as immutable things? Why can’t somebody make something—let’s just say it’s a song—but that song can now be made by multiple AI models? If they make the song, there’s a reasonable claim that, even if they don’t have the copyright, more people will want them to perform the song than some random AI. Can’t you make a living without having the copyright?

That is the choice of the artist. Some artists are very well known for not wanting their art to exist in some mediums.

As a perfect example, the Rolling Stones for a long time thought they would be sellouts if they had their music used in commercials. And when they did “Start Me Up” with Windows, that was a really big concession from them. That’s up to the artist to decide.

You make a valid claim: Yeah, you go on tour and make more money. But that’s the artist’s decision, not the technologists’ or the people stealing their content. And by the way, $20 million a year is 100 highly paid, $200,000-a-year journalists and fact-checkers at the New York Times. They’re going to get 10 of those deals, and it’s going to create a golden age of journalism and content. We should be happy.

I told you this example, Jason, but at Beast, we did a licensing deal of our content to allow OpenAI to learn—

Yeah.

—to run training runs on our videos. And at the board, the thing that we kept talking about was that I was really concerned. Let’s just do a couple-year deal, max.

And the reason is, we have no idea what this looks like in—

—5 or 10 years. And there’s just as much chance, to your point, that we get it wrong as right. Now, that was about 6 months ago, and the intuition that I had back then was maybe we should keep the deal term as short as possible. But now, when I see how important AI is in the global landscape and what China is doing, I think, on the margins, that this idea that these copyrights will mean something—in my mind, I am underwriting the value of these things going to zero, and I’m asking myself instead, for my businesses, how are we actually building a real defensible moat and not a piece of paper that we can use to sue somebody?

Okay, Friedberg, you want the last word here? We’ve got to move on to some other topics.

I just want to be the last word. I just want to be clear that nobody is losing their copyright. Copyright is the right not to have your work copied. And if an AI model produces outputs that copy or plagiarize your work, then that’s a violation of the law.

I think the president specifically said that we’re not allowing copying or plagiarizing. The question is whether AI models are allowed to do math on the internet.

Pattern recognition.

Basically, that’s what it is. And, J-Cal, I think you’re conflating the two. I don’t want to be interrupted. I just want to say this.

I understand the distinction.

I think that this idea that I can’t, for example, go to the library, rent a book, read it, and then learn some of the good techniques on how to write a good book should be restricted to humans in this AI context. This is exactly what they’re doing. They’re identifying patterns, and then they’re building predictive algorithms that allow them to output stuff that starts to fit within different kinds of variable settings.

Do you guys think it’s possible that, if you allocated enough compute to the problem, you could write Michael Crichton’s “Jurassic Park” de novo without ever having read it?

Yeah, me too. I think—

I don’t know what that would mean. Like—

Well, this is my point. I know who Michael Crichton is, and I know what “Jurassic Park” is. I don’t know what it means—

This issue.

I don’t know what it means to say, “Can AI write that?” But you guys remember the Ed Sheeran lawsuit. Do you remember the lawsuit?

I did. But let me just make one point here, because you’re saying I don’t understand it. I spent my career in it. I understand it much better than you do, and I understand it from lawsuits and being in the weeds on it. I understand it from first principles, which you do not.

I will say this: What we’re talking about here is the definition of a derivative work, and the output matters. So if you were to take my knowledge and then create a derivative work from it, you used a percentage of my work, and that’s where this will get into the nuance: What percentage of the original work is used in the derivative work, and under what context—a commercial context or a noncommercial one? This is clearly a commercial one.

If OpenAI were a nonprofit right now, we’d be having a distinctly different discussion, because you wouldn’t be competing with me as the copyright holder to use this new medium and create the derivative works. And it has to change substantially. So if it’s a CliffsNotes—

When China has the only models that are able to meet your stringent definitions of copyright—

Well, no. Here’s the thing: I think the China fear—the China fear, I’ll be totally honest here—just because China steals IP does not mean you get to steal from Americans.

In America, we have rules. And when you go to China—and, by the way, we spent the last 30 years—the major issue with China is not Taiwan. It has been the technology industry itself. Let me finish. The technology industry itself has leaned on our government for 30 or 40 years, including Microsoft and Google, to make sure our trade secrets are not stolen, our IP is not stolen, and our movies are not stolen. That is the key issue with China.

So just because China’s a thief does not mean American companies get to—

Have you seen the latest batch of Chinese open-source models or open models?

They steal everything. Does that mean you should be able to steal Windows? Should you be able to steal Jason? We don’t think it’s stealing.

Elon has said this pretty clearly, but Grok 5, and for sure Grok 6, will not use Common Crawl. It will not use the internet, okay? It’ll just be an enormous amount of synthetic data. And back to what Friedberg and I just agreed upon, if you synthetically go and try to generate all this content to learn across, you’re invariably going to produce something that’s already been created.

That’s like some sci-fi level—

I understand. That’s what’s happening now. It’s happening now.

If somebody—

What do you think happens to Grok 5 or Grok 6? Is that violating copyright? It didn’t even know that it existed—

On the output?

Yeah, that’s fine. If, on the output, it created a similar work, they would need to take it down. That would be a really interesting new space we’re going to have to contend with. If it does happen, it’s a new concept that we would have to address in a new way.

I’ll give you a science-corner example. There’s this EVO2 model that they published at the Arc Institute, which Patrick Collison, you know, is the name—

There’s this EVO2 model. They just ingested all the DNA data they could find in the world—trillions and trillions of base pairs of data that they ingested. Then they looked at patterns in DNA, and that’s it. They had no context for what the DNA represented. They had no context for the concept of genes, none of the structured understanding of what that DNA does or what it is.

And you know what it did? It fed in a BRCA gene variant, and the thing output a warning saying, “I think that this is a pathogenic variant of DNA.” This is the breast cancer allele, and it didn’t have any knowledge. It wasn’t trained on that at all. It had no knowledge that there are pathogenic variants for cancer, and it identified that this was a genetic variant that can cause some sort of pathogenic outcome in the organism.

That’s a great example where there’s a lack of understanding at the human level of what really drives some of the patterns in nature, the patterns in society, and the patterns in behavior—kind of emergent phenomena, perhaps—that these AI models are starting to identify. I think, to Chamath’s point, we may end up seeing this in things like entertainment as well.

All right, this has been an amazing debate. We got to move on. And you know what? We're going to have more amazing debates September 7th through 9th in Los Angeles at the All-In Summit. The lineup is stacked. Alibaba's co-founder Joe Tsai, Thoma Bravo co-founder, ARK Invest's Cathie Wood, Uber CEO Dara, Sequoia's Roelof Botha, YouTuber Cleo Abram, and many, many more coming. You get the last word here. Go—

I was just highlighting this tweet that I saw talking about Chinese open-weight models, basically open-source models. Basically, all the leading American models are closed source, and all the leading Chinese models are open source. This is kind of the way things played out.

It’s a pretty good technique for catching up, because then you’ve got the larger open-source developer community helping you out. The point is just that these open-source models are catching up pretty fast. We’re ahead in many other aspects. Our chips are a lot better, our data centers are better, and so on. I’d say our closed-source models are better. But they have this one area of open-source models.

Again, if you hamstring our AI models’ access to data by creating a whole bunch of new requirements for contract negotiations, we could really lose the AI race. This is a really big deal. It’s not a made-up concern. I don’t know why you think it’s made up.

I never said that it was made up. I think it’s an opportunity for America to actually have a distinct advantage, which is that $20 million from Amazon alone is 1% of the New York Times’ revenue. That’s going to go directly to the bottom line. It’s going to allow them to hire more journalists.

Then that protected site will be giving something in real time. These language models are going to have to hack and steal that real-time data. That is going to be a distinct advantage for Gemini, OpenAI, Amazon, whoever chooses to do it. And we can create—

You have these nostalgic, quasi-romantic notions about journalism and the need to save the New York Times.

It’s also art. I mean, you can say all the derogatory things you want about me personally, Sacks. That argument doesn’t work. No, no. You just said I have this whole nostalgia thing, whatever.

Yeah, you do. You’re nostalgic for journalism as it used to exist.

When I know I’ve beaten you in the debate is when you make it personal like that. It’s not personal. I’m not being nostalgic. I’m trying to create a sustainable advantage for America. And you are our public servant, and you’re learning AI. You will take my feedback.

You will take my feedback.

We’re going to ignore your feedback. Take your feedback. Public service. Throw it in the trash. No, you take it, and I will be showing up at the White House for my tour.

You have this crazy idea that we’re going to win the AI race by tying one hand behind our back so that you can subsidize journalists and subsidize movies. You’ll get more content. You said before you want more training data—pay for it. Pay for more training data. You’re the czar. Take it back to POTUS. All right, let’s keep moving here.

We have to keep moving. This is a great debate. It’s not going to stop, folks.

It’s just you yelling, saying things that don’t make sense.

Okay, you can say that.

You only have like 3 topics going on. You can personally attack me.

You know what it is? It’s like, we’ve got to let in more immigrants. Number 1. Number 2, high-skilled immigrants.

AI is going to put everyone out of work. By the way, no sense of perceived contradiction between those 2 things. Number 3, we need to subsidize here. The audience says the same thing. When the 3 of you guys attack me, Jason, when the audience gangs up on me like this, when the 3 of you gang up on me and personally attack me, the audience comes up to me and they say, “Wow, you really nailed and beat him.”

Have you done that today?

No, not yet. Not yet. But a little bit of the baiting.

Yeah, that’s true. Let him eat. He’s emaciated. He’s 11% body fat. Let him cook.

We should talk about this AI privacy issue that Sam Altman mentioned.

All right, that’s a great segue, because I saw that as well, David Sacks. As our civil servant working on AI, this is something where you can have an additional contribution. There’s more work we can give you.

AI user privacy is becoming an issue because friend of the pod Sam Altman says there is no legal confidentiality when using his product, ChatGPT. Here’s a 30-second clip—again, friend of the pod, Sam Altman, on the pod.

Sam Altman

People talk about the most personal things in their lives to ChatGPT. Young people especially use it as a therapist, a life coach: “I’m having these relationship problems. What should I do?” Right now, if you talk to a therapist, a lawyer, or a doctor about those problems, there’s legal privilege. We haven’t figured that out yet for when you talk to ChatGPT.

So if you go talk to ChatGPT about your most sensitive stuff and then there’s a lawsuit or whatever, we could be required to produce that. I think that’s very screwed up. I think we should have the same concept of privacy for your conversations with AI that we do with a therapist or whatever.

Okay, Sacks, this is bringing up something super important. What’s your take on it?

I think this is an interesting topic because, like copyright, this is an area where we have existing law, but it does make you rethink whether those laws are truly applicable or make as much sense in this new world. The existing example is search history. The government can get a copy of your search history. They can subpoena it.

Yeah. Every true-crime story starts with a person’s search for, “How do I kill my husband slowly with poison?” And then they—

Right, exactly. The point is, though, that I think Sam is right about the legal treatment right now, which is that your chat history isn’t any different from your search history in the eyes of the law. But it is much more personal. It’s much more interactive than your search history.

You are using it, like you said, as your doctor. You could use it as your therapist. You could use it as your lawyer. And so the ability for the federal government to be intrusive is so much greater than with your search history. I don’t know what the right policy should be yet, but I will say it does make me uncomfortable.

Yeah, there’s a market.

Can I make a recommendation to my AI czar?

Yes, please.

Why don’t we let AI models get bar-certified and medically certified? If the AI models turn out to be more accurate, more thoughtful, more responsive, more reasonable—whatever metric we’re using—and they pass the same criteria one would need to pass to qualify for the bar or to qualify for medical certification, why don’t we do that for AI?

If that happens, then the same privilege applies to AI as it does to the individual human who does it. And if you extrapolate from where that takes us, if we’re suddenly giving AI the same sort of privileged rights that we give to privileged humans, where’s that going to take us, ultimately, with respect to the overall rights for AI?

Well, and they have responsibility. Hold on a second. I just want to point out here once again: you have a mind-blowing concept here. I’ve never heard anybody vocalize that. Could they actually be certified in that knowledge? And if they pass the test, it makes sense they would. But then you also get responsibility.

So, with great power comes great responsibility. I will tell you this.

You can turn this stuff off. But this is an opportunity. I’m going to send a note to you. And it sounds crazy today, but I guarantee if you put it on Polymarket, there will be a date when this happens. Polymarket—shout-out to Shayne. Let’s get that up there.

This is an opportunity to create the Signal equivalent of an LLM. All of your chats should be encrypted. All of it should be encrypted by default. Encrypt it by default on Grok. Make it so that Grok can’t even see it. They don’t have it.

So when you try to subpoena it, you can do what Tim Cook does, which is say, “I don’t have it.” If you want to try to backdoor it, you can. That’s a market opportunity. I can tell you, I only use the Brave browser and Brave Search for this reason. I don’t want my search history saved somewhere or whatever. You can take control of this as an individual, but the defaults matter, and you have to do the work. It’s a great market opportunity.

Chamath, I don’t even want to know what you’re talking to ChatGPT about. What’s in your ChatGPT logs? What’s in there, Chamath? How to extend? How to get the extra centimeter? What’s in there?

I keep asking it to find me a moderator.

Oh, great. I keep asking it to find me a participant who isn’t a douche. My God, you are so deep in your villain era, and you’re leaning into it. I’m so here for it, Chamath. I love your villain.

You know why?

Why are you going into your villain?

I am so risk-on right now.

It’s liberating, actually. It’s amazing. Is there any blowback to how outlandish you’ve become this year? Any blowback at all? Has it had any negative consequence on business or hiring or anything?

No. But outlandish? How have I been outlandish?

You’re just filter-off, and I think it’s great. I think you’re over 2 windows back. It’s absolutely fantastic what we’re seeing here.

I asked ChatGPT about my future and my IQ. It’s very interesting when you ask ChatGPT to analyze you.

I suggest everyone do it.

Well, actually, yeah, when you just ask ChatGPT, “What do you know about me?”

And it’s scary how much it already does.

It’s scary. There’s this great personality test. You can put this personality test into Grok, and this guy made this prompt that tells you all your personality traits based on your Twitter and X history. It is wild how accurate it is. What does it say about you, J? I’m actually curious.

It says the same thing about all of us. We’re all overworked narcissists, ENTJs.

Yeah, your chat history. But I like your mind-blowing concept there, by the way, of them becoming certified in some way.

Fresh economic news: it’s time for the administration to take its victory lap. GDP growth was 50% higher than expectations in Q2, as the Fed held rates at 4.25%. In Q1, GDP declined 50 basis points. That’s probably due to the imports; people were stockpiling goods.

That’s the most pointless chart ever.

Okay. Yeah, it is. I agree. It’s a little bit distorted. I wanted to have both as bar charts.

Totally on drugs. Just say it. It’s okay.

What drugs are you on? I’m not. I had coffee, and now I’m out. We’re all friends. You can tell us. Is it really just out? All right, that’s it. I’m taking it out. Oh, my God. I took it out. And now let’s get back to the news here.

Okay. The Fed kept rates unchanged for the 5th straight meeting. This time, 2 out of 11 Fed governors dissented from Powell’s decision. The 2 dissenters were both Republicans nominated by Trump, so it seems like the Fed is becoming a little polarized now, too.

It was the first time in 32 years that more than 1 governor dissented. Even 1 person dissenting is rare. Here’s a 25-second clip of Powell explaining how GDP factored into the decision. Nick, please play the clip.

Jerome Powell

Recent indicators suggest that growth of economic activity has moderated. GDP rose at a 1.2% pace in the first half of this year, down from 2.5% last year. Although the increase in the 2nd quarter was stronger, at 3%, focusing on the first half of the year helps smooth through the volatility in the quarterly figures related to the unusual swings in net exports.

The PCE index for June dropped on Thursday. PCE is the Fed’s preferred gauge of inflation over CPI. PCE rose 30 bps in June, in line with estimates. If you remember, we talked about this in a previous episode: CPI rose a bit 13% or 30 bips from May to June.

So we’re not close to the 2% target, and that’s what the Fed keeps saying: We’re not there yet. The economy is al fuego, Sacks. You noted—I don’t know if you noticed this—but people are talking about the Q2 GDP print, which was amazing. You were talking about it a bunch on the socials. Powell keeps referencing the first half, so he’s trying to blend those 2 together, I think, because of the tariff differences, or maybe to smooth it out, as he said. What’s your take on this? GDP boomed at 3%, which is pretty great.

The problem Powell has is that he’s trying to smooth it because it allows him to justify his political decision. But the reason why you have to segregate Q1 and Q2 is that Q1 was before tariffs and Q2 was after tariffs.

If you look at the run rate from Q2, what you’re probably going to see in Q3 and beyond is more similar to Q2, which is to say a large surplus, good GDP expansion, and moderating inflation. So why does the Fed not cut? At this point, not cutting is the only thing that you can do to slow the Trump administration down going into the midterms, if you wanted to politicize the job.

If, on the other hand, you just take the data as is and ignore Q1 because it was pre-tariff, then start to look at Q2 and project forward, if you inject a 100-basis-point cut into the economy, this thing is going to go gangbusters and Trump is going to look like an economic genius going into 2026. So, again, in the absence of politics, you cut.

Okay. Sacks, what’s the take from inside the administration and around it? I know you’re not speaking for the president on this issue, but you’re in the administration, so I’m assuming you’re—

Look, I’m not speaking for anyone, but obviously the 3% number is way ahead of expectations. It’s a fantastic number. It just feels like everything’s humming on all cylinders here. One thing you didn’t mention, but I think is relevant, is the new trade deal with the EU.

We’re about to get to that, by the way. That’s the next story.

Oh, okay. Well, I would include that because I think it was a deal that just got announced where the EU is going to open its markets to U.S. products. There will be no tariff on U.S. products, but they will pay a 15% tariff coming into the U.S. They’re going to be investing $600 billion in the U.S., and they’re going to be buying $750 billion of U.S. energy.

There was also some very large number—I guess they didn’t specify a number—on defense products, basically American military products, hundreds of billions of dollars. That’s the follow-up to their commitment to raise their contribution to NATO to 5% of GDP, up from, I guess, around 2% before.

Okay, include it. Yeah.

This is a huge deal for the United States. I think it’s a huge win for the Trump administration, and the deal is so good that what I’m seeing from European sources on X, European publications, and just commenters is that they were outraged. They felt like they got taken to the cleaners.

You see a lot of that on the European side on X. A lot of the European leaders are saying that Ursula chickened out. All those stupid TACO memes are going away now because people are realizing that Trump’s willingness to raise tariffs on these countries as a threat to renegotiate better trade deals is working. It’s working extraordinarily well.

Just this EU deal, 1 way to think about it is that you add it all up, and it’s about $2 trillion. It’s effectively $2 trillion of stimulus into the U.S., but without money printing.

Yeah, over the next 3 years, so it’s not inflationary. It’s not insignificant. Friedberg, your thoughts on the Fed and the GDP print? Then maybe you could get into the granular details of that print. If you pull up the schedule of data—

This is the National Income and Product Accounts data from the Bureau of Economic Analysis. This is where the inflation print comes from. I think there are 2 lines worth taking significant note of.

The first is the furnishings and durable household equipment line. In June, the cost for furnishings and household stuff jumped 1.3% month over month on an annualized basis. That’s almost 15% year over year if it were to continue at that level.

The second is recreational goods and vehicles, which jumped 9% month over month. Neither of those categories has jumped that much in recent history. Part of the argument being made is that what we’re seeing in these jumps is actually some of the first effects of the tariffs and the cost of goods being imported, because these are largely imports, having an adverse effect on the consumer.

I think this is a wait-and-see moment on some of these categories that are predicted to have a tariff price effect starting to show through. This is where a lot of folks are keeping a close eye, and it provides a little bit of support for the economists who are saying we should keep rates steady. If we are seeing a significant inflationary effect here, it’s worth noting that we need to be thoughtful about rate policy.

I think this is a really good point. If you look at this debate, which is obviously highly political, we’re at 2.567% inflation. Spending is increasing. Obviously, the stock market is at an all-time high. Unemployment is trending down again, so we’re at around 4.1%. People are YOLOing into crypto, and they’re doing sports betting. Bitcoin is at an all-time high.

I think the Fed is now in a position where cutting rates seems like putting kerosene on the fire. If Trump had tanked the economy in Q2, he probably would have gotten the rate cuts. But now, as you’re saying, Dave, I don’t think it’s reasonable. The reasons not to cut are building because the economy is on fire.

Maybe the shock and bore approach to tariffs, which is now becoming a playbook, is working. I had a nice talk with Lutnick about this, who I love, by the way. He really described to me how they’re doing this. The shock-and-awe playbook is basically that Trump says something completely outrageous, shocking everybody. The media loses its mind, and business leaders lose their minds.

Lutnick told me that what he does is set the table and propose something reasonable. Now I have direct contact with all of the administration. Sacks, thank you for that. He described it this way: Trump comes in, sees all the stuff, and then starts making these micro-tweaks.

It’s on the finish line. It’s in the red zone, at the 5-yard line. Trump comes in, sticks it to them again with 3 or 4 extra asks, and then they wrap it up. This is becoming really effective. It was chaotic at first and seemed nonsensical, but they’ve put the Fed in a really bad position because they’ve never seen this before.

So now they’re going to be in this defensive position of, what if we cut and the market rips? To your point, Chamath, you just said the market will rip the second they cut that. The cynical view is that the market rips as we go into the midterms, which is the same claim the Republicans made about the cuts Biden did in September, right before the election.

There is some level of politics and gamesmanship going on here. But you have to hand it to the Trump administration for what they’re doing with this sort of 2.0 playbook. If this was Sacks’ premeditated plan and we all just didn’t understand it, fine.

The outcome here is that this administration has to live or die by the results of the $600 billion from the EU and the $550 billion in investment from Japan. You put those 2 together, and I asked Lutnick at the event, “Is that going into the sovereign wealth fund? And how does that get spent?”

He said it’s at the discretion of the president, and he’s advising him to spend it on putting more nukes in. That’s fascinating. We have $1 trillion now that we can put into nuclear power plants and these small modular reactors. That’s what Lutnick said he wanted to spend it on, and he’s going to advise the president to spend it on that.

Now we’ve got them investing in our country. It’s absolutely brilliant. If it works out—we’ll see if it works out.

April 2nd was Liberation Day, and the media went crazy. They were predicting a Black Monday and a market crash. They basically tried to spook the markets and create fear. They said we were going to go into a recession or depression.

Now look at where we are. It’s just a few months later, and all the markets are at all-time highs.

Trump has extracted trillions of dollars in these trade deals that people know.

Premeditated. Tell us the truth.

Premeditated.

Hold on. President Trump has extracted trillions of dollars from other countries using powers that other presidents didn't even know they had. One hundred percent, one hundred percent. Was it premeditated? Because it was chaotic. The market made those moves because of the media, by the way. Those moves happened because people were scared. And we just had a 3% GDP growth print.

What I think happened is that President Trump saw an opportunity here that other people ignored. It's like when a new CEO comes into a company that's been mismanaged for a decade, but it's got wonderful assets on its balance sheet and a market position that's still very strong but has been underutilized.

He came in and understood that the United States had tremendous leverage in all these trade negotiations. Actually, they weren't even trade negotiations then; they were trade relationships. He was able to essentially renegotiate all of them. And look at the results. I mean, they're just staggering.

Everyone said, "Oh, Trump's going to chicken out. He's not going to hang tough." It's all these other countries that have folded like lawn chairs. I mean, they have all capitulated.

Yeah, yeah, they followed.

It's really remarkable.

But you're not answering my question. Was this premeditated? Give us some insight here.

I don't know what this is. What are you talking about?

When they came out and said, "Oh, 100% tariffs, 200% tariffs," the market wasn't making that reaction based on the media. They were making it based on what Trump was saying. So, was it premeditated this shock and bore shock and reasonable negotiating strategy or do you not know? You're not privy to—

Look, I'm not speaking as an insider here, but we said at the time that all of that was happening, and Larry Summers was on the pod preaching doom, that all of that was an opening bid. It was all a start to a negotiation, and we had to see where it ended up. The administration still had to stick the landing.

Okay, but I've got to say, based on the European Union, Japan, and South Korea, this is looking really good right now.

Well, listen, it's the top 5 that are like 90% of the negotiation. As Trump said, there was another little note he made in the keynote when he drifted into the different things he wanted to talk about, where he said, "I don't even need to know about the bottom countries. I've never even heard the names of some of these countries."

He's just got to nail the top 5, the top 10, and we're done. This administration has to stick the landing as well, because these are handshake deals right now. They have to be inked. They have to be approved. So there's a lot more work left to be done.

I said as well, there's one other piece of it. We talked about the fact that Europe has 0% tariffs on American products, but even after this deal, European products coming into the U.S. will have a 15% tariff. We're not including the $600 billion of European investment in the U.S. We're not including the $750 billion of sales of American energy to Europe.

Just talk about the tariff—the 15%. What we're seeing now across the board is generating about $300 billion a year of additional tariff revenue that goes to help balance the budget.

Yeah.

So $300 billion a year over 10 years is $3 trillion. That is a big number.

It's incredible. Yeah. It's got to—

I don't know if that completely satisfies Friedberg, but that's a big help.

Friedberg, do you think there's a chance that inflation is going to tick up because of all this? This is a lot of money being pushed into the system again. So could we see a 3-handle on inflation in the next 6 months, or what's the probability of that in your mind?

That's the big concern everybody has. I don't know. I don't know. I think the big question, if you look at each of these categories, is how much margin is the seller making?

If they're making a 30% margin and we charge a 15% tariff, does their margin go down to 15%? Or do they take their margin down to 20% and raise the price by 5%? What's the right balance?

What will happen is that the market, with this effective tariff—which is a sort of tax on the system, a tax on the market—will find its new equilibrium, where the buyers are willing to pay X and the sellers are willing to sell at Y. I think every market is going to be a bit different.

In some of these categories, we will see significant inflation where the seller has a very thin margin. In some of the categories where there's a monopoly and they have a big margin, they're going to eat it because they don't want to have competition, and they don't want to see pricing competition emerge. So I think we'll see it vary by category, and we'll see how it goes.

Great job, everybody. Great job to everybody. It's a class.

Great job, everyone. Even JCAL. Even J. Even Jason.

Actually, I want to thank Friedberg, because Friedberg did most of the work to organize the AI summit.

Let's give him a big shout-out. There's me.

Great job.

Guys, can we just make a note here? One of us can run for president as a Manchurian candidate in 8 years. Look at me and the president. I put on the red tie out of respect. I put my blue suit on out of respect for the president. Does it not look like I'm running for president? Jason.com.

That photo could be like that famous photo of Bill Clinton meeting JFK. That could be the thing that—

I'm in, like, that famous image—

—that propels you to the presidency.

I'm in. Thank you for giving me that and for putting me in touch with each member of the administration directly. Thank you for that. And we had a wonderful tour of the White House the next day. What a wonderful tour. Some of us had a tour at the White House the next day.

Did you?

No, I was taking the pictures. That was my joke, because all of you guys were giving each other a tour. We could have gotten you a tour.

I mean, listen, I love J.

Did you ask for a tour?

I'm not the kind of guy to ask. I'm the guy—

Some of us have actual meetings to do, bro.

It's all good. I've got a lot going on. I've got a lot to announce and a lot happening in the coming weeks. But, Sacks, do take us behind the scenes here. I thought it was hilarious, so I don't mind getting trolled by the president. It was great. But how did that go about behind the scenes? How did he nail that joke?

Don't tell him. Leave it. Leave it.

What did you do? Because that looked like it was workshopped. Or is he just naturally—I mean, he's obviously naturally comedic—but did you put that in with him? Did you have to clear that with him?

Well, they asked me for the names of my co-hosts so they could do shout-outs. So I gave him the list.

Oh, no. And I said—I put in even JCAL.

But I mean, he went for it.

No, he got it. We went through it.

He got the joke. Okay.

He got the joke. We went through it.

He got the laugh. He got it. He heard the laugh, and he doubled down.

I thought it'd be funny. But no, we went through everyone's names beforehand.

I mean, talk about EQ. The guy's EQ is off the charts. His timing is great. I suggested the name JCAL, and he's like, "No, no, give me his full name." He thought it was more courteous.

He's actually a very courteous man. He wanted to use your full name, not just your nickname.

I think what he probably realized was how much it meant to my parents, who were just over the moon. So thank you for that. It meant a lot to my dad, who's—

That's lovely.

Yeah, he's been struggling a bit, and it really—it gets me a little choked up here, but my dad's been struggling a bit. I got to see him in Brooklyn after that, and we were on a tech stream. It meant a lot, because for a kid from Brooklyn to get a shout-out from the president of the United States is—

You made it. Your father should be really proud of you.

Thanks, man. Appreciate it. I appreciate it, boys.

For your Sultan of Science, the amazing David Friedberg, who put that event together in 10 days and then jumped right in—he's got to run Hallow at the same time—I just want to give our MVP of the week—

We should give a shout-out to the Hill & Valley guys for partnering with us. Jacob Helberg did a great job.

Love Jacob. I love Jacob.

And Delian and Chris. Thank you, guys. They were our partners on the event.

Hill & Valley did a great job. Yeah, I love those guys.

I'm giving the MVP of the week, of the besties, to you, David Friedberg. You put a lot of work into this, and we appreciate it. You're running Hallow and then went right into working on the All-In Summit, which we'll be at in a couple of weeks.

Chamath, thank you for buttoning up. We're getting a little criticism from the HR department about the buttons, and so we've now renegotiated that.

I'm going to unbutton 3 buttons now and walk around Forte.

Perfect. And Sacks, I will see you at the White House. JD and I will be in the commissary. So, we'll invite you to lunch with us and JD.

It's called the Navy mess, actually.

In the mess. Yeah. And you know what? Lutnick's joining us as well. And who's our energy guy? Chris. Chris said he wanted to jump in on that, so maybe you can join us. I'll invite you now that I am deep into the administration. Thank you for tuning in everybody. Allin.com events. The scholarship tickets are up. So, if you want to try to get one of the very few scholarship tickets, we always like our up-and-comers. Please, if you're of means, don't apply for the scholarship. You won't get it in. But if you're up and coming and you're part of the audience and you want to get one of those discounted tickets, we have a limited number of those available. All.com/events. Love you besties. Bye-bye. Love you. Bye-bye. Let your winners ride.