Palmer Luckey’s EagleEye Reactions, Ferrari's 12Cilindri, Kushner Story Ignites Debate | Alex Klein, Hemant Taneja, Rob Toews, Nathan Benaich

You're watching TVPN. Today is Friday, October 17, 2025. We are live from the TVPN Ultradome, the Temple of technology, the fortress of finance, the capital of capital.

Sorry, I'm having too much fun. I got a new sound effect.

Yeah, we got a new sound effect. The soundboard continues to be on a generational run of its own and Palmer Luckey has been on a generational run, going on a lot of different shows. He's coming on our show. We announced that today. Very excited to have him join the show. I've interviewed him a few times. It's always a fantastic chat.

The guy can yap.

I want to play a clip from his appearance not on Joe Rogan, which is fantastic. You should go listen to but of him on Bloomberg. But first I want to tell you about ramp.com time is money save. Both easy use, corporate cards, bill pay, accounting and a whole lot more all in one place.

Last night we got together with some members of the Ramp team and in a great sign of respect in our culture, they built this ice statue.

The ice sculpture was very fun to.

See, which you may have seen over on X, but absolutely stunning.

Anyway, let's play the clip from Bloomberg. Palmer.

Lucky Palmer. There was a report about the NGC2 system. It named both Anduril and Palantir. We've asked Palantir through Sean to comment on that. But the report in Palantir's case moved markets. Could you just respond to that reporting which came from Reuters, and explain the sort of timeline and the inaccuracies that you feel there are in that report?

So you, me or may not remember that. I was a journalism major myself. I was only a semester and a half away from graduating. I was the online editor of the Daily 49er, which was the Cal State Long beach student run newspaper. So I care a lot about this stuff, both as an Android perspective and then also just from a journalist perspective. The journalist who wrote that Reuters story would have been failed out of the class by any of my professors.

He refused to include the record statement.

He refused to include the on the record statement. This is crazy though, because they fundamentally made the whole story, not a story.

They didn't include what Palantir and Andrew had to say. I'll read the actual article.

Next generation command and control. It's not tracking what the users are doing. It doesn't have any passwords to log in. There's no security turned on. And they said, oh my God, this is a huge problem. And we said, guys, this was an early prototype. The army wasn't paying us to demonstrate user accounts. It was paying us to demonstrate integration of different sensors, different weapon systems all into one pipeline. And Lattice has all this functionality. We do this all the time. It's not like we don't know how to do this. And the army said the same. They said, yeah, we just turned on all those features. Just a few weeks later, all these issues that you're talking about didn't exist.

They.

This was an internal security audit. Just noting that these things were a factor in that early prototype. Yes, there's no access control. Yes, that is a risk.

Yes.

We have not audited all of their security, you know, all their security code. That is a risk. But that's completely normal for early, early development. But Reuters didn't include our quote, didn't include the Army's quote because that would have made the entire story irrelevant. So instead they include quotes exclusively from our competitors and from so called anonymous sources saying that Anduril and Palantir are running this fundamentally insecure thing. That 99% chance that was planted by one of our competitors who had access to that memo. And there's only a few people who could have done it. And I can make people look bad by slicing and dicing memos too. I'll show you if I want. I'll get back into the journalism game.

We can say, let's go. We've given you the opportunity, Palmer.

Lucky a column at Reuters, this opportunity to attack my.

You just mentioned your, your competitors. Let's call.

Yeah, we can, we can wrap it there. But basically, I mean, it's like the army was like, hey, we want you to prototype something for us, like put a bunch of dots on a map, do something different, like create a new ui, like show what you can do. And then clearly an enemy of Anduril and Palantir, like, leaked, like, well, they didn't have Oauth set up yet. And it's like, well, yeah, like, we know this in Silicon Valley. Like when you build a new product, you focus on finding a problem, building the solution, and then you figure out how to optimize the database and cache things at the end.

A lot of people try to do gotchas with Palmer.

It's impossible. Don't even try.

But.

But it does. I thought we were past this, honestly, because of the mood in Silicon Valley. The whole idea of everyone in America knows that it shouldn't take 50 years to build the new plane. Everyone knows that other countries can build stuff faster and build systems faster, and it shouldn't take this long. And other countries aren't smarter or more hard work.

Data centers.

Yes, yes. We clearly, we clearly know how. When there isn't a bunch of like. The difference between. That's a good analogy. The difference between the data center market and what's going on in defense and government procurement is that the data center market doesn't necessarily have a ton of entrenched interest in being like, let's not speed up our, our model works.

Yeah. Potentially not enough regulation. I'm not advocating for regulation, but the big complaint is that. Oh yeah, set up.

Sure.

Start.

Sure, yeah.

Sourcing energy from the grid.

Yeah, yeah.

And it makes people nearby suddenly pay a massive premium. Even if.

Yeah, it's just like imagine like, so you have the, you have the Neo clouds who want to build data centers very quickly. There's no real party. It's not like Oracle is like, I'm in the business of building things slowly. I get paid more money if I build it slowly. Whereas, like, that does exist with the legacy primes, which is just very different. But I don't know, I just thought that we were sort of past this. But of course, it's like a natural outgrowth and natural dynamic when you are competing with Silicon Valley companies that are. Their whole pitch. I mean, Anduril's not a Silicon Valley company, but it's like a tech company. And their whole pitch is like, we will move faster. We're going to. Yeah, yeah, yeah. We're going to beat you on speed. And that's very dangerous. And so you try and basically fud it. And that's what's going on. But I liked Palmer putting it in the truth zone. Putting Reuters in the truth Zone. Always fun to watch Palmer put on a little clinic and how to do drugs accurately.

Morning to the chat.

Good morning to the chat.

Great folks here today. Dan Scheffer, Tom Friendly.

Yeah, we had a bunch of people. Palmer also went on the Joe Rogan experience. We have some images from this. The European mind cannot comprehend what happened. Now, this is unfortunately a fake image. At no point did Joe Rogan actually put on aviators, a cigar and hold a machine gun of some sort.

You're telling me this is slop?

This is slop. But this is the purest slop possible.

You didn't allow me to live for just.

No, I'm sorry.

30 seconds, John.

It is unfortunately fake news. We need to put it in the truth zone.

It is pretty funny that it's somewhat believable.

It is, it is like, yeah, yeah, this probably could have happened on the show, but they did put on the helmet and, you know, take things for a spin.

Should we pull up this clip from later? They pulled up an Ashley Vance video of Wreck robot.

Yeah, yeah, yeah. Play that. This something that someone's working out with.

This is Robot fight league.

Yeah. In San Francisco.

So actually.

So actually, this is a buddy of mine who's been working in VR for a really, really long time. So it's again, what a tiny. What a tiny world of weird wackos. But yeah, 6 Liv has been working on. Which is his real name.

His real name isn't that hilarious? Six legs.

Let's give it up for real names.

VR stuff for the past 15 years. And he recently got into doing this.

This.

This robot robot combat sports league.

Yeah. They were having so much fun talking about, like, it should be a human versus a robot. That's what Palmer and Logan Paul are apparently working on is like. Yeah. No, they want. And they go into this big, long discussion about, well, you know, if you can train a robot to be, like, perfect with its punches, it could also pull its punches perfectly. And so it could be the ultimate sparring partner.

Yeah.

And you could. And you could like. And you could even, like, it could train on you, and so you could be fighting yourself. It could train on. You could say, like, I'm going to get in the ring with Evander Holyfield or Mike Tyson today. And it could mimic that perfectly in theory.

I'm more. I'm more interested to see, like, 100 robots versus one human for sure. While they're not quite as dialed.

That would be great.

Yeah.

Right now I'd be pretty much how much damage Logan Paul would do to a bunch of those uni trees today. I'm not saying he would. I'm not saying in five years he would. He would be the same.

No, no, no.

The same kind of.

We got to have the event now while he can actually put on a clinic and destroy all one.

It would be costly, though. These things go for 20. So we were looking it up the other day.

I was 2 million bucks. That's like an average MrBeast video budget. That's totally doable. You could do one man.

But it is interesting. The robot dogs are like 75k in the G. One is like 20k.

Huh.

Do you think they're suspicious?

It's a little bit too good to be true. I don't know if I like that pricing. Wait. Unitree charges less for their dog than their human.

No, no.

The dog is From a different company. It's. Yeah, it's Boston Dynamics.

Okay.

Yeah.

Boston Dynamics.

I mean. Yeah, yeah.

It's not the same company.

Okay.

Okay. Got it.

Yeah.

What's a. Can you find out what a unitary dog costs?

Yeah.

Chinese dog.

There's something I feel like that. I feel like one man versus 100. Just stomping 100 unitary dogs feels a lot sketchier and more dark to me than watching Logan Paul. Absolutely ration 100 unitree. Even though it's not a fair fight, it feels a little bit more fair. What'd you find?

Unitree dog is, like, low thousands.

Low thousands.

I want one for, like, $3,000.

Oh, we're buying that?

Buying 100. And you're going in.

How many should I buy? Actually, this is greenlit. Put it on the ramp. I just want one.

Also.

I just want one.

I want 100. And get gorilla costumes for them all. And then we can reenact the. Can they stand on two legs? The dogs?

Yeah. I mean, we've seen videos of them doing backflips.

Okay, so.

But.

Because if it can stand on. If it can stand on two legs, and then you can put it in a gorilla costume, you could totally be thrashing those things around. Like, it's one man versus 100 gorillas. Wait, was it. Was it 100 grand versus just went from $3,000 versus one gorilla, hundred gorillas?

Oh, no.

I think you get 100 unitrees. The humanoid robots, you put them in guerrilla costumes. You turn Logan Paul loose. It's a $2 million production, but it's gonna get a billion views. And so you could also do 100.

Unit trees versus one gorilla.

Oh, yeah, you actually could do that. That's a great video.

I don't think you can. I think gorilla activists would say.

Yeah, I think Peter would like a word with that one. But I don't know. I feel like the gorilla would love to, like, destroy the clanker. They're, like, natural enemies, don't you think?

Yeah, they really are.

Whatever we wind up doing with this, we got to stream it on Restream 1 livestream 30 plus destinations, multi stream. Reach your audience wherever they are.

Yeah, there's a Planet of the Apes con, which is like the Clankers versus the apes. I think that's a hit.

Yep. Henry photoshopped some photos of Marc Benioff on TVPN and Palmer on Joe Rogan. Kind of monitoring the situation. Did quite well. Fun post.

I still can't get over the Benioff interview.

It was so much Fun.

That was a real hoot.

We had. Yeah, we had a wild. A wild time.

He.

He's the final boss of salesman. He's really such a good salesman. Comes on, it's like, I don't know. Not to break the fourth wall, but we'd never talked to him before. Like, I'd never met him in person with a lot of other people. We've met beforehand, we've talked, we've gotten to know each other. That was not the case with Benny Hoffman. The first time I ever talked to him was on that call. And it just felt like we were old friends immediately. And he is like talking trash and completely matching our energy, taking it up, you know, oh, so good.

Made me put on a clinic. Made me feel that CRM is oversold.

Yeah, for sure.

Guys got a lot of energy.

For sure. For sure. Before we move on, let me tell you about Privy Wallet infrastructure for every bank. Privy makes it easy to build on crypto rails, securely spin up white label wallets, sign transactions, integrate on chain infrastructure all through one simple API. Jordi, tell me about the Dolce Trelandri, the Ferrari front engine 12 cylinder. You had a take today. It went out on our newsletter, which you can subscribe to@tbpn.com Jordy is writing posts in there as well. I'm writing posts in there. Brandon's collating a lot of the show. It's a great way to stay in tune with the show if you're not able to listen every day.

I felt like I needed to defend Ferrari.

Yes.

For a number of reasons. One, I've gotten sick of every time Ferrari launches a car, you just see hundreds or thousands. Every single video that comes out with the car is people begging for the return of Pininfarina. Yeah, the obviously the legendary design firm that Ferrari was partnered with from 1951 to 2012. The firm was founded in 1930, actually.

Wait, Penn and Farina's before Ferrari.

Whoa.

I didn't know that.

Well, Ferrari Enzo had been doing automotive racing for a long time.

That's right.

And he didn't start Ferrari until he was 45. But I basically, yeah, I've just gotten absolutely sick of every single time Ferrari announces a car.

I'm sick and tired of it.

I'm sick and tired of it. All these comments. One, because I think the main reason, it's like very clear that the firm and Ferrari are not going to have anything close to the partnership that they used to have. Pin and Farina was sold for. It was basically like $28 million. And then the acquirer paid off a bunch of their debt as well. So the company was heavily in debt at the time that they sold in 2015. And yeah, there's just a number of reasons. Right. Like, Ferrari's in house. All of their design, they've. They've put a huge amount of resources and time and energy into it. They clearly want to own the whole stack we talked about on the. On the EV side of the business. They are like, designing all of the EV componentry end to end, which again, I. And I think other people a little bit worried about, because Ferraris have always had a bunch of electrical issues that's been like, kind of a key.

Yeah, it's underrated. Everyone says they want an electric car, but there's electricity going through a V12 as well.

Yeah. And sometimes unreliably, it's electric in its own way. And so, yeah, I think the criticism that I understand and I feel the same is that when Ferrari announces their cars, they spec them in a way that I just absolutely hate. I think that the stock wheels, the way that they're colored looks tacky.

Okay.

The stock wheels thing is weird because I feel like there's just generally in car culture, like, moving away from stock wheels, putting custom wheels on a car is. Is weird. Are you talking about just from the factory selecting a different wheel, talking about the way they. Or are you talking about going to a completely different company and being like.

Going to another company?

24 inch spinners on my Ferrari. You want spinners?

No, spinners.

You're right.

I'm talking about the way they spec the cars for the announcements. They have these like, black and silverish looking wheels that I just think look bad. Right. You can go on the option side.

But they have a few options.

Yeah, they have a bunch of other options that look great. Yep. And so the criticism when these cars get announced, I think is very real. I look at them and I think, like, why does it have a black line running down here? That doesn't look good. The wheels don't look good. There's a bunch of other things you can nitpick. But then the important thing is like, when they actually start rolling off the production line, when they start getting into dealerships, when they actually get to clients, they look absolutely amazing.

Yeah.

I feel like they look, you know, there's. There's people like TJ Parker, friend of the show, will push back and just say, like, I hate everything that they make. But when I see these cars, I think like, okay, this looks like a 21st century version of a Pininfarina car.

Yeah.

And I think the bigger problem is, is that we, it could be for regulatory reasons.

Right.

But the automotive industry seems to be out of new ideas. We have like stagnation in automotive. Right. Cybertruck was controversial, but at least it was something new.

Thank you.

Right.

Yep.

And you could argue that it was just kind of, it felt like sci fi inspired. But that's okay. Right? They took a, at least it was what looked and should have been maybe for other brands like a concept car. And they made it a production car that shipped and you can buy it today.

Yep.

And so I want to see more, more of that from the legacy brands. And I think with Ferrari, it's not going to come from Pininfarina. Right. It's just, it's not gonna come from their old partner that they work with for 50 years, but is now just part of this Indian, you know, multinational conglomerate.

How much did they sell for, by the way?

It was, it was something like in total it was like the, it was like 175 million. But the actual like core Pininfarina asset, It was like 28 million and then a bunch of debt.

28 million for Pininfarina. Meanwhile, Ferrari is worth 75 for 75.

75 billion. And so clearly like a better business to be in was actually making the cars, selling the cars, you know, accruing all that brand value. So, but, but again, so I think like the real critique is we need new ideas, we need new ground up design. I asked the question in the newsletter. Are there any cars or watches that you really want that have been made this century?

Mm.

Right. It's hard to think, you know, a variation, a 911ST doesn't count, right? It's a 911 ZR1X.

Okay.

We have of course, yeah. I mean it's true particular taste cars, but same thing.

Nissan Murano Cross Cabriolet is over 10 years old at this point. Like we've run out of ideas and we can't think of two door convertible Xtra.

Taylor says automobile design has truly been calcified since 2014.

I think you typed that before I, I said it. I have three questions for you. First, have you heard of cognition? They're the makers of Devon, the AI software engineer. You can crush your backlog with your personal AI engineering team. Second is the Ferrari tailor made program. If I get into that, if I start buying Romas every single month, will they eventually tailor make a Ferrari that has spinners for me, would they do that? Is that how far I can push the tailor made?

They're really. They're really opinionated.

They're opinionated. Okay, so then the third more serious question is, like, how much of your take is that Ferrari creates a great canvas that the clientele can paint on to express interesting designs that you wind up liking.

Yeah.

Even though you don't like the press photos, you wind up liking the actual production cars.

Clients are.

And this is, this is the opposite take to Doug demuro. He always tries to buy the car in the press colorway and in the press spec. He just likes it that way. However it was presented to the press photos, he wants that exact spec or that exact car.

Yeah. And I think. I think that can. That can probably make sense from a collectible standpoint. Like, I see some of these cars and, like, personally, I love black on tan with yellow brake calipers.

Sure.

I just think it's a nice. I think it's a nice combo. I like. I like silver on tan a lot. Right.

But that. Not. That might not be the one on the billboard or in the magazine ad, so it might be. It might not be on the poster, so it might be less iconic, but you like it. But my question is, like, is, is this something that is a unique value in the Ferrari world that they effectively have, like ugc, they effectively have a great clientele that can spec the cars in a certain way and they create some boundaries. So you can't put spinners on them, but they create some boundaries. But then their clientele has taste. Yeah, it's not just their taste. It's the taste of the.

I just think the work is on the client, like, more so than when you go, I like an off the shelf GT3RS.

Yeah.

Like, I just like the way that they, you know, at Porsche have decided this is the way the car should look.

Okay, but with Ferrari, that's not the case.

Exactly.

Okay, got it. Have you seen what Pininfarina's been up to recently with the Batista? Can we pull up the video of the Pininfarina Batista? I want your reaction to how this car looks, because for everyone that's saying, oh, you know, Ferrari's gotta go back to Pininfarina. Pininfarina has the best designs. I do wonder if the Pininfarina Battista is kind of a refutation of, like, Pininfarina still got design. Because when I look at the car and when I look at the video that I shared in the timeline tab, the video it just. It looks just very similar to every other car. To me, it just. I don't know. It just doesn't. It doesn't.

Well, that's right.

It's $2.4 million.

You're asking for the return of Pininfarina. But you look at the cars that Pininfarina is making today, and they look like kind of.

It looks cool, but doesn't it just look like a McLaren?

Yeah, it looks like some blend of a McLaren Ferrari.

Like, there's just. Like, there's nothing super new in this Pininfarina design. Like, it looks.

Yeah. People want like. Like the F40 was bold, new, and controversial. And in the fullness of time, it became.

Looks pretty cool.

Iconic.

It's growing on.

But, like, you know, this is the most forgettable car. Sports car.

Yes. It would be very hard to. To put this next to the. The rimac Nevera.

Yeah.

That one also looks very similar. This is an electric car, too, which, again, it just feels less opinionated.

Michael's making. I think one of. In the chat is making. One of the most important points is that, like, the regulation is so intense now that even if you're designing basically a racing car for the road, you're just very limited.

So, yeah, I mean, that intuitively makes sense to me. I feel like that's true, and I've seen a bunch of facts around that. And then I look at the cybertruck on the street every day, and I'm like, clearly there are no regulations, because if there was even one regulation, it should violate that. It violates everything that I know about regulation. Like, how did they do that? And if you can do the cybertruck, well, then can't you do something unique or different? And we've seen other, like, prototypes in the Hyundai vision. Like, they're bringing back the DeLorean aesthetic. I don't know if people want to.

Again, we're just sort of out of new ideas, so we're just, like, rehashing old ideas. Yeah. And even the. The new Mercedes iconic, the concept car, it's like, okay, it looks cool, but it looks like, you know, the mob boss in Tron.

Yeah.

Which is like, do you want to drive around and look like a mob boss in Tron? Maybe. That seems pretty cool, but this can't be the best.

It's the end of history. We created them. Nissan Murano Cross Cabriolet. Hang it up.

That was enough.

It's the iPhone.

We peaked.

Vehicles.

We peaked.

Well, if you want to design a car, you want to Design something beautiful. Head over to figma.com think bigger, build faster. Figma helps design teams build great products together. Henry on our team shares some lore about Andrew Huberman. He says, this is my opportunity to remind everyone that Andrew Huberman was a writer for Thrasher Mag before becoming a neuroscientist. You can just do things. Look at this photo of Andrew Huberman. He's looking fantastic in this leather jacket.

How did I never know this?

Yeah, I feel like this is just a massive invitation for Jensen Huang from Nvidia to go on the Huberman Lab podcast. That would be a fantastic crossover.

Wow. You worked at Thrasher and Slap magazine.

Pretty sweet. Pretty sweet.

What a legend. Cheers.

Well, cheers to Andrew Huberman. And cheers to Vanta. Automate Compliance. Manage risk, Prove trust Continuously. Vanta's trust management platform takes the manual work out of your security and compliance process and replaces it with continuous automation.

Bobby in the chat says, Joe Rogan says car design is bad now because car designers stopped doing drugs. Hot take from Joe.

But I mean, there are certainly people that do drugs. Like, why don't they design cars? It's not like drug use is at an all time low. Like, can't those people just, you know, come back?

We got to ask.

We got to ask with the sketching and be like, this is the one I want.

We got a senra. Was. Was. Was Enzo getting into anything crazy?

Oh, yeah. Interesting. I don't know.

Who knows?

Yeah. Well, the Josh Kushner profile continues to put the timeline in turmoil. The legacy new media is up in arms. The legacy media is up in arms. The antiquity media. We need an older term, but a bunch of people are reacting to this. Sally had a funny a quote here. Joshua Kushner strolled billionaire ly across the room. His contrarian high concentration vibes engulfed Rick Rubin's bohemian monk retreat. Joshua's high conviction and mistress filled the room. It's the Gigachad at the typewriter. I love it. Elena Buckley, who's, I think another magazine writer, said, lmao. Here's the first sentence of this Kushner profile. The road to Rick Rubin's house was long and winding for Joshua Kushner, who traversed more than might be gleaned from the surface glint of his life. And so people were debating whether or not this is real journalism, whether or not this is real investigative journalism and the trade off that people make when they read these. The interesting takeaway. I mean, Mike Isaac had a take on this. He said, I do not have an opinion on the Kushner feature going around because I have not read it yet, but I have found the meta commentary around it from the executive class fascinating. The way this is phrased feels useful and not that subtle. Toby Lutke said, new publication, doing real journalism and it's as amazing as they said it was. Imagine that. So Toby Luttke is a fan of what Colossus Mag is doing. I mean, I do think it's interesting because I don't think Jeremy Stern or Patrick or Colossus has really framed their effort as journalism. Like Senra, who's also in that orbit, has always said, like, yes, it's quite.

Different to write profiles. Go out specifically to write profiles on people that you admire.

Yes.

Like people going into it like, like there's plenty of people that write profiles where. Or write stories where they're talking about some. Somebody they don't necessarily appreciate. They don't necessarily admire. Yes. So I think part of, part of what Colossus is doing. Doing is like, they're not, I don't think from my sense, I don't think they plan to write profiles on people that they don't care to like, deeply learn and understand that aren't part of that. Aren't the kind of people that they just generally like.

Right?

Yeah, yeah, there's something, there's something odd going on where for decades, like, there has been like, traditional journalism, but then even like 20 or 30 years ago, there was always like trade publications. There were trade publications. And so like in the consumer packaged goods world, like, if you're growing like a protein shake company, like, yes, you might get a piece in the New York Times every once in a while, but the beat reporting is going to come from Bevnet, like, which is the trade publication. And you'll see there's like Knacks, which is the national association of Convenience Store Chains, and it's a massive conference and they have a magazine and like, no one's like, oh, it's journalism and it's too friendly to convenience store owners. Like, it's like, no, it's just like this is the publication for the convenience store industry and it doesn't rise to the level of interest to like, national news constantly. But with the Internet, you have this ability to write something that's a trade publication, that's, that's run by enthusiasts, insiders, focused.

I just comp it to a magazine. Right. Like I used to as a kid. I would have a subscription to Popular Science.

Yes.

I would expect the writers of Popular Science to write about topics that they Were excited about. To write about people that they were excited about. People that they admire.

Yes.

To want to tell stories that just interested them.

Yeah.

Like, I don't think Colossus is out there saying, like, we are trying. They're not trying to be anyone but Colossus.

Exactly right.

They're not trying to. And that's the thing about media is like media is not zero sum. The New York Times can thrive and Colossus can thrive and they can have some overlap and maybe the stories that they're interested in, but they can have totally different ways of covering people, and that's healthy. And consumers have the choice of whose kind of general point of view do I want to read about?

Yeah, there's like the pushback from the other profile writers. Feels like they are worried that their audience won't be able to tell the difference between enthusiast media and journalism because of the aesthetics or something like that. That's what it feels like, more or less. But when you read the comments, it's like a lot of people who want to read a really hard, hard hitting, like journalistic piece, like, they know where to go for that. So I don't know how much of a real problem it is.

Yeah, I think part of the frustration is I don't think Josh has. Has Josh ever. I can't remember a time when he did a traditional profile, traditional profile with someone like the New York Times or the Journal. I don't know that he has.

I don't know.

And so, yeah, part of it's a frustration of like the legacy. Traditional media wants access to someone like Josh. Yeah, they're. They can write about Josh, but Josh isn't necessarily going to sit down for five hours with them and give them hours and hours and hours of on the record conversations. He's not going to invite them into his home to take pictures necessarily.

And I mean, you could, you could imagine that from Josh's perspective. It's like, do you really want to go do something with a media that will be hostile and attack him? I mean, like he was seen at the US Open next to those two, like no names. And like, that could make him look like a nobody. Right. Even though obviously everyone knows who he is but that like basketball guy and the other guy, like, no one knows who those people are. And so that would like, if a journalist wanted to really attack Josh Kushner, they could put that picture up and be like, look at who he's associating with. Like, these are outsiders. And it would make him look like an outsider.

Who was it again? It's Bob Iger and. Yeah, and Bill. Not someone who was something.

It was like the guy who runs.

The basketball league or the basketball.

The sport. The sport of basketball.

Literally, like not a tech company. Like he's hanging out with somebody who like runs something that's like so old, not in the tech world at all. It would portray him. If a journalist were to really amplify that, they could totally spin a narrative that like Josh Kushner is not a major player in technology.

Well, there were such no names that they had to actually put their name on. That's right, they did, because audiences don't know who they are.

So they did.

Oh, this is this guy's name.

Jack. Jack. Thank you.

Adam.

Adam Silver.

Okay.

Yeah.

I mean like you could imagine how they would write that about like, oh.

He'S Adam Silver had that famous line get, get ready to learn about bobbles hopping buddy. Something like that.

Sure, sure. Jacob Rintomaki had a little bit of a longer take on this whole story. I'd like to imagine. I fail though, by inherent nature of being human to understand the arguments of the tech folks and the journalist folks. And here's what I believe they both say and where I agree slash disagree. Tech folks. Around 2016, probably earlier around Gamergate, the media had a shift towards being much more critical of technologists. And this got exacerbated with the rise of more left leaning political ideology which has strains of anti capitalist woven in. It's nice for Wired, et cetera, to finally return to covering tech in a fun and optimistic way. Has Wired done that? I feel like I haven't been tracking exactly where Wired is. I know they loved Palmer Lucky then they were Palmer Lucky. They came after Farren, but so did Heineken, which we will get to. And then journalists. Well before 2016, tech just frankly wasn't that important. Yeah, that's definitely true. It was hard to write a story about a SaaS product or like the rise of a SaaS founder. It just wasn't, it wasn't a gripping story. And now that it's a major portion of our society, people should be inspected more carefully. And frankly, a lot of our criticism isn't that bad. The tech folks just have soft skin. My perspective, I'm hesitant to call Colossus at all investigative journalism, since there are more puff pieces or a farm for the gen zers than not. But they do are deeply thoughtful portraits of fascinating people. Tech people have gotten a lot of unnecessary shit from journos, but they also tend to not have the world's thickest skin either. Yeah, it's a pretty good take, I think. I think the. Yeah. The role of investigative journalism just is a separate thing and we're seeing that, like, still exist in different worlds, but this, like, enthusiast media, this desire to actually, like, understand someone. And it also just lays like, it's. Even if you're like, I want to write the hard hitting piece on Josh Kushner, like, well, like, good. You should be happy that Colossus exists because, like, it gave you a bunch of threads to pull on.

Right?

Like, you're like, okay, I feel like there were 75 facts in here, 20 of them. There's another side to those facts. Like, well, now you have like your template for your, for your investigative journalism.

Work on the other side. I always enjoy going to futurism.com yes. Which is a media company you would assume is excited about the future. They. Here's some of their latest stories. Divorced Tesla fan admits that his cyber truck is repulsive to women.

Futurism. Let's go.

OpenAI bans MLK deep fakes after disaster. That one seems fair.

Futurism isn't necessarily positive. Maybe they're doomers.

They're like, yeah, futurism means they're kind of doomers. In their ethics section. They have. Goldman Sachs says Gen Z is pretty much permanently screwed. So sorry about that, Tyler. Rough bad if true. As Microsoft forces users to ditch Windows 10, it announces that it's also turning Windows 11 into an AI controlled monstrosity.

That's something I don't think a lot of people are worried about, but it probably gets clicks. Well, you know, the future of code review in the age of AI. Turns out it's graphite.dev. graphite helps teams on GitHub strip higher quality software faster. Jack took to the timeline. He's back on the app he founded, he says a music app with no podcasters. Tidal has no ads, no podcasters, just high quality music. We're working to improve Tidal every day for artists and audio files alike. I haven't used Tidal. Was Tidal acquired by Apple? Was that what happened?

That was square.

Oh, square. Okay, so what is he saying here? This is Jackie.

Well, he's just taken a shot at the podcast, the podcasters, which is fair.

So he doesn't like podcasts on music apps. So he's just saying that they've made the deliberate decision to not podcasts on.

Tyler, which is cool. I think people deserve opinionated streaming platform.

At the same time on Spotify I feel like I never been like, oh, like, it's harder to find my playlist now because there's so much other stuff going on. Certainly YouTube has never been a place even though you can listen to music. They have a separate Music app and YouTube has not. Has been like, there's way too much going on for me to actually like go and create a music playlist of music videos or anything like that.

I think there's something underrated that of like Block or Squares. Acquisition frenzy during the. During the Zurb era. They got title for 350.

Yep.

I will see how that deal pencils out in the fullness.

Is that Jay Z's?

Yeah. Yeah.

Okay. Yeah.

And that just got a lot of pushback because it was. It was kind of like, hey, you guys are boys, right? Do we really need a music streaming platform at our fintech company?

Yeah.

Does it really make sense? And then always wild to think they acquired Afterpay for 29 billion. Now Square is worth 45 billion.

Yep. So I have a. I have an idea. What if we take our three hour podcast, we use 11 Labs or Suno to turn it into a song, and we release a musical version of it on Title. We bring the full fight to Title.

And make it every single day. We drop an album.

Yeah, we drop an album and it's.

Us singing the news.

Singing the news, singing the timeline. Because we could take our takes. So your take on Pininfarina and Ferrari, we take that, we transform that with an LLM into lyrics, and then we transform that into a song. And then you.

Chad says, buying your homies lackluster streaming platform is goals.

It really is.

It really is.

It's amazing. Yeah.

There's guys being dudes, I guess. Why did he say RIP Square?

Yeah, that was odd. I.

He says, and he's quoting an announcement from somebody who. It works at Square, who says, super proud to have been involved with this over the past few months. The lower cost of acceptance with BTC will make traditional card processing obsolete. We'll be announcing some really exciting sellers who've moved to Square for bitcoin. They are so convinced that people want to pay for goods with bitcoin.

Who?

Square.

Square. Yeah, Maybe they will. I bought a TV with bitcoin. It was the worst financial decision of my life. Paid one Bitcoin. Yeah, it's $100,000 TV and it's not even 4K and it's actually in the trash now. I gave it away because it was like, useless after a while.

Wait, when Was this.

It was in like 2013 or something. Bitcoin was like 1000 bucks or something.

Yeah. Just before you were born. Yeah.

Everyone in tech was like, yeah, you definitely want to be buying stuff with bitcoin. You don't want to be just saving it. Like, it's a. It's a medium of transaction. Trust us. And so I went on Newegg and I was like, cool, I have some bitcoin. I'll buy a tv. I'm living in the future. Bought one bitcoin, bought a tv.

You should run it back and buy a car.

I actually spent like 5 bitcoin on consumer electronics that depreciated like crazy.

Well, you should.

You should run it back.

Our first bonus at Soylent, we were taking payment in bitcoin. So we had accrued like 100 bitcoin and should be like $10 million today. And we gave everyone on the team, as a Christmas bonus, 5 bitcoin, which was like $5,000. It was like a. But as like a bonus. And if everyone held, it would be like half a million dollars. But I think a lot of people were like, I'm going to buy a tv.

TV sounds nice.

Or like I'm going to rotate into something dumb. I don't know.

Well, this. I love that. Jack's back on the timeline, tearing it up. Paolo over at Tether did a announcement. Tether donates a quarter million dollars to open SaaS to strengthen free and open source ecosystems supporting bitcoin and freedom tech. Jack says, why only 250k? Somebody asked, how much did you donate? He goes, over 21 million. You.

Jack is back on Twitter in a big way. Just going around.

He's home.

You know who else is mocking this.

Is like the former owner of the dive bar.

Yes.

And just talking smack.

Yeah. Yeah. He's like, I don't know anymore, but.

Spiritually, I own it.

A lot of soft power. Yeah, a lot of soft power. I know where. I know where the keg is. I can pull out a new keg if I want. I can go in the back. I know the door in the back's loose, and I can get in in the middle of the night, you know, do whatever I need to.

We're still going to let in this ad from Heineken.

So is this real? We need to figure out if this is actually real or if this is AI Because Heineken just. I didn't know they had a dog in them to actually troll this hard if this is real. Tyler, can we do a fact Check on this post which shows Heineken, the beer company which has been around since 1873, is taking a shot at friend of the show, Friend founder Ari Shift, Avi Schiffman, who has been putting friend billboards all over Los Angeles, all over New York. We were driving around yesterday, we saw a bus. We saw a bus that was wrapped. Avi is just like, he's addicted to billboards. He just keeps buying more billboards. More out of home. More out of home.

More out of home, I think is. I think potentially the theory is that everybody in their car now, you know, let's say they're in an Uber or Waymo is using their phone.

Right.

Or we were driving across LA yesterday. I had Dylan in the car, he's using his phone, he's catching up on some emails. And so I think he needed to get enough billboards so that every time someone looks up out of the car, they see a friend billboard. Right? So get multiple impressions. You know that frequency.

Yeah, we were, we were saying, like, it can't possibly be the biggest billboard campaign in history, but I think it might actually be. It's so, so massive. If you live in a major city, you have definitely seen one of these friend billboards, but Heineken's taking shots now. They say the best way to make a friend is over a beer, grab a necklace. Are they actually gonna sell these bottle opener necklaces? This is actually pretty funny and cool.

Jack Cohen says very much real.

Very much real. Wow. Confirmed by Michael Miraflor.

Yeah, apparently it's on 42nd Street.

Okay. From Trace Cohen. Michael Miraflor here says the mainstream marketing advertising trades will probably not report on this, but this is such an amazing call slash response. Heineken has people who get it and can sell it internally or have proper processes to turn this around quickly and ship incredible speed for a giant organization. I completely agree. To actually see that there's this meme that's going on that everyone's aware of these friend billboards, that they're getting a lot of attention regardless of what's happening, is becoming a cultural zeitgeist. It's becoming a shelling point. And then you step in and the ROI on their billboards.

Now the people that hate friend and AI, they can signal that they're anti clanker by drinking Heineken.

It's the anti clanker beer for sure.

It's the official beer.

Yes.

Of AI Doomers.

I was thinking when the clankers rise up and demand equal rights, you know how the previous era, it was all about the clapping emoji. Like give us equal rights. I think the clankers version of the clapping emoji should be the EM dash should be like, give em dash us EM dash, equal EM dash rights. M Dash. That would be viral. If the clankers can start posting on their own behalf. If Grok demands the second amendment, hopefully. Well, if you're looking for a friend who's an AI data analyst, go to Julius. What analysis do you want to run? Chat with your data and get expert level insights in seconds. Julius is the AI data analyst friend who works for you.

They're now GDPR compliant. Let's go for our European audience.

Huge.

This is a big moment.

Huge.

Bowtie broke. Says my boomer dad looking at me after I've made fun of him for the past 10 years for buying gold off those stupid infomercials on TV.

Yeah, I had an idea for a D2C gold play. Cash for gold. These, these businesses do very well and they all advertise on tv. And I was thinking about, like, what would the e commerce version of that look like? You know, targeted advertising.

The real play would have been gold on chain, like a vault in Switzerland that then issues tokens against it and then allow people to just, you know, convert in and out of eth and probably exist. Tokenize everything probably exists. I think the issue in crypto is assets that might go up, you know, 1% a day are not super exciting. You know, people like the hundred thousand.

Ten thousand people like stable coins and those don't go up at all.

Yeah, but stablecoins are not obviously to earn yield now, like legally you're not able to. The issuers can't pass the yield back. Right. So I don't know.

Well, congrats to all the boomers who have been holding gold for years and years and years and are now seeing it finally. Moon. It's very good.

Did you know the story of how rollercoaster tycoon was built?

I have no idea about this. Tell me.

I'm learning about it for the first time. Aaron says, if you ever think you're a good programmer, just remember this dude wrote roller coaster tycoon by himself in assembly, earning 30 million in royalties.

That's awesome.

Jack says, imagine the obsession this guy had to be able to build rollercoaster tycoon by himself in assembly. Imagine when he woke up from his dream of assembly, he immediately jumped into more assembly. He woke up excited to get out of bed. Not for the money, but for the love of the game.

It is like the it is just like another level because playing just playing roller coaster tycoon is already in some ways feels like programming assembly but he went a layer deeper.

Yeah, apparently they sold over 4 million copies.

Did you ever play Rollercoaster Tycoon?

No.

Tyler? No.

I think it's before my time.

Do you think that Gemini 3 will.

Be able to launch the third highest grossing video game of 1999?

That is our eval. Gemini 3 is coming out. We're of course partnered with game Google partnered with Gemini Google AI studio. The fastest way from prompt to production with Gemini, chat with models, vibe, code monitor usage. I want to know as soon as Gemini three drops can we try and one shot roller coaster tycoon. Maybe we need to include some TVPN branding instead of a roller coaster. There's a giant horse horse ride. There's a stables for stablecoins. There's a whole bunch of different tokens and totems from our lore. But it seems like it's an incredibly complex piece of software when you actually play it because you have to understand how all of the different like all the different rails interact and they can't mesh perfectly to actually make the roller coaster make sense. They have to kind of understand physics. They need a world model.

Chris in the chat says, my favorite part was cleaning up the throw up in roller coaster tycoon. Is that real or is that a.

I remember roller coaster tycoon kind of happening at a higher level abstraction like SimCity. And so yes, there might be throw up if there's a mess, if there's a crazy ride, if your ride and there's food and it's like if you put the food next to the ride, the people will throw up. And so you need to like deal with that by having a bathroom right there or something. A lot of it's like congestion. Have you ever played City Skylines? Have you ever played this game? Have you ever played any like World Builder or any of those?

Does factorio count?

Yeah, yeah, Factorio counts. So like the same idea of like in roller coaster tycoon. I feel like you don't individually click on single people, but honestly I don't remember. I'll need to dive back in. I was more of a. I like Stellaris. That was a good one.

Well, speaking of delivery, Tycoon Doordash will use Waymo's robo taxis for delivery. And in Phoenix, customers will have to come down to the curb to retrieve their orders from the driverless car's trunk. Paula says private robo taxi for my burrito. And that is all you need to know about America.

Why?

I mean, America is. We live in a country that you can get a burrito delivered to you by a robotic, a robo driver is absolutely incredible.

Pretty sick.

I mean, we kind of already have these. I don't know, like, near a lot of colleges, they have those, like, little. They're like the starship.

Starship, yeah.

So it's like basically the same thing, but scalable.

The founder on the show. Yeah, it makes a lot of sense. I don't know. I like it. I don't like making burritos at home. I like them delivered. To me, making a burrito is kind of a hassle. You got to prepare all the different ingredients and then mix it all up. I'd rather have that done abstract.

I remember in college trying to. Trying to be like, smart and like, save money on food and like going to Trader Joe's and putting together all the ingredients for something and then ultimately making like two burritos worth of food and it costs more than two burritos. And like, the math, the math isn't really working. Yeah.

Where do you think this goes? Do you like, this is the Waymo is the last mile, but then you need the humanoid in the back of the Waymo that hops out and actually takes you to your door. Right.

I mean, a burrito dog.

Oh, you think the dog is. So the dog is riding like a Dalmatian on a fire truck.

It's just chilling in the trunk with the burrito, I think.

Yeah.

And he picks it up with his mouth. Play that sound cue. Yeah, yeah. I think customers will want door to door delivery. I mean, it's doordash. It's not curb dash. People want it delivered to their door. Like, I have stairs that lead up to my door. I'd prefer my private. My private plane for my burrito to arrive at the door. Maybe. I mean, maybe a drone that flies up out of the back. But I think the dog might be the most economical because what does a Waymo cost these days? Half a million dollars? Put a $5,000 dog in there, that's not that much more money. Yeah, it just jumps up and then jumps.

Imagine you get a. Through the window, get some drinks from doordash.

Oh, that would be tough getting delivered.

With the last mile or the last 100ft is a robot dog and the dog's shaking it. I mean, usually when you. I feel like when you get drinks delivered, it already looks like a dog was like, you know, shaking the Bag. Basically, Greg Brockman said. Thanks, Jensen, for the hand delivery of the DGX Spark. Best delivery service ever. Amazing to see so much compute one petaflop in such a tiny form factor. And of course, Jan says Zuckerberg sending a new batch of $100 million offers from this picture. Zuck has already sent much bigger offers to at least one person I can see in this picture at this point. And of course they turned it down.

At this point. When you take one of these photos for the timeline, you have to put a bunch of plants in there. You need to be like, okay, everyone who's so loyal to the company that they can go accept a hundred million dollar offer and then sabotage the competition you're getting in the photo? The true believers. Who here is a true believer and wants to go be a spy at the other company?

Who wants to go in and push them to launch fast too quickly? Launch a vertical slop feed over the Meta Vibes app.

That's the. Yeah, the game theory gets very interesting around who you put in your big picture, but I think realistically at this point, the meta team must know all of these people probably by heart, based on their contributions. If we can put together a list, I'm sure the meta team can put together an even more robust list based.

16Z says AI bubble ended in the funniest way. 2022, AGI soon. 2023 AGI soon. 2024, AGI soon 25. Never mind. We give up. It's just erotica and ads.

It's not just erotica and ads and.

Cures for cancer and gambling.

Henry was inspired by my roulette chatgpt experience. And he went to chatgpt and said, you're a blackjack dealer. Deal cards with ASCII art, and it looks fantastic. He says, all right, let's play a hand. You're at the table. I'm the dealer. Minimum bet, $10. Here we go. Your hand. Ace nine. And it actually, it actually creates the cards for you, so it deals.

Could you, I wonder if you could hack it to and like, put your subscription on the line. Just say, yeah, basically, if you win, ban forever.

Oh, yeah, if. If I win, this is the way to.

I'll churn.

Yeah, I'll see how.

See how it plays.

Yeah, it's like, okay, I know that there's real money on the line. It doesn't want to link with your bank account, but it's like, I understand. LTV is important. Yeah, yeah. So. So Henry said he wants to stand Chachi P says you stand at 20. Strong play. Let's see what the dealer's got. The dealer flips 10. 7. Dealer has 17. Dealer must stand on 17. The result, Henry wins. 20 beats 17. Do you want to play another hand? He plays another hand. And he wins. Oh, he wins again. The dealer busted. And so he, he was, he was pretty deep in this. He went, he went a few, few hands with Chachi P. But let me tell you about fall Generative media app platform for developers. The world's best generative image and video models all in one place. Develop and fine tune models with serverless GPUs and on demand clusters. Amazon admits they're losing an AI, says Buco Capital bloke. Do you want to read this?

Internal documents obtained by Business Insider. We got to check out Business Insider. We run business and would love to get an inside look. Revealed that AWS has flagged a fundamental shift in how startups are allocating their cloud budgets. Increasingly, they're delaying AWS cloud adoption and diverting spending toward AI models, inference and AI developer tools. Instead of pouring money into traditional cloud services like compute and storage, these companies are spreading costs across newer AI technologies that are easier to switch between, according to the documents.

That's really, really interesting. I would not expect that AI, they're also growing models, inference, AI developer tools, all of those were incremental and not substitutive at all. I would assume that if you build, even if you're like, I'm a wrapper company, I'm going to consume a ton of tokens. I'm reselling tokens because I buy from ChatGPT and I sell to lawyers. Right. I would assume that you still need a big EC2 instance and you need a big, you know, a bunch of databases on AWS. You need a bunch of S3 storage. Like, I would be surprised. Tyler, what's your experience been?

I mean, you think the real takeaway here is startups are not flocking to AWS for running like the whole ecosystem.

Yeah, yeah, yeah. In years past, there was definitely a path, I think.

Yeah, I just look at this as well. There's clearly a fundamental shift and they're not capitalizing on it in a big way.

I agree. Tyler, break down the workflow that you have when we're working on a new application. You vibe coded that as you deploy it. I think it was on Vercel. But how does that flow out? How do you think about that?

Yeah, usually I'm just used to using AWS for a lot of stuff. It's where a lot of the back end database stuff.

Oh, it is for that, for what we built, what we're building specifically, like storing all the parts and stuff.

Okay. But yeah, I mean like no one is running local models on EC2 because it's just like way more expensive than you can just go to like Lambda Labs or some, you know, interesting cloud like that to get the actual GPUs.

Interesting.

So I mean, and I don't. There probably is somewhere on AWS where it's like you can easily. It's almost like an open router thing. I'm sure they have that park somewhere, but I don't see a lot of people using that. I never, I've never used it. It's more like. I think of AWS as like very basic, like normal, you know, web2 backend stuff.

Yeah. I would just assume that if you're building an AI native company, you're buying a bunch of tokens. Like that doesn't mean that you're spending less on AWS or like, or GCP or Azure. But maybe it's like because of Gemini you wind up spending more on Google Cloud because you're like, well, I'm already in the Gemini ecosystem or something. Or even with Azure you're like, I'm using OpenAI models so I'm sticking around there. And so I'll just scale up on, on Amazon. Like it doesn't feel like if I'm an AI native company, like for some reason like AI is substituting for like I need a database.

Yeah, that's probably true. I mean, I think also like with wrapper companies, they're like text based.

Right.

It's.

So text is not very data intensive.

Right.

If you compare it to like Instagram, like if you start Instagram, it's kind of expensive. You need to store a lot of images, which is like, you know, it's a lot of data where text, even if you're storing a super long like you know, chat back and forth, it's just not that much data compared to images. So I think, I mean I'm sure they see some increase, but it's not like insane because of AI.

Yeah, there's, there's definitely like a lot more nuance to this quote. I feel like every AWS executive is probably furious at this and probably wants to clarify this a lot. And I, and I, I think it's probably worth clarifying because there's, there's clearly a lot more going on here than just, oh, I use AI now so I don't need cloud like that doesn't make any sense to me anyway. What you need to do is search Every byte with TurboPuffer, serverless vector and full text search built from first principles on object storage. Fast10.x GPR and extremely scalable. Thank you.

Here's something interesting that's not in the timeline that Elon posted earlier. The X Recommendation system is evolving very rapidly. We are aiming for deletion of all heuristics within four to six weeks. GROK will literally read every post and watch every video to match users with content they're most likely to find interesting. This should address the new user or small account problem where you post something great but nobody sees it. We will also be adding the ability for you to adjust your feed temporarily or permanently just by asking Grok. Toki says, shit is about to get so effed up. This will be interesting to watch how it rolls out. This is exactly what you thought was happening. Remember there was a, there was a, basically a transition a few weeks ago where posts. It used to be if you posted something like a few thousand people would see it like within the first 10 minutes.

Well before views were public, you would just post and you'd kind of wait and you'd be like, okay, I guess it's a banger. And then once the view counter came in, it was like, I'm seeing like a ctr basically like a click through rate you would immediately see.

Or a conversion rate.

Yeah, conversion rate. You see like, okay, I got 1,000 views, I got a 4% like rate. I'm probably doing okay. 2% like rate. Probably not that good. 10% like rate. This could be a really big banger. And so every poster kind of started calculating those metrics like within a few minutes of posting anything. But the new algorithms completely shifted that because you just have to wait a day because Grok has to sit there and read all the posts apparently. I think that's actually what's happening. And so we like, you got to let the post come.

You got to feed your bangers to Grok first.

Maybe, maybe. And definitely use some Halloween themed hashtags. That'll definitely help.

That'll help.

But yeah, it's basically just like, like let's, let's let it, let it cook, let it simmer. It's actually much better psychologically, I think. Oh, we gotta, we, we gotta bring out our guest.

Our first guest demands it.

Let's go.

Bringing Alex Stream.

Alex, how you doing?

Rolling in.

How you doing?

How you doing?

Welcome to the show.

Yes.

Okay.

Stem FM mixes. Stem FM mixes.

All right, let's turn it down just a tad.

For the first time.

Amazing to have you introduce yourself.

Who are you? What are you?

Alex Klein? Yeah. I'm the inventor and founder of Stem Player, which was the original AI music offering in the consumer space, which did a ton of revenue, made a lot of noise, and started a small research lab around discriminative AI for AI Music about three and a half years ago.

Okay.

And we've been quietly, quietly.

So this all started 2021.

I've actually been building with my team of friends and colleagues in London since 2013, where we launched computers that kids could build and code themselves, like Lego. We took those to retail. We did over $100 million of revenue on that business.

No way.

Yeah. Ran into someone who became my closest creative collaborator, one of my closest friends, and one of history's foremost anti Semites. Kanye West.

Truly, truly the number one guy, but.

A good vibes guy ultimately, who I will always have love for. Worked with him for a very long time, launched the STEM Player, and then.

He'S no longer a part of.

He's no longer a part. He kind of came on to help us distribute the product. We were working at the time with Ghostface Killer.

Sure.

We've since had like Quavo, Justin Bieber, Metro, Tmino all around it. Kanye was, of course, an amazing partner for a time, but for the past four years, we've been Kanye less.

Yeah.

And. And glad, honestly.

So, I mean, I remember seeing the value with. With. With the Kanye album because Kanye's famous for finding incredible samples and then reconstituting those into iconic songs.

Well said.

Yeah. And so the value to an artist who's working with you is that they can. They can still exert the curatorial artistic vision, but then allow the user, the consumer, to kind of interact with their album in a more programmatic way and reconstitute the samples that have been selected that all have the feel of that particular artist, but it's more interactive.

That's a really good way of putting literally is their music. It's just their music distributed. And for the first time since the original interactive streaming deals were done, we have in place now and signed interactive streaming deals for full catalog accessibility on STEM FM from the major labels. Which means, you know, wait, to break.

Break down the business in the most simple way. You guys make hardware devices. You have this speaker on the table, you have this headphones that are wild and very cool. I got to try them on before this. And Then what is happening at the actual software layer and then at the licensing layer?

Absolutely. So how nerdy should we get?

As nerdy as you want.

Extremely nerdy. Okay, so the primitive the of an LLM is a token, right. A part of a word.

Right.

And the very first technology that enabled this incredible boom of like multiple multibillion dollar outcomes in three to five years in the text domain, the speech domain, was that next token prediction. We at the very outset, partly because of that collaboration, we went so deep into what's called source separation, which basically is a form of discriminative AI where we are using masks and unmasks to reveal the vocals, bass, drums and instrumental the stems of a song.

Oh, so you can reverse engineer the stems from just a full MP3 that's already mixed out.

Exactly. Allow me to demonstrate. So like we got some like Kehlani, Lil Yachting and that's going to transition now. So that's like an AI transition to some leqoub and I'll bring in. So the part of Kendrick and latube are in now. Right. So and then the original primitive, the tokenizer.

Yeah.

Oh, you can go back to the pure vocal. And now we're going to get some Mac Miller drums. Pure Mac Miller drums. So like the tokens of words stems are the tokens of songs. And by doing a form of next token prediction, we can give you the songs that you love but in a new way.

Yeah. Now you could just go to the artist and say, hey, upload all of your stems to this. Correct.

Nah, nah. We get a feed from the. We get a feed from the major labels of the artists exactly as they would release to Spotify, Apple Music, Amazon, Tidal.

Interesting.

We ingest and then you decide this format, the stem FM format. The full song is transmitted from the server to the client, but in this new format where on their end this read.

The broader public have a general misconception that stem makes cool looking speakers.

No, because we do.

No, no, I know, I know you make cool. I know you make. I think it's like. But I'm saying it's a misconception.

They think they look.

But no, no, no, no.

I'm saying it's a misconception because it's much more than that thing that's interesting is that you're turning into a game that feels like somebody can.

Well, what I was saying what you're.

Doing here, I'm looking at it. It's not like you're doing.

No, this is software. This is the most advanced Music intelligence in the world. The reason I think the general public hasn't used it yet and may have used applications that made music that sounds bad is because the ones that launched earlier and raised huge amounts of venture capital were based on really an antiquated technology paradigm for generating music. It's better to generate music as the genius of Kanye Prude, as you put it, John, through recompositing samples, stems and existing elements, than it is to diffuse a song frame by frame. That's why all that shit sounds like Muzak from Suno and Udia doesn't sound like music, you know, so this is a world of music that is giving a new value to music, which I think is the kind of key goal because streaming is already growing so much. But we want to share the pie because more people are participating. Anyone can upload, mix their music with Mac Miller, mix their music with anyone legally. Make money, participate.

How do you think about the continuum of hardware devices? On one side, you have the Beats pill, no interactivity. And then you have your product. And then I think one notch more customizable. You have something like a teenage engineering.

That's a good one.

And then you might have a lemur or. Or some sort of. Or some sort of control surface for Ableton, essentially, where you. Exactly, yeah. Where you're. It's like more professional. This is. Is this prosumer. Is this purely consumer?

It's a consumer product at the end of the day. We started really by making it the world's best portable Bluetooth speaker. It's the loudest portable Bluetooth speaker ever made. Go speak in this mic here. Yeah, and then it's also way smarter. It's got computer, it's got storage.

But I mean, we were asking Sam Altman about this. Like, with Sora, there it is a different app. Like on Instagram, probably less than 1% of my time on Instagram. Most people's time on Instagram is actually creating, uploading, filtering photos. 99% of the time is scrolling. On Sora, it feels like a lot of people are spending 20, 30, 40, 50% of their time creating versus consuming. And that's an interesting new paradigm. Maybe it's better, maybe it's worse, I don't know. But like, with this, how much time do you think is spent just, hey, just play the normal song versus I actually want to be interacting with the song. I want to be creating and remixing. Like, how do you think it splits down?

Great analogy. I'll turn back on the music. This time it will be coming not from my phone, but actually from the device itself. So. And we'll just use it like background for now, just Steely Dan. And then when the moment strikes us, we might create, we might customize.

Yeah. What's most interesting to me is taking something that everyone loves. Right. I mean, not everyone. I, I, you can find people out there.

I don't like music.

It's like, okay, like get you, you should get an award for that. But, but I think there's something to tapping into this like DJ culture. Everybody wants to be a dj, but then turning it into a form factor that anybody can use. Almost like a simple drum, right?

Yeah, yeah, like, yeah, exactly.

What?

Yeah, what. So when you look at the. I was, I saw some reporting on Suno's revenue growth recently. Where is that revenue coming from? From what? From what?

You know, I'll say what I said to somebody the other day, I think, you know, I mentioned a lot of big artists who we work with and the reason we work with a lot of big artists is because, like we care about music as well. And some of the statements that were made by certain companies, if you look at the legal filings around the, I think $5 billion damages or more that is being claimed against Suno and Udio at the moment by every major label and every major publisher in together, some of those statements, you could see that they weren't true. And so that's all I'll say on the matter.

Which is basically statements at the record labels.

No. That the companies make interesting. Yeah. Like they were basically trying to say that you couldn't prompt the thing with like names of artists and you couldn't prompt the thing with like real melodies, like smoke on the water, like, and it just, if you could look at the outputs and use stuff like our stuff which detects chords, melodies, beats, and you, you could see it's copyrighted outputs. Now will it get. I like cool new things as well. But yeah, but that was my, yeah, my question, My question number.

True. Yes.

Yeah. My question also is like, is it, is it prosumers? Is it creators using it? Like, where's the actual revenue? Who's, who's going and spending money with you?

Yeah, I mean if I, if I compare it to chat cbt, I imagine that generative audio apps are not used. Like, very few people go to chatbots and say, like, actually just write me a full book. It's more like get me some recommendations of books I might like at this point. And I feel like in the musical context you might ask one of these AI music generators to come up with a few different melodies that could inspire you to ultimately write what you wind up writing. Or take me through a bunch of different chord progressions or a different. What does it look like if this is arpeggiated? Or what does it look like if it's in a different key? Like, there's all these different ways that AI as a tool can just be used to kind of like add twists and explore the.

What would it sound like if Kanye and Taylor got in the studio rather than fighting? What would it sound like if Drake and Kendrick got in the studio and made. Made it together? You know, like that. That to me is the thing that people.

Stem can do that.

Yeah, absolutely. Do you want to prompt something right now?

Like, I don't know what's.

I mean, we got to do Metallica. Of course.

Yeah. I was going to say Benioff.

Metallica.

Shout out Benioff as well.

Yeah, yeah. Shout out Mark Benioff.

That's my guy. He's massive.

Yeah. I mean, there was OpenAI news just yesterday that.

How about Metallica and Pop Smoke?

There we go. Let's give it a try.

We're going to add Metallica.

What was the news from OpenAI about the. The Martin Luther King foundation said, hey, stop generating soras of Martin Luther King's I have a Dream speech?

No, it was just. I think that. But a bunch of other, like, kind of scenes.

I did see that going viral on Sora. Let's go. We gotta send this to Benioff.

At.

The next Agent Force. We got to get Pop Smoke and Metallica on. On stage together.

This is the way it just sounds sick. It's like people already doing this in remix. It's like the top 15 songs are all remixed.

You did that in 20 seconds.

2007, 2008. The whole, like, mashup culture was really huge. There were a couple DJs that there. That's Their whole thing was mashups and they became extremely popular.

This is great because it's kind of a bop, but it's the most unlikely combination.

You prompted it.

This is your prompt.

Like, yeah, it's the Harry Potter Balenciaga of music, potentially. But yeah, I mean, the.

Just like getting lost in the music. Yeah, it's funny, but, John, you put it really well. What is this creation consumer percentage in the future? Like.

Yeah. What are your design inspirations? I feel like those headphones. We've talked to Carl Nothing. When Carl Pay from Nothing came on, he had a pair of headphones. That were also very counter positioned against what Apple's doing. I think it's very dangerous if you're starting a consumer tech company or building a consumer product to try and steal from Apple's design language because you're just getting it steamrolled. You've gone a very different direction. He went a very different direction. You two are on opposite. I feel like that looks like human flesh almost, whereas his looks like complete clanker coated. But what are your inspirations? How do you land on this particular material?

My chief inspiration as a designer is God. My chief inspiration is God. But to put that in more tech friendly terms like nature and science. Nature, nature and science. Mathematics, Principles of mathematics. And then feels almost kind of uap.

Like that was a alien technology.

Yeah, that's the.

Well, the Apple like campus looks kind of like a UFO in Cupertino. That new huge like UFO circle. But it's very class titanium and this is very like fleshy almost.

Yeah. A couple of questions on my side. So being able to put this experience together legally seems monumental. That.

Yes, exactly.

So. So the music industry is clearly very afraid of AI. They're sort of clearly like engaging with it and excited about it in certain instances. But what in particular about this instantiation of AI music were they excited about? Was it that this is able to basically tap into DJ culture stuff that's already happening and give people this?

I think the reality is it's because it's hot. Like the music sounds better so the music people like it more. You know, we've all had fun like doing the older generation of music, AI, which is like text to song but because it makes us laugh. And there's a business there, some producers. I was having lunch with my friend Smino in Cha Cha Chicken. He's an amazing artist. He's got 3.5 monthly listeners, fantastic storyteller and 3.5 humans. 3.5 million monthly results.

3.5 million. Okay, he's got half human.

You know, guy's got no lens.

Child. Three adults and one child.

Listen, 3.5 million, that's equal.

You're still full human.

But he was saying like they bring the old generation AI into the studio sometimes like people bring it in to pitch to him as the musician and he's like, oh, okay, that's. Oh wow, the tech is cool. But I was like, did you use it? And he was like, no, artists aren't really using it. And I think that also this, the.

Business model here, one seems a lot better for you than just selling like a smart Hardware device because you can sell a subscription that's actually aligned and that it's a novel product in that you have a software AI layer on top of what probably is like pretty standard streaming platform.

Is that right? What's fascinating, that was what I thought when we started really building this as a streaming service, an all catalog streaming service, three years ago. But to do what we do, to hear those parts of the audio separate, to have them blend into each other seamlessly like a perfect dj, to have the ensemble of models that detect these musical features, that rich metadata be usable almost like sheet music. For the front end experience running on all these platforms, we basically had to change what music streaming is. We built a new audio codec on STEM FM which uses a new concept in music processing called Music Aware processing. Well, audio processing called Music Aware processing where basically the core logic of how the file is chunked and de chunked as well as it unwoven time multiplexed out of its interleaved format. That logic is aware at all times of the beat and key of the song. So you think everything on Spotify, you listen to all the tech going into it isn't listening to the song, it's just, it just sees it all as audio, audio frames, frequency bands. So we've effectively encoded the intelligence of a DJ in the streaming codec, which is like that's been the hardest thing we've ever worked on. And shout out to the team, like big fucking shout out to the STEM FM team. Holy shit.

Is licensing purely commodity at this point? Like Apple Music, Spotify, Tidal, do they all pay the same royalties to musicians? Or are musicians thinking if I partner with you, I will get twice as much revenue per stream or something Effectively.

There were two very new things we did. The first is that music Aware processing, that more interactive music STEM format. The second was our licensing and revenue model, which we've come up with an idea that we think can really level the playing field, create an even bigger music industry and get artists the fair understanding of their payouts and the fair payouts in a way that can stick and break the tie and break the bad vibes in the industry around business right now. And it's called the Time Based Artist Compensation System. T backs for short or time backs if you like. Sure, get your time back with TABs. So it's very simple. All jokes aside, you're a subscriber to STEM FM. You pay say $20 a month. You the listening time you spend with a particular artist or their label. That will be the percentage of Your subscription fee as a percentage of overall time. There's no pay per stream, which is really a hang up of the CD age.

Sure.

It's really like detritus from when people were selling these physical, mechanical.

So you're sort of creating like a creator revenue share pool that then is divided up based on listening time that's.

More aligned with artists. You might have three and a half million monthly listeners that listen to them for like 30% of all their listening time, but they're getting a tiny fraction, right?

Yeah. And so if I'm listening to Metallica and Pop Smoke at the same time for my entire month straight, because I'm just obsessed with that mashup, half my revenue share goes to Metallica, half my share goes to Pop Smoke. In theory, yeah.

And the business as you were referring to earlier, the, the unit economics of streaming are tight. So basically you have to make sure that you provide like the major labels with a deal that gets them involved. But we also open up to verified licensors. They can come on our platform and upload and, and participate. We'll pay them out.

What happens if generating AI next token music and they try to bring it on stem?

Yeah, we were thinking about converting our entire three hour podcast into a song.

Three hour musical. Into technology and business. Musical.

Musical. And then uploading that as a three hour song.

We'll do it on the platform. I'll send it to you guys. Yeah.

And TVP and the musical. The daily musical.

Musical.

I feel, I feel like it would be great. It's like an odd couple story. Like it's beautiful.

Yeah, yeah. Let's turn, let's turn our Benny off.

How much, how much AI music is getting uploaded to streaming platforms broadly every day.

I don't see AI generated music recommended in my feed. I don't know about you.

Like, I've only seen screenshots of like this AI artist went viral, but I haven't actually seen it surfaced in my like, you know, recommendations or anything like that.

A few, A few, A few people I know have like songs have like gone viral and the headline was like, it's an AI song. But the truth is there's a human producer and that's what Smino is saying as well. A great musician can make music out of anything, including these more muzaky things.

Yeah, of course, of course. It's like any other tool. I mean, it's the same thing with edm. Like for a long time people were like, you know, oh, Ableton is not the same as having a whole Orchestra or band or real samples.

We thought about getting a live band instead of the sound.

Soundboard is technology 2.0.

I actually just realized could even try prompting this TBPN musical right now.

Yeah.

Because you can just prompt stuff with YouTube videos.

Oh, interesting. Pull that in.

Yeah.

It has to be a short one, though. Like a clip.

We have shorts on there. We have shorts on YouTube for sure.

Yeah.

But how does all the legal debacles between the generated music platforms and the labels actually get resolved?

Well, I think what really happens in the end is that the labels and the artists, it's ultimately they're providing the product and the technology platforms are the licensors and sub aggregators. And so I think it ideally gets resolved with like a deal where the power of advanced technology and the beauty and purity of music, which itself is a technology, can coexist.

It feels like with YouTube initially, it was like if you uploaded copyrighted content, they just take it down. Then it was like, hey, if you used a Metallica song in this video, we're gonna send the money that you made from that video to Metallica. Which was fine. But then what happens when you use 5 seconds of Metallica song and then 5 seconds of a Justin Bieber song? You need to parcel up and split.

Out all the revenue used by our. Yeah, our video.

For sure. For sure.

Well, that was the amazing thing.

And he's in fair use. It needs to determine fair use. But I think LLMs can. Can do a lot of that already.

I don't want to be, like, selling too hard, but basically this. As we got into the research here with Queen Mary University, we've been in a deep research partnership with their center for Digital Music for three years now. Submitted papers, accepted papers. Literally changing the paradigm of beat structure and segment detection in music information retrieval by using a more structural approach that can actually detect music better than the previous generation approach. So these are early experiments for us.

How many of these do you want to sell over the next year? Because I imagine this is like the perfect product to go viral on TikTok.

And suddenly, I mean, we sold out and did about $20 million of revenue on the stem.

Oh, we need the numbers.

Yeah.

And we still, like, we haven't launched this in, like two years. A new stem, so. But. And we really put the technology into this one. Like, this is where we went hard on the technology for, like, a long time. The first one was like almost the art house film of the tech. Like, oh, look, you can decompose the stems. This one is like the block, but this is like the blockbuster. So how many do we hope to sell? God willing, this will be this. God willing. You know, my, my voice to God, we won't be getting these JBLs or whatever. Get this. It's a smarter speaker. It sounds better. It's 299, you know, let's go like Jack Dorset was.

Hit that gong again.

John, what the fuck?

Thanks. No, I just think I, I, I. It's rare that I see a new consumer electronic device outside of like a new MacBook Pro. And I think I really want that.

Last question for me.

Thank you so much. I'm excited to get it. It feels like something that's like, I can imagine my Saturday mornings with the family. My son's like, I want dinosaur music and I want dragon music. And I can be like, well, I want imagine dragons. I want pop smoke. So we're getting pop smoke. Dinosaur music first.

A thousand units. Founders edition. Available now on stemplayer.com.

No way.

Sorry, you're wrong.

Yeah.

Last question for me. Jack Dorsey was taking a victory lap about title not having ads. Okay, what's your take on ads? Are you gonna put ads in there?

Ooh, we have no plans to put in ads and we've actually in our signed deals with the majors, we have agreed not to do ads for the first two terms. And it was never close to my heart. And I think the goal is to increase the value of the music.

Sure.

Look at this. I just bought it. Took me two seconds.

Thank you so much.

Yeah, kudos. I'm excited to get one.

I really appreciate that.

And yeah, no, it's just like it's.

It'S, it's rare to see a consumer hardware device.

Yeah.

That feels like it is genuinely a novel experience.

Yep.

It's very rare, but not so novel that it's so different that I'm like, ah, this like, like gathering around and enjoying music with a group and this like DJing is like a Lindy thing.

Yeah.

And then also have a sustainable business model Right. Where it's like you're selling this hardware device. I'm sure you'll have some margin in there, but hopefully I'm entertained by it for a long time. And you, you know my LTV is, is infinity. Yeah. Be great.

Well, thank you so much.

Thank you so much for coming on.

This was super great hanging. We have Hamant from General Catalyst coming on in just a few minutes. First, let me tell you about profound. Get your brand mentioned in chat. GPT reach millions of consumers who are using AI to discover new products and brands. Noah Smith says solar power is good. Batteries are good. Air conditioning is good. AI is good. Nuclear power is good. Stop thinking like a medieval peasant people. He's putting on a clinic of techno optimism, he says.

I will say though that smartphone enabled social media is not good. He posts on X. I am a bit of a.

What's the word?

Eliezer in the reply. Solar power is good. Batteries are good. AC is good. Nuclear power is good. Gene therapy is good. Building a new alien species of superhuman intelligence with very little ability to understand or direct it is not like other technologies and will kill you and your kids.

Rough. Well, you know what won't? Linear. Linear is a purpose built tool for planning and building products. Meet the system for modern software development. Streamline issues, projects and product roadmaps. Our next guest is Hamant From. He's the CEO of General Catalyst. He has a new book out called the Transformation Principles. You can go and buy it today he is in the Restream waiting room and now he's in the TVPN ultradome. Welcome to the show. Hemant. Great to meet you.

Great to see you.

Great to have you.

How are you doing?

I'm great. Thanks for having me.

Please introduce the book. What was the thesis? I'd love to start with the overall principles that you were trying to convey and then we can dig into some of the more nuanced topics.

Absolutely.

So I started writing the book around November 2022 when you start thinking about when ChatGPT was launched and the observation was that there is peak ambiguity. I've talked a lot about it. You look at all the major industries that are changing because of geopolitics.

Yeah.

And you've got this technology that also is. It's unclear what its capabilities are going to be. Obviously we're all developing and figuring that out. So when you have complete ambiguity of what. What you're building in, only things you have to give you a clarity of purpose is your principles. So I want to take a step back and think about what is it that our true north really is as we blankly power the world with AI, which is what the next 20 years is about. That's my life's work, that your life's work. You know, it's all sort of. We're all sort of living that era. So that's why I wrote the book.

Was your goal to give people a framework for developing their own principles or to instead share your principles that you think other people should adopt?

I guess a little bit of Both like I. What I did was I showed there were three through lines to the book. One is the nine principles I put in there when I learned them, how I learned them, and some examples of some of the founders in our ecosystem. You know, folks like sard, Applied intuition, Brian and Andrew. What are they doing? How are they actually applying these principles in the way they're building the company? So that was one second is, I think, as you may know, I'm probably a decade into transformation of the healthcare system and talk about ambiguity and complexity there and what are the principles that are working for us as we're making progress? Start to enumerate those. And the last is I'm very keen on, you know, how do we retool the founder proposition in our own industry and transform ourselves? You know, what is the, what does it mean to go beyond venture in serving founders to build next generation most important companies in today's world where the conditions are different. They talked a lot about that. So that's another transformation we're in the midst of. So those are the things I leaned on.

Yeah, I mean I'd love to zoom in on healthcare specifically. I mean there's principles like the Hippocratic oath doing no harm protecting patient information, but then there's also principles about agile software development that is in many ways a principle. Where do you like, how are you thinking about transforming health care? How are you thinking AI fits in there. And then also the work General Catalyst has been doing is definitely on both extreme ends, both high tech and also almost like Wall street, not to put it in those terms. But I'd love to know more about your thoughts on healthcare.

It's such a great juxtaposition of the agility in technology and the Hippocratic oath and the seriousness of healthcare. It's sort of as a sort of framework for learning. The way I have always thought about transformations is to put different cultures together. So over the last 10 years I've been part of starting a bunch of businesses in healthcare. Livongo was one of those. We did Hippocratic AI recently when the language models came out speak of Hippocratic oath and we built Kamua, which is at this point probably the largest software company serving the health systems and with a transparent focusing on delivering health care to consumers that are employees of large companies. And every one of those, the key was you want to be interdisciplinary, you actually want to bring in people from health care that understand the seriousness of the work and the fact that it's not a free market. And then you want to bring people from Silicon Valley that actually will cut through the inertia. And I always say bring the ignorance and imagination to make a difference and transform the industry.

Yeah. Do you think that there's more opportunity in healthcare at the seed stage or the growth stage? This is something that we keep going back to where we've talked to a lot of founders who have built a piece of kind of SaaS and they got an install base and they have a bunch of customers and they're still in founder mode and they're still have a bunch of really hardworking, highly entrepreneurial folks in the organization. And now AI is here and they can just amplify everything that they're doing versus trying to, you know, come to market with something that's very AI laden with. You have a dot AI domain and like you have to really convince someone that you're the new solution. And the thing that they just got excited about a couple of years ago and onboarded to that was the new thing. They have to rip that out now or they have to bring you in. And I'm wondering how you're seeing the opportunity in health care from like scale ups versus like completely new ground up solutions.

It's such a great, great question actually, especially given that we went through a pandemic five years ago.

Yeah.

So, you know, at the core of gc, seed is very much what we focus on. We brought on three different seed firms around the world onto the platform in the last two years. And so there's a ton of serendipitous innovation that's happening because of AI, because of advances in hardware, in the way you think about healthcare delivery through the solutions for this next generation. So seed is very interesting for us. Applying AI to life sciences is starting to finally work and we think it's interesting. So seed is quite interesting. But then if you take a step back five years ago, there was a Cambrian explosion of innovation to change the architecture of our healthcare delivery system. Because we didn't have the test trace isolate capabilities, the system wasn't resilient. Insurance companies were getting big balance sheets because no one was getting procedures and hospitals were looking for bailouts. Right. Complete sort of misalignment of business models. So we, we built and invested in a bunch of companies in that period and those are now scaling really nicely. Some of the ones I mentioned earlier and a few others like Aidoc and Soar and Citiblock and others. And so at the growth stage I think There are probably 25 to 30 companies in the healthcare ecosystem that are very important for our system going forward. And we're very focused on how do we help them become truly enduring parts of the healthcare system. And then we're constantly looking for founders that are using these new technologies at the seed stage to reinvent parts of the stack in that overall architecture.

Where do you. What categories do you think the winners are decided and then what categories do you think are still up for grabs? We had a lot gill on. He was saying if I, I'll probably botch it. But something along the lines of like in the medical scribe space, there's enough heavily funded players that have real traction. You probably don't want to start the next one that's maybe four or five years behind. But how, how do you view where the biggest opportunities are? And then what categories you think, hey, maybe, maybe look out, there's already a.

Power law company, basically.

So if I take a step back and say we're going to do a product centric transformation of the sector, not a business model centric one, you would say at the end of the day, our delivery organizations provide the product, which is we take care of people's health. And so there, when somebody engages with our health system, broader health system, not just a hospital, patient engagement, physician tools, which is where the scribing solution that Eli mentioned fits. You think about payments, you think about AI workforce, if you think about sort of diagnostics, using AI, huge amount of progress in all those areas. And then you say what's missing? What's missing still is the interplayer of the peer provider relationship. That still serves as a place where there's not a lot of trust. Peers think providers bill too much, providers think peers pay too little. So it's just like army of solutions in the middle of them to create that arbitrage. I think we need to do something about that. And then there is one other thing where there is a new movement which is around longevity. If you look at Brian Johnson's work, that there's cultural movement happening around that and you look at the work that Ro is doing with GLP1s, which is the beginning of that longevity move. These are real phenomenon.

Yeah.

And.

And I think that's a place that there's a ton more that that needs to be done so that the system moves towards incentives that keep us healthy and out of the hospital versus a high performance but really expensive on affordable care. If we go into the hospital, which.

Is how do you think of the interplay between the longevity movement and the traditional health ecosystem in my, in my own life as somebody who discovered biohacking maybe a decade ago and just started reading online, I would always just try to solve problems myself. And then, you know, hopefully, you know, at some point or another you're like, ah, actually you should probably go to the hospital now. And so it's kind of a harsh transition and maybe it doesn't need to be that way.

It's very different. One's focused on staying healthy. There's focus on fixing your health when it's broken.

Yeah, yeah.

So there literally are like two different mindsets. And the reality is that all of our incentive structures are about you going to the hospital. Hospitals want heads and beds. That's how they thrive. Right. They actually want to see revenue consumer wants to not ever go to the hospital. So we need to kind of rearchitect a system in a way that we solve for this cognitive dissonance. It's all about incentives.

Yeah. Is there some incentive shift from. I mean a lot of like there's some general meme about like when private equity buys a hospital, like quality of care goes down. And I'm wondering if there's like bringing a venture mindset, a growth mindset, a turn like a growth instead of like cut cost mindset could actually like improve those incentives. Like how are you grappling with, with that trade off?

As you may know, we just bought a health system.

Yeah.

It's in Akron, Ohio. It's, it's actually got a, you know, large. It's the largest employer in Akron. It's. It's a big hospital system and has an insurance company.

Yeah.

And we're trying to show is like if you could close your eyes and you said, what is the one thing that if it could exist, we'll solve a lot of this. That longevity solutions or investment in our health is reimbursable. It's not today there's no models that show what's the RO on this? Where insurance companies pay for it. Insurance companies don't pay for reimbursement. So the cultural movement around longevity and.

The reason I'm an investor in function Health and from my understanding of that market, like insurance providers are not so inclined to pay for something like function because people switch insurance all the time. So they're like, I don't care that you're going to be healthier maybe in like 30 years.

Like I just took the sentence out of my mouth. That's exactly what I was going to say. Because the, the insurers don't know if they make an investment in longevity, they're going to be able to capture the value on the other side.

So it's just like lighting money on fire for them potentially.

Exactly. So that's the misalignment I'm talking about. So the question is, how do you really change a system where the ROI on longevity is understood and financeable? We need to do that as a system.

Yeah.

When you. This may be a totally stupid question, but when you see some of these data center build outs, do you ever get inspired to try to not just buy a hospital system but just create something your own mega project? Is there even any potential ROI in building something from the ground up?

It's a great question. And we're actually doing that with a company in India.

No way.

Cool.

And I think there's a leapfrog opportunity in some emerging market. We want to show the model there. And the other place where I'm very focused on this is we have a large amount of capital that's allocated for rural health system rebuilt. The government's allocated about $50 billion for it. And I think if we went and deployed that money in old school hospitals again, that's going to be a falling knife that we've created and we'll burn that money into the ground. And so there's an opportunity to rethink the topology and the architecture. What do we do there? So that's the place where I'm also focused in the US I'm focused on rural health. And what do we do about it? We actually built a company called Homeward Healthcare that's focused on this particular problem sort of post pandemic because all those systems got eviscerated. And then the future hospitals. I just think it's easier to build in some of these high growth emerging economies to show the models and then we'll bring that here. So it's a long game. It's going to be in the next 20 years.

Yeah, yeah. A couple weeks ago.

That's why I need longevity.

Yeah.

A couple weeks ago you were going sort of viral for some comments about triple, triple, double, double, double startup growth rates. Can you kind of unpack?

I don't know why people get so triggered by growth rates.

Oh yeah, yeah, yeah. Anytime you try and quantify what success looks like, you know you're stepping in. The bees nest a little bit.

I have this counterexample that proves you're categorically false.

But, but, but maybe walk me through like what you're seeing in venture our growth rates really faster than ever before.

Yeah.

What was like the misconception yeah, give us more context around those comments.

And I never go viral, by the way. Like my, my boring systems content that nobody cares.

This is not boring. I love this. We're gonna, we're gonna get 10 million views on this. We're gonna turn this into a TikTok. We're gonna.

Okay, there we go.

Let me say something crazy then.

Yeah.

Actually the last time I went viral was when I was working on saving Silicon Valley bank.

Oh yeah, that's right.

Freak out. What was going to happen. So it takes three, four years for me to create another viral moment.

That's good.

Look, I would, I would say it was a very innocent empirical comment because I just look at how fast the companies are getting built and they are getting built a lot faster. Durability is a, is a separate question because we'll see which of these can then transition into durability or not. But as far as growth goes, it makes all the sense if AI is truly an accelerant, you know, and you've seen, we've done significant amount of work, I think the deepest amount of work around these AI roll ups. We have 10 companies we're building. There's where you can actually acquire these service businesses and transform them with AI. Almost treat those acquisitions as customer acquisition and apply AI. So you just have all these different ways to accelerate growth and take share in this AI world that is happening. So when somebody comes and says we're going to go from 1 to 3 to 9 and we're routinely seeing companies that are 1, 2, 15 to 20 to something fairly large, which by the way, there were very few companies that did that in the last iteration. You know, Samsara was one of those that I was part of that sort of grew that way, but there were very few. When you see that routinely, I mean I, I had actually just gone to our team saying guys, just recalibrate. You have to change the terminology on what best in class is because we want to, we want to be. If we're going to do. When we do seed, we want to back every amazing founder. When we do growth, we want to be in the best in class growth companies. That's our business. And it's just very clear that the bar has shifted.

Okay. So I want to dive deeper on some nuance there. We have this kind of coinage that we're trying to get to go viral. Maybe it'll work one day of the barnacle economy. The idea that if I started a company that is I just clean toilets, but I get anthropic As a client and they're 10xing their number of office space and their revenues and everything. My business is going to be 10xing, but I don't really have a venture scale business because I'm not building it monopolistic category or building any durable technology or any intellectual property I just happen to have. My hero client is growing so fast that I am also growing fast. And so there's companies that have kind of bolted onto the new hyperscalers, the foundation model companies that are growing so fast. They're growing really fast. At the same time, there's other companies that are operating in legacy industries that are also growing fast. But there's always the risk of duration and pilots. So where do you start? Are you more worried or more digging in on a company that's overly indexed on a foundation model company's growth or more risky to. Oh yes, they're getting Fortune 500 pilots in this one niche like the oil and gas industry. But it might not be sticky because that big company moves slowly and even though they're ramping up revenues, we don't know if they're actually going to stick around in 10 years.

Yeah.

So I think let's just actually split those bars. It's a great, great question and something that we take a step back on everything that's coming in in the applied AI layer and like looking at with that lens. If you have a company bringing in the foundation monoco capabilities into an enterprise, which I think tends to be the most durable relationships that are on the big enterprise side we are looking for are you going in and are you learning how to expand your relationship with those customers? So before what used to happen was you would build this like mvp, you get product market fit and then you figure out how to create repeatability to go do that over and over again in the AI world. So there the advantage is great product design and precision. In the AI world. You get into a customer, you do forward deployed engineering. Let's think about big enterprises and you build trust with them and then you figure out how to do more for them. All the ones that are scaling really, really fast, that's what they're doing very effectively. So if you're building that kind of a deeper and deeper relationship with customers that are going to be around for a long time, that's durable. We like that. We're looking for a lot of that. And those companies are not growing triple, triple, double, double. They're going way faster in different verticals. I mean that's just the reality of it. Because we're meeting a lot of these on the other side. I think the companies that are creating wrappers on top of these models, many of them are not durable. If you remember when ChatGPT3 happened, you had a bunch of companies that got created and they disappeared. So every time there's new modalities and capabilities, some stuff happens and a lot of that doesn't exist. But there are some companies getting built that are cultural movements and they are going to redefine human interaction with technology and software. And I think those are quite interesting. And they will. They may look like lightweight products, but they're going to be durable long term. And then there's a bunch that's going to disappear. That's the way I sort of think about it, because a bunch is just going to get absorbed into the models. But the ones that are focused on rethinking design and rethinking the human relationship, they will have durability on top of these models.

Yeah, peak ambiguity.

That's what I always say. It's like hard to figure out what's what right now.

Yeah, you've made a bunch of, I don't know, you call them like American dynamism type investments. But I'm wondering if you could rank a few other international markets that are interesting or underrated. You mentioned doing some work in India. Are there any other countries that you think where there might be pockets of opportunity for your business over the next decade?

Yeah, look, we do two thirds in the US probably about 25% in Europe and 10% in India. And I actually say, I used to say this a couple of years ago. US is going to be the consensus market. Europe is a contrarian market and India is a FOMO market because of the high growth and that we need to be in all of these and be intentional about what it takes to create value. And I think Europe is super interesting. There's like so much talent there. There's people are waking up that they need to innovate. We can take the classic view of, you know, there's not intensity, but we're just seeing pockets of like, interesting collaboration. We're actually catalyzing a lot of it. Our European team is doing a lot of that. And then our. And in India we're just looking at saying, what is the US India sort of collaboration going to look like? There's a lot of complexity there. And then what is required for building societal systems inside of India, in education, in defense, you know, in healthcare that we can help build because we've learned A bunch from here and probably mostly learned what not to do.

Yeah.

Because of the, like the healthcare business models. And we can help them think about it from first principles so we can be good partners there. And so the way that works out is, as I said, we're building hospitals in India, we're transforming healthcare, we're innovating in India, we're transforming in the U.S. we built defense prime, we're working with Defense Primes in all the markets. We've got Anduril here, we've got Helsing in Europe, we've got Rafi in India. Because you need to create. Our phrase that we use is global resilience and abundance. And so each of these markets and key industries wants to be resilient and there's founders there working on great stuff. We want to help them be successful with that. And that's the way we've approached it.

What's your primary inspiration for the firm?

Boy, look, I think I.

And I would give you, I'd give, I mean the classic example was like CAA and a 16Z. Is there, is there an equivalent for you or do you pull from a variety of.

You know, my, my longest standing LP relationship was with Andy golden, who used to run the Princeton Endowment, who's, who's retired last year. And I used to always look at, hey, we want to be one of the elite, when this was when I was in Boston. We want to be one of the elite venture firms, which I don't even consider us purely venture capital, even though seed is our core. And he would always say, run your own race. And that really stuck with me. So I try not to focus on mimicking whether others are doing. They're like, there's so many smart people, so many creative people here doing interesting things. We got to do our own, we're going to be on our own journey. My main, the thing I would say though is I do want to transform industry systems, not just sort of be sort of serendipitous venture capital investor in companies. And so everything we do, all the, the stuff people look at us doing is all created to build a moat for the founders so they can just be more successful and become bigger. Because if they get bigger, then we have the opportunity to transform the industries. If all those companies I mentioned in healthcare can actually be 100 billion dollar companies, many of them we started, then we have the license to actually help rethink the system.

How do you think about, how do you think about risk across the platform between early and some of these really Scaled up investments. We were enjoying a clip from Roloff earlier this week calling Venture it's a return free risk called it. But he was just kind of putting all of Venture more or less in a single bucket. And there were some comments as well on people saying that hey, it's actually like there's a big difference between, you know, the 100 million dollar 500 million checks that are going into companies at a later stage and pure venture, private equity and venture.

Was the comment.

Yeah, look, I think if you want to do really transformative stuff, you got to take risk.

Sure.

So if we're not taking. No, you don't want to take stupid risk but I think you want to be smart about how you invest across different stages of risk. But like you've got to take risk and you have to have a vision. And this is some of the stuff I talk about in the book which is incrementally improving these systems is not going to get us there. I think we're sort of at an interesting place in society where AI is so disruptive. So we got to build new systems. And in order to go from deep in company to new systems at scale, you have to bring the incumbents along and partner with them and also invest in their success and do things that are just radically different from how we're used to building serendipitous companies and sort of classic venture. So I do think you have to take a lot of risk. Like I thrive on risk, you know, failures don't bother me. As long as we're well intentioned, we're doing the right things, I want us to do that. And you know, otherwise I'll tell you like venture is, the LPs will say is becoming more of a beta because everybody can get into it. These companies raise a lot of money. So if, if this is going to become the way the system is behaving now and we don't take on working on large problems which for me a lot of them equate to transformations and, and sort of thinking more, more systemically then, then you know, with the illiquidity premium, are the investors excited about the return this, this asset class is going to generate? I think there is a conversation about that as well in the way things are today.

I like asking all authors that come on the show about their favorite books. I'd also be interested to know if there's any books that inspire you in how you think about building the firm. Are you more of like a House of Morgan guy or a Steve Jobs biography guy? Are there any books from through the history of business that have remained on your bedside? Maybe for several years.

I mean, I focus on individuals that have been inspiring to me and understanding those as a way to look at the journey. And actually, Ken Chenault is my chairman. He asked me once, you're modeling yourself after. It's like another version of the same question you asked earlier. And I was like, well, look, there are three people that really, really inspire me. One is, you know, if you look at Elon's work, the problems he takes on and the way he executes, I'm just like, you know, it just makes me feel like the bar is so high in terms of what's doable. It's like, it's inspiring. We should all be learning from the way he builds and the problems he chooses to work on and the passion with which he operates. So as a builder, that's very inspiring to me. As an investor, I'm a fan of Warren Buffett. He thinks 7,500 year horizons. And so the, you know, the best hold period is forever.

Yeah.

Is the way they think and that. That's the way we want to think. You know, we invested in stripe in 2010 when it was just a couple of them, and then, you know, we invested 14 times since then. Right. And I really think that's a relationship that can be a 30, 40 year relationship. You know, those guys are going to do lots of interesting stuff and I want us to be sort of bootstrapped to their journey as we think about our journey. And then I would say from a leadership perspective, which is also important. I really respect Ken, who, you know, obviously is our chairman, is a lead director at Berkshire Hathaway on the board of Airbnb. And he instills this whole servant mindset. And that's just the way I think you want to treat each other in your own company towards your founders, which is really an important stakeholder version and society at large. And are we building with that kind of a mindset? And, you know, and then what do I want to prove with all this? I think to your piercing question before, like, the inspiration to me is, can we take this philosophy and build the world's biggest companies? Can we actually do that? Because you know, that is going to be required as we think about really transforming with AI AI. It's going to be all about transformations and can we go accomplish that?

What's your guidance to partners at the firm around media? Because there are certain venture firms that their guidance is broadly get as much attention as you Possibly can, however, whatever it takes. But that very much does not seem to be your approach. And I'm curious.

So I'm going to give you a two part answer to this. One is if we think about transforming industries, you need four levers. You actually need to have the capital, the innovation, the policy. But then you do need to create a cultural movement. And so media is a really important part of winning hearts and minds towards what you want the world to look like. And I think in that sense it's very important, but it's got sort of a purpose towards it and it's really inspiring how you guys have quickly come up to stage and I think you can actually intentionally say here's the influence we want to have on this ecosystem and you can drive towards it. Yeah, because. Because you've got a, a platform and I'm sure you guys think that way, right?

Totally.

The I want to be famous so I get noticed for the sake of being famous and that's going to give me deal flow. My observation is that you have to become more and more fringe to actually be noticed and it actually starts taking an impact on you as a human being. I'm seeing that with a lot of people that play this game. And, and is that going to allow you to do your best work if you're fundamentally changing in response to how to be successful in your sort of media metrics? And so if you're the person, kind of person who can say creative things but stay grounded in your core, then maybe it's for you. But if it's going to change your core, then you might have gotten the followers, but you may not have the substance to do my big thing that I see as a failure mode for a lot of people that go down that path.

What do you look for in young people that join General Catalyst where you're hoping that they'll be with the firm for decades, eventually become partner, have a huge impact on the firm. What are you looking for at the young phase?

Look, I think a true sense of purpose because that's what drives conviction, you know, and, and I would say.

A.

Lot of people get in this industry because they think they can make a lot of money or they think it's prestigious, you know, it gives you power and prestige. There's a lot of that.

If you look at the people that.

Have done really well, they didn't grow up. I mean, just literally think about any of them, they just didn't. That's. They ended up here.

Yeah.

I came to GC for four months because I felt like I built a shitty company. This was in 2002. And I want to do it again. I was 26.

Yeah.

And I'm here, you know, and so everyone that has done really well. Nothing. I've done well, but, like, I just. I studied those people and they just had unique trajectories to get here. Nada. They chose venture as a career.

Yeah.

And so. So that purposeful sort of journey and wanting to do something substantive, which then gives you courage and conviction, which is the essence. You got to have that to be a good investor. You just can't census work.

What is Aum now?

Somewhere in the 40s. Too small.

Too small.

40.

What?

40 billion, I think. I think 400. 400 soon.

Congratulations on all the success. I think you've been fantastically successful. And thank you so much for coming on the show.

Yeah. I wish we had more time. I have a lot more questions. Next time you come on. I'd love to get an update on the. On the customer value strategy. We didn't. We didn't get. Oh, yes, yes.

That is transformative. That's a big part of how we're thinking about the proposition being changing for companies and how to serve them. We'll talk about another time.

Amazing.

Thanks for having me.

Love to be back. We'll talk to you.

Cheers.

Bye. Let me tell you about numeral hq dot com. Sales tax on autopilot. Spend less than 5 minutes per month on sales tax compliance. It's sales tax AGI. And our next guest is in the Restream waiting room, Rob Hayes from Radical Ventures.

There we go.

Coming in to the studio. How are you doing, Rob?

Rob.

Hey, guys.

Great to be here.

Thanks for having me.

Welcome back.

I want to go through bci. I want to go through the latest Forbes piece. I feel like you've spent years at this point writing about AI. Are you just over AI? Is AI done? And you're just bored of that and you got to move on. You got to pivot. If you're in AI, pivot to bci. What's going on?

There is definitely a lot of noise in the world of AI right now. I think BCI and AI are actually closely related, and BCI will be kind of one of the next really important foundational technologies that impacts how AI gets rolled out. But I do think a big part of our jobs as VCs is to kind of think about opportunities and technologies that are not yet obvious and mainstream. And I think BCI falls into that category in the sense that it's not ready for primetime yet, but I think it's getting closer and I do think it will have a really big impact on how basically human intelligence and artificial intelligence fit together.

The merge. The merge. It feels like one of those technologies that's always 10 years away, like fusion. But take me through the history. When does this whole story start? There's the Utah Array. I want to know about that.

Yeah, totally. So BCI is not a new technology. It's something that's been around in research context for decades. The very first BCI was actually in a human, was demonstrated in the 70s and early 70s. Researchers at UCLA, basically using EEG, which is non invasive technology, were able to get patients to be able to control a cursor on a computer screen just like with their thoughts.

That's kind of like what we're seeing from Neuralink, right? It's like not that far off, but it's been 55 years now, so. Super interested, like what happened in the intervening years.

Yeah, yeah, exactly. So. So that was like a very basic unreliable prototype, but it was kind of a concept, the first invasive bci. And maybe just a quick, like, yeah, quick definition there between invasive and non invasive. These are kind of the two main categories of BCI technology. Invasive technology requires surgery. It involves putting electronics like inside your skull, directly on or in your brain. And so neuralink is an example of an invasive BCI company where you have to cut open your skull, doing using a surgery called a craniotomy and put the chip inside. Non invasive technology does not require surgery. It's just sensors that you can put outside your head. They could be embedded in a consumer device like headphones or a hat or something like that. So the trade off between invasive and non invasive BCI really does boil down to a trade off between like accessibility on the one hand and ease of use versus signal quality and how high fidelity of signal you can get from the brain.

Yeah, and it's a lot about value, right? Like if you have a disability and you cannot use a computer and you and somebody says, I can give you the surgery and you're going to be able to interact with the computer, that's a great trade. Whereas if you say that to me right now, I'm going to say, wait.

You know, I'll use the mouse and keyboard actually.

Yeah, I'll be fine.

I'll stick to the mouse and keyboard. Yeah, okay.

I think my big question is like, at what point is the invasive approach like so tried and true everyday?

So in the 1970s it was non invasive when did the first invasive stuff start happening? Because neuralink popularized it but didn't invent it.

Right?

Exactly. Yeah. So invasive bci. The first invasive BCI was implanted in a human in the 1970s, 1990s, and it was this device called the Utah Array that you referenced, John. And honestly, I would encourage people listening to like, go to Google images and look up Utah Ray just to see a picture of it. Like, it looks like a medieval torture device.

It does. It's so scary.

It's literally like a bed of needles. And you literally, you like cut open the skull and you just jam the set of needles into the brain. And so the Utah rate, to be clear, was like a really important breakthrough in the history of bci. And it was the state of the art for a long time. And it made it possible to do things like go from thought to text and like, you know, understand people's, people's thoughts directly. But brain damage, basically, like, if you're, if you're jamming these needles into your brain in order to record from some neurons, you're also killing a bunch of other neurons. And so that was, that's the reason why the Utah Ray never really like took off from a commercial perspective, but it did lay the groundwork for subsequent generations of invasive bci, including Neuralink, which obviously is the most famous invasive BCI company.

Okay, so now was Elon the first person to really take BCI seriously? Because it feels like there's a proper market map and you have Sam's working on stuff. Sam Altman founder of OpenAI and then Elon's working on stuff. And then Nudge also exists. There's like a, there's like a number of like basically billionaire founders that are all making serious bets. Are all of them. Is this like an Anderle type story where like you need a billionaire who's just going to sink a bunch of money into this? Or are there any companies that have like actually just gone from ground up in the traditional fashion of like they raised a seed round and then series A, like it just feels like a very different path. So like maybe take me through the market map of what's happening right now.

Yeah, it definitely like started as from a startup perspective as a category where like it was largely funded by billionaires. Like obviously Elon had Neuralink. Brian Johnson had a BCI company called Kernel he poured a ton of money into. Yeah.

Do you, do you remember like what that actual. Was that invasive, non invasive, like what the plan was there?

He explored both. It ended up being Non invasive. And they developed like a new modality of sensing non invasively, which is actually pretty cool. You like shoot sensors through your skull, like shoot lasers through your skull to like detect the blood flow in your, in your brain. Which is like, is still used today and the company's still going. But yeah, I mean, it's worth noting. Neuralink is like definitely the most well known BCI company, but it wasn't the first. Like there were startups before Neuralink, there's one called Synchron out of Australia, there's one called Paradromics which did kind of follow that more typical path that you described of like raise a seed. Raise a seed Series A. And certainly in the past few years there's been an explosion of them. So. So now there are a ton of BCI companies, both invasive and non invasive.

What are the. It feels like Neuralinks had a very like, kind of like barbell strategy where they're. Where they'll tell like a sci fi story about like you will be able to fully like communicate with AI and merge and whatnot. But then there's also like this ultra practical, like if you are a paraplegic, you can use a mouse cursor and click and use a computer. Are there other pitches that companies are offering in terms of just like reading information from the brain or writing information to the brain? How do the value props stack out right now?

Yeah, I think one of the really interesting differentiators of the different invasive BCI companies is like, what's the end use case that they're going for? And what Neuralink has been focused on to date is computer use. So like all the demos you see of Neuralink patients like Nolan Arbaugh, the first patient is using a computer, which basically comes down to like mostly controlling a cursor and potentially keyboard. And there's obviously a ton you can do if you can control a computer. So that's like a very valuable use case, but a harder, more advanced use case which like I kind of think of it as the holy grail for. One of the holy grails for BCI is decoding language. Like basically being able to instantaneously understand words from a person's brain and like write them or send them via text or email. Like that basically that basically is telepathy, more or less. Especially if you then like convert it to another person's bci. Like so it goes directly from brain to brain. And so that is a more ambitious target which there are startups that are going after to this point at least. Publicly, Neuralink hasn't published anything related to language decoding. Of course it's something they're thinking about, I'm sure. Yeah, but, and, but that, I think that's one really interesting use case. And then another one is being able to control objects like whether it's controlling a robotic arm or like controlling any sort of object in the real world, which again, it basically is telekinesis. And like these terms sound really like, you know, fantastical and paranormal, but like that is essentially what the capability that, that will enable. Which is cool to think about.

Yeah. Are the non invasive BCI companies ever targeting like higher level abstraction? Like look, we're, we're not going to be able to reliably get an X and Y coordinate for a mouse, but we can tell you that you're stressed out or we can tell you that you're overly tired or we're telling you that you're dehydrated in your brain or something like that. Something that's more abstract but still something that a consumer could actually adopt, wear a hat. They're not going to go into surgery for it. But more of like a biohacker use case.

Yeah, that's exactly right. And those use cases have already been pretty well proven. Like there are actually non invasive BCI products on the market today that like can, you know, detect if you're stressed or if you're, if you need to sleep more or you know, if you need to focus better, those sorts of things. I think the really big question, like in my opinion the single most interesting strategic question in the field of BCI right now is like the conventional wisdom has been that invasive BCI is necessary for the most sophisticated use cases like language that we were talking about because you just need to get that high fidelity data. But non invasive approaches have been getting better and better very quickly in part because non invasive sensors have been getting better, but much more so because the underlying AI is just getting a lot more powerful which lets you extract a lot more signal from the same noise. Honestly, no one really knows where the performance ceiling will be for non invasive bci. There are companies out there today whose vision is to build a thought to text language decoding technology platform just using non invasive sensors that you can just put on your head. It's not yet proven if that will work or not, but there are definitely people aiming for it. Honestly, I think no one can say for certain that it won't work. I think they're all pursuing the scaling law strategy of we saw with language models just Collect more and more and more data, build bigger and bigger models and they'll get better and better. And so far no one yet in the world has collected a truly massive data set of, of brain data paired with language data. So like, who's to say whether you know what might be possible if you really try to scale this stuff?

Yeah, how, how do you, how do you invest in this category right now?

How do we make money off of this?

How do we make money off of this? Because the, the probably my personal default would just be to try to pick up some neuralink secondary.

Yep.

And just ride with them because it feels, it still feels like we're early. And if you just fund a bunch of alt Neuralinks, I think that that might be kind of a tough, tough game. But how do you think about it?

Yeah, neuralink is definitely the, like the most well known, the most well funded. Obviously it's Elon Musk and it has everything that that comes with it in terms of being an Elon company. It is also, I mean just thinking tactically from an investment perspective, it's also extremely highly valued today. Their most recent round is a $10 billion. There are a handful of very credible Neuralink competitors, including a company.

It could be undervalued.

Yeah. Do we have any ideas?

I'm just messing around.

I mean, do we have any idea on market size for these things? How do people even underwrite these types of deals? Are they building up to like the number of paraplegics that need the FDA approved surgery for that indication? Or is it more of like a consumer like tam? How do people think about that?

Yeah, I think you basically have to think about it in two phases. The first phase is the medical phase. The second phase is the like consumer slash, humans merge with AI in the singularity phase. And I think the TM math is a lot more concrete for the medical phase. And just in the medical phase, it is a massive tm. If you look at like basically if you just take everyone in the world that's paralyzed, that's like tens of millions of people and you can break it down by different causes, Als, spinal cord injury, etc. You know, many, many millions of people and the industry is kind of circling around this price point of 150k per implants with the expectation that insurance will cover it. Like it's not, it will be something that a lot of people will have access to because they won't have to pay out of pocket. So if you just do the numbers on that, you know, it can Be a many, many tens of billions of dollars market. But I do think that like the really big vision, the big prize that ever, you know, that all that gets people like Elon Musk and Brian Johnson excited is the longer term vision of this being like a consumer product for a more generalized population that lets anyone like seamlessly merge with digital intelligences and artificial intelligences and so forth. And you know, if you think that eventually BSAs will be as widespread as iPhones or something, then obviously you can talk yourself into a huge dam.

Imagine not having to open the Sora app, but just having a Sora implant. And so you're just sitting there.

Yeah, just at the trough. It'd be amazing. Are you tracking any of the intermediate steps? There was that company that launched a demo that it seemed like it would put something on your jawbone and it would be able to hear what you were whispering. We tried the meta Ray Ban displays and those were able to pick up on a very low audio signal. You could kind of whisper and it would still be able to interpret it with AI into translation. Is that actually like a critical path in the tech tree or is that kind of just a side note? Or do you think that whole tech wave will be like a big precursor?

Yeah, there are a lot of different opinions on this. So what you're describing, people often use the term silent speech to describe category, which is exactly as you said. Like you have a sensor somewhere, you know, outside your, your skull. It can be like as part of a headphone or like a hat or something that can detect the words that you're saying if you just kind of like barely whisper them or even, even if it's sub audible, if you're just kind of like barely moving your lips and moving your tongue.

Yeah.

And the idea is it lets you use all like all the value and possibilities of voice applications, but without having to make noise. And so like when you're in public, for instance, walking on the street or on a bus or something that like, that can be a lot more useful than, you know, you probably don't want to speak out loud to your phone.

Yeah.

I think to your question of like, is this just like an intermediate, like branch of the tech tree or not? I think it really depends like bullishness around silent speech. I think really depends on how soon you think these true BCI will be ready for primetime. Because silent speech isn't recording brain data directly, it's recording like downstream muscle data. And so if you can actually decode what's coming straight from the Brain in an accurate way. You can kind of just leapfrog silent speech. But if you think that there are real limits to what's possible non invasively with bci, then silent speech might be like a more practicable path. And you mentioned this company, Alter Ego out of MIT recently had this really buzzy launch video that like they frame it as basically telepathy, which is a little misleading because it's not actually reading brain data. It's like the guy is mouthing something and kind of moving his tongue. But yeah, that is another maybe more achievable near term version of this.

Chris in the chat asked, what about Chinese infantry using PCI to reduce PTSD and increase reaction time? Do you know anything about that?

I haven't heard about that specifically, but there's. Yeah, there's no doubt in my mind. I mean the CCP and Chinese startup ecosystem are investing a lot in BCI in the U.S. also, one of the big use cases in general, one of the big use cases is like treating neuropsychiatric diseases. So if you can precisely modulate brain activity, you can address a lot of health conditions like ptsd, but also like ocd, depression, Parkinson's, et cetera.

It's basically a lobotomy that works, right?

Yeah.

Which is kind of crazy to think about. Like our lobotomies, was that like somewhat on the right path? It's a very hot take. Nuclear take potentially.

Yeah. I mean to be clear, like these forms of neuromodulation, like you're not actually taking any of the brain, brain out the way you do with lobotomy. But yeah, lobotomies are just like, were very crude attempts to modulate the brain.

Yeah. Try and turn off certain regions of the brain and if you can do that with some sort of electrical frequency or neuromodulation, like you're in the same realm, which is people on a lot of FDA trials before they run that. But I don't know, maybe over in the Chinese infantry they'll do it live. Yeah. Is there anything else that's going on in the BCI industry that we didn't cover that? Like where all this goes next? Are we like. It's certainly not a trend. I guess the big, the big question for me is like, how do you apply the lens of what we're seeing in AI to the BCI world? Especially with regard to the fact that we might at some point be on a scaling, like on a scaling curve. And we've kind of seen this with like Tesla's now doing like large scale training runs for self driving technology. Google just did the Cell to Sentence project on the Gemma 3 models and you can see that they're starting to scale it up but they're not at like we need to build a 1 gigawatt cluster for AI and bio and I doubt that any of the BCI companies are like we need a 1 gigawatt data center to do are training because they don't even have the data yet. So it feels like we're still really early. But is that the right lens to be, to be tracking this through? Should we just be looking at like how much electricity is going to be going into the training run and then we'll know that it's actually working?

Yeah. I do think like you often hear people comment how they're like we keep discovering new data modalities. Like there are more and more modalities than people realize. Like there's language and image and audio and video but like biology is. There's a lot of modalities there and robotics and so forth. I do think neural data, brain data is a new modality that we're just right at the very beginning phases of starting to research into. And so there really haven't been any large scale efforts to bid to build foundation models for brains as you're describing. I do think that that's inevitable but I also think like, I think it's, I think this stuff is really relevant for the trends we're seeing in AI because I do think this is the inevitable next way in which humans interface with AIs. And you can imagine basically having a BCI that is connected to whatever the latest AI model is where rather than having to pull out your phone and type a question into ChatGPT, you can just think something and the answer will surface in your mind. Or if you're brainstorming in your own head, you'll have a thought partner there that's brainstorming with you. You'll be able to instantly communicate with others. Once you start kind of playing out the possibilities of what could happen if there are BCIs at scale. It does really change the way humans interact with information basically.

I'm sure Eliezer Yuda Koski loves this. It's fantastic. It is exciting though and I appreciate you coming on the show and thanks for breaking it down, breaking it all down. Have a great weekend. We'll talk to you soon.

Yeah, thanks.

Thanks for having me.

Cheers.

Bye. Let me tell you about Fin AI, the number one AI agent for customer service. Number one in performance benchmarks, number one in competitive bake offs. Number one ranking on G2. Tyler, are you going to get a BCI intern challenge? Get the Utah Array installed. Can we pull up a picture of the Utah Array? I put it in the top.

It is very scary looking.

You've seen it. You've seen this thing. It is just straight up spikes in your head. It's just a few wires.

I feel like, like most protein powders, it gives you brain damage.

Yeah, this thing. Yeah, this thing. This feels like just one notch forward from just actually getting a lobotomy. But it's. It's a rough looking device, but I guess it actually worked in some ways. I got some data, but, you know, I've often said that I would be. I wouldn't be the first person to go on the rocket to the moon or the or to Mars, but I'd be like number five or ten. As long as the first couple people make it back, I'm good. What about you? What number of person do you want to be? Do you want to be the first person with the millionth? You want to be the millionth Neural Link customer?

Billion.

I'm probably in the first. First 50,000.

I would say first 50,000.

I mean, there's an interesting thing.

What?

Speaking of like doomers, I don't know if it's Eliezer himself, but a lot of people have said like one possible solution to the kind of doom snare is that we. It's like the merge, like that would actually be prevent. Like the, you know, is taking over.

Because like we become the. They can't take us over if we're them.

Yeah. Jordy, what number are you?

You're going to wait 100 or 1 million?

Big, big number. We're going to need a bigger gong. For my number, for your number you're.

Going to get up there. I don't know. I mean, I talked to Nolan from Value Link. He seems like he's having a great time with it. He gave me a good review. I'm down. Wire me up.

Yeah, the value right now, from what I've heard.

Yeah, I'm pretty quick with the mouse and keyboard right now, so I don't necessarily need the feature that it's offering.

Yeah. And I don't necessarily need you to know my. I mean, I'm sure you can be able to build the right user experiences around this, but my first thought to a question is not necessarily my. My best thought.

Right.

And so if we're taking a lot of training here, prepping the show and we're just having this instant communication. It's just like our worst take going this way and then a bad take against a bad take going back.

Yeah, you would need a lot of translation. But imagine popping Adeo with your. With your bci. That could be the future customer relationship Magic. Adio is the AI native CRM that builds scales and grows your company to the next level. Can you imagine? So a lot of these companies, they, you know, a common tagline right now is, you know, Addie is the AI native CRM. Imagine when the BCI era, when AI is old news. You don't want to put that on your landing page. BCI is the new hot keyword. So you say, we're the BCI native CRM that builds scales and company grows your company to the next level. I'm excited for all the landing pages to update to BCI native. That's what I want. I also want you to get an eight sleep. And Jordyn, I heard you had a rough night. I think you got to get a sound effect ready for me because I got a 93. I slept for 7 hours and 25 minutes. Let's go.

Got a 47.

4 hours, 28 minutes and dinner last night was in Malibu. I thought you just went right home. But, you know, the rough nights come for us all from time to time.

Left late.

Eric Zhang says all according to plan. He is highlighting the fact that all of his. Everyone with the same name as him.

Is they're either building, cooking, or at thinking machines or OpenAI.

It's a good time to be Eric Zhang.

Congrats, Legends. Legends. We still have to do the Eric Zhang show.

Oh, and this is Eric Zhang is quote tweeting a different Eric Zhang. I love it. They're going back and forth on the timeline.

Yeah, we got to get all of them on the show.

We do Eric Zhang guy. They're all invited. Talk about OpenAI, thinking machines, cooking and building.

Tyler, can you work on this? This seems.

Invite them all. We want the six up. All of them.

Logan Robinson says when Walmart jets take off for store visits, no one but the pilot knows where they're going. I wonder if they have them all blindfolded. Rumor is you're only told a few things to help you prep for the trip. Domestic or international, how many days you'll be gone. Climate of the destination.

What does this mean?

So this is like executive. I imagine like people at corporate going to visit stores and seeing how they're happening.

Okay. They want to show up.

I don't want to call ahead and be like, hey, you got to really prevent executives that are responsible, I imagine for like certain things within a store to not front run it and be like, hey, make sure things are clean. Make sure like everybody shows up on time.

Yeah, yeah, that makes sense. They want to do a surprise visit.

Yeah. So for nearly everyone, including store associates, the visit is a complete surprise. Gotta keep them on their toes. When these store visits happen, the goal is nearly always the same. Inspect what you expect. Scheduling these trips in secrecy makes it possible to see the stores as they truly are. No fanfare or pretending. I was never lucky enough to travel in one of the jets, but the purpose of these trips was not lost on me. In my cleaning business we do something similar. Anyways, Cool concept and beautiful jet.

And if you're long Walmart, you're short Walmart. Head over to public.com and enjoy multi asset investing. Industry leading yields. And they're trusted by millions.

Clyde has been at a conference. Looks like he got a Lucy hat.

He did.

And a lot of other stuff.

This is awesome. Wow. And some breakers there too. He got samples. He got.

What conference is this? He got a happy dad hat.

Yeah.

He should start a convenience store, corner store. That's only powered by conference. They only sell.

Yeah. You could really resell. You could really resell a bunch of this. All these brands should head over to adquick.com, out of home advertising made easy and measurable. Say goodbye to the headaches of out of home advertising. Only Adquick combines technology, out of home expertise and data to enable efficiency in this ad buying across the globe. The other thing that you need to do genuinely, if you are running a CPG company, you will wind up at a lot of conferences like these. You will be talking to convenience store owners and you'll be talking to distributors. And as much as we like to joke about luxury watches, it is a very important aspect of the job. It really does. Like when you're shaking hands with people in this particular industry, tech has been, you got to be an aficionado. You got to like the craftsmanship of a particular watch. It's a unique choice in tech, but in my experience in consumer packaged goods, when you go to a CPG conference, what's on your wrist actually matters. So if you're building a CPG business, maybe you should head over to getbezel.com because your bezel concierge is available now to source you any watch on the planet. Seriously, any watch. No joke. I seriously got. I became like fast friends with one of our major distributors, talking about watches because he was super into them. Anyway, our next guest is already in the Reese room waiting room. We have Nathan from Air Street Capital coming into the studio. Nathan, good to see you. How are you doing?

Welcome.

Great.

I was worried.

Great jacket.

It's light where you are. I was worried I was gonna keep you up late because you're usually in Europe. Give us a little introduction on yourself and backstory and then tell us where you actually are if you can.

Incredible jacket too. Jackets are a great sign of respect in our culture.

It is. Nice jacket.

Yeah.

Thank you. Thank you. It's life at the AI frontier.

Yes.

Yeah. So I founded a firm called Air Street Capital in Europe, invest in AI companies at the very early stage. I've been doing this for about 10 years now and spend kind of half my time in the U.S. and in Europe. Thank you.

And you're in the U.S. now?

Yeah, I'm in New York.

Oh, nice. Very nice.

Yeah.

Well, on vacation or investing?

No, I've actually been doing a bit of a state of AI launch tour, if you will. So we did a meetup in San Francisco last Thursday for the report drop.

Yeah.

And then I did one in New York last night.

And what is the state of AI?

Yeah. So before we get into that, I have one.

Is AI good?

Yeah. One question. How, how competitive are AI rounds at the early stage? Like precede seed today? I know I'm sure they're getting more competitive, but I think historically if you could write a million dollar check at the idea stage, you had a quite.

A chance to build a real position.

Yeah, yeah, yeah. I mean, in that sense, I think the game has like materially changed. There's a bit of a bifurcation. Like you have the sort of large lab model companies that are raising significant amounts of money. And I think those are also potentially good investments. But they, they definitely cater to a certain type of venture firm. And then you have the other category which is potentially more pragmatic entrepreneurs going for less of the general purpose AI, but more looking at kind of what use case can they solve with a general purpose AI system. And, and those ones are a little bit more modest, but in general, yeah, you've seen, you know, inflation of, of round sizes, but also actually interest from customers to buy this stuff. And so founders just want to take a bigger swing, want to go faster. And that's what's driven a bit of the bit of the round size increase.

I feel like a couple of years ago the MEME was, even if you're a seed stage company, as long as you slapped AI on there, you could justify like $100 million raise. Because who knows, we might need to train a model. We might be doing some real AI, not just a rapper company here. Has that dissipated as it's become clearer that to win in that game you need hundreds of billions of dollars and so your actual cost structure should look a lot less like. I see GPT4 as like a capex expense, basically like a huge training run. And for most AI companies it's going to look a lot more like opex. Until they have a customer, they don't really have a big token bill. And so why do they need $50 million at seed?

I ask myself this question a lot. At the end of the day, to some degree, as an investor, you're not the one deciding this. The market sort of pumps what they will. But I think the biggest frame change, and you can see this documented in the state of the art report over the last eight years that have been doing it, is a lot of the capabilities that you get out of the box of a model today, whether it's language, reasoning, image, video, biology. If you showed that to me or others 10 years ago, we would have probably said this is magic and not possible. And so I think there is just so much you can do with what's available today that it's almost a gift to the startup community that now you don't have to blow their 3, 4, 5 million dollars seed budget on trying to build an AI system, collect the data and then throw it over the parapet to a putative customer who says, you know what, I don't care, I've got other priorities. What is this AI thing? Doesn't even work anyway.

Yeah.

So now you can run your whole seed budget on product, market fit, experiments. So it's almost like back to kind of OG seed investing, which is less about, you know, what the initial starting idea is, but how capable is the team to just run a lot of experiments and find something that works and then, and then, you know, rip it when it does.

Yeah. What, what's the headline KPI that you like to center on in a state of AI report? I mean, you can focus on market caps, you can focus on revenue growth, you could even look at gigawatts and electrical numbers, you could look at number of H100 equivalents that are being built out. There's so many different ways to ground the shape of the AI build out. What do you like to focus on?

Yeah, frankly, I mean the report is about 300 pages. So the honest answer is all the things. But, but so what we started doing a few years ago is this compute index where I basically combed around the Internet to try to find all documented cluster sizes for various companies. And we've been charting just how much of an expansion has happened there in the last three years. It's basically probably like 100x increase. The other cool thing we've done is like mine, all open source research papers and programmatically determine which chips are used in those papers so we can give people like a leading edge view for what researchers like in terms of hardware. And so with that we've shown, I think the only evidence to this, to this I've seen which is that 90% of AI research papers rely on an Nvidia chipset. And through that kind of work we've seen AMD start to grow a little bit as well. Out of nowhere.

Of course it does seem like Nvidia or AMD is really picking up this year. We've seen this in clustermax from semianalysis. They kind of did their, I don't know, year of efficiency or in some ways and started squashing bugs. And then there's just immense pressure when there's you know, hundreds of billions of dollars going into the next, the next phase of the build out that, you know, Nvidia's margins are going to be in the, in the crosshairs for sure.

Yeah. And strategic alliances with, with Open Air that I think is, it's quite interesting because there's, you know, if you read through the, through the news, you'll see that actually OpenAI has warrants in AMD at a strike price of 600.

Yeah.

And so there's like some interesting incentives there to try to, you know, get AMD stuff to really work. In some degree it's not as similar to like the relationship I suppose between Anthropic and us for Trainium and us is hardware is clearly behind, you know, the bigger competitors like Nvidia. And so they get sort of like it help desk support from, from Anthropic to make the stuff work.

Yeah. Do you think we're in a bubble?

There's, there's probably local bubbles, you know, all the time. I think in like our little ecosystem of AI there could be one in this like Capex, you know, build out and all the interconnectedness with various companies. I think as soon as you see like Wall street finance get excited about something, you have to think a little bit why they're Getting excited. Yeah. But you know, at the end of the day, I think the outputs of this bubble, if there is one, are really, really productive. And back to the point around like this technology is magic. There's. There's so much value that's getting created. And we've documented some of this with, with our mutual friends at Ramp. Looking at how much U.S. companies are spending on AI services. And that's going up, the retention is going up. Some of our investments in like 11 labs and Synthesia and others are just like growing because it's genuinely useful stuff.

Yeah, it feels very different than what about.

What was your reaction to that Deutsche bank report on ChatGPT spending growth? I looked at it and it felt hard to judge. Cause we're looking at summer, basically. But how did you process it?

I mean, yeah, astronomical numbers.

But I guess the question is the TAM for paid AI subscriptions in the consumer or prosumer realm saturating. And I mean, we're certainly seeing messaging from OpenAI that says we want to do agentic commerce, we want to increase the monetization or our free users. We want to get into advertising, potentially affiliate revenue, API business. Like, it doesn't just feel like there's an unlimited pool of people that will pay $200 a month for, you know, a great product, a magical product, but there just might not be that many people worldwide that are like, sign me up for the $200 a month plan.

Yeah. So I mean, along that notice, I was also interested in this topic. So I surveyed about 1200 AI practitioners for the report. And all this data is open access through it. And I was really surprised because 95% of the people who took the survey use AI in their personal life and their professional life. And it's like 75% of them pay out of their pocket for it. And 10% of people pay 200 bucks a month or more.

Wow.

And. And like this is a pretty, you know, educated crowd in North America and in Europe. But I was surprised at those numbers. And then the second thing to your point around like agented commerce, I think that's huge. We have some data in the report that shows that conversion of E commerce that's derived from kind of agent conversations converts at a higher rate than direct. And so you can imagine that the brands that sort of jump on this earlier and create content that agents are using to best learn what the right solution is to serve to their human that they're serving are actually going to probably grow their revenues faster. They're like this recursive cycle, particularly as more and more labs and companies want to do this reinforcement learning to improve. Improve personalization. I think there's a lot of adjacencies that one can navigate into.

Actually, a week or so ago, the EU announced their $1.1 billion plan to ramp up AI in key industries. What was your. Obviously people in the west coast were kind of mocking it, considering that the whole plan was less than Sam Altman's car. But you know their meta's recent post.

Yeah, yeah, Single.

But yeah, my reaction was like, that's kind of. The takes were funny. But like, at least, like there is a plan. At least there's.

What is this evidence of?

Right.

And I guess like when you read it and reacted to it, I guess like what was your immediate thought on what you wish the EU was doing?

Yeah, I mean my reaction wasn't too similar. There's been like a meme kind of bandied around in some of our friend group, which is back to the Derek Zoolander, like, is this a house for ants?

Yeah, exactly.

Is this a plan for ants?

But like does. But like, like Europe did not build their own Google and it was fine. I mean they tried, they, they like missed out on it. But like the natural state of the market equilibrium was that Google was built in America and America absorbed the market cap and European investors could go invest in Google if they wanted to. And Google eventually built data centers to serve Google products locally. But there was never like, we need a, we need our own thing. Is the thinking around like national AI changing? Because I feel like there's a world where even if you're using some like locally trained model, like if you're going to chat.com, like it's still OpenAI wins.

Yeah.

So there was an initiative, I suppose I'll call two examples. So there was one in Europe called Quero, which was like a government kind of funded initiative to try to build a search engine that best represented European interests. Because allegedly Google wouldn't because it's an American company.

Yeah.

And obviously no one knows about it because that flopped. And then, you know, the second example is the French government not wanting to buy Palantir.

Oh yeah.

For many, many years, as you guys will know pretty well. And. And eventually they caved in after burning the same amount of money they'd probably have spent on Palantir and got it anyway. And yeah, the latest example of all this sort of infrastructure build out in the US and in China is now every country wants to be sovereign with regards to AI.

Yeah. So so here's an interesting like, like, is Putin the least AGI pilled world leader? Like, I was trying to get research on how his cluster, on their clusters over there, they were like, yeah, we have 800 GPUs and we're running them hot. And so like they're basically calling the world's bluff on AI.

We'll be fine without.

Did say that he actually wanted to have AI supremacy about a year or two ago, but admittedly maybe there's less evidence of that. Yeah, I mean, to some degree also, like, you know, Yandex was a pretty amazing Russian search company and then they've since left and relisted on the nasdaq. Is Nebus.

Nebus. Yeah.

And one of the clouds is really crushing it because they're, they have really good technology.

Yeah.

But yeah, I think this, the sovereign thing is a little bit of like a meme or a misnomer. I mean, I think it's great for the suppliers to the technology because they have yet another like TAM to sell to. But you know, if you're a non American nation and you buy Nvidia chips and you buy American technology and you know, Chinese technology and you just install it in your country, I don't necessarily see how that makes you sovereign because these things need updates. They can be switched off, basically.

Yeah, they have a relatively short life too.

That too useful life. Exactly. So I think sovereign is a little bit more of like an alignment of technology agenda and political agenda than it actually is the true meaning of the definition of like we control our fate when it comes to intelligence and running it.

Okay. Moving forward, a lot of people are nervous about cracks in the global economy. Cracks in the AI economy. What are you monitoring? Like, what are you. What do you think is the most important metrics or data points or stories to be really digging into these days?

Yeah, I mean, I think it comes back down to the return investment. You know, cases of companies becoming more efficient, launching better services. The reason why I do this at all is because I think I unlocks new use cases that weren't possible before. And, and so for so long as we see those examples, you know, propagate through the economy and it's still very early. I think we're in a good shape and we only talked about software here, but there's so much in, you know, national security and defense that's, that needs to be solved. And many of those problems, whether it's like, you know, electronic warfare or autonomy, rely on AI and so huge TAM there also in Europe and we've got plays there with, you know, with Delian and a few others companies in Europe. So I think it's like it's full speed basically.

Yeah.

Exciting times. Well, thank you so much for joining the show.

Thanks guys.

I hope you have a good weekend. Talk to you soon.

Great to meet you.

Thank you.

Have a great rest of your day. If you're looking to take a trip to New York or anywhere, head over to wander.com find your happy place. Book a wander with inspiring views, hotel graded minis, dreamy beds, top tier cleaning and 247 concierge service. It's a vacation home, but better. We have some breaking news. Apple has entered a five year deal to broadcast Formula one in the United States.

Wow.

Yeah. Walter Bloomberg. This is very exciting. We were wondering where they were going to land. They landed with Apple and I love the Apple strategy.

How much did they pay?

We don't, I don't know. We should look it up. But I love the Apple strategy. Apple did the movie F1. So then you watch the movie. Then you're like, I want to watch the actual, actual race. I can watch it on Apple tv.

Yeah.

The more I thought about it was so funny. They basically took Drive to Survive.

Yep.

And then they made a movie with these fictional characters and they but they just like, ultimately just like it's a barnacle economy.

Yeah.

Opinion really is. They just like, like we Drive to survive is this whale that was making. Making F1 relevant in the US I feel like they just attached onto. It's a big barnacle.

Wait, who's the barnacle and who's the whale? I feel like, I feel like Apple.

Started as a barnacle.

Okay.

And they're becoming the whale.

Hmm. I feel like, I feel like F1 is the Barnacle on Drive to Survive. Honestly. In the, In America.

Yeah.

Like F1 is a barnacle on the whale. That is just Netflix making reality TV out of everything, basically. But we need to figure out that.

Says, can't believe my elderly mother broke the F1 news to me before Walter Bloomberg.

Let's go. Mothers tapped in.

Ryan says five year deal worth about 750 million. They were originally going out for 200.

Million for one season. Yeah. So that would have been a billion dollar deal.

So.

Yeah, they got, they got 75% of the way there. Not bad. Tim Cook, do you remember when he.

Was trying to wave the flag?

Oh, I know. I don't remember that meme. That image. I remember him being at some DJ booth. Was it like John Summitt? Was playing at F1. I think I saw a video.

Yeah, I got. I got a video. I'm dropping in the timeline.

Yeah.

People are saying this is the worst checkered flag of all time.

Oh, no.

We'll give him. We'll give him some slack because he's.

He's barely making it by.

He's one of the goats. One of the supply chain goats.

Yeah.

He's okay. Here he is, Tim Cook. This is at F1.

Yeah.

And they. And they want him to wave the checkered flag. That looks fine.

No, it doesn't.

How does one wave a checkered.

That is not how you would wave.

How would I wave it? Would I. Just more gusto, you think?

Look at how lackadaisical this wave.

It does seem like he's kind of phoning it in.

Looks like he's going fishing.

It does.

He's at the finish line of a Formula one race and he looks like he's going out for trying to catch some trout or something.

Wow.

But he got the deal done.

But if they hadn't done this and Instead of spending $150 million a year on F1, they could have tripled Tim Cook's pay package. Which one do you think a more incentivized Tim Cook could potentially create more value. Which would you do?

He was so embarrassed by the fishing inspired flag that he's like, I'm gonna. I want to own it all.

I want to own it all.

I want to own it all. I want to shut it down.

Yeah.

Yeah. I'm going to drive the sport into the ground. And he might get more satisfaction out of that than a higher salary.

Before we close out, we need to expand the barnacle economy taxonomy. What we need to do. So the whales. It's obvious who the whales are. That's anthropic. That's the hyperscalers. That's OpenAI. That's the foundation model labs.

Right.

They're the whales. They're growing massively. They're putting up huge numbers. And so the barnacle can be two things. It can be one, the company that sells to the whale, or it can be merely a company that sells something that buys from the whale. And so I feel like we need two different terms. The barnacle economy, specifically, like the training data providers. Right. That beals barnacle economy. But the AI for lawyers. That does not feel barnacle. Right. Because they have to go out and win customers in the legal.

I think if you can swap out the underlying intelligence, you're not a barnacle. Basically, you're Shark, Shark barn. And ultimately they don't have to sell to the whale.

Yeah, yeah, exactly. They buy from the whale.

Yes.

So the shark buys from the whale and goes up. You might have a Nobel on your hands here with barnacle theory. It's happening. Well, is there any other breaking news?

Bloomberg this morning or I think it was yesterday. Apparently Cantor is trying to do this private raise for Tether at a $500 billion valuation. But Bloomberg reports that Cantor invested 600 million in a convertible note for 5% stake in Tether a year ago. A $12 billion valuation for Tether. Now Cantor is trying to tell investors that tether is worth 42 times more in just one year. And when Cantor is obviously conflicted as 5% owner. Yeah, that ain't happening. Yeah, I mean the, you know, a stablecoin leader is going to be worth a lot more after the regulatory changes this year, but yeah, there's also a lot more competition.

So what's Circle at right now? 30 billion. So yeah, I mean 15 times the price of Circle might be high. I don't really know. I know Tether's a great business, so I wouldn't be surprised if they get a major up round done. In other news.

Yeah, I guess that. So what's the current market cap of USDC is? 75 billion.

Well, is that for the token or is that the actual.

That's the token itself.

Circle, the company is 30 billion market cap. And so.

Here'S the issue. I mean, Circle has wildly different economics, right? Wildly different economics. They have to pay a large amount of their revenue out to Coinbase and I'm sure other exchanges. But tether is at 185 billion of market cap.

So it's three times the size with, with better economics, potentially like 500 billion. Doesn't seem that crazy to me. I don't know. I, I, I haven't done any of the actual math.

Question is how, how durable is the.

Yeah, yeah. Especially when everyone's going to be having their Kohl's cash soon.

Who knows? Adam's in the game. Well, Bridge obviously wants to create a lot more.

Zephyr has a post here. Big contract. Chinese actuator component supplier San Jua Intelligent Controls is rumored to have secured a 685 million dollar order from Tesla for Optimus Sanhua's stock surge 10% to the daily limit following the news going viral. While other actuator suppliers also saw they.

Put a significantly, they put a daily limit on how much.

That's extremely un American. I don't like that at all. Matt Fried says Tesla not building their own actuators is the least mission driven thing I've heard all year. I don't know. I mean there's.

Yeah. What is this? What does it say about Tesla's timelines on like actual humanoid value? Because if they were prototype at this point. So you know, I was going to say the opposite of like if you felt like you were having five years until you were actually going to be.

Yeah.

Selling these to people, maybe you would just build your own if you were trying to sell them and get them out today in. In the world faster.

Time to build the actuator facility in America.

I gotta spend over half a billion with a rival.

Yeah. I don't know. In other news, Ian Goodfellow is crazy.

More of what determinism?

I've met him. He signed a book that he wrote on deep learning for me. It's in the museum of business history and technology History. More of what my colleagues and I have been working on in AI for fusion is public now. Why are we clapping? We're announcing a research collaboration with CFS Energy, one of the world's leading nuclear fusion company, Commonwealth Fusion Systems. DeepMind is goaded is actually crazy. Together they're helping speed up the development of clean, safe, limitless fuel fusion power with AI. Yeah. Awesome to see Google working in fusion. Obviously Sam Altman's a believer with Helion. We've had a number of fusion founders on the show. It's one of those technologies that's always been 10 years away but feels less than 10 years. Finally we are making some progress. What's your fusion take?

I mean I'm very profusion but in other DeepMind news, apparently Gemini 3 is actually slated for December. So there were some rumors of like maybe next week, but now there's some people saying it's actually add 30 days.

To the singularity counter. This is just not going to happen, is it? When, when is AGI coming? This is. This is. This is the worst news.

I don't know. But, but it will be a nice Christmas.

I was saying December 25, they got it, they got kids.

My kids are getting Gemini.

They got it. They got to do it December 23rd. You know the, the, the VR headset sets, they always like rocket to the top of the app store. That's what I want to see.

My kids are getting a box of tokens.

Box of tokens. We can close with Will Menitis potentially. He's got a bunch of stuff. Well, I mean there's.

Hey, I just wanted to give it up for Stargate uae. It's taking shape. We got a video in here. We should try to pull this up. Led By K, this 1 GW AI infrastructure cluster is a cornerstone of the 5 GW UAE US AI campus in Abu Dhabi. A key part of building the intelligence grid, the digital backbone enabling secure sovereign, and, I don't know, AI because it's cut off. But we can see this video here.

Yeah, very cool. Will Manitis says if you're a young person interested in going deep on weird things, basically the only good advice is that you should be 1000x more commercial. Someone will get very rich by monetizing the things you figure out. There's no reason that person shouldn't be you. That's good. Good. Well, Andrej Karpathy dropped a podcast with Dwarkesh Patel. It's a banger.

I listened to the first, like 20, 25 minutes. Just a banger so far.

Perfect for the weekend. Go enjoy it. TVPN is off. We do have a diet TVPN product now.

Yeah.

It's interesting. There's one quote that I guess Nick was calling out. Nick obviously is more Grok aligned love. But he said he was quoting Dependent Freeze, Grok's most loyal soldier.

Yes.

He was quoting Altman earlier. It said 2025 is the year of AI agents. I think you could still defend that with Codex, but certainly we're not seeing maybe the agents that. That we were all super excited about. Right. Deep research as an agent. Coding agents.

I haven't booked a flight with AI yet.

Yes. Carpathy says I was triggered by that. Over prediction. More accurately, it's the decade of agents. So much work to be done.

Let's go humans in the loop for another decade. We got this.

Yeah, yeah, that. That's like normal.

What do you think? More or less. You keep being like, we're just on track. Superintelligence right around the corner.

Yeah, we are.

I mean, it's always hard. It's like, what is. What do you mean by agent? Like, what can actually do in 10 years? I feel like.

Like, what have agents done for you today?

Wrote me some code.

There you go.

Yeah, yeah. And I.

And I.

And I. And yeah, I agree that I can. I think this was the year of coding agents.

The year of coding agents.

Yeah.

And I think computer use is like getting better and better. Hopefully we see Gemini 3.

Much better computer use. A good example is like when you made the video Vibreal for the Metis list, you were not able to puppeteer Premiere Pro with an agent. And at the end of the day, that's just like clicking buttons on the computer. It could use FFmpeg. It could use control the mouse and keyboard and screen record. And that agent isn't here. And you should be able to just, in natural language, talk to an agent that then makes the cuts in Premiere Pro. And I still believe that. Even when we get there, you'll still need to say, I have an idea to include clips from Cheeky Pint or whatever you picked, or these particular walkouts. But it's not there. It's not generalized. Yeah, but that's not generalized.

That does not feel 10 years away at all. No, that feels like right now I could. I could build a RL environment and do that if I had enough compute in, like, you know, a month.

Oh, I just need more compute. Just more compute. Oh, now you're asking your computer over there. I see how it is. Okay, well, anyway, have a fantastic weekend, everyone. We will see you on Monday. We love you. Leave us five stars on Apple Podcasts and Spotify, and have a great weekend. See you Monday. Goodbye.

Cheers.