Trump's First Week: Inauguration Recap, Executive Actions, TikTok, Stargate + Sacks is Back!

We had a dinner where Friedberg, J-Cal, and I were invited to have dinner with the folks from another top podcast, Off the Record. There was an interesting moment that even Friedberg had to recognize. We went around the room and said, “What podcasts are you listening to? Which ones are your favorites?” We went around the room, and I’ve got to tell you: Do you know what their favorite podcast is? Bingo: the All-In Podcast.

My wife thinks I’m bullshitting all the time. I say, “This is mine, too,” and she doesn’t believe it. She thinks we’re all—

Why do you think I waited to make my worldwide podcast premiere for All-In? All the ankle-biters called, and I said, “No, I’m just going to wait. I’m waiting for the besties to call.” Can we just say it, Nick? You got to it when a podcast called, and you were like, “Nope, can’t do it. I’m waiting for the real deal.”

We don’t need JV. Let your winners ride, Rain Man David. Instead, we open-sourced it to the fans, and they’ve just gone crazy with it. Love you, Queen of All.

All right, everybody, welcome back to the number-one business, finance, technology, and science podcast in the world. This is the place where we call balls and strikes, where we tell the truth, where there is zero censorship. Every week, we tell you about the truth behind the most important stories. We point out who the heroes are, who the villains are, who are the delightful ones, and who are the disruptors.

With me again on this week’s program is the cackling chairman dictator, my guy Chamath Palihapitiya. How are you doing? How was your victory lap?

Really, really, really fun. You look like you were in your afterglow, and I have to say, it was looking pretty good.

We have some pictures. Don’t worry, we’re going to go through them behind the scenes. Armani came through for the amore—the Armani Amore collection is coming this fall. Armani Couture is firing on all cylinders. The other one, which you’ll see, is a great picture of this incredible jacket from Louisa Beia. Two shout-outs: Armani and Louisa Beia.

I was just thinking that Armani really has been making it happen.

Now that we’ve talked about that, we’ll talk about my wardrobe when the time comes.

Just get through the rest of the introduction.

All right, all right. Of course, here is your sultan of science, David Friedberg.

Move along, J.

Let’s go, let’s go. He had a wonderful—

He’s very bitter today. He’s got a lot of anger.

He is the barista of bitterness today, serving up bitterness all day long. He had a great time at the—

Maybe I’m a little upset about the Lex Fridman triangle we’re in right now. I think that could be part of it.

We also have a special guest here, Thomas Laffont. He’s the co-founder of Coatue Management. They have $50 billion under management, public and private. Before that, he worked at Creative Artists Agency for 7 years after college.

He does have a college degree. He signed Captain America, which is Friedberg’s favorite Avenger.

Chris Evans.

I have a client, by the way.

Yes, and you’re still in touch with him?

I am not.

You’re not?

I watched with wonder at what he’s achieved after I left being his agent. It’s been an incredible career.

Now, behind you, we see this incredible view. You’re missing the one part, which is that Thomas was one of the highest-rated speakers we had at the All-In Summit.

That’s true. I did really great last year. I crushed it, crushed it, crushed it. Of course, I think I’m in the seventh-best position and am hoping, potentially after the podcast, to move up by 1.

All right, so there you go. Let’s see what happens here. What’s this view behind you?

Los Angeles.

That definitely puts you ahead of J-Cal, Thomas, so you’re up a spot.

My God, he really is serving up the bitterness today.

He’s so mad.

You know why he’s mad?

I’ll tell you why. Jason has been complaining all week in the group chat that he has influenza A. He’s sending pictures of himself getting IVs. Friedberg, the conscientious guy and co-founder of the podcast that he is, says, “I will moderate.” He invests his time. He’s got a lot of things going on: 3 kids plus 1 on the way, and he’s running a company. He prepares, and then you wake up after a dose of Tamiflu and decide, “No, I’m ready to moderate.”

Of course, you can’t just do that because it’s not nice.

Can I say something? I said, “Hey, Jason, how are you feeling?” You said, “Horrible. I have the flu. I’m getting an IV tomorrow morning.” To me, that was, “There’s a 50% chance he misses the show.” So I reached out to Friedberg and said, “Hey, Jason’s really sick with the flu. Maybe you should prepare yourself to moderate tomorrow.”

Now, when did I find out about this? I found out about this this morning. I found out about this about an hour ago.

I think that was a smart decision, but you, sir, are a small, bitter man. What you did was try to blow up Friedberg’s preparation.

How are you feeling?

I feel 80% of a normal J-Cal, which is 110% of any other moderator.

Get yourself Andrew Ross Sorkin.

I’m sorry, he’s genuflecting at Davos to a bunch of mids.

We’re going to get Lex on.

We’re going to get Lex. That would be great.

Hello. I am Lex Fridman. Welcome to the number-one podcast in the world. Today on the program, we have Thomas Laffont, investor in such great companies as—

All right, listen. Enough shenanigans. We had so much fun tearing D.C. apart. Let’s start with this. Thomas, did you watch any of it on television? Did you think about going?

There were so many technology people in Washington this weekend. I saw the livestream and talked to Sacks last night to get the download. My takeaway, just from watching the executive orders, was democracy in action. Democracy self-corrects; dictatorships double down.

To watch, in real time, a country of our scale and a democracy decide to make a change—I think that’s why we’ve been great for 250-plus years, and why we’ll remain great. It was fun to watch. I’m really curious to see what your takeaway was.

Let’s go behind the scenes first. It’s hard to see, but this was me dressed up for Sacks’s ball. I really brought the heat. This is by Luca Rubinacci, the tailor. It’s hard to see because it’s blue, but this is a very refined green and purple glen plaid—subtle, and hard to bring out.

I will say this was the most popular shirt over the entire weekend. This is just a simple Brioni tuxedo shirt. There were 200 versions of it. I felt bad when I saw that because normally I don’t like to wear things that other people are wearing.

What did Michael Saylor think of this when you saw him at the—

I met Michael’s CFO, a wonderfully nice guy.

So this is us at the crypto ball. I went for a white tuxedo on the first night. I’ve never had 3 tuxedos before, but David looked really great.

I really loved Sacks’s tuxedo. It was more crypto-themed. It looked like the opening scene from The Matrix. There was this cool print.

You can see my green in there a little bit. Thomas, you’re in this one.

Yes, I got it.

There’s Skye and New Money Sunny Madra, the president of Groq, and Supermadra from X.com. Here are the besties.

I’ll be honest, guys: even with all your bitterness, Friedberg, you felt the love at this moment when we were all back together, did you not?

Minutes before this, when the 3 of us took a photo, it was great.

Great. Sacks, let Friedberg moderate if you want. If you’re that bitter, he’s doing a great job.

Keep going. We’re 18 minutes into this and we’re still doing fashion.

Vinod Khosla, Diego Berdakin, the co-founder of CloudKitchens—so handsome. This was the crypto ball.

This was the crypto ball. This was Sacks’s event. It was incredible. There were a large number of people there. I was shocked at the number of people at the crypto ball.

It was huge. Here’s a little poker aftermath behind the scenes with Pinky and Travis Kalanick, the founder of Uber.

Look at who’s got the biggest stack. It’s the founder of Uber next to the third investor in Uber, and the founder of CloudKitchens next to the investor in CloudKitchens.

This is at Peter Thiel’s house. This is Katie Haun and Kyle Samani, my wife and Kyle’s wife. Amore, amore. Peter Thiel had this person going around taking pictures.

This is not good lighting for you.

There’s better. This was great, also at Peter Thiel’s house.

I have to say, putting aside that Jason likes being a Republican, here’s a dinner that we had. There’s Lex Fridman, Julius Genachowski, the former chairman of the FCC, and—

A little-known fact: Julius was a former roommate of Barack Obama.

Following this dinner, all the streets were blocked off and there was no way to get around. It was about 10 degrees Fahrenheit, so we walked out.

It was terrible. You were going to die after 2 seconds out there.

We started walking to try to get our car, and after about half a mile we said, “There’s no way out of this downtown area.” Vinod was shaking. He said, “I’m going to die. I’m going to die.” He turned around, ran back to the restaurant, jumped in, and basically huddled down there.

After about 10 minutes, there was still no way out. It was like being in a military zone. He said, “I’m going to go for it.” He ran out, jumped on a scooter, and disappeared into the distance. I swore I would never see Vinod again.

I think it’s very smart that Lex wears the same outfit—the black suit with the white shirt and black tie. I really like the consistency of it.

It looks really good.

He was getting stopped every 2 minutes. It was unbelievable. Here’s the jacket. This is Louisa Beia. Here’s the box from Armani. Thank you, unboxing.

Wow.

Here’s a huge shout-out to 2 guys I had really wanted to meet in person. I finally got the chance at Zuck’s party. This is Brendan Carr, the incoming chairman of the FCC. Actually, this picture was taken right after he had been confirmed, so this was his first day as chairman.

This is Jared Isaacman, who is now going to run NASA. He’s the founder and CEO of Shift4 Payments, but he was also the first civilian to conduct a spacewalk. Both of these guys are pretty epic. You should follow them on Twitter.

Brendan was really nice. I met him at Peter’s event, and he told me he’s a big fan of the podcast. He listens to the show.

Amazing. If you want to see some really amazing tweets, Brendan had some real bangers. He’s a very smart guy.

I’ve had Brendan on the show a couple of times. He’s the guy who fought for Starlink over wasting $20,000 per home to install broadband.

This guy’s a hero. He did that when he was in office under Biden, despite all that crazy corruption and grift.

This is a picture of Mike Johnson and his wife. There’s a crazy story behind this. There was a G7 summit in Italy, and Mike and his wife were seated beside my father-in-law, Nat, at a dinner. They got to know each other. As a proud grandpa, Nat showed Mike and his wife pictures of Nat, myself, and our kids, and how we met.

When we met this weekend, Mike’s wife said, “Oh, my God, I recognize you. You’re Sergio’s son-in-law. You’re Sergio’s daughter.” It was amazing, and we had a cool conversation.

This is Bill Pulte, who has been nominated to run the FHFA, which, for people who know, will be in control of the conservatorship of Fannie Mae and Freddie Mac. He’s also a co-investor with me in MrBeast.

You’re an investor in MrBeast?

There you go. This is where J-Cal is at his peak. Go ahead—play it.

I don’t even know what this is.

What is that? What did you order?

I’ve given it some thought, and I’m going to order the $95 dosa.

The $95 dosa? Your tip’s included in that?

We were having such a great time tearing it up. I did order the $95 dosa.

This is us walking into the Capitol for the inauguration.

You basically have to get off the buses that were taking us in because of security, and then you walk underground to get into the Capitol.

This is Bobby Kennedy and the guy right there.

That’s C. Everett Koop.

Another great American. Make America healthy again.

Here are a couple of my picks. This is me with somebody who’s going to be in the administration. I’m not big on politics, but Tulsi Gabbard came to our party. She is incredible.

She came up to me and said, “J-Cal, I am a huge fan of the podcast. I love what you do. I love the balance you bring to the program.” She really liked new details about the podcast, so I wanted to get a picture with her.

As a joke, I said to her, “Do you want to take a selfie?” She said, “I was going to ask.” I said, “I’m preemptively asking.” We took it together. I think she’s going to be great. I like her a lot. She’s such a star.

Hello. I am Lex Fridman. Today, with love and peace in my heart, I am interviewing Idi Amin, who has eaten half of his enemies. We will share love, kindness, and a fatty liver. Enjoy.

That was us tearing it up after the YouTube party, where we hung out with Sundar and Neal from YouTube. They’ve been very good to us.

Here’s one more, I guess. One more, and golfer Bryson DeChambeau. There’s me and the president of Twitter. This is the podcast crew at the Free Press party.

Let’s talk about key takeaways. Chamath, what is your big takeaway from the weekend?

My biggest takeaway is that if you look in the past, there was very tight coordination between private industry and the public sector—the president and the people who were in charge of managing huge swaths of resources in America. Somewhere along the way, that became unfashionable and uncool, particularly under Democrats.

What I saw was a very broad-based embrace of businesspeople, because I think the president understands how important it is to make sure that economically America is firing on all cylinders. In the inauguration, you saw the CEOs of Meta, Google, Apple, Amazon, and Tesla. Founders were there. Bernard Arnault was there. Mukesh Ambani was there. Businesspeople from the entire world were there.

I think it says that America is going to turn a totally new page. We’re not going to ostracize people, and we’re not going to play favorites. Elon was there at the beginning, but all the CEOs of the major companies were there. It’s not like he excluded Zuckerberg because of their past issues, or Sam Altman.

That was the most notable one: to exclude Sam Altman.

The president doesn’t play favorites. This is Team America. It was a projection of tremendous power to the rest of the world. This is what America is about. I loved every minute of it, and I think this is exactly how the American government should be working: hand in hand with private industry to set the pace for the rest of the world.

Thomas, did you have a takeaway after the weekend?

I did. I’m curious what you guys thought, but to me what was notable was that Scott Bessent, the new Treasury secretary, was asked in one of the exchanges about the green-energy race with China. He reframed the debate incredibly well. He said, “We’re not in a green-energy race with China. We’re in an energy race with China.”

He made the point that China is adding 100 coal plants and adding nuclear. It’s adding a 137-gigawatt hydroelectric facility, which is unbelievable. It was such a good framing of the debate, because I think, to Chamath’s point, it’s about reframing the right issues.

That little piece of dialogue exemplified what we’ll see over the coming year, and it was a really important takeaway.

Friedberg, did you have a takeaway coming out of this?

There was just so much. It was impressive to be there. I echo the sentiment about seeing the strength of America represented by the entrepreneurial engine that has driven this country for 250 years and the government that’s meant to serve the people working together to take America forward. It was really inspiring.

The biggest observation for me, which was unfortunately a bit of a disappointment, is that I feel like DOGE and cost-cutting are going to be a lot harder to realize than people assume. Nearly everyone I met who works in government or is entering government had, if not disdain for DOGE, a concern that it doesn’t align with the political objectives of politicians. They need to get more stuff for their constituents in order to get reelected, and they’re not going to vote programs away. They’re not going to vote themselves out of a job, which is effectively what they would do if they cut programs.

I was watching Brooke Rollins’s confirmation hearing this morning. Mitch McConnell was given the opportunity to ask her questions. His question was, “In the last Farm Bill, I was able to get $60 million to build a research lab at the University of Kentucky, and it hasn’t started being built yet. It’s been 4 years. What are you going to do about that?”

This is literally the intention of so many of these conversations if you sit and watch them. I watched a similar conversation with a senator from Minnesota. I won’t get into all the statements, but every conversation is about, “What can I get for my constituents? How do I get more jobs and more money?”

Ultimately, that incentive is what gets people elected, and I don’t think it’s going to change. That was my biggest shock. I thought there was a moment when people understood the importance of cost-cutting, deficit reduction, and debt reduction. I was saddened to see that there’s no stop in the train.

That was my takeaway. It’s a little bit bitter. I hope I’m wrong. I really do hope DOGE is effective, and I really do hope policymakers start to recognize the importance of this. By the way, Russell Vought has a new book coming out. We’ll talk more about his book once it’s released publicly, but it speaks very clearly to the challenge the United States is facing with respect to spending.

I know I’ve talked about this, and I know we all felt like there was a big reprieve with this election, but for me it was a little bit bittersweet.

I want to double down on that first topic. One of the things some people said—and it makes sense when you think about it—is that when you look at the confirmation hearings, there are certain people for whom resource allocation is relatively modest inside America. They tend to do much better in general.

If you saw Marco Rubio’s confirmation process, it was 99 to 0. People who allocate internal spending in the United States get a lot more scrutiny because there’s all this horse-trading. Politicians want some of this money allocated to their specific areas.

It is going to be a very complicated thing. One thing that changed—I don’t know if you saw this—is that DOGE was meant to be outside the government. It wasn’t actually a government department. In the 11th hour, the executive order that established DOGE established it as a governmental agency.

We’re going to get into that in just a moment. To build on your point, Senator Sanders had a tweet today. The discussion of who should take the brunt of these cuts is starting to happen. He’s pointing out, as he should, “What’s going to happen with the VA? You’re putting a freeze on hiring. We have to make sure our veterans are taken care of. We have a shortage of doctors and nurses.”

People were responding to him, saying, “How are we going to pay for all this?” Even if it’s going to be hard to do, it’s going to take courage because you’re going to be faced with situations like this. Does anybody want to see the VA’s budget cut? These veterans coming back should get all the services in the world. These hard discussions are now happening, and they’re not taboo.

The Overton window is wide open to discuss even the budget of the VA and where the money is going to come from. This was a sacred cow, and now we’re talking about it. I think we’re going to make progress.

To Bernie Sanders’s point, I don’t know if you saw that there was a community note pointing out that Trump said none of the cuts should impact veterans. I thought that was a good use of community notes.

The one comment I’ll make about Bernie Sanders is that when Russell Vought, the nominee to run OMB, went through his confirmation yesterday, Sanders asked a similar question around Medicare and Medicaid. What was interesting is that Russell tried to answer the question and give details about where there could be waste, fraud, and abuse. Sanders didn’t want to hear it. He just wanted a sound bite.

I think these confirmation processes are performative. They’re hugely performative.

I agree. This applies to both sides. When Democrats are eventually in power, I think that when you’re given a clear mandate as president, you should be able to pick your team. The American people voted for an agenda. It’s not like he hid that agenda. The people who want to implement that agenda should be put on the field.

The point of these hearings should be to determine whether somebody could be compromised or whether there’s something really serious in their background. Otherwise, let people build their team, and then they have to own the results.

Tell us about your feelings and takeaways from the weekend.

Well, it’s obviously mixed. I was a never-Trumper, and now I’m rooting for him wholeheartedly to do great work.

You kind of like him. Come on, you do kind of like him. Anybody who met or interacted with Trump—

Don’t go to your diatribe. Just say it. Just say you like him.

There are aspects of the platform I like. I like probably 80% of the platform. I’ll tell you today that there’s 20% I don’t like. I’m always going to call balls and strikes.

But you kind of like him. Just say you kind of like him.

I like 80% of his platform. As a person, there are things I really don’t like, and I’ll talk about them today. Let’s put aside the personal thing. What matters is what he does for the American people.

I know you can’t say it yet, but deep down inside you like him.

I had a lot of fun. Republicans are fun. I will say that. The fashion was off the hook. America was beautiful again. No more jogging pants and white sneakers. Get rid of it. Burn it to the ground.

My takeaway, as somebody who had concerns—and I’m not going to make the show about me—is who Trump was going to lead with this time around. Was it going to be the people from the 1.0 movement or the 2.0 movement?

There wasn’t a 1.0 person to be found. There might have been 1 or 2 on the margin, and they were not the focus. Who you put in the front row speaks volumes. Who did he put in the front row? Sundar, Elon, Zuckerberg, Lauren Sánchez. He’s got his priorities straight.

The priorities were very clear as to what’s important for this administration.

I’m fully behind both of those priorities. All of those priorities seem incredibly important to me.

You’re such a troll. I don’t want to troll too much, but that’s the big takeaway for me. If that’s indicative of where he wants to go—with business first, pride in our leaders, and pride in innovation—I think he’s sticking to it.

Did you see him at Davos today? He has incredible energy. Trump zoomed into Davos and dunked on them for 45 minutes.

He destroyed everybody at their dying conference as they genuflected and begged him to come next year.

Did you see the photo of the half-empty conference room?

Yes. Most of the events had empty chairs. It’s irrelevant. These people are irrelevant now.

Thomas, do you remember when Davos used to be a really big deal?

I think it’s worse. It’s a counter-indicator now.

It is. There was a good conversation yesterday. Graham Allison, who spoke at our summit, was on a panel, and they basically said, “We lost. They won.” Their group at Davos had lost. It was over. They were no longer on the right side of history.

They said, “We had a point of view on the future. We believed we were going in the right direction, and everyone told us in the last couple of months that we were wrong.”

The gentleman who spoke after Graham was also very good. He said this was the greatest comeback in political history for a politician, and therefore Trump thinks he can do anything. We need to factor in not only who won, which is Trump, but who lost—which is us.

He had a long monologue on why they lost and what they lost. That elite, if you will, has been replaced with a populist vote and a populist leader.

Let’s get to our docket today.

We did our inauguration livestream, and it was awesome and fun. Special thanks to Aircall, Abra, and Hims. Thomas, if you’ve ever got an issue with weight or other issues that men have, you can go to Hims and use the code All-In at Hims.com/All-In. I want you to write that down, Thomas.

In our first topic, President Trump smashed the record for day-one executive orders: 26 in all, beating the prior record of 9 by President Biden. No other president signed more than 1 executive order on day 1 since the Federal Register started tracking this in 1937.

Let’s go through them. If you want to comment on one, we’ll stop and pause, or we can run through them. DOGE was officially established. People are making note of the DOGE SWAT team. DOGE will assign an agency head to each federal agency, and the agency head will build a 4-person team. This will include an engineer, an HR specialist, and an attorney.

It was separately announced that Vivek Ramaswamy was leaving DOGE to run for governor of Ohio.

They suspended the TikTok ban for 75 days. There’s a lot more to get into there, so we’ll break that down later.

The January 6 pardons: Trump pardoned 1,500 January 6 participants, including some who savagely beat police officers. That’s very controversial, and we’ll get back to it.

Ending birthright citizenship: This is another controversial one that’s being fought legally. Those born in the United States to illegal immigrants would no longer be considered citizens if this holds up. Birthright citizenship is codified in the 14th Amendment, and 22 state attorneys general have already sued over the executive order, so it might be a performative one.

“Unleashing American Energy”—we’re going to jump into that one for sure.

Federal employees: a hiring freeze and a full return to office. The hiring freeze doesn’t apply to the military, immigration enforcement, or public safety.

There was also a regulatory freeze on regulations. Gentlemen, where do we want to start? January 6, TikTok, DOGE? Thomas, you’re our guest. Where do you want to start?

I’m happy to chat about TikTok.

As we mentioned earlier, Trump extended TikTok’s grace period by 75 days. He said he wants to get a deal done. He also said at a press conference that he’d like the United States to own 50% of TikTok and that it could be a trillion-dollar asset.

He was positive about the idea of Elon possibly buying it, or Larry Ellison, or anybody. He wants to make a deal. What are your thoughts on what should happen with TikTok and what will happen?

I don’t know what will happen, but I do want to share with you guys that we are investors in TikTok. We invested back when it was obviously just a Chinese company, and investing in China was very different back then.

What resonated with me when I met the founder was that he had a very simple idea. His view was, “If I have a piece of content and it’s a great piece of content, how can I have it be shared?”

His view was that if you shared content on Instagram, the only way people saw it was if you had a lot of followers. He had a really simple idea: “What if I take every piece of content uploaded into the system and show it to at least 1 other user? No matter what that content is, at least 1 person will see it, even if you have no followers. Based on whether that person views it or how long they view it, I might show it to another user, and then a third and a fourth.”

It was such a revolutionary idea, and it really resonated with me at the time. It shows how we’re in an idea business. When you have a truly revolutionary idea, it can get really big.

The genesis of TikTok came from that simple concept: Your content will be shown, and if people like it, it will be shown to a lot of people.

If you look at TikTok in the United States, we can look at the math in a very rough way. Meta, or Facebook and Instagram combined, has about 200 million daily active users in the United States.

The interesting thing about TikTok’s time spent is that its daily active users are about half of the combined number for Facebook and Instagram. We get about 100 million daily active users in the United States, but the time spent on the platform is equivalent to Facebook plus Instagram.

It has incredible scale—not just in daily active users, but in time spent. So what would that mean for the value of the business?

Meta’s market cap is roughly $1.5 trillion. Internet investors generally assume that roughly half the market cap is the United States, so if we take $1.5 trillion, maybe that’s $750 billion.

We know TikTok’s time spent is equivalent, but the daily active users are half, so maybe we cut the $750 billion in half. That would tell you that the maximum value, if you monetize the way Meta does and run it the way Zuckerberg does, might be $375 billion for the U.S. asset alone.

From there, you apply a couple of discounts. If that’s my long-term value, what kind of return would an investor want to get to $375 billion? Is it 50%? Maybe you cut that by 50%.

Then the devil is in the details. What exactly are you getting? Are you getting just the users? Are you getting the algorithm? Are you getting the data? Maybe you apply a more severe discount to that.

Put it all together, and is $100 billion a reasonable scenario? I think the math proves it. Could it be a trillion-dollar asset, as Trump mentioned? In a world where Facebook is worth $1.5 trillion and TikTok’s penetration in the United States is 50% of what Meta’s is, I don’t think that’s unrealistic.

Let’s see what happens. I don’t have any particular insight into what will happen, but I do know it’s an incredibly valuable franchise.

Let me ask the obvious question. People are looking at this and saying, “Wait, you’re going to take 50% of your shares?” In this case, we actually have somebody here who has shares. Are you comfortable with the government seizing 50% of your shares?

I don’t know. First of all, there’s no deal, so it’s a little hard to say.

I can tell you that our analysis was always based on the value of the Chinese business, specifically because it’s hard to know what the TikTok asset ex-China would be worth. Is it worth zero because it’s so compromised regulatorily? Is it worth a lot more? It’s very difficult to know.

We always took a conservative view and looked at the Chinese and some of the other assets as the core value of ByteDance, the holding company.

Ultimately, when you try to sell a company, what’s the value? It depends. Is there 1 bidder or 5 bidders? Is the government the only bidder for that asset? In that case, it indicates a value of X. Or are there other companies that could be acceptable bidders to both sides?

Trump’s deal is that the United States government and the citizens of America would get half of the company. That would be the equivalent of getting shares in the U.S. company. What do you think of this proposal? It seems unique in the world.

If, in order to spin it out and maintain half your shares, you were willing to give the government and the citizens of the United States half your shares, I’d say that the alternative might be that it gets shut down and is worth zero.

In one sense, anything greater than zero is by definition better. Ultimately, it will come down to whether the government wants to be the only bidder for the asset, or whether it would be comfortable with Elon owning it and merging it into X, Microsoft owning it, or somebody else owning it.

Can you tell the audience a little bit about how you guys have been involved as investors in China for a long time? You made a fantastic investment in ByteDance, the parent company of TikTok, and you’ve made other incredible investments in China.

Are you still investing in China? If not, walk us through what happened and why, and what goes on today.

If you look at most—virtually all—of our investments in China, they were Chinese companies catering to the Chinese market.

There was a massive trend of a very significant economy adopting technology. That included smartphone companies like Huawei, companies like Meituan doing food delivery, and Tencent doing gaming.

Our approach was based on these companies. A lot of them were public, which is what people don’t remember. Chamath, you may remember that Tencent was once a public company listed in Hong Kong with a market cap of $200 billion.

A lot of that trend played out through public markets. Baidu was an early IPO. The trend of Chinese companies catering to the Chinese market played out across public and private markets for more than 20 years.

What has significantly changed over the past 5 years is that technology really became a matter of national security. When we were early investors in Meituan, we didn’t necessarily view it as a national security threat. As companies got bigger and AI came to be, it became obvious that there were national security implications.

Our view as investors is that we don’t set the rules of the road. We rely on governments and regulators to do so. It was clear that the regulatory regimes were changing in both countries, so we were going to follow the rules and follow the governments.

My only regret through this whole thing is the quality of the entrepreneurs who came out of that era. It was an unbelievably strong cohort.

When you think about Wang Xing, the founder of Meituan, and how brilliant he was and what he’s done with that franchise—not just food delivery, but also the software operating system for restaurants—it’s truly incredible.

Pony Ma from Tencent was another one. It was an incredible group of entrepreneurs. They were a lot of fun to be around, excited and passionate about bringing technology to their country.

Is it over?

It’s hard to know. If you look at the capital markets, we haven’t seen a tremendous amount of innovation or new companies coming out.

The key question is whether it’s a pendulum. Does the Chinese regulatory regime move one way and then another way, or is it fixed in this state?

Is it fair to say that the Chinese Communist Party wanted to stop the extraordinary wealth creation by a subset of individuals in the country? Was the system becoming too capitalist and enterprising and challenging some of the foundations of the CCP? Is that what was going on, or was there some other security concern?

I don’t really know, because as an investor it’s not a regulatory regime that you have much interaction with. You’re talking to entrepreneurs and reflecting on their energy and business ideas.

We would meet businesses and founders, and if the idea was interesting to us, we would fund it. In some ways, it’s a black box. As a market to invest in, you only have a limited amount of information.

After Jack Ma disappeared for a couple of years, the message was sent. If you’re an entrepreneur, the message was, “Don’t do this.”

The message was received. Then you start thinking about that and what Trump just did. Trump put all of our top people on stage. It’s the exact opposite of China.

The message is very clear, and it’s what Thomas said: Technology is a national security imperative. We’re going to embrace these people. We’re not going to play favorites. We’re not going to pick 1 and exclude another. We’re not going to have summits and exclude the founder of an entire category.

That’s patently stupid and immature, and that’s what the Biden administration did. It picked favorites. That’s dumb, given the importance of technology.

I want to make 3 points about TikTok. First, I’ll give you my sense of the TikTok situation, and then I’ll make a prediction for the future.

In 1984, the way LVMH began was through a deal Bernard Arnault architected. A business owned by the French cotton king Marcel Boussac had Christian Dior inside its empire. His empire was crumbling, and Arnault did a deal where he bought that dying business from the French government for 1 franc.

Fast-forward 40 years of hard work and tremendous execution, and that’s now a $350 billion public company. Imagine if, back then, the French government had said, “We’ll sell it to you for 50% of whatever you build. Take it for $1, but we’re going to own 50% of the upside.” They would have an extra $175 billion today.

That’s important in terms of what’s possible today. I like Thomas’s math. I believe it. I think TikTok is about a $100 billion asset. But the president was clear that it is completely and entirely worthless without his permission, and he wants to own 50% of the asset.

If you’re a buyer of something, you’re not going to pay $100 billion if the person selling it controls whether it can exist. You’re basically going to pay today’s equivalent of 1 franc, which would be $1.

I’m not saying it’s going to be $1, but if the U.S. Treasury wants to hold 50% of an asset that could be worth $100 billion, $500 billion, or perhaps $1 trillion, it would make sense for the Treasury and the American people to own it at effectively zero. The incentive is there.

The president has essentially said that he’s going to get the best deal in the interests of the United States, which I suspect means a price as low as possible, approaching $1. Ultimately, won’t this be the Chinese government’s decision? They could always say, “Let it die.”

It may not be $1. It could be $25 billion or $10 billion. My point is that I find it very hard to see how it gets to the actual value it has today.

That raises a question about the future: Does it make sense for the American government to be smarter about how it allocates incentives and resources?

The American government is still the largest single landowner in America. We give concessions to private companies to drill and build things. If you apply this example more broadly, wouldn’t it make sense that when you create incredible economic incentives and upside for private investors, some portion of that gain goes back to the American people?

For example, suppose you said, “We’re going to accelerate permitting for all these data centers. We’re going to make any form of energy available.” That was the opposite of what Biden’s executive order said on his way out the door. It had all these conditions on the sources of energy.

If President Trump’s executive order on data centers says, “Any kind of energy, but we want to own a 5% royalty,” and we allow you to put the data center on federal lands, would it really change the underwriting that Blackstone and all these other companies do? I don’t think so.

Would it enrich America? I think so. Would it make it easier for people to feel like they’re participating in these incredible gains? I think so.

My prediction is that this becomes more of a template for the future. It will be less about permitting and more about creating incentives, then allocating those incentives in exchange for a share of the upside. I think there’s a strong economic argument for America to do that.

To build on that, people don’t remember that Tesla got $465 million from the Department of Energy. It was called the ATVM program, the Advanced Technology Vehicles Manufacturing program. It was a visionary program at the time.

It was visionary, but it took no equity.

It didn’t take equity, and that was dumb. They should have gotten some portion of equity—10% of Tesla, for example.

The president has sniffed this out. You may not think these things are related, but when he said he was considering eliminating the federal income tax and funding the government from tariffs, and when he invited every company to make things in the United States because he would create economic incentives to make it cheaper here, all of this is part of a broader economic realization.

Unlike 2008, 2009, and 2010, when we were allocating money, or 2020, 2021, and 2022, when we were sending money out the door with no consequence and no accounting, this time he has incredibly sharp businesspeople.

That’s the upside of having people who have succeeded in the real world working inside government. They’ll help him get to this realization, and I think it’s a good one.

My only pushback would be that I think it’s dangerous for the government to start picking winners. If it owns 50% of TikTok, does that disadvantage Meta, for example?

The value-capture mechanism for the government is essentially taxes. The way you capture a portion of the value created by these companies is through the tax system. We have to be careful about the government suddenly picking different companies to win.

I think the right approach is to have good incentives and a level playing field, let the competitive dynamics of different companies play out, and then capture the value through taxes.

I agree with that part completely. There’s a way to architect this. You have requests for proposals that, by design, have to be published openly and be available for anybody to bid on.

My only point is that when there’s an ultimate winner, why not have a small stake in the upside? Look at the amount of money the DOE gives in grants. One company I helped start a few years ago to make advanced battery materials received a $150 million grant.

I’m very appreciative of that grant, but if they had also asked for 5% or 10% of the equity, I would have said yes.

Of course you would have. It’s a great deal. Why not give some upside to the taxpayers?

I agree with that. A parallel would be spectrum. Wireless infrastructure was built after the government auctioned off spectrum and allowed private companies to bid on it. There are a lot of different mechanisms you can use.

I’ll wrap on TikTok. I worry about the precedent being set here with respect to foreign government actions against U.S. companies in their countries.

Imagine the CCP tells Tesla it has to sell its Chinese operations and give 50% of those operations to the Chinese government because of security concerns about Tesla tracking Chinese citizens and where they are driving. You can quickly see how this presentation of the security risk we face with TikTok could create a counterreaction that hurts American companies.

You could see this happen to Google, Apple, or Tesla, and even to food manufacturing companies that have operations overseas. Local governments could say, “For local security purposes, we need to take ownership of your business.”

One of the things that has benefited Americans in global trade is our ability to sell and export our technology, goods, and services around the world through open trade. We’ve used this TikTok security issue to justify doing this as a one-off, but it opens the door for other countries to say that their actions are also a one-off.

I’m generally wary about the TikTok situation opening a can of worms.

You’re 100% right. The CCP has already done it. They did it to Apple. When Apple introduced iCloud in China, the Chinese state-owned data company, Guizhou-Cloud Big Data, told Apple, “We get all the data.” Apple had to roll over.

For every citizen in China who uses an iPhone, the data is owned by the government. Period. Full stop. Apple had no choice but to do that. If you want to play in a certain market, you have to play by the rules.

The thing to keep in mind is that I hear all these theoretical qualms about how it’s unfair, but this stuff happens all the time. It’s just that we don’t benefit from it.

During the pandemic, the Federal Reserve stepped in and started buying commercial paper and corporate bonds. It owned a lot of Ford bonds so those companies wouldn’t default. There’s all kinds of action where the federal government is involved in private companies.

My point is that if you’re going to be involved, have some equity ownership on the back end so that if it really works, it’s broadly value-creating for more Americans. That could help de-escalate the feeling that a few people are winning while everybody else is standing still.

We should talk about Stargate and the AI project, because it connects to one of the other key things I heard a lot about in D.C., and I know Thomas is passionate about it, as I am: energy and electricity capacity in the United States.

Building AI infrastructure means that we need to build energy infrastructure.

We’ll get to that after we wrap up these executive orders. I also wanted to talk about January 6.

JD Vance had said that the people who committed violence on January 6 obviously shouldn’t be pardoned. There’s a parade of Republicans coming out against how President Trump handled this. I think you all know where I fall on it.

I come from a family of police officers, and I feel like he betrayed the blue on this one. If there are people for whom he should have taken his time, it’s possible that some sentences were too long. The justice system is imperfect, and the pardon power is there to do a granular job of looking at each case.

I think we have to look at pardon power generally after what we saw the Biden family do, and now we see Trump doing this. The pardon power is not being used as intended. It’s being used politically by both parties.

I’m trying to call balls and strikes, but as somebody whose family is in law enforcement, and whose chosen career was to be a police officer and an FBI agent, it’s heartbreaking to see people beat police officers and then be treated as heroes.

Some of these people are very dark. The Oath Keepers and some of these groups are extremely violent. They’re coming out now and saying that they’re going to double down, buy more guns, and seek retribution.

You have to be careful with violent people. If you give them a coronation and tell them they were justified, they may go do more violent things. I understand that Trump made this promise to people. I think he should have been very granular. I would have liked to see him do this thoughtfully.

I knew this would be important to you, so I took a few minutes to collect my thoughts. I’ve been walking myself through it since yesterday. This isn’t an explanation for the public so much as an explanation to you, my friend, about how I think we got here.

Before you can look at January 6, I think you have to go back to the COVID lockdowns and what we started to see in liberal cities with the BLM and Antifa riots. Think about what happened in San Francisco, Los Angeles, Seattle, and Portland, the lawlessness in places like New York, and the decarceration movement.

What started to happen was a feeling that, for the same weight, there were different measures. If you look at some of the deportations happening now, it’s shocking. We’re talking about incredibly violent offenders who were walking randomly down the street.

One of the people being deported was an illegal immigrant from Haiti who said he was a gang member with 17 criminal convictions in recent years. He said, “You feel me? Biden forever, bro. Thank Obama for everything he did.” There was also a rapist picked up in Boston and somebody charged with assault.

There was a feeling building for a long time that the law was being applied in very odd ways. Depending on what you did, you would be adjudicated in totally different ways based on your political affiliation or leaning, or based on the desire of a district prosecutor to go after one thing or another.

I agree that January 6 is a stain. There’s nothing to be proud of there. But I think what we’ve found out since January 6 is important.

First, there were people whose convictions were very speculative. The Supreme Court ruled last June, by a 6–3 vote, that at least 350 of those convictions should probably be thrown out.

Then there were cases where a misdemeanor was trumped up to a felony, procedural motions were denied for the defense and approved for the prosecution, and sentences didn’t match the crime. There was also a president who had been the subject of lawfare himself.

A more methodical approach would have been better. But I think what’s happening here is a moment when we can close this entire chapter and get back to observing the law and adjudicating it equally for everybody.

There were pardons that were 100% obvious. There were others that had to be done because the court was perverted in how it dealt with people. Then there’s a smaller number of people who did really bad things.

There were also informants, and the informant issue has to wait until more comes out because it’s in conspiracy court for now. I’m trying to explain to you how we got here. I appreciate you understanding my passion for it, but I think we have to put a pin in this and say that both sides should now have enough experience to apply the law equally to everybody going forward.

Let’s move forward and not look back. I think that’s the best thing we can all do.

The Proud Boys leader Enrique Tarrio said this about getting out: “I’m happy the president’s focusing not on retribution but on success. But I’m not going to play by those rules. They need to pay for what they did.”

These are seriously bad hombres—the Proud Boys and the Oath Keepers. Don’t be surprised if they do something worse.

I want to talk about the citizenship issue and provide some color. I’m sure many people listening have employees who may be affected by this. To be clear, this is going to the Supreme Court.

The 14th Amendment says that birthright citizenship applies to people who are subject to the jurisdiction of the United States. For a long time, that phrase wasn’t really considered in how the Supreme Court administered the 14th Amendment.

I think what this executive order and the lawsuits filed almost immediately afterward will do is send this back to the Supreme Court very quickly. People will opine on what it means.

If it means nothing, then if you’re here, however you’re here, and you have a child, that child will be an American citizen. If the qualifier is now part of the administration’s interpretation of the 14th Amendment, then what President Trump has written could take effect.

If you’re here illegally, it may not apply. If you’re here under a visa, that may mean you’re still subject to the jurisdiction of somebody else, so it may not apply. The Supreme Court is going to decide this in short order.

Let me make a quick comment on January 6. I generally don’t like the power of pardon. We saw what Biden did—thousands of pardons.

The original intention of the pardon power in the Constitution was to protect civility in the Union at times of insurrection or rebellion, when the country needed to come back together and realign itself. It has been abused far beyond that original intention.

It almost gives the president the ability to rewrite the law, allowing actions from the executive branch to supersede the judicial branch, which is meant to protect and execute the law of the nation. I think it’s gone too far. It feels like an amendment is necessary to address the ability of presidents to preemptively pardon people for things that might be discovered in the future simply because they’re close to them.

Alexander Hamilton wrote deeply about the pardon power in Federalist No. 74. He emphasized that the justice system, while designed to be fair, may occasionally result in overly harsh or unjust outcomes, and the pardon allows acts of mercy to correct those situations.

Many of the actions we saw from Biden don’t fit that definition. Hamilton also wrote that in seasons of insurrection or rebellion, there are critical moments when a well-timed offer of pardon to insurgents or rebels may restore the tranquility of the commonwealth.

To counter my own point, Hamilton’s argument is that when the nation is divided, a pardon may help bring the nation back together, as may have been the case with January 6. It may not be about justice or retribution; it may be about restoring the tranquility of the Union and allowing the nation to move forward, even if it doesn’t seem just.

I’m not defending the action or the pardon. I don’t like pardoning as a tool when it’s used in a hundred different ways. It certainly seems to require change.

As a programming and research note, 6,500 of the 8,000 people Biden pardoned appear to have been convicted of marijuana possession.

Then the law should have been changed, and the courts should have overturned those convictions. It shouldn’t necessarily fall on the executive branch to take that action. The courts and legislators need to do their jobs.

I don’t disagree, but how long do people have to wait for that?

How long do you have to wait? The people convicted of selling dime bags or having an ounce of marijuana were Black and brown, while all my white friends who started cannabis companies are taking them public in Canada.

Shout-out to Trulieve.

Let’s keep moving. I think we did a good job on reconciliation on the podcast. Where do we want to go next?

The Stargate project.

Stargate and energy. According to OpenAI’s press release, Stargate is a new company that intends to invest $500 billion over the next 4 years building new AI infrastructure for OpenAI in the United States.

Friedberg, we have to find out what’s going on with the Stargate movie and television series, and whether they got permission to use the intellectual property.

SoftBank and OpenAI are the lead partners, with Oracle and MGX also participating. Oracle, NVIDIA, and OpenAI will build and operate the infrastructure. Microsoft is somewhat involved, although we’ll see about that.

Masayoshi Son was at the announcement. He’s going to be the chairman. Gavin Baker and Elon Musk both called it out as nonsense on X, saying they don’t actually have the money. Gavin did some back-of-the-envelope math, and as always, Sam and Elon have been spicy in the replies. Sam is hurt, and Elon is dunking. Their relationship is very fluid.

Microsoft CEO Satya Nadella jumped into my thread with Elon saying he was good for the $80 billion, and that Microsoft was ready to build other products and services. He said that was what was important.

Thomas, are you an investor in OpenAI? Or in xAI? Let’s establish that so we know whether you’re talking your book.

I am an investor in OpenAI.

There are 3 points I want to make. The first goes back to the origins of this podcast. One notable difference between public-market investors and venture investors is that venture investors tend to be very tribal. If you’re in 1 tribe, by definition you’re not in another.

There are domain-specific reasons for that, but because of my legacy as a public-market investor, I’m comfortable being pro-OpenAI, pro-Tesla, and pro-X. That’s my first caveat.

My second point is a slight detour on Masa. I’m curious to get your guys’ opinion: Is Masa the GOAT?

When I started in the growth business, I met Yuri Milner. He said, “Thomas, you get in the Hall of Fame if you can have 1 deal that gives you a $1 billion return”—meaning the investment plus what you take out is $1 billion or more. He kept a tracker in his mind of who had made those deals.

Masa is in the $100 billion club. Not only has he done it once, he’s done it twice. He did it the first time with SoftBank and Alibaba. At its peak, that was a $200 billion win, and I think he ultimately netted more than $80 billion.

He did it again with Arm, which he bought for $30 billion and now has a gain of approximately $135 billion. He could have done it a third time with NVIDIA if he hadn’t sold.

There’s nobody like him.

You’re 100% right. You have to give the man his flowers.

I’m never betting against Masa. You have to give him credit for what he’s done.

Now, on OpenAI, we do own some OpenAI, and I’m bullish on it. I suppose that’s now a counterintuitive view. My thesis is really based on ChatGPT.

OpenAI has different businesses: the model business, the API business, and the ChatGPT business. If you look at downloads, it’s interesting to watch the market share of ChatGPT compared with Google Gemini and Grok, both in the United States and internationally.

It’s a dominant franchise. It maintains more than 80% market share. J-Cal, if this were ridesharing, you’d say, “This company is the winner.”

The quality of the ChatGPT franchise is incredible. They disclosed 300 million weekly active users and more than 1 million enterprise users. It has become a core part of my workflow. I have it on my phone and my desktop, and I use it all the time. I can’t quantify the value it brings. It’s that important.

It keeps getting smarter with every query I ask. It learns more about me and more about our system. I think the ChatGPT franchise inside OpenAI is one of the greatest and fastest-growing success stories.

On Stargate, the real question isn’t whether they have $500 billion. That money could come from equity, or from Masa. If you look at Gavin’s tweet, he caveats it by saying it assumes Masa doesn’t want to sell Arm and SoftBank.

There’s leverage, and they could use debt. The real question is whether we believe there’s a return on that $500 billion. If there is, people will want to fund it. You can do it data center by data center, even at the GPU level.

The financing is available. The question is whether the market is big enough and the opportunity is large enough for investors to get a return on $500 billion.

You framed it perfectly. Since you bring up ridesharing and the 80% market share, it’s important to pause there.

In ridesharing, you didn’t have Uber being challenged by Google, Microsoft, and Elon Musk at the same time in a digital product rather than a real-world product. When I use Gemini, xAI, ChatGPT, and Claude, I use them all daily. The difference between the products is very narrow.

The equivalent would be somebody with 100 million cars on the road already competing with Uber, which didn’t exist at the time. I look at this as a different kind of race.

I do believe the ChatGPT franchise is worth something, but when I talk to enterprise customers, they want the open-source product most of all. It’s still anybody’s game, and I don’t know how you get a return on $500 billion invested in this. I’m not sure that’s the real number.

The way these things happen is that you don’t fund all of it with equity upfront. These things get funded facility by facility and data center by data center over 4 years.

It’s not just equity. There’s site-level debt and other financing, so it’s a complex capital structure. We live in a world where I was looking at the cloud-capex chart. Total 2025 capital expenditures for the top 5 players in the United States are expected to be $312 billion.

That’s in 1 year. We can frame the $500 billion over 4 years as being at least in the same order of magnitude as what others are spending.

Another interesting point is that U.S. internet companies spend 20 times what Chinese companies spend on capital expenditures. As we think about competitiveness, it’s powerful that we’re spending 20 times what our competitor is spending on these advanced technologies.

Value and money are correlated in some industries, like real estate or heavy industry, but I don’t think they’re very correlated in technology.

Last week we saw a model released by the Chinese under a completely open-source MIT license that you can run on a laptop. It turned out to be as good as OpenAI’s o1 model. It was a couple of versions old, but the point is that this model cost millions of dollars and competed with a model that took billions of dollars to train.

My takeaway is that these costs are falling precipitously. Spending more money doesn’t necessarily get you further along if the goal is progress. That’s particularly true in AI, because we’re learning as we go.

It raises the question of what the point is in announcing these grandiose numbers. If an oil company came into the Oval Office and said it was going to spend $500 billion, you’d know it was real because you know what those things cost.

If the oil is deeper and the rock is harder, you know you have to spend more. Those numbers are grounded in reality. If you’re a rocket company building a rocket to Mars, you can also be detailed about the costs and assume some reductions from scale.

But in this world, there are advances that let you do things more cheaply almost overnight. You can spend one-thousandth of what you spent 6 months ago. The spending is more of a gimmick than a technical commitment.

If they want to spend the money, go ahead, but it doesn’t tie to whether they’ll be able to make custom vaccines or cure cancer. I thought all those things were connected.

My takeaway wasn’t a commentary on Masa, Larry, or Sam. I think all 3 companies are very good. It was more a comment that you have to protect the president’s legacy. If I were them, I’d make sure the announcements are further down the technical spectrum and are actually real.

If they achieve these things but it costs $1 billion and they hire only 50 people, there’s going to be a little bit of egg on the face. The spending and the technical progress seemed decoupled. It seemed like marketing and sizzle that was hastily put together.

It definitely felt put together hastily. It would be great if OpenAI builds another incredible model—whatever comes after o3, o4, or o5—but it’s not clear that you have to spend $500 billion to do it.

I think people are playing to what Trump likes, which is a big number and his influence on growing the number. Masa said, “I told you $100 billion. You challenged me to get to $200 billion, and I came back with $500 billion.”

That’s the Trump ethos. You can be critical and say he’s a BS artist, or if you’re from Silicon Valley you can say, “He’s moonshotting. He’s going to shoot for the stars and get to the moon.”

If they end up spending $50 billion or $100 billion and become more competitive, more power to them. My question was whether Elon had any insight that this was happening.

Probably not, based on his reaction.

I have no insight either, but if you were going to criticize Trump for putting his thumb on the scale for any one company or individual, this shows that he’s not. It’s an open platform.

The Trump administration is an open platform. The highest bidder gets to come to the White House. Whoever spends the most or buys the most military equipment and ammunition from us gets time in the White House.

This is a capitalist country—the greatest one in the world. It’s an equal opportunity for anybody.

The obvious consequence of this data center and chip infrastructure effort, which is going to happen with or without government involvement, is increased electricity demand.

Thomas, I know you have some slides. Do you want to pull them up? I’ve talked a lot about the difference in electricity production capacity between the United States and China, historically and prospectively.

The United States is paying roughly 1.5 to 3 times the price per kilowatt-hour for electricity compared with China. We have roughly half the electricity production capacity of China. China is adding more, and its cost to add new capacity is between one-tenth and one-half of our cost.

Everything is moving in the wrong direction. If AI and automation are the critical factors for economic growth, and electricity is the critical input, we’re hugely disadvantaged and aren’t going to catch up.

Thomas, your analysis seems to indicate the same thing. Does it make sense to walk through it now?

Yes. If you think of the GPU as the atomic unit of the future and a key competitive advantage, you can’t power a GPU without a data center, and you can’t use a data center without power.

At the end of the day, it comes down to the grid and power. That’s constraint number 1.

The other important thing is that in an inflationary era, you don’t want to increase electricity prices for consumers. We need to add capacity without making consumers’ bills go up.

If we run through the slides, the data supports your point. We did not invest in our grid for a very long period of time, while China did.

That wasn’t always the case.

During the postwar boom and the information age, the United States did invest. Then, around 2000, we stopped.

Globalization is why we stopped, right? We didn’t have factories, so we didn’t need as much power.

Exactly. We outsourced a lot of our hard-core manufacturing and industrial production.

No factories meant no need for additional electricity. The new factories are H100 factories. We need AI factories.

Exactly. You can think about it almost like a subscription business on a net-add basis.

The data on net additions makes the contrast even starker. Since 1985, the gap between China and the United States has grown to almost 7 times. In 2000, the United States and China were both at roughly 1,000 terawatt-hours. Today, the United States is at roughly 1,600 terawatt-hours, while China is at approximately 9,000 terawatt-hours.

The Biden administration issued an executive order last week allowing power plants on federal lands to power AI data centers. It started as an intelligent executive order, but the second paragraph began talking about DEI and requiring certain percentages of energy to come from certain sources.

Today, it looked like President Trump overturned it. Now it’s full steam ahead: If you have the permit, get the power however you need to get it. I think that’s the smart decision.

Can we do this without nuclear?

No.

Can we catch up without nuclear?

No.

One interesting public-market stock is Constellation Energy. Amazon and Microsoft are signing direct agreements with companies like Constellation to secure energy access behind the meter.

At the end of the Biden administration, federal agencies also announced contracts directly with Constellation to power some facilities. The number was small, but the market took it as a clear indication that even the Biden administration acknowledged that nuclear is the best path forward.

We expect the trend toward nuclear energy to continue. It has been proven in China and Europe, and it will be a key part of the AI arms race over the next 10 to 20 years.

This is the story in one slide. U.S. nuclear capacity has been basically flat for more than 25 years.

Since the 1990s.

China has added an enormous amount of nuclear capacity over that same period. When Scott Bessent was asked whether we’re in a green-energy arms race with China, he was looking at the chart on the right. It shows how much nuclear China has added. It’s clear what the answer is: China is leading the way, and we need to follow.

We’re in a different era of nuclear technology than we were during Three Mile Island or Chernobyl. We have new Generation IV technologies that are meltdown-proof and use a totally different architecture.

There isn’t the same setup for runaway heat and a reactor collapse with a release of radioactive material, which is what happened at Chernobyl. New systems, like the pebble-bed reactor that has been producing electricity in China, are incredible technological architectures.

They’re here and running. China is rolling out dozens or hundreds of them, while the United States is rolling out zero. That needs to change.

We have only a few months to stand up the engine that will allow us to make the materials and production technology needed to deploy these stations. If we don’t, we’ll be hugely disadvantaged on energy costs and on our ability to deploy AI competitively.

You can also look at France, my home country. It’s not known as a technology leader, but 70% of its electricity comes from nuclear.

France held the line when Europe tried to deprecate nuclear power. Germany did not, but France held the line.

It’s too important. It’s more than 70%, and it’s doable. Now that AI is a matter of national security, the need for nuclear energy dovetails with fabs—semiconductor manufacturing facilities—which are another strategic area that we need to ramp up in the United States.

The thing I noticed about the chart is the trend line. Look at how fast China is building these facilities. I don’t know what government incentive we need to create or what red tape has to be moved, but if some states can’t build multifamily housing, nuclear power is going to require a dramatic change in how we operate.

There’s a lot of capital and intellectual capital in the United States that’s eager and ready to get moving on this buildout. The only thing holding it back is regulatory challenges and roadblocks.

If those are removed through emergency action, I view this as more of a national security threat than the border. We need to increase energy production in an accelerated way for the United States to challenge China in manufacturing and AI in the decades ahead.

Emergency declarations may be needed to eliminate the regulatory roadblocks that are keeping electricity production from proliferating. It’s time for that kind of action.

We have the technology, the physics, the will, and the talent. We have a blocker, and the blocker is regulation. Leadership in government needs to remove it.

Am I summarizing correctly?

That’s right. You have an energy czar who I think is aligned with that intention in Doug Burgum. We have a major AI arms race, an automation and manufacturing demand that will massively increase energy demand, and a deregulatory administration.

The stars are aligned. This could happen.

Thomas, what’s the downside to removing the regulations and building this extra capacity? Can we define any possible downside?

The downside is the amount of time it takes to build these things properly. China has had decades of practice building them safely and quickly.

We can build things, but we haven’t had the practice to build them safely or quickly because we haven’t built any. That’s a complicated issue.

If there’s resistance to this, what would theoretically be at the top of the list? How could anybody want to block it?

One other element is that we’re blocking China from accessing the latest GPUs. The whole point is that we don’t want China to use them to develop systems against us.

But if China someday has 10 times our power capacity, it won’t matter. Even if its GPUs aren’t quite as advanced as the latest GPUs Jensen and others provide, it will have so many more of them that its industrial AI capacity will be much larger.

In order for our strategy to be effective, we need to build this power base. It’s a very astute point.

I don’t know if you’ve been watching Netflix, but I’m a shareholder and delighted. I don’t know what’s going on over there, but they seem to be getting their subscriptions right. That disastrous fight seems to have driven the stock price up, and some strategy appears to be working.

Chamath, can you riff on it for a moment?

I have the opposite take.

Netflix is obviously a very good company and has been incredibly performant, but I think it offers a dark commentary on American society and productivity.

I asked Nick to put up 3 charts. The first is Netflix’s stock price from 2010 to 2020. It’s only gone up since 2012.

The second shows the percentage of Americans taking SSRIs over the same period for depression and anxiety. That has also gone up.

The last chart is from the U.S. Bureau of Labor Statistics, showing what happened to the labor-force participation rate over the same period.

I’m not blaming Netflix, but I think it’s a commentary on what’s happening in society. It’s a leading indicator. As the company does better, it wins by being a sink for people’s time.

Netflix and chill is a phrase we all understand. The dark underbelly is that there are a lot of people who are more dejected, and a lot of people who opt out and decide not to work. You could graph the fertility rate, and it has gone down as well.

It’s all part of a soup in which people are living a life that unfortunately doesn’t exist. It’s a well-run business, and I wish them all the best. But at some level, I think it’s a commentary that as this company does better, more people are living a shell of what their lives could be.

Cause and correlation aside, these are correlated, not causal. Netflix isn’t causing depression.

I’m curious how much of it is a substitution effect away from traditional media. The measured time spent watching television hasn’t really increased over 20 years.

To me, it’s more a reflection of people substituting away from traditional media. You could overlay the stock charts of Fox, Comcast, and CBS, and all of them have imploded.

The question is whether the Netflix product is fundamentally more unsafe or detrimental to someone’s health than prior broadcast television.

This isn’t a Netflix-specific commentary. Online products versus linear products have the benefit of algorithmic targeting and personalization.

Netflix and other products have figured out that they can hypertune content to what people think they want. What people think they want isn’t necessarily what they need.

With linear television, you had to take what was given. You might not have loved Cheers, but you watched Cheers. You might not have loved St. Elsewhere, but you watched it.

Now you can go somewhere that feeds you exactly what you want. It gives you a visceral feedback loop, and you can binge-watch for an entire weekend. The question is what you didn’t do. You didn’t necessarily work out, eat well, or spend time with friends.

I’m not saying I don’t do it. I do it too. When Jason or our friend Goldberg says a series is amazing, it’s hard not to start watching it and then finish it. It’s addictive.

I have to check myself, because for 2 days I’ll be tempted not to leave my bedroom. My kids ask, “Where’s my father?”

I’m not picking on Netflix. Societally, I think we want people to be vigorous and engaged in the real world. There needs to be a balance.

I wholeheartedly agree. You have to schedule these things and make them a permanent part of your life: playing poker on a certain night of the week, doing things with your kids on the weekend, and having certain rituals.

The real damage to our kids isn’t that screens are damaging them. It’s that screens are replacing time. This anxious generation, as Jonathan Haidt described it when we interviewed him in the All-In interview series, is substituting TikTok, Netflix, and YouTube for time with friends and physical activity.

You have to limit screen time so that those other things—and boredom—can emerge.

We’re going to be having a conversation about dopamine in the coming years. The algorithms have become very good at firing your dopamine every time you swipe on TikTok, reach a cliffhanger in a series, or watch a MrBeast video. MrBeast is a master of it, too. These platforms are all very good at giving you the hook and the amazing moment.

Musicians are doing this now with hooks, and then the hooks go on TikTok. When you get burned out late at night from getting those dopamine hits, you feel depressed the next day. Then you go to a doctor, and the doctor says you have depression and prescribes Wellbutrin or whatever the medication of the moment is.

I’ll tell you what I focus on with my family: sleep hygiene, diet, exercise, and meditation. Do those 4 things and you can reverse these trends. You don’t necessarily need medication. Medication is a difficult issue for kids, but try the first 3 things.

When I was growing up, my father made us sit there at 9 p.m. every night. My sisters and I had to say a Buddhist prayer. Then he made us close our eyes for 15 minutes and meditate.

I didn’t know what meditation was, but I was forced to keep my eyes shut. I would just daydream. I suspect there was some grounding value in it.

I understand what you’re saying. I just have such a negative connotation from that experience that I’ve never been able to come back to meditation.

I would daydream too. What I dreamed about was getting out of that house. I thought, “I’ve got to manifest some success here, because this sucks.”

Thomas, thanks for coming on the show. It was a great first appearance. We’d love to have you back, unless the king returns at some point.

I’d love to come back.

You should come back to the All-In Summit as well. I know we still have great topics, like Broadcom, that I want to get to someday.

When I come back, I’d love for us to double-click into Hock Tan. He’s the CEO of Broadcom.

For people listening, we were going back and forth about Hock Tan, and I was excited because Broadcom is the only trillion-dollar company nobody talks about. Everybody knows NVIDIA, Amazon, and Meta, but Broadcom is a $1.2 trillion business. It started through a series of roll-ups.

This isn’t the Broadcom we’re all thinking of from the telecommunications era, is it?

It is. Henry Samueli.

When Hock bought Broadcom, to your point, Broadcom was the province of Orange County, founded by Henry Nicholas and Henry Samueli—legends in semiconductors.

It meant a lot to them that the combined company, Avago, which was buying Broadcom, kept the Broadcom name. Hock got a little bit of a price discount to name the company Broadcom. That was a great deal for him.

The ticker remained AVGO. I asked Hock why the ticker was still AVGO. He said, “They forgot to negotiate for the ticker.”

Never forget the ticker.

Interesting. When he started, what was the market cap?

It was $3.5 billion. It wasn’t just an orphan; it was a spinout of HP, and then a spinout of a spinout of HP—a double orphan.

It has grown by 8 times over the last 5 years. NVIDIA has grown by 20 times, while Intel is down 70%.

Do you think this is a function of investing in the right things, or of making long-term investments and thinking down the road?

Hock is an M&A master. It’s less about a single product winning and more about portfolio approach and construction.

Did he buy well ahead of the curve? Did he buy over the horizon, in a way that Intel missed?

Intel’s problem was a complete abdication of corporate governance at the board and CEO level.

We have a breaking news story coming across my feed. There’s a video trending on the internet. Apparently, an executive order has just gone across the president’s desk, delivered by none other than David Sacks.

David Sacks is going to be forming an internal working group to make crypto make America the world capital of crypto under his leadership.

That’s right.

Look at this. He’s using a Sharpie to sign the executive order. Sacks gets to keep the pen. There it is.

I love how he holds them up.

I wonder if Sacks gets to keep it.

They might not be excited, but we’re going to make a lot of money for the country.

Thank you, sir. So is David.

Chamath, what’s your reaction?

I think this is great. We’re going to make crypto great again.

To be more specific, I think there are a handful of ideas that, if implemented correctly, could be revolutionary. I think it’s good to have a stockpile of crypto, although I have less of an opinion on that.

A set of stablecoin rails that makes payments instantaneous and nearly costless would be an enormous accelerator for GDP. It would cut fraud. I think David is going to figure that out.

You’re talking specifically about stablecoins?

Stablecoins. I talked about this at the beginning of the year as one of my predictions. I think U.S. stablecoin rails will be hugely disruptive and value-adding.

David Sacks will have help from the head of the SEC and the Treasury secretary. I think Sacks is already off to a hugely fast start.

The entire country knows we need to win the AI race. The numbers are the numbers. You put up a chart like that, Thomas, and there’s only one way for us to win: build more nuclear power plants.

Let’s see who tries to stop it. I don’t know if AOC, Elizabeth Warren, Bernie Sanders, or anybody else wants to jump in front of this Trump train right now, but I think you get slaughtered if you do.

If only there were somebody who could comment on this executive order. If I had the Batphone—if I could phone a friend—who would you call?

I don’t know. My Saxy-poo was always on speed dial. We used to have great battles on the podcast, then play chess and have drinks together at the Battery.

That was a different era. I don’t have my boy anymore. He’s busy.

Oh, do we?

I’m back.

There he is. He’s back to reclaim his seat. It’s yours, homie. We missed you, buddy. You look so handsome.

I’m going to cry.

It’s been 61 days. You’re not my bestie anymore.

We miss you too.

Were you just in the Oval Office?

I was just in the Oval Office.

Let me ask you a question: Was that your first time in the Oval Office?

Technically, it was my second.

How’s it been so far?

It’s been pretty incredible.

Tell us what the vibes are like. What’s it like to go into the Oval Office and have an executive order signed? Tell us what the moment was like, and then tell us what’s in the executive orders.

We actually did 3 executive orders today: one on crypto, one on AI, and one on PCAST, the President’s Council of Advisors on Science and Technology.

We can start with crypto. The main thing that executive order does is form an internal working group with the goal of making the United States the world leader in crypto. It’s not 1 specific action. It authorizes and creates a working group with the goal of achieving that objective.

The members of the working group will include me, the head of the SEC, the Treasury secretary, and all the agencies and departments that have an interest in crypto. We’ve formed this working group, and I’m chairing it.

The goal is to identify and make recommendations that the departments can then execute, with the goal of making America the world capital of crypto.

The president set off Davos today. What are the other 2 executive orders?

The next one is AI. The number-one thing that executive order does is rescind the Biden executive order that we discussed on the podcast many episodes ago.

It was a 100-page monstrosity that imposed burdensome regulations. The president had promised on the campaign trail to repeal it, and he did. There was a massive executive order on Monday that rescinded dozens of Biden-era executive orders, and this one provided more detail.

It said that the Biden order was unnecessarily burdensome, and that we want to be the world leader in AI. In fact, it uses the word “dominate.” We want global dominance in AI.

Similar to what we’re doing on crypto, it directs the creation of an AI action plan. That study will be led by 3 people: the national security adviser, Mike Waltz; the director of the Office of Science and Technology Policy, Michael Kratsios, as soon as he’s confirmed by the Senate; and me.

The 3 of us are going to work together on the AI action plan. The idea is to figure out what the industry actually needs to make us number 1 globally in AI. It puts more substance behind what this executive order is about.

The last one is to build a science council. I don’t know if everybody is familiar with that.

Could you describe what that is generally for the audience?

PCAST has been around for a long time—decades. The goal is to assemble a group of top science and technology minds to provide the president with advice.

It was already announced, but Michael Kratsios and I are going to co-chair it. This executive order provides the authorizing language to set it up and gives direction around what PCAST is going to do.

Some of the language may appeal to Friedberg. It talks about the need to return to truth-telling in science and get away from woke science. The AI order discusses this as well. We don’t want AI models to be ideologically biased or carry an agenda. We want them to be as politically unbiased as possible.

If you need a recommendation for somebody for the council, I know somebody who is extremely well-versed in science and has a broad array of scientific knowledge.

Hear, hear.

We could have 2 besties in and around the White House.

Thanks for including us in so many aspects of the weekend. The audience wants to know what the chances are that we get Sacks back on the podcast from time to time in 2025.

Now that confirmation hearings are underway and your role is being formalized, will people be able to get back on social media and talk about things they were restricted from talking about over the last couple of weeks?

If you’re working for the White House, there needs to be a reason for you to be out there. You don’t want to make news you’re not supposed to make.

In this case, the president signed executive orders on crypto and AI, and they wanted me to explain them. It makes sense for me to go out there. I have an appearance on Fox Business in about an hour to talk about the executive orders, so that may come out before the podcast.

You guys are the first people I’m actually talking to. There needs to be a good reason for me to come on the podcast, but I’m going to try to come on every chance I get.

This is our appeal to President Trump to let you come on and chew the fat. Maybe there’s some upside to it for him.

Do you miss being on the podcast?

Of course.

We miss you so much.

I’ll be back at some point. Don’t worry.

What are your favorite memories from inauguration weekend?

The inauguration ceremony itself was spectacular. I was there in the Capitol.

Yes, we were there.

J-Cal and I took our seats in the back, in the peanut gallery. I thought that was where I was going to be sitting. Then somebody came up to me and said, “Mr. Sacks, we have a seat for you up front.”

They put me with the secretary of state and the Treasury secretary.

You were with the big boys.

It was a huge honor. We were behind the Trump family. You may not have seen me because the Trump kids are pretty tall, but the whole family is very tall, so I disappeared a little bit.

They put me on the dais, which was crazy. I didn’t realize that was where you were sitting.

There was a shot where David and Jacqueline were in the front. They looked amazing and were looking around for where to sit. I thought, “This can’t be right. There must be a name on the seat.” It’s good that it got sorted out.

What was the vantage point from your side?

You’re looking out at the audience. I could see the faces of the Democrats as the president came out. President Biden and Kamala Harris didn’t look too happy.

I wish there had been a fixed camera on them the whole time. I wanted to see their reaction.

I’ve seen some of the reaction shots.

The inauguration speech was incredible. First, it was completely consistent with everything the president had said during the campaign. You felt he had a mandate because this was everything he had said during the campaign.

He won the presidency, the popular vote, 7 out of 7 swing states, the House, and the Senate. It really feels like he has a mandate. He reiterated that mandate in his speech.

The cherry on top was that the losers had to sit there and listen to it.

Kudos to them for coming out and keeping with tradition.

When he talked about the corrupt establishment that tried to put him in prison through lawfare, he was talking about some of the people sitting behind him. It was a remarkable moment.

Take us back to how you ended up getting the offer for the role and deciding to take it. You haven’t talked about this publicly. Anything you want to share about the path to getting here?

I don’t know that there’s much to report. I was helping with the transition, I was offered the role, and I decided to do it.

It wasn’t something I was expecting or planning. As you know, I didn’t even know there was going to be such a role. They decided to create it and offered it to me, so I decided to do it.

One thing I should clarify is that there are still things I can’t do yet because there’s a government ethics process. I’m in the middle of that process.

Signing these executive orders gave me certain responsibilities and authorized me to do certain things. That process will begin as soon as I complete onboarding. There are things I can do and things I can’t do.

I can be in listening mode, but I can’t necessarily shape policy yet. That will all be worked through over the next couple of weeks.

Sacks, I have to ask: First time in the Oval Office—what was going through your mind?

My head was on a swivel. It was one of those moments when you can’t quite believe you’re there. It does feel a little like you’re on a movie set. It’s so famous.

Is it bigger in person or smaller in person?

It seemed about the right size. I know people say it’s smaller than they expect, but I didn’t feel that way. It also didn’t feel bigger. It felt like the Oval Office you’ve always seen.

The staffers who have been in there 100 or 1,000 times are obviously used to it. It’s just where they do their work. But the first time you go in there, you’re pretty awestruck.

You mentioned that Vice President Vance had a post about this. Did you see it?

I saw the video of Mike Johnson walking in. He was in shock because he had never been there. He became a senator only a few years ago, and we had a Democratic president, so he had never been in the Oval Office.

He said he was honored and humbled to be there. That resonated with me. I think every American would feel that way the first time they enter the Oval Office.

The president said some nice things about you. He said, “There’s nobody like this guy. How did you get David Sacks? How did you do that? He’s doing it for the country more than anything else. We appreciate it, David.”

He’s been really nice to me. It’s incredible. He’s been just incredible to me. He gave me the pen from the signing, which was pretty cool.

Can we see the pen? What brand is it?

It’s not branded. It says “The White House” on one side and has his signature on the other.

Is it a Sharpie?

It is. You’ve seen him sign. I assume it’s what he likes.

Have you seen all those ASMR videos of the president signing? You can hear the pen on the paper.

Exactly. There are all these ASMR videos of that.

Sacks, Friedberg wants to know if you can sign his new Tesla with that pen.

I think only the president is allowed to use this pen.

Put it away. Keep it safe. Don’t lose it. You have to put it in a glass box.

There are thousands of commenters begging for you to come back. Do you have a message for your legions of commenters?

I thank them. I’m back right now. I know maybe it’s not enough, but soak it in.

It would help if you guys did a better job on the podcast so they weren’t complaining all the time.

We do the best we can without you.

Get back to work. You’re on the taxpayer’s clock. Go back to work and save America. Make America great again.

We wish you great success. We miss you greatly, and we thank you for your service to the country. We wish you all the best as you handle 2 of the most important issues for our country—AI and advising the president on cryptocurrency.

And PCAST.

And PCAST, of course. Love you, Saxy. Hopefully we’ll see you on Tuesday.

Thomas, thanks for coming on.

Thank you. Great podcast. Love you, boys.

Let your winners ride, Rain Man. Instead, we open-sourced it to the fans, and they’ve gone crazy with it.

Love you, Queen of All.