Loading summary
Transcript1,155 lines
- [00:00]
A
You're watching TVPN. Today is Friday, October 24, 2025. We are live from the TVPN Ultra Dome, the temple of technology, the fortress of finance, the capital of capital.
- [00:11]
B
And we're here. We have a special guest friend of the show.
- [00:14]
C
Happy to be here. Thanks for having me.
- [00:16]
A
Thanks for stopping by. We wanted to have you on the show for a few reasons. We are going to run through the mansion section and get your opinion on what whether these homes are worth passing up or an absolute steal at something like $50 million. So we'll be going through that in a minute. We also wanted to hear from you. What's new in Feed Me World? Do you have any scoops for us? Do you have any breaking news? Do you have any exclusives? We invited you here.
- [00:43]
B
So relaxing.
- [00:45]
A
We want some news.
- [00:46]
C
Well, I've been in LA for a few days and I got some local LA news from some of my readers because I did a pop up for a few days to sell these new hats. Which LA readers got first. Some of the people that came by had TVPN merch, which was cool.
- [01:02]
B
There we go.
- [01:02]
C
They're fans so it's good to have crossover. What news do we have? I'm hiring an intern, which you guys are saying not to say.
- [01:10]
A
Interns. Breaking news. You heard it here first. That's why you come to us for scoops like this.
- [01:15]
B
This has never been shared. She did not post about this included in the newsletter.
- [01:20]
A
It's not breaking news until TVP uncover.
- [01:22]
C
That's right. That's right. What else can I say? Feed Me's first podcast is going live in two weeks. We get to push back the live date and wait.
- [01:33]
A
Run through the thesis of the podcast. I thought it was good.
- [01:35]
C
Oh, it's like. It's a restaurant podcast. It's hosted by my restaurant critic. I think it'll be fun. We have this idea that there's sort of a white space in kind of. Wait, you guys don't curse or you do.
- [01:47]
A
We don't curse, but you're welcome to.
- [01:49]
B
You're welcome if you feel like.
- [01:51]
C
I'm not going.
- [01:51]
A
If you want to be Joyce, the first person to say a swear word on our show.
- [01:55]
C
Yeah, it's sort of. Everybody likes restaurants and anybody can talk about restaurants. So we have some interesting people coming on to talk about restaurants with our restaurant critic, which will be a blast. That'll go live in two weeks. Thanks, Substack, for supporting that. Some other things might come up that'd be fun for the next few Minutes.
- [02:14]
A
Yeah, I like that. It wasn't just like the Feed Me show. That was intentional. Right. Because that would be the logical thing.
- [02:20]
C
Is just, like, to do, like, Emily's show.
- [02:22]
A
Exactly. Or, I mean, it could.
- [02:24]
B
Feed Me would kind of be a good name for a restaurant.
- [02:27]
C
I know, I know. I was talking to somebody the other day.
- [02:31]
A
Would you have a different brand for it?
- [02:33]
C
It's called Expense Account.
- [02:34]
A
Okay.
- [02:34]
C
Which is the same name as the restaurant column.
- [02:36]
A
Is Ramp sponsoring it? They need to.
- [02:38]
C
They should.
- [02:39]
D
I'm having some.
- [02:40]
A
They're sponsoring us. Ramp.com Time and money. Saving money. Save both cards, bill, pay, counting, and a whole lot more all in one place.
- [02:46]
C
Yeah, they. They should.
- [02:48]
A
Where is my single, by the way? Is it over here? Where? Oh, which one am I on? Am I on this one? There we go. Okay, now I found my single. Perfect. Thank you.
- [02:57]
C
Some people read my substack who watch this. That's so good.
- [03:00]
A
Yes.
- [03:00]
C
That's so great.
- [03:01]
A
Only David Senra can curse on the show. You're keeping it.
- [03:04]
B
That's true.
- [03:04]
A
Okay, well, we are about to understand the landscape of the restaurant world thanks to your new podcast.
- [03:13]
C
Yeah, it's gonna be great.
- [03:14]
A
Which is called Expense Account. Expense Account.
- [03:17]
C
Expense Account.
- [03:18]
A
But the real reason that we wanted to have you on the show was to help us the media landscape.
- [03:24]
B
There's been a lot of debate. Right. There's been a lot of talk. Paramount, Skydance, you have, you know, the free press starting a conversation. You have media. People like to talk about media. So today we're going to talk about media also.
- [03:39]
A
I think people have just simplified things way too much. They try and break it down into a binary. Are you legacy media or new media? Right, right. Are you legacy media or new media? We all know it's way more complicated than that.
- [03:49]
B
And people like to say if you can't explain something in simple terms, you don't understand it.
- [03:53]
A
Yes.
- [03:54]
B
That doesn't apply here.
- [03:55]
A
Today we will do it.
- [03:57]
B
We are going to break it down. We have a market map.
- [03:59]
A
Market map? Yeah, sort of market map. I'm going to kind of walk us through a little bit of it, get feedback from Emily and Jordi and Tyler. And if we get anything wrong, take it up with Tyler, because he's the.
- [04:10]
B
One that created this, made this graphic. So if there are any errors on it, you know where to go.
- [04:16]
A
We have to start with the mainstream media. What is the mainstream media? Arguably the best.
- [04:20]
B
What is.
- [04:21]
A
I think it's the best stream, because if you were to pick a stream to do media in Would you want to be in the main one or some alternative one? You want to be in the mainstream? Ideally, I think everyone is on the march to become the mainstream media. But mainstream media is broken down into several sub sections. You have the legacy media and the traditional media.
- [04:41]
B
And why don't you explain the difference? Because a lot of people would say they're the same thing.
- [04:45]
A
But a lot of it comes down to the aesthetics. The aesthetics of. You're seeing serif fonts, the New York Times, the Washington Post. When you go over to Fox News, BBC, cnn, you're getting something that feels like it is fighting for its life.
- [05:03]
C
It looks like tv.
- [05:04]
A
It looks like tv. It looks like tv. I need a little bit more on. My lab is really doing a number. Here we go. Okay. Is that better? Can you guys hear me? Okay.
- [05:14]
E
Okay.
- [05:16]
A
I got really caught up.
- [05:17]
B
This is your Sam Hyde bit. Sam Hyde's TED Talk. Yeah, you know, you know.
- [05:21]
A
Okay, so there is a continuum of. There's a gradient that exists between legacy and traditional. Cnbc. Right in the middle, you have Forbes, you have the Financial Times. And you can kind of tell that as you get towards Business Insider, cnn, you're more in the legacy world.
- [05:38]
B
You're kind of in the danger zone.
- [05:39]
A
You're in the danger zone. David Ellison might be kind of licking his chops saying, oh, maybe I'll take you over at some point. David Ellison, he's not going to make a play for the Wall Street Journal, for the Washington Post, for the New York Times.
- [05:53]
B
Part of traditional means that you can actually get it via newspaper. Right?
- [05:57]
A
Or yes, that is key.
- [05:59]
B
That's a factor.
- [06:00]
A
A lot of people will put the split divide of like your new media if you have a blog or your legacy media, if you print it out. But we forget that there is a whole host of, of mainstream media outlets that do not print out their work. They do not ship a newspaper. And so most of the newspapers you'll see the Wall Street Journal, the Washington Post, the New York Times, those are in the traditional media.
- [06:24]
B
Economists.
- [06:25]
A
Another one.
- [06:25]
B
Yes.
- [06:25]
A
The Economist. Another one. Atlantic. Yes. So this is really the heart and soul, the beating pulse of the mainstream media. But you also have offshoots from each of these areas. So you have the post trading.
- [06:37]
B
Let's talk about post trad and why I see this little number right here, Bloomberg.
- [06:41]
A
Bloomberg is moving over into post trad media. Obviously that's on the back of Tracy Alloway, Joe Wiesenthal with the Odd Lots podcast. You can think of that as the classic post trading What?
- [06:53]
C
And Tracy.
- [06:54]
A
Yeah, I said Tracy.
- [06:55]
C
Oh, you did?
- [06:55]
A
Yeah. Tracy Alloway and Joe Weisenthal at the Odd Lots Podcast are a great example of post trad media where they come from traditional media, but they have been so early to the podcasting game. They're what, a decade in. Isn't it a decade?
- [07:11]
B
I think so, yeah.
- [07:11]
A
A decade into that. And so the joke Bloomberg was really.
- [07:15]
B
An experiment, you know, from Michael Bloomberg. What if I could just be the news? What if I could create the news?
- [07:21]
A
Yes.
- [07:22]
B
And so there'd be some frustrations with traditional media and you know, saying, why not me? Right?
- [07:27]
A
Yes, yes. And so you can see him like he's slightly less traditional than the New York Times, but still, you know, kind of in that.
- [07:34]
B
Squarely in that post trad region.
- [07:38]
A
We also have proto media. I'm not exactly sure why Digg and Hacker News wound up there. More just like early online efforts that weren't really trying to play in these weren't trying to be traditional media corporations, media organizations. They weren't telling a story about being new media in any way. They were sort of the program.
- [07:57]
B
But you could almost put Reddit here. We're missing Reddit in some ways, but Reddit was trying to be the front page of the Internet, curate stories.
- [08:05]
A
Yeah, I mean, you could probably go all the way up here and add social media and TikTok and stuff, but we want to stay on the media outlets, the folks who really proclaim that they are the media. Right. And so over here we have the post legacy media. And this is a crew of former legacy media folks who have now ventured out into substack and YouTube, substack and substack.
- [08:29]
C
Would you argue that a few years ago Barry, like the free press would have been there too? Post Legacy moved?
- [08:34]
B
Well, I think, I think Barry's way down. If I remember correctly, she's in more of the legacy. She was new media.
- [08:42]
A
Yes.
- [08:42]
B
But now she's in the legacy new media bucket.
- [08:45]
A
So, so you're. So if you're. What are these guys? Post legacy. So. So post legacy is where you were up until recently, like a part of.
- [08:55]
B
The legacy or traditional media.
- [08:57]
A
If you worked for the legacy media in the 2000s, like Alex Heath at Verge, which isn't exactly legacy media, but at Forbes, Legacy New Media and obviously Bloomberg with Ashley Vance, if you were working with the legacy media, you didn't jump ship in the 90s, you didn't jump ship in the 2000s, you didn't jump ship in the 2010s. You rode it all out the mid-2020s. And then the mid-2020s, you were like, okay, this legacy media thing isn't working for me anymore. I got to go out.
- [09:25]
B
That's where we got Ashlee Vance and Bloomberg here. So this kind of crossover.
- [09:29]
A
Yes, yes, yes. But the free press. The free press. Barry left the New York Times famously, you know, crazy dust up. And that was in the pre post legacy media era right before.
- [09:42]
C
Right before that started.
- [09:42]
A
Yeah, exactly.
- [09:43]
B
So she used to be in the new media bucket.
- [09:45]
A
Yes.
- [09:46]
B
But current now she sits in the legacy new media.
- [09:48]
A
And that's because there's a whole new wave of new media. And so Andreessen Horowitz is leading the charge here. They have a team that is just called new media, which of course asked the question what were they doing before?
- [10:00]
B
Which was their marketing team.
- [10:02]
A
Yes. So they were doing media. It's not like they just started doing media. No, they just started doing new media. Their old media is now known as legacy new media. And so legacy new media also includes the information buzzfeed, Wired Vice organizations that came out and said, we are going to be doing the new media thing. But they said that a decade ago. And so now it's a little bit legacy. Then you also have of course, alt media. Vice kind of bridges these two gaps. You have the trappings of a new media organization, but with a little sharper edge over here. You get into the darker media, dark media zone. You get the Sam Hyde, the Curtis Yarvins, the passage press. Of course, we still haven't figured out where Feed Me goes.
- [10:45]
B
We'll do that at the end.
- [10:46]
A
This at the end. But we can move over to neo alt meet.
- [10:51]
B
You have the Adam Fried lens.
- [10:52]
A
Yes. So you still have some of the edginess, some of the rebel spirit of your vice magazines, your red scares, but obviously newer products, newer media products launched more recently. Then you also have new media with the nu. That of course is media that covers new metal. So anyone who's. If you're writing a blog about Limp Bizkit or POD or Slipknot, that obviously goes in the. If you were to build kind of like the free press for corn fans, you would go in new media and you not new media.
- [11:26]
B
And of course we have to talk about the post corporate media category. Chapo Trap House, obviously fitting in there.
- [11:33]
A
It's a very funny choice to put that there. But obviously anti corporate, sometimes socialists, sometimes communists, sometimes left wing, but distinctly anti corporate. You're not going to run a lot of ads on a post corporate Media property.
- [11:50]
B
But you might curse some people, you might hire some people on.
- [11:55]
A
That's possible. Now we do have the post Neo media. So as you think, you think traditional media, legacy media, regular media, new media, Neo media is the new Neo tradition. Post Neo media, you can think of it as the most accelerated, the most aggressive, the final, the most cutting edge media properties that are out there. We have Martin Shkreli. He just turns on a live stream. It doesn't.
- [12:21]
B
He goes for hours and hours and hours and hours.
- [12:23]
A
It's not a media company, it's not a, it's not a TV show.
- [12:27]
B
And he's moving, he's moving markets, right?
- [12:29]
A
Exactly, exactly. Then you have. I show speed, Mr. Beast. Other examples of Neo meet post Neo media creators. You of course have had the expert media sphere.
- [12:38]
B
Andrew Huberman, Lex Friedman, AI expert, Joe Rogan, UFO expert, and Andrew Heubman of course.
- [12:44]
A
And then we have the Neo factual media. More of the analysts focused on the actual truth. One of the really the only truth seeking organizations on the map in general. Where are we? Let's move over. Zoom out to. So traditional media, we know traditional media. You're printing things out, you're making traditional content, you're instantiating your ideas in traditional ways. And the Neo trads are doing this in a new way.
- [13:12]
B
So this is new media that cosplays as traditional media.
- [13:18]
A
Exactly.
- [13:18]
B
So that's tvpn, right? It has some of the aesthetics of an espn.
- [13:22]
A
Yes.
- [13:22]
B
But it looks and feels.
- [13:23]
C
Good job, guys.
- [13:25]
B
Thank you, thank you to the production team. I'm glad you guys like the show. And so Colossus is another good example here. They're obviously Neo. You know, if you look on this map, what is Colossus? There was a debate last week, right? This is part of what started the debate. What is Colossus? Are they writing profiles? You know, is this journalism? And we just immediately thought this is obviously, you know, neo trad media. There's no other way to look at it. This is traditional print media reborn for the modern era.
- [13:54]
A
Yes. And specifically the. You wind up getting a lot of attention in neo trad media from the old media, from the mainstream media, from the legacy media, because you are wearing traditional media clothing. So arena magazine kind of sitting squarely.
- [14:12]
B
Between post trad and Neo trad.
- [14:15]
A
And so if you are the editor of Forbes or Fortune magazine, Fortune magazine has done a cover story on Josh Kushner. Colossus magazine just did a cover story on Josh Kushner and they printed it out. And it's a physical Magazine. And so it feels much more like a direct attack and a comparable product as opposed to. If Colossus was just a substack, the editor of Forbes, the editor of Fortune, they would be able to say, well, that's just a substack. What makes us special is we print ours out and mail it.
- [14:49]
B
Would you ever do a daily print newspaper?
- [14:52]
C
Daily print? I don't think so, but I would do weekly daily print. There's so many hands involved and like material getting it.
- [15:00]
B
You just geofence it to New York, Manhattan. You could have little people in feed me outfits.
- [15:05]
F
Yeah.
- [15:05]
C
Or I guess the other thing people do. I don't know if you guys remember this paper during COVID called the Drunken Canal, which was like a downtown gossip paper that they put in the news newspaper stand boxes around the city. But that was maybe four times a year or something. But that was really fun because you couldn't really find the content online. It created some tension around discovery.
- [15:25]
B
What are they going to publish?
- [15:26]
C
Yeah.
- [15:28]
A
So then we have Neocorporate media. These are like obviously from a corporation that does something that's not media. You have the AWS Developer Tools blog. Let's give it up for them. Some of the greatest content. You have the OpenAI podcast, Cheeky Pinterest, which sort of dresses up as Neo trad in many ways, but also is fully owned and operated by Stripe, a corporation. Then you have the Neoconglomerate media. This is whatever David Ellison is working on, we believe. David and Larry, I want you to take me through.
- [15:58]
B
We forgot we could have added UFC as part of this new Neo conglomerate.
- [16:02]
A
But then we also have the East Coast.
- [16:05]
B
Yeah, the east coast underground. You'll have to explain everything in this section.
- [16:09]
A
Yeah, we weren't familiar with this.
- [16:10]
B
What's going.
- [16:11]
C
You guys know New York Magazine? Come on.
- [16:13]
B
Well, yeah, from the. Andrew Huberman.
- [16:15]
A
Is that. Is that the New Yorker magazine?
- [16:17]
C
No.
- [16:18]
A
Is that the New York Times Magazine?
- [16:19]
C
No.
- [16:20]
A
Is that the New York New Yorker? It's New York Magazine.
- [16:23]
C
New Yorker. Part of Conde Nast.
- [16:25]
B
Okay, okay.
- [16:27]
C
New York Times Magazine is part of.
- [16:29]
A
The New York Times.
- [16:30]
C
Correct. And New York Magazine is something else.
- [16:32]
A
Who's it owned by?
- [16:33]
B
And it's. Is it supported by like the.
- [16:34]
C
It's under. It's under Vox.
- [16:37]
G
Well, Boston.
- [16:37]
A
We should nationalize it. Mamdani. Let's do it. Yeah.
- [16:40]
C
I mean, that's a big one. You guys should probably know about that one.
- [16:43]
A
I feel like I love New York Magazine because to come out of the magazine, the New York Times already has A magazine. And the New Yorker has a magazine. And you're just like, you know what? New York needed a magazine. That's something that we would say.
- [16:54]
C
It's the best one. That's the one where I worked.
- [16:55]
A
Yes, it does seem cool. And I feel like when you see a New Yorker, a New York mag profile, it is great.
- [17:01]
C
Oh my God, it's great.
- [17:03]
A
They have their own special flavor. That's not quite the New Yorker. It's not quite the New York Times Magazine.
- [17:08]
C
Right. It's the New York. It's New York Magazine.
- [17:09]
A
New York Mag, NY Mag. Okay, so why are. Why, why are these. So you guys have never heard.
- [17:15]
C
Okay, you've heard of Puck because you've been on Dylan's podcast, but Tyler, our.
- [17:20]
A
Intern has not heard of Puck and he's heard of all of these other.
- [17:23]
C
You haven't heard of Hellgate you wouldn't know about. That's more of like a New York left leaning.
- [17:28]
A
Tyler, do you know Semaphore?
- [17:30]
H
No.
- [17:30]
A
Okay. Do you know Status?
- [17:33]
H
No.
- [17:33]
C
Status was started by Oliver Darcy. So. And he came from cnn, I believe. So, like we should move him maybe over there. But something that's interesting that's happening is some people are logos and some people are faces.
- [17:44]
A
Yeah.
- [17:44]
C
Like you guys chose to put TVPN as your logo.
- [17:47]
B
Yeah.
- [17:48]
C
And I'm curious, I'm happy that Feed Me is not my face and it's my logo.
- [17:53]
A
You want the brand to.
- [17:54]
B
To grow, but Feed Me is starting to become more of a platform.
- [17:57]
A
Platform company.
- [17:58]
C
Bingo.
- [17:58]
A
Yeah, I would say.
- [18:02]
B
We left off Punchbowl. Punch bowl.
- [18:04]
C
Oh, Punchbowl.
- [18:04]
H
Yeah.
- [18:05]
C
I mean, they're ex politico.
- [18:07]
A
Right. Okay, so what I'm actually hearing is that east coast underground is really sort of a further post trad. Post legacy. Maybe they're post legacy.
- [18:18]
B
Yeah.
- [18:18]
C
I think a lot of them are post legacy. I think the difference between them and like an Alex is a lot of them have raised like tens of millions of dollars to get their things off the ground. But their newsletter.
- [18:28]
A
But. So they're thinking less independent.
- [18:30]
B
We needed a category for heavily VC backed. You really do.
- [18:34]
C
You really do. Because we're going to be watching how those play out in the next few years or months.
- [18:39]
A
Yes. Well, are there any. Are there any major, major corrections that you think we should make? Do you think we got any of this wildly wrong?
- [18:48]
C
No, I mean, it's really interesting that you guys. The East Coast Underground thing, because those are the publications that people are like wringing their hands. Is that like going crazy over In New York City.
- [19:02]
A
Sure.
- [19:02]
C
And maybe your audience is.
- [19:05]
A
I definitely think these folks need to be over here.
- [19:07]
C
Yeah, I think we can redistribute the underground.
- [19:10]
A
Yes. Because you can basically see it goes like.
- [19:13]
C
I mean, New York mag, there's like.
- [19:14]
A
Less tech, more tech here, and then politics is kind of running through this way. But then over here you have some tech, and then over here you have some tech.
- [19:24]
C
Yeah, those can be sort of redistributed.
- [19:26]
A
Next to each other status.
- [19:27]
C
I would call Post Legacy, but.
- [19:29]
A
Okay, but it's almost like there's some difference because these were started 5, 10 years ago. Correct. PUC is not this year. Puck didn't start this year.
- [19:40]
C
Puck started, I think, like three years ago.
- [19:42]
A
Oh, three years ago. Okay. That is pretty recent.
- [19:44]
C
And they've raised 20 million.
- [19:46]
A
It's not quite as recent.
- [19:47]
C
And they also just acquired airmail, which was green Carter, sort of.
- [19:50]
B
Can you hit this? Can you hit the gong for puck for raising 20 million?
- [19:59]
A
But.
- [20:00]
C
But the big story with Puck right now is that they just acquired airmail.
- [20:05]
D
Okay, what's air mail?
- [20:07]
A
Who's that?
- [20:07]
C
Graydon Carter, the great editor of Vanity Fair.
- [20:11]
B
Okay, there we go. So he's more than airmail in the Post Legacy media.
- [20:16]
C
Airmail is something else. I mean, it still is. It's still in existence, but it's sort of like, you go, it's very luxury. So it's like this is what happened in this resort in Sweden. This is like the crazy thousand word story about it. And every, you know, sort of 20 words, there's a big Hermes ad. So it's like. Yeah, but it's subscribers.
- [20:45]
B
Zach Meyer in the chat says, I think this needs to be three dimensional to better illustrate their.
- [20:50]
A
I agree, I agree. Okay, well, that concludes.
- [20:53]
B
Well, I'm glad we could clear up the media landscape because there's been so much debate lately, and sometimes you need to, you know, expand in order to really help.
- [21:03]
A
I mean, we were joking about this being completely schizo, but we got to have something. Yeah, we got to add barstool. Okay, let's make some notes. We'll do a V2 before we release the next iteration, but we'll get something there. Let's go through the mansion section. I want some advice on what is a buy, what is a sell. Before we do, let me tell you about restream.com, 1, livestream, 30/ destinations, multistream. Reach your audience wherever they are. Well, you said that you were interested to talk about Carl Icons nyc, Penthouse. Is that correct?
- [21:35]
C
Are you Familiar with Carl Icon pasta at the restaurant. I think it's like two blocks from his house. Yeah, there's this restaurant called, I think it's called Iltanello. It's an Italian restaurant and they have like a $20 Carl icon pasta. I looked this up recently.
- [21:50]
B
20.
- [21:51]
C
Yeah, because it's not exciting. It's like, it's like a very simple pasta. But he ordered it a lot. So I think one of the things if you buy there, you can sort of like always take friends to that place and sort of get into, well, you know who used to own my apartment?
- [22:07]
A
Who?
- [22:08]
C
Carl Icahn. You can do that every time.
- [22:10]
A
Oh, if you live there.
- [22:11]
C
Yeah, yeah. That's, you know, neighborhood. You can have maybe rename the pasta if you learn if you move there, you know.
- [22:19]
A
Icon has also entertained numerous corporate giants at the apartment, including computer titan Michael Dell, Apple CEO Tim Cook, Netflix co founder Reed Hastings. He would extend. He. It is. He was hanging out with Dell. It was pretty fun. It's $23 million property.
- [22:37]
B
How many square feet?
- [22:40]
C
People are liking the lifestyle news. You guys should talk about restaurants more.
- [22:43]
A
We should. The apartment has views around the city including Central Park. Dramatic double height foyer, has a circle staircase. 23. 250 units. I'm not sure how big this is. Oh, okay. It is 14,000 square feet and it's a combination. Why is that sad? That's huge for Manhattan. That's massive. 14,000 square feet is the combination of three apartments that were purchased over 40 years. And so he slowly built this like.
- [23:13]
C
Larger and larger 53rd.
- [23:15]
A
Does that make a difference? Well, is that material like do you know Manhattan that well to know?
- [23:21]
B
I would hope so. I would hope. You know it that well.
- [23:24]
C
I know it that well.
- [23:26]
A
I don't know the difference. So it is on. The apartment atop Museum Tower on Manhattan's West 53rd street spans the 51st and 52nd floors. It has 14,000 squares.
- [23:37]
C
That's a buy.
- [23:38]
A
You think that's a buy. 23. So Carl Icahn. Carl Icahn has said the billionaire is unsure why his beautiful apartment on West 53rd street hasn't found a buyer yet. You think you agree with that? Take. It doesn't make sense. Somebody needs to step up.
- [23:55]
C
Yeah, somebody will step up after this story.
- [23:59]
A
Okay, so you could get one 53rd Street, 14,000 square foot apartment or.
- [24:05]
C
I like that they say in the story.
- [24:06]
B
I mean, so here, here's the thing. So he had a Hindenburg research report on him, on, on his apartment, on his empire. And so I think. And Puck has an article here. I just searched Carl Icahn fall off. And Puck pulls up Carl Icahn on how to lose 20 billion. So I think people don't want to buy the apartment because they're worried about negative aura.
- [24:29]
C
But you call a priest. You call. You get some sage, you just sort.
- [24:32]
B
Of sage it up.
- [24:33]
A
Okay.
- [24:33]
C
That's every apartment in New York.
- [24:35]
A
So for 23 million, it's fixable. You can make it work. Okay. What about this Del Mar Beach House? For $50 million, this one was actually sold. I can hold it up here. A beachfront compound with a ping pong pavilion. Are you pro? Ping pong?
- [24:51]
H
Sure. Yeah.
- [24:52]
A
Rank these kind of fell off Rank these paddle sports. Ping pong. Padel. What's the other one called?
- [25:00]
C
I don't like sports that were invented in the world.
- [25:01]
A
Pickleball. And then tennis.
- [25:03]
C
Yeah, tennis is Strong.
- [25:04]
A
Rank those 4. Tennis is the top.
- [25:05]
C
Yeah.
- [25:05]
A
And then ping pong.
- [25:06]
C
Yeah.
- [25:07]
A
And then Padel and what's the other one? And then pickleball's at the bottom.
- [25:11]
G
Yes.
- [25:11]
A
Pickleball. It just has a terrible name. The word doesn't sound good. Oh, here's the picture of the house. So 50 million. It has 6,000 square feet, seven bedrooms. How would you rank this? I will keep. I will keep giving you extra information.
- [25:25]
B
Basically, Del Mar. Del Mar is in San Diego County. It is.
- [25:28]
C
That's really not my kind of place, but no. Yeah.
- [25:31]
A
What's kind of the Hamptons of Los Angeles?
- [25:33]
B
Kind of a brutal view.
- [25:34]
C
Don't say that.
- [25:35]
B
It's kind of a brutal.
- [25:36]
A
Look how brutal. Better. Hampton. Don't say that. Like, if someone was like, from first principles. Let's just make the Hamptons better, they would come up with Del Mar.
- [25:44]
B
But look how brutal the view is from the pool.
- [25:46]
C
I don't think this was the best photo they could have. Ran.
- [25:48]
A
You can see the beach. Yeah.
- [25:50]
B
Oh, yeah. I can see the beach behind the staircase. Like through this little sliver.
- [25:55]
A
Oh. So are you an infinity pool guy? You want the pool straight out to the ocean? Okay. Okay. Well, the seller is a limited liability company tied to Sandra Napsker, whose family has been major cattle ranchers over a century in Arizona, California and Oregon. Land, land.
- [26:14]
B
Family landmaxing land is the most underrated asset.
- [26:18]
A
Completed last year, the oceanfront compound first came on the market for 75 million, but the price has been reduced. It finally sold for 50 million. Listing agents said it's remarkable sale for San Diego coastal real estate agent. The Property includes a 4,500 square foot main house. I agree.
- [26:37]
B
Somebody would have to pay you quite a lot to live there.
- [26:40]
A
It is crazy to me.
- [26:41]
C
That's not my kind of beach, and that's not my kind of house.
- [26:43]
A
It's crazy to get. To get half the space. And you're in Del Mar, which is cool, but you're not in Manhattan. You would think that Manhattan would just be more expensive. And you're paying twice as much for half the space.
- [26:53]
C
Yeah.
- [26:53]
A
I say take Icahn's property. That makes no sense to me.
- [26:56]
D
What else?
- [26:57]
A
What about Hawaii? What's your take on Hawaii been once? It's the Hamptons of the Pacific. Do you agree?
- [27:04]
C
The only thing I'm curious about with Hawaii is I really want to visit a Zuck's ranch and meet those cows.
- [27:11]
A
Okay.
- [27:11]
B
Meet those cows.
- [27:12]
A
You want to meet those cows specifically?
- [27:14]
C
Yes.
- [27:15]
A
You don't just want to see how big the property is? The cow specifically.
- [27:17]
C
I'm curious about what he's sort of doing there with the. He's obsessed with, like, breeding the perfect cow.
- [27:22]
A
Yeah.
- [27:22]
B
We missed. They had Zuck burgers at the Meta Connect event that we were live from, and we missed. We just got really busy. Podcast.
- [27:31]
A
It's basically impossible to eat while you're streaming because you have to eat talk on the microphone.
- [27:36]
C
None of them could have put one aside.
- [27:38]
B
Did you guys have any?
- [27:38]
A
Yeah. No. No. No one had any.
- [27:40]
H
Wow.
- [27:41]
B
Missed opportunity.
- [27:42]
A
Missed opportunity. So Catherine Conrad spent a decade perfecting the gardens at her Honolulu home. Overlooking the ocean. The estate has granite pathways that wind around a Carrara marble fountain. I don't know what that is. Actually an antique pergola. Cost about $40,000. Her favorite view of the garden is at night when the cypress monkey pod and flowering Tacoma trees are illuminated.
- [28:08]
B
She's trying to pitch the monkey pod as a feature. Sounds like a bug.
- [28:12]
A
You have a monkey pod. I guess that's like. It's like a cat tree for a monkey.
- [28:16]
C
Visit you in Hawaii, too. Like, you have to.
- [28:19]
B
The time zone is so brutal.
- [28:21]
C
Yeah. Or you have neighbors you love, maybe. Or you meet new friends. But if you're buying in California or New York, someone's always going to be totally.
- [28:29]
B
I also feel like, as for the celebrities out there, it really puts a target on your back when. Anytime there's. Every time there's been, like, a natural disaster, it's like, oh, yeah, you know the rock because you took all our water. Yeah, well, that. But there was the tsunami warning last year, and they were getting really canceled for that and then canceled for the tsunami and then the fight. No, because apparently they had like probably private roads that every, you know, everyday citizens were backed up in traffic trying to escape the tsunami and they, they were not inclined to open their private roads.
- [28:59]
C
It's always going to be more controversial to buy in Hawaii.
- [29:01]
B
Yeah.
- [29:02]
A
Okay, so 0.8 of an acre it was. The first lot was bought for 7.55 million, then 4 million for the adjacent parcel. So you put two, two properties together, around 11 million in land value, then 12 million in renovation. They're asking 34. It's a four bedroom, 5,200 square foot house, ocean views. It also has a Pilates studio. Is that a feature for you?
- [29:31]
C
I like to leave my house for Pilates.
- [29:33]
A
You want to leave your house for pilates? So at 34 million, are you buying or are you selling? No, you're not buying, you're passing.
- [29:39]
C
Passing.
- [29:40]
A
What about this house in Puerto Rico? I haven't been to Puerto Rico in a very long time. Before we move on, we got to do another ad. Do you have the ads list? I don't have it today. Let me pull it up. Here we go. So privy wallet infrastructure for every bank. Privy helps John makes it easy to build on crypto rails, securely spin up white label wallet sign transactions and integrate on chain information infrastructure all through one simple API.
- [30:09]
B
So yes, Puerto Rico was very much a thing during 2020.
- [30:16]
A
That is crazy.
- [30:17]
C
Their invite, they're saying, come, hang out.
- [30:20]
A
I have a friend who. Yes, yes. They want people to come visit. I have a friend who would refer to that as a gulag because you get all your boys to come and visit and hang out and they all sleep in bunk beds, but you really could sleep with 12 people there or something. So this mansion has 17,000 square feet, eight bedrooms, a basket basketball court, a pool. It is for sale for $65 million.
- [30:44]
B
Is it in that zone? I think. I don't know if it's the Four Seasons or there's another hotel company that has effectively set up a mini city in Puerto Rico. You know about this?
- [30:55]
A
There is information about the Ritz Carlton in here.
- [30:58]
B
Oh, it's the Ritz.
- [30:59]
A
Let me read it. Simon said he and his wife had little intention to buy in Puerto Rico before stopping by the Ritz Carlton Sales office in 2015. They quickly became enamored with the idea of building a house there. Buying a one acre parcel of land on one of the community's golf courses. The parcel overlooks the ocean. It was completely impulsive. There were a few false starts. The phallics got the keys to the new home in 2017, just before Hurricane Maria hit, closing down the resort community and resulting in minor damage. Water damage to the home. Soon after the community recovered, Covid hit. In recent years, the phallics have spent long weekends and holidays at the property, as well as several weeks in the summer. They are selling because their grandchildren are older and don't spend it much as much time on the property. Puerto Rico's luxury housing market has boomed in recent years thanks to tax incentives and the rise of remote work. It's become very attractive.
- [31:53]
B
They undid a lot of those incentives or they. They. You. There's some other in. But now they're just sort of trapped and in the Caribbean.
- [32:01]
A
Yes. And so this is a lot. In the Dorado Beach Resort. Yeah, Dorado, which is a Ritz Carlton Reserve Resort. 65 million. Buy or pass?
- [32:12]
B
Pass.
- [32:12]
C
But maybe try to get. I don't know if those Disney posters are coming with the house, but I feel like the posters in the film room, you could flip those.
- [32:22]
B
Yeah. Split them off, you know, break it up. Break up big Puerto Rico. Sell it for parts. We got to talk about J.P. morgan's new headquarters. Michael Dell shared it. I have a post pulled up here.
- [32:34]
A
Oh, yes.
- [32:35]
B
This is peak performance and I wanted to get your reaction to this setup. Can we pull up this post in the timeline?
- [32:43]
A
It would be great to be able to pull up posts both in the timeline and there. That's the future.
- [32:47]
B
Well, that's what. Can we not pull it up there.
- [32:49]
A
While we're pulling that up? Let me tell you about cognition. They're the makers of Devon. Devin is the AI software engineer. Crush your backlog with your personal AI engineering team.
- [32:57]
B
All right. Pull up this post from Litquidity and.
- [33:00]
A
We'Ll put it on there. We don't need to put it on the. On the back screen. We can actually just pull it on the. Yep, yep. And while we're doing that, let me tell you about figma.com. think bigger, build faster. Figma helps design and development teams build great products together. You can get started for free. Did you see the. What else is going on here?
- [33:26]
C
Would you guys buy anywhere besides California?
- [33:29]
A
No.
- [33:29]
B
New York, eventually.
- [33:31]
A
California has so much. I feel like I just want to, like, grow the map of places I go that are driving distance.
- [33:38]
B
We're both. We're both excited about Santa Barbara. Yeah. The Montecito to La Marin, but even Big Bear.
- [33:46]
C
So you just want to buy houses.
- [33:48]
B
Close to each other. Houses that are less than an hour.
- [33:51]
A
Away from each Other in California, in LA specifically, you can get in a car and in two hours, be in a completely different biome. You can be in the desert, or you can be in the ocean, or you can be on an island.
- [34:02]
B
So here's. They got video footage of the JPM analysts going to their new workstation. This is a post from Liquidity, but pull up the post from Michael Dell that shows the actual setup here. And we need a rate. My rate. My office.
- [34:17]
C
Did you guys hear that employees have to pay for the gym?
- [34:20]
A
What?
- [34:20]
B
At jpm.
- [34:22]
A
Why?
- [34:23]
C
There was a story about it.
- [34:24]
A
I mean, I guess that's just.
- [34:25]
B
Does that inspire them to grind it?
- [34:27]
C
I mean, employees were not happy about it. They were like, why do we have to pay to go to this?
- [34:30]
A
I guess it's just such a big organization that if you don't put some sort of limit, people will, like, kind of all flood right on January 1st or something.
- [34:39]
C
And I'm sure that they have.
- [34:40]
A
It's just kind of like a full.
- [34:41]
C
I'm sure that they have, like a health account or something towards it.
- [34:46]
A
Yeah. So this is what peak performance looks like, for sure. This is beautiful.
- [34:52]
B
I actually. I saw this setup and I was. And I thought, I want to recreate this here.
- [34:57]
A
100%.
- [34:58]
B
I'm down to be in the trenches.
- [35:00]
A
Yes. This is what the trenches look like.
- [35:02]
C
I like that they have the thing that new hotels have now where you just. You have the outlets right there.
- [35:07]
A
Yeah.
- [35:07]
C
And the desks. That's nice. We're not.
- [35:10]
B
Imagine having ChatGPT add up numbers for you while sitting in that setup.
- [35:16]
A
So what I'm interested is, like, when was this taken? Because I thought they worked hard.
- [35:21]
C
I don't think that's a real photo. That doesn't really look like a real person.
- [35:24]
B
Michael Dell just posted this and said, congratulations, JP Morgan, on opening your new headquarters.
- [35:29]
A
He sold them all the monitors because the monitors are Dell monitors.
- [35:32]
C
I think this is a rendering.
- [35:33]
B
I don't know why. This is real.
- [35:35]
A
I think this is real.
- [35:36]
B
You don't think that jawline, that's. That's pure JPM power?
- [35:40]
C
I don't know. I don't know. I have to see it for myself. I emailed their team.
- [35:44]
B
Even Michael Dell replied to his own post and said, we are monitoring the situation.
- [35:48]
A
Yeah. You don't do that on some slop. I think they finished building out the office, Michael.
- [35:54]
C
Looks like slop.
- [35:57]
B
Look at those beautiful Dell monitors.
- [35:59]
A
I will fight you on this.
- [36:00]
B
I need this.
- [36:01]
A
This is.
- [36:01]
C
I want to go. Maybe that's where My shoot is.
- [36:03]
A
That is where you're shooting. This is the spot. We're hop. Skipping a jump from the people that will actually.
- [36:09]
C
I need to go. I need to see. I need to see it for myself.
- [36:13]
A
Yes, yes, yes.
- [36:14]
B
Okay. Is it like you seen it last thing we have to.
- [36:19]
A
No, it's beautiful. It's a whole new building and it has like crazy led at the top, right? It has a crazy light show. It's cool.
- [36:26]
C
There's like rock. There's like a big story about like the rocks that they brought into Dwayne the Rock Johnson.
- [36:32]
A
You should bring him in. Have him do it. I feel like when you, when you set up a new, a new fortress of finance, you should have like an inaugural, an inaugural lock in and you should bring in like a, like Warren Buffett should come and like lock in and type the first row in a spreadsheet and be like, yes, like, we've opened the office. Jamie Dimon himself has come and written the first cell of the spreadsheet and now the office is open for business. It's like smashing a bottle of wine across a boat. I like that. You know, you need to christen a new office.
- [37:08]
B
Let's pull up this post from none other than the New York Post, which I don't think they made it on the media landscape, but mobsters ran rigged poker games at Kylie Jenner's Luxe NYC pad while she still owned it. Apparently Travis Scott also was on the owned a piece of it too. What's going on here? You think this all ties back into the whole.
- [37:36]
C
Isn't this part of the story that came out yesterday about.
- [37:38]
A
I just got an email from you 19 minutes ago. You're sitting right here.
- [37:42]
C
It's crazy, right?
- [37:42]
A
It's crazy.
- [37:44]
C
Isn't this part of the story with the NBA gambling?
- [37:48]
A
I haven't been following this too.
- [37:49]
C
Yeah, the headline yesterday on the Times about the NBA coach gambling. There was in the headline it also mentioned mafia. So unless there's two mafia poker rings going on, which is totally possible in a city like New York, I think they're connected. And I wrote about it yesterday and one of my readers emailed me. I'm not going to say who it was, but he was like, if you want to talk about this, I was there. Or the game was held at gunpoint. What if you want to talk about it? And I said, sure. And be the neighborhood.
- [38:20]
A
What is game held at gunpoint? Like someone's forcing you to play poker.
- [38:23]
C
Well, you guys don't play poker.
- [38:25]
A
Sometimes that happens what do you mean that happens?
- [38:28]
C
Guns, rugs.
- [38:29]
A
But in what scenario?
- [38:30]
C
Mafia.
- [38:32]
A
But under what conditions am I like, I'm really enjoying playing poker with you. So you will die if you.
- [38:38]
C
I think it's more like they're just.
- [38:39]
A
Having so much fun.
- [38:40]
C
Yeah, you do a little. You have a little too much fun and somebody like pulls something out they're not supposed to pull out. And yeah, I've never had a gun at a poker game, but I've heard.
- [38:50]
A
This is why I'm not leaving California. This, this New York seems crazy to me.
- [38:54]
C
There's none of this here.
- [38:56]
A
This is, this is not.
- [38:57]
B
There's a post here on the. On the sports gambling arrest Mike over on X with 2 million views on this post. The NBA is very concerned about the multiple arrests made as a result of the sport FBI sports gambling raid. Use code code multiple arrests for a 30% profit boost on a three leg parlay with DraftKings, the official gambling partner of the NBA.
- [39:17]
A
I love it.
- [39:18]
C
I am glad that people are still playing poker in person as opposed to just only playing online. That's good.
- [39:23]
A
That's kind of a white pill for sure. Touching grass.
- [39:27]
C
Yeah. I used to have a poker podcast.
- [39:30]
A
Really?
- [39:31]
C
Yeah, it's called Chips and Dip and it was about poker and chips. Like potato chips.
- [39:34]
A
Potato chips, yes. Were you, were you like a good poker player?
- [39:37]
C
I am, I was. My ex was like a good poker player. And then our friend the third guy is also very good and he's borderline addict.
- [39:47]
B
He's gambling addict. Cool.
- [39:49]
C
He's doing okay. He might start writing about poker.
- [39:52]
A
He should go work at a hedge fund or something. Yeah, put it to work.
- [39:55]
C
For sure.
- [39:56]
A
He should go lock in. He should be. He should do the inaugural gamble. The first gamble at the new. The new Morgan Stanley hq. It's Morgan Stanley, right?
- [40:05]
B
No, it's jpm.
- [40:06]
A
Oh, it's jpm. Okay. Jpm. Yes, yes, yes. The first fat fingered trade. This is important. Anyway, thank you so much for stopping for having me. Guys, we know we have to get you. You have to get out of here.
- [40:19]
C
Really fun. Gotta go to the Hammer Museum.
- [40:22]
A
Breaking.
- [40:23]
B
She will be the Hammer Museum. I didn't know this is Lodge.
- [40:25]
A
You could go by and say I.
- [40:27]
C
Play it when you don't say that.
- [40:28]
A
Be careful.
- [40:29]
C
Okay.
- [40:29]
D
Bye guys.
- [40:30]
A
Bye.
- [40:31]
B
Great hanging. Cheers.
- [40:35]
A
While we rearrange the seats, let me tell you about Vanta.com Automate compliance manage risk, prove trust continuously. Vanta's trust management platform takes the security and compliance process. The manual work out of your security compliance process and replaces it with continuous automation.
- [40:52]
B
Alex Heath says Snap is in talks to raise at least 1 billion for AR glasses. Multiple investors involved, including Saudi Arabia's sovereign wealth fund plan is to move specs into a new independent subsidiary. Subsidiary. Spinning it out and structure similarly to Waymo and Alphabet.
- [41:09]
A
I want to get Evan on the show so badly. I think it would be a very. Working on conversation on it.
- [41:15]
B
It will happen soon.
- [41:16]
A
Who? Samsung just launched a Vision Pro competitor, right? Samsung Galaxy XR, the headset. And Google's announcing it today. This is 16 minutes ago. The Samsung Galaxy XR is powered by Android xr and it's supposed to be basically at the Apple Vision Pro level in terms of fidelity, I think. But they obviously have a YouTube app because it's a Google partnership. And so on day one, they have a lot more partners because Android's been kind of a better steward to the developer community than Apple. And I think a lot of the obvious developer partners that would go and build apps for the Apple Vision Pro are taking a wait and see approach because they don't want to. They want to have more leverage over Apple. Instead of just showing up and being like, okay, I'm going to be paying the Apple tax on, on day one, they're saying, hey, like, why don't you go get the audience first, Apple, make sure that there's actually 100 million DAUs of this thing like the iPhone before I build an app for this.
- [42:36]
B
Yeah. Speaking of smart devices, Kohler, the toilet manufacturer, is making a new toilet camera that provides health insights based on your bathroom breaks, which is insane because Throne, you know, the startup Throne, is doing this. I really thought nobody would touch it. No, nobody would go. And I thought it was like an insane idea. Yeah, but at least no one else would go and compete. And then Kohler, you know, obviously major manufacturer decided, I'm sure they were somewhat inspired by Throne to go get into the game of toilet cameras health tracking. So we'll see how this plays out.
- [43:18]
A
It's very funny. Let me tell you about graphite.dev code review for the age of AI. Graphite helps teams on GitHub ship higher quality software faster. You can get started for free.
- [43:29]
B
Gabriel is posting. There was an article from the Times that said female spies are waging sex warfare to steal Silicon Valley secrets.
- [43:40]
A
This is a funny story from the Times. Is this the Times of India? I think it's the Times of India, but this has been the meme of like the CCP spy girlfriend. The Times of London. Okay, so we Had a whole, there are a lot of different times is. There are a whole bunch of different like iterations of that, of that viral concept of like is, is the girlfriend a spy? And then the backlash to like, well you shouldn't judge the girlfriend because she might be a spy. And going back and forth, back and forth on that. So it's interesting that the Times of London. Is this just something where Europe takes two months to get to the current thing on X? Because this, this is a, this is an article that literally just says what X was talking about four months ago. Right?
- [44:30]
B
Yeah, I think there's, I mean this or is there new information, an ongoing thing? I think maybe what they've discovered and you know we're not factual media so we don't need to get into kind of the truth here. But they're saying some of these spies are even marrying and having children with their targets, which is really, you know, what a good.
- [44:51]
A
Have you seen the Americans? That's a great, great TV show. The Americans. Great.
- [44:56]
B
It's Atlas says, geez, I would sure hate for a beautiful Chinese spy to wage sex warfare against me for my production secrets. My production secrets are super important. I would hate for this to happen.
- [45:07]
A
Did he go mega viral for this?
- [45:09]
B
Yeah, 15,000 likes. He got Elon response.
- [45:12]
A
That's great.
- [45:14]
B
Near has been going on its hair highlighting Andreessen backed companies and there's definitely some that are specifically SpeedRun. These are a 16Z speedrun company which is under Andrew Chen and it's more.
- [45:29]
A
Of an incubator or accelerator.
- [45:30]
B
It's an accelerator.
- [45:31]
A
It doesn't have the same weight as like David George came in with the growth team and like actually like did some analysis.
- [45:36]
B
It's like they're ripping checks but it's out, you know it's out of their Speedrun run by Andrew Chen.
- [45:42]
A
Sure.
- [45:43]
B
And yeah, so this one is Cheddar. It's the, the TikTok of sports wagering. Unlike other platforms, they allow players under 21 and in 46 states.
- [45:54]
A
Is this from a deck or from their website or something?
- [45:56]
B
I think this is from the Speedrun.
- [45:58]
A
How did Nier find this?
- [46:00]
B
I think this is from the Speedrun website.
- [46:02]
A
This is ridiculous. Okay, wait, there's another one. I need to talk about doublespeed. I need you to explain this to me. The founders in the chat going back and forth in the replies. So Nir says a 16Z back double speed lets you control thousands of social media accounts with AI ensuring they look as human as possible. Quote, never pay a human again. Control is all you need. I can imagine some. I was about to steal this, but it's really hard. I can imagine that puppeteering a thousand devices might be useful for something or other in like a data generation.
- [46:43]
B
So Double Speed, I think they're trying to position it as this is an alternative to influencer marketing, but it does look and feel more like an influence operation. So I would be this just. This is the kind of thing sloppy.
- [46:57]
A
And it feels very.
- [46:58]
B
Nation states typically take this kind of thing on where they build, you know, a phone farm.
- [47:03]
A
This feels like bot farming. It's the other one very much against the. The terms and conditions.
- [47:08]
B
The other one that was.
- [47:09]
A
Hold on. But like near went through this whole flow where near is just quoting them, which is interesting because near is not really dunking here. Near is just saying a 16Z back. Double speed lets you control thousands of social media accounts with AI. It's basically just the same as what's on their website. Copy. One of the founders or the employees automating attention at Double speed. Zuhair Lakhani chimes in and says, let me know when you sign up. We can help you grow your company. Orin and Nir says, it wants to charge me 5k before I try it, it seems. Zuhair says, sorry, we don't do free trials. Let me know if you have budget to allocate and we'll get you onboarded. Day says you can do 2,000. Zuhair says, yes, join our discord. But the founder. Wait, is this. This is a founder of Speedrun Cohort Foundation 5. I don't know if it's the founder. Bessar Kopa says, people are calling this, like morally lame and impoverished and Bessert Coppa says, how so? There are already a million bot farms around the world. They wreaked havoc on our elections not so long ago. Shouldn't the best ones be built in America and pioneer the latest tech and rules to govern it?
- [48:26]
B
Wait, so he's saying they have wreaked havoc. Yes. On the world, but so we should build it in America?
- [48:31]
A
Yes. Yes. It's basically like, oh, yeah, you know.
- [48:34]
B
How I would like this? I think Double Speeds should pivot to being a defense tech company and just work with our government to run foreign operations.
- [48:43]
A
Yes. This is good.
- [48:44]
B
America has such a long history of running, you know, various influence operations in far off regions of the world that.
- [48:52]
A
Spam from first principles.
- [48:53]
B
Here's. Here's another one. Edgar is bringing real money engagement into Free to Play Sweepstakes. America's next number one social Casino for slot lovers.
- [49:03]
A
These are insane.
- [49:05]
B
I can't believe this is real. I had to go check the speed.
- [49:08]
A
It is wild comparing these to the Nat Friedman cohort that he did, which Julius was a part of. Julius AI. What analysis do you want to run? Chat with your data and get expert level insights. What was that?
- [49:23]
B
So the last one we have to cover is covered, which is bet on your bills onlyfans. This is so dark. I can't even bet on your bills with covered OnlyFans, child support and last night's Uber. Wipe them from your credit card by playing your favorite casino games. All from the comfort of our app. No vpn, no crypto deposits. And so my only my reaction here, I mean, John, you're gonna be impressed by this. A 6 million views on this post and only 292 likes.
- [49:59]
A
So bad. It means that someone like nuclear ratio.
- [50:06]
B
But I mean, such an insane missed opportunity to say bet on your debt. Right. Like they say, bet on your bills. Yeah, like bet on your own debt.
- [50:15]
A
Well, it's not debt.
- [50:16]
B
This is potentially the most.
- [50:18]
A
Because you've already paid it. Or maybe you're in debt.
- [50:22]
B
No, but you can wipe them. If you bet right, you get them wiped from your statement.
- [50:28]
A
So bad. It's so funny that just a quote tweet from Hero Thousand faces that just says societal breakdown gets more likes with 30,000 views. The like ratio is insane on this. Wow.
- [50:43]
B
Trey says this is incredible. Double your bills or nothing.
- [50:47]
A
What's funny is there was a guy who was a really cool kind of funny designer who was designing stuff like this, little modals. And one of them was Apple pay with a double or nothing button. And people were like, oh, that's really funny. But to see it actually productized is like, oh, we were joking about that.
- [51:05]
E
Yeah.
- [51:06]
B
The interesting thing is a lot of these kinds of ideas, kinds of apps, end up being massive businesses. But you don't typically see, you know, venture funds backing them. They typically boot crap. But this is one of the, this is one of the challenges of scaling a venture platform, right? It's like, you know, there's tons of partners at Andreessen that probably would have passed on these deals. But, you know, individual partners have autonomy and they can do what they want.
- [51:36]
A
Well, stick to the infrastructure, baby. Go to Turbo Puffer search. Every byte serverless vector and full text search. Built from first principles on object storage. Fast 10x cheaper, extremely scalable.
- [51:49]
B
Did you see this? A gummy bear. Basil says a gummy bear company casually dropping revolutionary battery technology for $25 on Amazon. The lightest ever of this capacity. And pretending nothing happened is one of the funniest things of all time. This is Haribo. Yeah, I actually think this is Haribo. I think this is a someone stealing the Haribo brand. Well hopefully it's the Gummy Bear company.
- [52:17]
A
So yeah, people are immediately wondering is this real? Is this actually the best power storage per dollar possible or per weight? And so Michael went ahead and bought one and here's his findings. He says Its weight is 286 grams. Just a little bit over what they claim the capacity is 83 watt hours. The printed, printed on the battery it says capacity 20,000 milliamp hours or 77 watt hours. So I'm measuring in excess of what is claimed. I think they're stating the voltage piece is what makes the battery capacity measurements nebulous. But watt hours don't have that issue. And I'm getting better than claims. Minimum charging current of 0.5 of an amp less than this battery pack turns itself off. Not able to recharge the battery via the built in cable. That's weird. So overall my conclusion is this battery is the real deal. It meets the claim capacity and pretty much hits the weight spec. Wow. So maybe you got to go pick one of these up. Very, very funny post.
- [53:19]
B
Oracle just secured a $38 billion debt deal to finance two new data centers. They said they couldn't let that anymore. They shouldn't, but they are not listening. 23 billion will go go toward a Texas site and nearly 15 billion towards one in Wisconsin. Ted Jang says here we go. The debt finance deals are coming.
- [53:39]
A
We got Chase Lockemiller from Crusoe coming on the show in about an hour.
- [53:43]
B
To break down his series E. The godfather of Stargate.
- [53:47]
A
Yes, Star Lord.
- [53:49]
B
The Star Lord himself. Citrini put this out a bit ago. According to Ted, something that has been a relative comfort in being able to stay long despite calls for an AI bubble so far has been the fact that most of the capex has been come from cash cash flow. You don't have a bubble really until the majority of the expansion is done on debt. We have yet to see significant financing, but that absolutely should be something to watch. On one hand that leaves another leg of the trade. Leveraging up still results in more money being spent. On the other, it sets the stage for consequences of not realizing ROI that we haven't really had to contend with in the existing paradigm. With Microsoft 10 year bonds trading at at a G spread of 3 basis points, there's still a lot of room for the hyperscalers to issue debt or tap private credit to fund continued spending on AI. However, this also induces a lot more fragility going forward. Monitoring the share of AI capex that is financed rather than funded from cash flow is probably the most important aspect of determining exactly where we are in the cycle. Again, there's there's been a number of players using debt. X Xai has been using debt Meta's already tapped the debt markets. I feel like Satya just generally is a lot less inclined to lean in and lever up too much. But there's a post here from Junk Bond Investor. It says just offloaded some data center construction debt to his school teachers 401k. My father is a retired school teacher, so hopefully he didn't get burned here. I have to be he's going to be a little frustrated with me if that ends up being the case. But I'm sure that's happening with Naxperia. There's been a little chaos on that front. Nextperia, the semiconductor company out of the Netherlands, they notified its Japanese automotive customers that it may no longer be able to guarantee supply, triggering renewed concern about upheaval at the chip maker. So some chaos over on the European front.
- [55:55]
A
Before we do the next post, let me tell you about Google AI Studio. The fastest way from prompt production with Gemini Chat with models, Vodka monitor usage. Do you think they everyone's talking about the AI bubble. They never talk about the podcasting bubble. Maybe we should be getting some more attention. Everyone's saying like wow, TVPN is feed me these businesses. If they don't work out, if they don't deliver, the entire global economy will collapse.
- [56:19]
B
Well, no, on a more serious note, we did hear of a podcast raising at a half a billion dollar valuation that is burning money. So that could be a sign.
- [56:32]
A
There might be.
- [56:35]
B
Packy has a post here. Never change Masa. This guy rocks. And it's a post that says SoftBank hunts for humanoid robot startups. SoftBank's Masayoshi Son has long predicted a future where robots would outnumber people and take over many jobs from humans. But some of SoftBank's bets and robotics over the past 10 years, such as robot company Pepper, have not panned out. Now the company is renewing its robot ambitions, particularly in robots that resemble humans.
- [57:02]
A
Can you imagine Brad Adcock reading this? He's got to be just like, oh, let's go like I want to be the one.
- [57:12]
B
SoftBank up 172% year to date does.
- [57:15]
A
It feel like actually what's crazy to invest in humanoid robots? I don't think it feels like a bad time at all.
- [57:20]
B
I think it's probably true that Masa and Softbank were too early in some of their robotics bets. Right. They bet really heavily a decade ago.
- [57:30]
A
This feels like a great time to be building in humanoid robots or robots generally. There's a ton of. There's a ton of interesting startups. 1x and there's smaller companies and there's supply chain stuff and it's like, it's one of those things where it's like, are you more worried about being too early or being too late? Probably too early on the. On the humanoid robot side because they're not quite, you know, getting deployed profitably. They're very much like research tools. They get bought by researchers. One off stunts here and there. There isn't really. There's no customer that's just like, oh yeah, they're buying these like every month. They're buying more and more and more because, like they're actually seeing increasing returns to scale.
- [58:19]
B
There was that story last week about Tesla buying actuators from a Chinese firm, right?
- [58:27]
A
Yeah.
- [58:27]
B
It was like a $675 million order, which to me signaled. That's the kind of move you make if you're actually trying to get into production, like in real time.
- [58:36]
A
Yeah.
- [58:37]
B
Or at least have product that people can order. My theory was that if they felt like they had more time, they might have done them themselves. Right.
- [58:46]
A
Well, our first guest of the show is in the Restream waiting room. Let me tell you about Profound. Get your brand mentioned in ChatGPT. Reach millions of people, consumers who are using AI to discover new products and brands.
- [58:56]
B
Be like MongoDB. Indeed. DocuSign ramp.
- [59:00]
A
Yeah. We have Sagar and Jetty from Breaking Points. Sagar, how are you?
- [59:05]
D
Oh, it's great to see you guys. Thank you for having me.
- [59:07]
B
Great to see you finally.
- [59:08]
A
Only after a million text messages every morning back and forth on the AI apocalypse and political crises, have we actually just put it together that we should just hang out on the stream. I really appreciate you taking the time.
- [59:21]
D
Oh, thank you, thank you.
- [59:23]
A
Give everyone like a quick shape of the actual flow of the content, like.
- [59:29]
B
The proper form of your business credit. He kind of created Neo Trad media, which is a new. Which is a new framework we're working on, which is new media that sort of embodies or kind of looks and feels like.
- [59:44]
A
So your show looks like a proper TV show. You wear a suit and so we call that neo traditionalist media, neo trad media. And we're firmly trying to do that.
- [59:52]
D
I love it. And I will say one of the only reasons I am here is because both of you are bringing suits back to the bones of Silicon Valley. And I know many of you are watching right now and you're Lululemon, your Vuori, and you don't look cool, you don't look better than John or Jordi or myself. You all are billionaires. Buy a suit. Stop. Just stop it.
- [60:13]
A
Stop it. That's great. I couldn't agree more. I think we're starting to have a small, small fact, but it's taking hold.
- [60:20]
B
A lot of people that visit the ultradome wear suits.
- [60:23]
A
Well, I mean, pretty much. Like there are going to be tons of AI folks, tons of tech people who are going to be dragged in front of Congress over this data center build out. I think you've clocked this correctly, but walk me through, is this a thesis that came from, you see this coming or you're seeing smoke signals already? Walk me through your thesis of like, how tech and the AI build out runs on a collision course with the average American.
- [60:52]
D
Yeah, the thing is, guys, and I really want to hammer this home for people who are watching because I, you know, shocker, I think many people in Silicon Valley do live in a bubble. One of the best parts of my job is I do genuinely is I interact. And we'll return to bubble conversation, by the way, but what I get to do is not only interact with, with a lot of people who are normal in politics, and I mean that at the very base level, not people who live in Northern Virginia. The vast majority of my audience don't Live in Washington D.C. many are Uber drivers, hotel clerks, et cetera. And I start to see bubbling stuff over the last five years over cost of living. At the same time, I'm tracking the electricity story and it doesn't take a genius to pair that with data center build out. So one of the things that I really want to flag, I think here on the show is that the coming data center collision at the heart of that is about electricity. It doesn't have to be this way. One of my favorite notes is there's an analyst I follow who was in China. He had a very different view of the, of this in, in China, even though of course, you know, they don't have democratic politics, but of course they do have some sort of democratic input. Is they said nobody's concerned here because electricity is, is considered a solved problem. We don't have a solved problem of electricity here in the United States. In fact, worse. You know, our grid is aging and actually things are getting worse. So with all of this capital expenditure in data centers, you are seeing a literal finite resource competition, especially in rural areas. So Virginia, the state where I live, 40% of all of our power is consumed by data centers. Oregon, for example, is some 33%. To be fair to many of the AI people who are watching, it's not all AI. We are the hotbed, right of. Of servers and Amazon and much of the federal government. But the point is, is that this is bubbling up at a very organic township level. So from Arizona to Virginia, data centers actually had a mention in our more recent gubernatorial debate. Both the Republican and the Democratic candidate said that they wanted to do something about it. There's Prince William county here in Virginia, where this is an active live issue and the public is overwhelmingly against it. The city of Tucson is moving. And so if we see this continued capital expenditure go up, and especially in more rural areas, we see headlines like 267% power increases for power bills in rural areas which are built near data centers, not to mention environmental concerns, etc. Then when you're going to see people start to get dragged before Congress because. Yeah, go ahead.
- [63:20]
B
Yeah. I'm just curious what. Obviously every area, you know, people are not going to just watch a data center get built, you know, in their backyard and not, you know, talk to their mayor or talk to the governor and actually start a conversation and a debate and ultimately create policy around this. Is there not some type of solution that would allow, effectively allow data centers to pay a higher rate than everyday citizens and kind of make up the. The difference?
- [63:50]
D
I think it's a great idea. Unfortunately, it's not how our power system works. The foundation of the United States is basically built on the fact that we're going to build this grid, this infrastructure, and we can all kind of consume the social benefits of it. There are pieces of legislation, Jordy, of what you're talking about on the table. I think in Oregon and a few other states, but these are few and far between. I also think we got to take everybody through the story of the last 25 years. Again, with great respect to the audience I'm speaking to, you can't be Amazon and be the second largest employer in the United States and then in your shareholder meeting come out and say that we want to fire 600,000 people or at the very least, not hire 600,000 fewer people. These are people who watch their towns get decimated. Then they watched Amazon. I mean, Nomadland, it's an Oscar winning film about this, you know, this, this.
- [64:39]
B
Yeah, I think, I think what's, what's, what's challenging, what's, what's so challenging about this issue is that you have these massive infrastructure projects with massive spend and almost zero job creation.
- [64:52]
D
Exactly, exactly.
- [64:53]
B
Like the plumbers and the electricians are going to get paid. And that's great. But we've even heard stories of the plumbers and electricians being flown around on private jets because their skill set is so and so that's not going to get people. But yes, if you're living in a little town and you're watching, you know, the mega, mega project gets spun up and you're getting not only, you know, no real benefit, you know, maybe, maybe you get more slop in your trough every day, but it's.
- [65:24]
A
Can you maybe take me?
- [65:26]
D
That's the point. Yeah, go ahead.
- [65:27]
A
Can you maybe take me a click forward into. Let's assume the trend lines kind of continue. Electricity prices go up. There's more, it's more more clearly attributable to data centers, et cetera, et cetera. Like, how do the different political coalitions reframe themselves around this issue? Like, I imagine that there's a right wing version of tech backlash that looks one way. There's the abundance lib direction. There's going to be the actual socialist left and maybe they just say like no AI at all. So how do you see the different coalitions, like grappling with what the, the v2 of the messaging might be here?
- [66:09]
D
There's going to be a horseshoe element here. I think the socialist left and much of the right is going to actually come in the same place of enough. And so a lot of the socialist left is actually Bernie Sanders. I don't know if you guys saw this right before I came on the show just called for a breakup of open AI. So this is huge news. He's the first US politician actually to do so.
- [66:26]
B
Now what would they, what is there to break up? They have chat GPT and then they have a bunch of other things that are subscale.
- [66:33]
D
How about the non profit profit arm or whatever the hell? You guys can explain that.
- [66:36]
A
Yeah, yeah. I think Sam's like, I would love to break it up. I'm actually trying to break it up. I would love to just have a.
- [66:41]
B
C corp again, break me up.
- [66:43]
A
Bernie, is this some sort of like 4D chess? And Sam's gotten to Bernie and told Bernie, hey, yeah, we, we need to break it up. That's really popular messaging. And he's like, yeah, you know, Bernie's.
- [66:54]
D
Probably, you know, he's picking from an old toy school books too much.
- [66:58]
A
He's basically just saying, let's pressure them. Let's make sure we have more regulation.
- [67:02]
D
Right, exactly. More regulation on the right. I mean, much of this is going to be about distrust of Sam and many of the tech guys themselves. I mean, when you have these people outwardly bragging about mass job loss, mass societal creation, there's a lot of us who are like, yeah, do I want this guy who's introducing AI porn to be the most powerful person in the world?
- [67:24]
B
I mean, the challenge is that the job loss, like vision is so good for fundraising. Because if people are like, maybe I don't have a job in 10 years, it's like, well, I want my money going into these companies, so at least I have a chance at escaping the permanent underclass. Question for you. Social media went through, you know, and continues to go. It's like a perpetual backlash at this point. People say they hate it, they don't like it, they say it's bad. Many of those people still use it. How do you think this, this AI development is different than that in that a generation will just log off and. Or do you think it's. It'll be the same type of dynamic where people say they hate it, they're, they're frustrated with the potential, potential impacts of it, but they still line up.
- [68:20]
D
That's a very student observation. And I think it's going to be somewhere in between. And so actually, Jordy, this kind of gets to the point of what I reacted to. You put out a tweet about AGI and pornography, and what I reacted to that is I was like, yeah, AGI is not coming, guys. What's coming is porn. And what that means is the Internet is just coming again. And so, you know, I downloaded the ChatGPT browser. It's fine, you know, it's fine. It's a browser. And so it's a browser that's going to serve me. I mean, literally, the ad read before I came in was about putting ads into chat GPT. Wow, what a business model. Somebody should have invented it already. They did. It's called AdWords in 2003. So are we just competing for the exact same type of unit, economics of screen time, of use, of pornography, of the very same things we, which as you said, Jordy, are making us miserable? Because that doesn't seem very cool to me. And it also doesn't necessarily justify, let's say 80% gain of the entire S&P 500 over the last 2025. This is the stuff actually that I text John about literally every day when I'm like, a bubble is coming. A bubble is coming. A bubble is coming.
- [69:27]
A
I think the steel man of the medium case. I don't know how to put it, but it's like if I told you, it's just like, well, yes, like AI is not AGI. You're not going to lose your job over it. It's not going to kill anyone. But it will help you do research better. It's better autocomplete. It's this nice tool that will help clean things up better. Spell check, better grammar check, you know, better odds. And if you're like, you're going to be a little bit more knowledgeable, a little bit more performant, a little bit more efficient and it's going to cost. We're going to need to, to build some data centers, but we're not going to build, you know, so many data centers that we all lose our light and heating and ac. I think that's, that's a reasonable like abundance.
- [70:10]
B
We're going to end up.
- [70:11]
A
That's where we should.
- [70:11]
B
The moment we are right now is that when somebody, you know, sees an AI video of, you know, the, the one that went super viral a while back was like people on the roof of an apartment building and it's a pool and the glass breaks and people flow off. It's like that video and videos like it are not worth for the average American an extra $100 a month, you know, an extra thousand dollars a year added to their electrical bill. Right. And so you're exactly right.
- [70:36]
D
And that's the political dynamic. And that's why John, I go, hey, all of that sounds great. Yeah, it's a little worrying to put my entire retirement portfolio based upon total, you know, a very different message. And when that's the reality of where we're going to end up a little bit better. Microsoft Teams is just really not the anthropic vision which is being sold to me. Okay. About, I mean, I talked to a guy who works at Accenture. I said, how do you use AI? He's like, yeah, it's great. It does auto meeting summaries for me. And I was like, wow, listen, definitely great. I mean, it means you don't have to hire some entry level guy. Although maybe that's not great. And what really ended up happening is that you end up using it for these extraordinarily mundane tasks. I, I have no issue with that as long as the attendant social fallout, government policy, power bills and all of that is tied up into it. Not to mention, I really can't belie the social cost because what you said, Jordy, is, is a central thing. There is a mass elite revolution right now against social media. And between. I think Jonathan Haidt might be one of the most public, like important public intellectuals in modern and like in a long time leading what I think is going to be at least a top middle discussion with all of these different school districts, states and others which are banning phones. The recognition is that they are controlling us and we need to retake control of our life. AI is kind of supercharging that message. And if you talk to the vastly rich, which you do, and I've spoken to a few of them, how many of them really let unlimited screen time for their phones in the way that, you know, the income of, let's say, people making less than 100,000 do for their children? It's just completely different. You know, they're very aware of that dynamic and that is permeating social culture at a way that I don't think people in tech are quite realizing.
- [72:24]
A
Do you think we. It always takes a generation of guinea pigs to understand where technology winds up. Because I grew up with the panic around video games and the idea that if you played Counter Strike competitively at a very high level and became one of the greatest Counter strike players of 2003, you were definitely gonna be crazy. Well, I did it. I was great at Counter Strike. I wasn't maybe one of the best ever, but I was pretty good. And I don't have any violent impulses whatsoever. I was fine. I was not. One shot by 360, no scoping people. I became a productive member of society. And I feel like we came up with the ESRB E, you know, R, like teen rated games, M rated games. And we kind of figured out like where the line is. Parents adapted, kids adapted, and we got through that social media. It feels like we're towards the tail end of getting through it. Where most parents know, hey, the young kids should not be on Instagram like constantly or TikTok constantly. And with AI, we're just starting to understand, but it feels like we'll kind of wind up in the same place, which is like.
- [73:31]
D
You might be right. But let me challenge you a little bit on the video game front, because I think that video John you also know my crusade against sports gambling. So Robin Hood, if you're listening, please don't allow sports betting. But one of the important things around gambling. You also know my crusader against marijuana is. Here's the issue with video games, with all of these vices and things that you're talking about.
- [73:53]
A
About.
- [73:53]
D
It's fine sometimes, but there is an attendant part of society which actually does lose it when immersed in this culture. So, you know, I've pointed. Unfortunately, part of my job, I cover a lot of mass shootings. We had two or actually three very recently. We had Minnesota, we had the. Also the Charlie Kirk Killer, and then we also had the ice shooter. All three were deep in these weird Discord gamers. The Ice Guys spent 20 hours a day gaming.
- [74:19]
A
But Discord is a social media thing.
- [74:20]
B
Discord, the platform.
- [74:22]
A
I don't think it is like a video game.
- [74:23]
D
The Discord platform.
- [74:24]
B
Yeah, yeah.
- [74:25]
D
But you guys can look at his Steam data he was logging. I don't exactly remember the number off the top of my head, but an absurd number of hours playing video games. This is somebody who's 31 year old, male.
- [74:34]
A
This is kind of a contrarian take. I feel like most people close the book on the video games as, oh.
- [74:38]
H
I'm back, your brother.
- [74:39]
D
I never left. Okay. Actually, no, I never left.
- [74:42]
A
I love it.
- [74:42]
D
And that's part of, that's part of what I think is important to say is that people say it works. I just dispute that. I mean, we have millions of people now who are addicted to gambling, who are addicted to marijuana, who have gone psychotic, people who are wasting their lives in front of these screens and it's like, did it work out? I mean, yeah, it worked out for the S&P 500. Did it work out for the United States of America? I would say no. I mean, if you put all those things together.
- [75:08]
B
Yeah, it was interesting. Ryan Cohen from Gamestop came on the show on Monday. He was basically. He was. He didn't say this explicitly, but he, he was basically. It's. He was seemed generally in favor of just stopping all AI research.
- [75:22]
A
Oh, I thought you were going to talk about his take on Call of Duty. He was like, I don't.
- [75:25]
D
Oh, what did he say?
- [75:26]
A
He was basically like, I don't want people playing Call of Duty. Like, I don't want kids playing Call of Duty. Which is crazy from like the GameStop guy. He's the CEO of the company, but he was. He had sort of a similar like anti video game take. Like maybe it's not as Safe as people think it is.
- [75:39]
D
Well, I mean, if you guys talk to teenagers and you're like, how do you guys hang out? A lot of them will be like, oh, we don't hang out very much, but we pro play a lot of video games together. Sorry, that's bad.
- [75:46]
A
All right.
- [75:47]
D
I mean, it's just, it's weird.
- [75:48]
A
It's weird. Wait, but what about all the dudes? I'm sure, you know, some of the guys who have been like, one shot by golf, where they're just. Golf is their whole personality. They just are. All they do all day is think about golf and then on the weekend they go golfing. It's similar, is it not?
- [76:02]
D
No, it's not similar because. Well, okay, when I said video games, what I said is they're at home playing online with their friends.
- [76:08]
A
Yes.
- [76:08]
D
Now, you and I, you guys, we're all relatively the same age.
- [76:11]
A
Yeah.
- [76:11]
D
Online play was not around that time. So what you would do is you would gather at a friend's house.
- [76:17]
B
It was, it was, it was, it.
- [76:19]
A
Was couch co op.
- [76:20]
B
It was a thing when I was a kid, but I lived somewhat in the country. And so before anybody could drive, it was like your, your parent wasn't necessarily. It was, it was tough to get around, basically. And so video games were a way that you could hang out with your friends and otherwise it would have just been hanging by yourself or whatever. Maybe that's. Maybe that's, Maybe that's better.
- [76:45]
D
Maybe. I mean, I don't know.
- [76:47]
B
What would get you to. What would get. Is there any. Can you imagine any scenario that would get you to change your mind on AI? Because I'm with you right now. It feels like AI is an expansion. Like the current state of AI in terms of products that people use and are an everyday part of their life is just an extension of big technology. Right. OpenAI has the aesthetics now of a big tech company. They launch products like they're a big tech company. They have a user base like they're a big tech company. They're very young, but that's what they look and feel like. They still have their nonprofit and they still have a research arm that is working on AGI. But it's hard for people to spend too much time thinking about AGI research when all the launches and the announcements are social media and new web browsers and erotica and things like that. But I'm curious, and this is something we are always trying to tease out on the show, is people that come on the show to talk about AI Typically have some AI to sell you or they have a startup equity to sell you that's sort of predicated on, on these sort of more utopian visions of AI and it's hard to parcel out. Right. And so you had the Carpathy interview recently, which was kind of throwing Fantastic interview. Yeah. And that didn't necessarily, like, update my worldview a lot. I think you put a lot of these ideas really well. Right. This idea of, you know, agents are not. The agents that we were promised, maybe this year are not here yet. And I think that's okay. But I'm curious what would, what would update your worldview? Because it's hard. It's like if we get massive cancer.
- [78:34]
A
If we get like an ozempic level drug success that was designed from AI, that probably makes it all worth it.
- [78:42]
D
I think what John is correct is just show me. Like actually show it to me. Stop telling me because it's creeping me out to watch talk about UBI and, you know, oh, everyone's going to be extremely happy. Even though I run the entire world. I mean, listen this. I got to put my own cynicism and bias on the table. Like, I'm literally an anti establishmentarian podcaster. When these guys open their mouths, I'm just like, you're lying. And most of the time that turns out to be true.
- [79:09]
A
Yeah. So. So I, I would, I would take the other side of this argument and, and say that it is possible that we never get some sort of big binary moment where we're like, oh my God, yes, like this. It did the thing that we wanted to do, but it's still like net good. I think about it like the invention of the credit card, it just speeds up the economy every little. It takes a little bit of friction out so that the corner store can sell a couple more muffins and then the person can pay with the digital cash stored in their bank account and they don't need to do the check writing. And it sort of puts a couple check writers out of business, but people have new jobs.
- [79:52]
B
Digital currency will track your, your CO2 consumption. No, no, no.
- [80:00]
A
But there are tons of, there are tons of examples of.
- [80:03]
B
Not to put on the tinfoil hat, but do you think it's. Do you think it's funny how for basically decades people were so against central bank digital currencies and now they're like, let me. Now, now, now everybody's like, I present to you stablecoins.
- [80:19]
A
They're a little bit different.
- [80:21]
B
Sure, they're different, but the potential risks associated with digital money are still there.
- [80:29]
D
Yeah, I want to pick up on what John said. I think it's such an important point which is. I think you are exactly correct but this is why I react so negatively to the rhetoric and guys, sure you have as well. One of my favorite books.com, you know, by Cassidy I could go through the list. Michael Lewis has a book on dotcom I've read through. I even I watched a few documentaries CNN and you'll never hear me say this actually has some decent documentaries about the decades and they have a decent documentary about the 1990s. Yes, shout out to the old CNN. Shout out to the old CNN. They had a great documentary about the 90s which I watched and the same level of enthusiasm they were talking about the information superhighway is quite literally matches much of the rhetoric of today. I'm too young to actually have experience much of it on the airwaves but the same way and the fervor which was fueling so much of.com just seems. I mean same thing like Ironically I use ChatGPT just before I came on here because I knew we were going to talk about this and that's what I was saying.
- [81:31]
B
That's what I was saying though. It's. It's one of those things like people like social media, they hate it but they use it well I don't hate.
- [81:40]
D
Yeah my finding was that we are mirroring almost exactly the same levels of tech stock growth from the year 1999 of the entire S&P 500 of stock gains and similar ones in terms of our gdp. If anything the GDP number is higher today. So the systemic risk which matches the rhetoric and that the risk itself of the fact is, you know John, what I've probably been texting you the most about is these round tripping vendor finance deals and you made a great point to me privately about ASML and how vendor finance itself is not necessarily a negative term in and of itself. But it's really hard not to go and read the story of Cisco and to say hey you know, if we have very similar type of downturn and reduction of expected revenue, I'm not saying OpenAI itself will go bankrupt, but that itself is fueling the entire retirement portfolio of the whole country. Not to mention gdp, interest rates, everything is banking on this one sector. I think that's really dangerous and especially that background article. You guys are going to know a lot more than me. But Sam's story about you know, going back to Masayoshi San which look no offense to the viewers. You know, I read the We Work book. I, I'm just saying, I'm not saying.
- [82:56]
B
I trust the guy. Credit credit on Masa with WeWork, Masa never really got a model markup. Right. Or maybe you got one or two, but not, not, not a lot on the real dollar.
- [83:05]
A
No, he knew he was buying the top.
- [83:06]
B
He top ticked it. The dialogue earlier this year was mas. Was, is Masa top blasting open AI and then of course, you know, OpenAI's up, you know, another one or something.
- [83:16]
D
Listen, his deal worked out, but it was more just the, to the extent of it's very similar kind of Adam Newman esque to me about like, oh, Sam, you need to go bigger. And that's exactly what encourages him to start doing so many of these deals. I don't begrudge them necessarily to do it. But you know, I really took to heart the too big to fail elements of this because I'm thinking about this.
- [83:35]
B
From a society just say you're not an investor in Softbank Group Corp. Up 173% year to date.
- [83:45]
A
So, so we have Jordan Schneider from Chinatalk joining in just a minute. But the last question I want is the update on, you know, obviously, you know, you believe there's an AI bubble. Is there a media bubble? Is David Ellison trying to build one? Like, what's going on in the media landscape? We were trying to map it out earlier, taking you through new media, neo media, traditional media, the legacy media, the mainstream media. There's so many different phrases like what's the most interesting story in Media in 2025? Because we've already had the story about like we're leaving the mainstream and going independent.
- [84:19]
D
Here's my counter take, which I really came home with the gambling story recently at NBA, there is a new media bubble and it's in sports. And it's entirely because FanDuel and DraftKings have propped up way too many people with opinions. And when there is even a modest pushback on prop bets and the rest of the bullshit that these sports books are allowed to send you. PM pardon my take. And maybe two or three others are going to make it. Everybody else is, is going to get wiped out. There are way too many sports sports podcasts if you're out there. And in my opinion, the biggest bubble is in sports media right now, entirely inflated by the gambling industry.
- [84:55]
B
Yeah. Do you have any, do you have any sense what percentage advertising revenues and in sports I think is gambling? But it has to be like 70, 80 to 90%.
- [85:05]
D
Yeah, 80 to 90%.
- [85:07]
A
I think the Pat McAfee deal was maybe 50% paid for by sports betting.
- [85:11]
B
Well I have a startup to pitch you. There's a new startup that launched that allows you to bet on your bills. OnlyFans, child support and last night Ubers wipe them from your credit.
- [85:20]
H
You'll love this by playing your favorite.
- [85:22]
B
Casino game to enrage soccer all from the comfort of their app. Are you.
- [85:27]
A
So if you have a big onlyfans, you can gamble it away. You can gamble it away, you know, double or nothing. Yes.
- [85:34]
D
The investor in me wants to stock. The investor in me wants to buy.
- [85:38]
A
That company vice clause on your own investments.
- [85:41]
D
Yeah, you're right. I mean I, I've said as much as I preach against this stuff. If I actually had to open a business it would call be called Game Day loans like instead of payday loans and it would float sports bettors who are out there who need like an 18 to 25% interest. So that's a free idea out there. Game Day loans. I'm telling you it'd be a billion dollar company.
- [85:59]
A
It's going to happen.
- [85:59]
D
It's happening. It's going to happen. If you want to sell it to fanduel, you're ready.
- [86:03]
A
Yeah.
- [86:04]
B
What, what's your last thing? What's your kind of like timeline on, on this sort of tension bubbling up around data centers and slop and, and the build out over.
- [86:16]
A
I have a very clear idea of how, of how like how bubble it.
- [86:21]
B
Can get more, it can get a lot more intense from here.
- [86:23]
D
It'll go national in 2026. The Democrats are going to win a decent amount of elections. Anti tech is already a small d Democratic issue. They are going to start hearing a ton of it at town halls. It's already happening and 2026, 2027 is.
- [86:37]
A
When it's gonna happen. And on the right you're tracking like Marjorie Taylor Greene or Josh Hawley like who's, who's going to be leading the front there.
- [86:43]
D
On the right it'll probably be Tucker Carlson at the top and from that point it will flow to the Marjorie Taylor Greens, the Josh Hollies. I'd be very curious to see where, where JD lands on this actually where David, David, I mean is a central figure in now.
- [87:00]
B
Yes, David Sachs Point. Sachs. Sachs's point I think from earlier this week was that without AI you have basically zero GDP growth. So I think he's taking side that's.
- [87:11]
D
The problem so well and listen that's not his fault. He's the AI Advisor, he should be making that point. But if I want to lead our country, that's not the sell I'm making. Because what we have learned over the last.
- [87:22]
B
We need to prop up the economy. So you can lose your job.
- [87:25]
D
Yeah, exactly. There you go. That is the end all of this entire conversation.
- [87:32]
A
Okay.
- [87:34]
B
This is the final boss of the anti tech wave. It really is potentially the final battle.
- [87:42]
A
Yeah.
- [87:43]
B
And it's going to be crazy. And I'm excited to have many more conversations.
- [87:49]
A
The singularity will be.
- [87:50]
B
Oh, I would love to.
- [87:51]
A
Yeah, yeah, we'd love to have you back. This is so much fun.
- [87:53]
D
Thank you guys for having me. I really appreciate it.
- [87:55]
A
Always welcome and congratulations on all your success, obviously.
- [87:59]
D
Well, I've been doing this shit for seven years. Nobody gave me a Sunday New York Times profile.
- [88:03]
B
I think it's time.
- [88:04]
D
I don't know what's going on here.
- [88:05]
B
We know a guy.
- [88:06]
A
Yeah, yeah. We gotta make it happen. Well, thank you so much. Have a great weekend. We'll talk to you soon.
- [88:11]
B
Cheers, Sagar.
- [88:12]
D
You too. See ya.
- [88:12]
A
Before we bring in Jordan Schneider from ChinaTalk, let me tell you about linear. Linear is a purpose built tool for, for planning and building projects and products. Meet the system for modern software development, streamlit issues, projects and product roadmaps. Let's bring in Jordan Schneider from chinatalk. Jordan, how are you doing?
- [88:29]
H
I'm doing all right. You should have kept me on with that last guest. That was great.
- [88:33]
B
I know.
- [88:33]
A
Yeah, we started doing a little bit of that earlier this week. We had like Brian Armstrong and Brian Chesky kind of next to each other for a little bit.
- [88:40]
H
Call them back in because the answer, answer is we will all be so distracted by our AI boyfriends and girlfriends that we won't be stressed out about losing our jobs.
- [88:50]
B
I think I actually, maybe that's the move. It's very dark, but I think that the world is so entertained now that almost nothing matters. And it's sad, but no matter how dark certain futures could be, it's. It's expected that everyone will have infinite entertainment constantly. And it's so much harder for people to stand up and protest and try to create change. If you know you can go back to your phone and oh, here's another funny video. I just saw Stephen Hawking's at the X Games and it was awesome and I'm gonna send it to a friend.
- [89:32]
A
It actually is awesome. It is sick. It's really sick. It's totally worth paying a lot for that.
- [89:40]
B
Totally worth the extra 2000.
- [89:41]
H
The duality of TVPN.
- [89:43]
A
I haven't even seen my dog. Just wait. Once I see my dog at the X Games.
- [89:48]
B
No, I think it's important. I think, like, you can't, you know, I continue to be a firm believer that the technology is a massive net positive for the world, but it has a dark side and it's important to talk about it, and it will be important to continue to talk about it.
- [90:07]
H
It's. It's numbing. There's like a. There's like a taking antidepressants lens to it, right, where you just kind of feel less if you get these dopamine hits. And we've, you know, we have not evolved to have the dopamine hits with, you know, within literally five seconds. Anytime you feel bad about yourself or bad about the world or something that happened around you or, you know, someone disappointed you in your life, it's like, oh, okay, like, instead of like addressing that or, you know, going to protest to Jordi's point, you know, we can be consoled by these things that live in our pockets and they're only going to get better at it, which is the sort of, I guess, awesome.
- [90:51]
B
This is the worst. This is the worst the, the dopamine feedback loop will ever be.
- [90:57]
A
This is the worst, worse, the infinite jest will ever be. Do you think China is less dopamine addled? Like, probably brain rotted because.
- [91:05]
B
Yeah, it feels a little bit.
- [91:07]
A
No, no to no brain rot. TikTok.
- [91:10]
B
Yeah, their TikTok is just science video.
- [91:13]
H
It's just absolutely not.
- [91:14]
A
Yeah, right.
- [91:15]
H
Absolutely not. I mean, look, like, who needs. Who needs opiates for the masses? More like a country with 20% youth unemployment or a country that has, like, what, 3.5% unemployment? I think the. The party has successfully kind of like.
- [91:33]
A
Wait, which country has 20% unemployment? Youth unemployment? Yeah. Which country are you talking about? Well, yeah, give us get.
- [91:41]
H
Let AI run for an extra year or two, I guess.
- [91:44]
A
But over the rookie numbers, if you told me that was America, I'm not that far off. I mean, I've heard that the numbers are going up. They might have spiked, obviously, but.
- [91:54]
H
No, I think, I think there is. I think this sort, you know, it's fun. You guys should have Jasmine son on. I asked her because she just came back from a trip to China. I was like, who walks around looking down at their phone? More people in the bay or China? And she couldn't really distinguish between the two identical. I don't just like, equally sort of like zoned in while like walking down the street. You are like simultaneously swiping.
- [92:23]
B
Ben sand in our chat says iOS Liquid Glass is the best line of defense against phone addiction. I totally agree. It's like the worst software Apple's released in so long and it just makes using your phone worse. So I'm sure, I hope my screen time is going down.
- [92:38]
H
Don't worry, Jony. I've will come in to save the day. Jordy, we won't be safe.
- [92:42]
A
That's not.
- [92:42]
H
That won't save us for long.
- [92:44]
A
Yeah. Okay, wait, zoom out. The original reason we wanted to chat was because of the. It's always fun to have you on the show, but obviously about the rare earth. Like just give us the high level update on like the trade war. This has been what, since the entire time? Since your very first appearance.
- [93:00]
B
Is it real?
- [93:00]
A
A year?
- [93:01]
B
Is it real?
- [93:02]
A
Is it even worth like thinking about like what the high level summary is of the status of the debate? Sure.
- [93:09]
H
I mean, I think like, like, look, we've seen this administration swing from side to side on lots of different things. You know, we were yelling at Zelensky in the Oval Office and now we're finally doing the sanctions that Biden couldn't bring himself to do around oil. You know, we had Trump trying to, you know, doing Liberation Day, taking it off, you know, giving them H20s. They don't want them now we're trying now, now they sort of tried to tighten the grip with the 50% rule from, from BIS, so sort of Huawei subsidiaries can't buy equipment and China came back with rare earths. I think this sort of the, the fundamental fact is that like both countries have the, you know, the sort of. Both countries have the ability to do significant economic harm to each other. And it's a bit of, you know, there's an aspect of playing chicken and there's this kind of question of escalation, dominance, like which side can take more pain and which side really cares more.
- [94:18]
A
Both of them have basically a ranked list of tools in the tool chest, chips on the poker table and something like we're not going to sell you rarer is extremely painful to America, maybe not that painful to China. And then we're going to tax Happy Meal toys or whatever is maybe something in the American tool chest and the Nvidia thing and is the process that we should think about the lens of this year more thinking about just every. Both sides are trying to actually understand the shape of how painful each tool in the tool chest is because it feels like we're very much like Turning these things on, turning them off to understand, okay, if Rare Earths are plus 60 points, Nvidia chips, we thought it was plus 20 points for the US it's actually plus 5. It feels like we're just trying to map all these out and then we will go to the meeting and actually say, okay, what's equal? Is that what's happening?
- [95:19]
H
Yeah. Well, the problem is, I think Xi has had a longstanding goal, which he's repeated in speeches numerous times, that he wants to make China more self reliant and he wants to make the rest of the world more reliant on China. And the problem with what you're seeing out of the Trump administration's policy towards this is what they're optimizing for changes literally week to week. So tariffs are for revenue, tariffs are for. This trade war is to bring manufacturing back. This trade war is to actually help Nvidia gain market share in China. The trade war is to crush the Chinese semiconductor companies. And the problem is it would be easier to get to some sort of equilibrium where each side is just going to run as fast as they can at their things. Like if this administration had a sort of grand vision that it was working towards consistently for. But I think the, the dance that we've seen now play out over, they've met each other like 15 times is it seems like the negotiators, every time they go are actually operating under kind of like different, a different incentive structure. So and I think this just plays to China's benefit because the fact of the matter is America has an enormous amount of power and leverage that it could bring to bear. But the first time that China bit back, in the case of rare earths, that kind of really freaked out the Trump administration and it showed and sort of the.
- [97:01]
A
What'll be interesting.
- [97:02]
H
Sorry, yeah, please.
- [97:03]
B
Yeah. To make this more tangible, like how is this netting out for China's AI industry? The messaging has been, obviously, you know, we're not, you're not going to buy Nvidia chips. Maybe we'll look the other way on, on Malaysia and Singapore data centers. We're going to invest as much as possible in catching up to the leading edge in our own, you know, domestic semi industry. We're going to dominate in power and in five to 10 years like we're going to be, you know, relatively, you know, you know, in their view, hopefully, you know, relatively equal in terms of like overall computing power. And then I want to get a sense of like on the rare earth side, how solvable that is. For the US in your view?
- [97:50]
H
Well, Chris Miller on chinatalk had a good line like, we're just, we're about to see what's easier to do. Build an EUV machine or build a rare earth's mining, you know, rare earth mines and refining facilities. So from the Chinese AI perspective, I do think that their decision to sort of say, we don't even want your H20s is fascinating. On the one hand, it may just be a ploy to get Blackwell to be like, oh, okay, like, we don't want this. We'll sell you even harder drugs to use Lutniks like addiction. Sure, phrase. But there also may be some sort of information mismatch because there obviously are lots and lots of sort of Chinese AI labs and hyperscalers who would love to buy Nvidia chips. But Huawei may have gotten into the ear of the State Council saying, yo, like, we got this. Don't worry, like, you know, but isn't there also.
- [98:48]
B
Haven't, haven't, you know, hasn't, like, Singapore and Malaysia, like, spiked quite a bit? Like, can that demand not just flow there?
- [98:58]
H
Yeah. So it turns into a bit of a bridge. And this is something that the administration seems to be, well, look, if you're, if you're fine selling the chips into China. Yeah, why not just sell them into. Into Malaysia and Singapore? I mean, I for one, am more comfortable with that as a kind of intermediate step. But the problem with that is, like, what you're doing is providing this bridge for all these competitors of the US AI stack to kind of like, keep developing and stay on the frontier, such that when Huawei really is able to scale, you know, potentially is able to scale to a level to compete with Nvidia, you have these firms still running. And the question I really want to know is, like, look, if we're, if we want to sell the AI stack to the world, why don't we just make. Why aren't we selling, like, all of it? Like, why is it cool that we're just doing the Nvidia stack and then, like, we're having, we're allowing Nvidia to do partnerships with like, Alibaba Cloud abroad and like, sell that into. Into Malaysia? I mean, it would not be that hard to say, okay, like, you want Nvidia chips, you should also be working with American hyperscalers and you should be buying from AI and buying Western models. That seems to me to be like a, Like a, like there aren't too many, like, alternatives on the table. The US does have the leverage there. And I think it's a shame that that's not the line we're pushing. We're instead just sort of like ending the monopoly of the AI stack at the chip layer and letting the clouds and models just, you know, go whichever way.
- [100:29]
A
I think America needs a deep seek for energy. An American deep seat for energy. Right. So what that would mean is like what did deep seek do? Like they got, they jumped to our first.
- [100:39]
B
Yeah. So Jordan Schneider is the cheap state.
- [100:43]
A
Yeah, really cheap, really fast. And. But they stole like our fundamental thing and like we need to steal how fast they can build energy over there and make it an American company. So how just like invent nuclear fusion?
- [100:57]
H
Yes, John.
- [100:58]
A
Well, it's not that they're doing right, aren't they? Aren't they install. They're growing way more energy. So what is the company that's doing that over there? What's the organization over there? What do they have that's special? Let's, let's copy that. Is that possible?
- [101:13]
H
I mean the answer is like it would like the reason I went to nuclear fusion is like it would be really nice to have a shortcut because the answer is like really boring and frustrating.
- [101:24]
B
But don't you think America will, don't you think America will just get lucky? Like we always do discover, you know, some novel, you know, method of energy. But we always, anytime we need a resource, we just find some of it. I feel like we'll have about this.
- [101:39]
A
Yeah, this is that this is the.
- [101:40]
H
Whole Tyler Cowen thing. Like do not underestimate the elasticity of supply. It's not just finding it right. It's like all of these, you know, global capitalism is now scared of their dependence on Chinese rare earths. And they kind of got a taste of that in April. You know, they should have learned their lesson in 2014 when China did this to Japan. But now it's really real. And now you know, all, all your VC friends that come on the show are going to understand that this is a, this is a much more investable, you know, finding like the solution to ytterbium and like you know, manufacturing like lab grown diamonds. Like people are going to be willing to pay a premium for that stuff in a way that they maybe weren't even six months ago.
- [102:25]
A
Why are lab grown diamonds important? There is actually a cyan banister company Long Journey is in some called diamond age.
- [102:32]
H
Amazing.
- [102:33]
A
I mean but is that important in the rare earth context?
- [102:36]
H
Yeah, it's an input to I think machines. This is actually one of the ones that the world is doing okay on. China has like 70% of the market. But yeah, it's part of the sort.
- [102:47]
A
Of.
- [102:49]
H
You know, it's used in fabs. It's part of the sort of manufacturing process.
- [102:54]
A
Yeah, that makes a lot of sense. How are you tracking the backlash to data center build out? We were just talking to Sagar and Jetty about this. Like, he thinks that like there's going to be a serious backlash. What's the deep state saying, Jordan?
- [103:08]
H
So here's the answer. You saw these, you may have seen these articles come out with the Secretary of the army talking to PE firms about how they're going to lease out land. It's like, all right, well, just stuff on. Yeah, just build this stuff on military basis.
- [103:25]
A
He's been on.
- [103:27]
H
No one's going to complain if it's just like in the middle of nowhere.
- [103:31]
A
Yeah.
- [103:32]
H
And, and you know, you can pollute all you want on those. There are all these sort of regulatory exemptions for these places. I think the money's going to find a way. The problem with this is like, sure, you know, the, the, the NIMBY stuff is useful when it's like, you're going. When it's like one neighborhood, right, or two neighborhoods. But like you can put this stuff almost anywhere and it is like 97% as useful as if you get to build it in, in Virginia. Right. So I'm not, I'm not all that pessimistic. I think, okay, maybe a few, a few, a few localities will be like loud enough to kick it out. But America's pretty big.
- [104:12]
A
There's a lot of places I like that.
- [104:17]
B
What do you think is going to happen on the humanoid front? We saw this really insane demo of a Unitree robot that was transitioning from walking to crawling like an insect. And now Unitree is selling in Walmart. And it just feels like we're going to repeat the DJI thing where we let one of every American have a humanoid in their home, a Chinese humanoid in their home. And I'm sort of surprised that we're running this back.
- [104:42]
H
It's interesting. I mean, there's, there aren't real competitors at scale for the consumer market in the US yet. I mean, there are very straightforward regulatory things that the Trump administration could do to ban them. There's also a handful of kind of dependencies that the Chinese robotics ecosystem still has on the US still has connected to the US So, you know, a lot of those chips come from tsmc. Like, there's, there's A sort of like Nvidia software layer. Maybe this is something that the Trump administration is thinking about when they're saying, oh, we're going to like ban critical software. So there are things that the US can do to kind of slow down the. The.
- [105:24]
B
Yeah. And I just, I just.
- [105:25]
H
If they really want to.
- [105:26]
B
Humanoids are not, you know, even in Uni unitree's robots current form, it's not like they're delivering incredible value to Americans. And by banning them, you're like, you know, reducing the productivity of businesses or anything like that. And so it feels like maybe in two years they're actually providing some value across the economy and it'll be much harder to go out and say we're going to ban the cheap ones and you have to buy American. And it just feels, you know, I don't.
- [105:54]
H
Well, what's really interesting is like in Ukraine, UGVs are a real thing now. So, you know, a lot of the, like sort of resupplying infantry that's on the front line is an incredibly dangerous thing. And now they just have these little track robots that go back and forth. You've even seen videos of like Cassievac, so people who are injured that they can't, you know, carry out in any other way, being sort of like rolling themselves onto some like bed and having a drone drive them out. So I'm with you, Jordi. This stuff, like, they're kind of cute and fun for now, but we'll get like, we'll get there pretty soon. It's not going. It's not impossible.
- [106:34]
B
Yeah, they're, they're cute. They're cute. But at the same time when it, when the humanoid starts crawling across your floor and you realize, and your, your, your human alarm, you have a biological alarm bell going off that says like this thing like looks like it like from all the sci fi movies I've seen, this thing looks like it wants to kill me.
- [106:54]
A
Yeah, it feels way more tractable to me. Like self driving cars like you know, you know, OCR text in an image. It feels like something that we can just like chip away at. Versus, like the novel scientist scientific discovery thing. That feels like very uniquely human, very complex, like much further out. This is just like make the motors work the right ways. It feels very simple. Like we're not trying to make someone superhuman at moving. Like it's just do the thing that any human can do and we have tons of training data for.
- [107:27]
B
So what's happening in Chinese markets right now?
- [107:32]
H
I don't know, man. I'm just the tech guy. I'm going to pass on that. Let me go back to you should.
- [107:40]
B
You should spin up a, spin up a small venture fund on the side. Then you'd be able to answer that question.
- [107:46]
I
Well, now.
- [107:46]
H
Well, the thing that I am spinning up that I'm looking, that I want your blessing from actually is so I've started a new series about war. I think it is actually just SportsCenter for war. We've been doing it once a week. The current name is Second Breakfast, but I think BPN would probably be a better thing. So I don't know what the licensing arrangement is here. Maybe this is a conversation for offline.
- [108:12]
A
I love the Second Breakfast concept. I love the, I love the topic selection. I would not recommend SportsCenter for war because the, because like there are just going to be episodes that are sensitive and it's hard to bring humor and levity to something that's really dark.
- [108:31]
B
SportsCenter is about providing an entertainment layer over existing entertainment. And like, I'm sure it's been a little dark this week with all of the gambling ring action. But, but in general.
- [108:47]
D
I don't think.
- [108:48]
B
You want your face across the New York Times being like Jordan schneider, Sports.
- [108:52]
A
Center for WAR yeah, it is tricky. I mean, people have, people have pitched us like, oh, are you going to do a trading card for the hostages being released or something in geopolitics and stuff. And it's like, it just feels like that's not our space. And when you recontextualize it, even if it's a positive story, it still feels like, oh, there's like, there's real lives at stake versus like, you know, just money at stake, which is what we talk about or just, or just points on the, in the imaginary sports world, like in the literal sports world, like it's, it is just a game. Yes, it's extremely high stakes for at the super bowl, but it is just a game. And so you can have a lot of fun with that and you can also have fun with just some VCs who are writing checks. But the war thing is a little bit, it requires a little bit more sensitivity, I think.
- [109:40]
H
Yeah, I mean it's interesting because like this community is just chock full of dark units humor. So, you know, I've told this to a few people and they just like think it's the funniest thing in the world.
- [109:53]
A
Even like going back and forth, sensational element of talking about the war history that makes so much sense. Like, for sure.
- [110:02]
H
I don't know, but. All right, point.
- [110:06]
A
Where can people check it out? I know it's in the Chant Talk RSS feed. Do you have to hit substack? How are people. People getting into the ecosystem?
- [110:13]
H
Yeah, it's just on the Chinatalk podcast feed for now.
- [110:17]
A
Maybe it'll go check it out.
- [110:19]
H
Metastasizing. But yes. Second breakfast. If you need more Venezuela content in your life.
- [110:27]
B
I've been out of the loop on what do you think happens with Venezuela over the next month? While we're here, you can give a little taster of your new show.
- [110:37]
A
Well, there's two.
- [110:38]
H
There's two paths.
- [110:39]
A
Where.
- [110:40]
H
Well, I guess there's three paths. They're on one, he's just bluffing and we're going to move on to something else in a month. But like, having this many boats in the Caribbean, sending a carrier makes me feel like it's a little more than that. So then there's path number two, where America fights like a covert war with a cartel. And then there's path number three, where we sort of like kill, exfiltrate, end up intimidating out of the country, sort of Maduro and his like leadership team. And I don't think this administration is necessarily going to be the if you break it, you buy it mindset. So, you know, then I guess the cartel takes over the country. I don't know. It is not clear to me that this has been entirely thought out. It just seems a lot like a sicario cosplay at like a giant, horrific. On a giant and like really horrific scale. Maybe there's a better idea out there behind all of this. But the, the sort of head of southcom for whom this operation would have been like the crowning achievement, decided like, said he was going to retire one year into his three year tour of duty, which has basically like never happened in the past 40 years or so. So I don't know, it's going to be really messy. The US Actually does have like a kind of nice case study in what in Colombia and what we were able to do to sort of like get the, you know, bring the like talk and fight the FARC out of existence over the course of the 2010. So that's the happy story that might manifest. But the thing there is like you were working with the central government. There were only 50Americans on the ground doing the thing. And like now we're trying to get rid of a cartel that apparently the president is saying is like one in the same with the government itself. So I don't know, Big Mess. Very far from like pivoting to China. But we'll get there eventually, you know, it's only a matter of time.
- [112:49]
B
I know. I think you gotta maybe start with updating China Talk. Maybe it's just.
- [112:54]
A
Maybe it's just Globe Talk. World Talk.
- [112:57]
B
World. Just World Talk.
- [112:58]
A
Globe.
- [112:58]
B
Yeah, Globe.
- [112:59]
A
World Talk.
- [113:00]
B
Globe Radio.
- [113:01]
A
I think just Jordan Schneider. I don't know. I mean, all these.
- [113:04]
B
Somebody in the chat said Saturday Night Live for human rights violations. So that's what you want to. That's what you want to avoid with your whole sports center for War thing.
- [113:12]
A
Because then you're rough. Although maybe if you just own it. I mean, like, you listen to Tim Dillon. He's like, extremely candid. Anyway, we do have a guest waiting, so thank you so much for coming on the show.
- [113:21]
B
Always good. Always good to catch up.
- [113:23]
A
Let's do it again soon. We'll talk to you later and have a great weekend.
- [113:26]
B
Good to see you.
- [113:27]
A
See you guys.
- [113:27]
B
Cheers.
- [113:28]
A
Before we bring in Chase from Crusoe, let me tell you about numeral. HQ.com, sales tax and autopilot. Spend less than 5 minutes per month on sales tax compliance. We have Chase Lockmiller from Crusoe in the Restream waiting room now. He's in the team. You've got Ultradome.
- [113:42]
B
Welcome to the show.
- [113:43]
A
You're looking fantastic. How are you, Chase? Congratulations are in order. Give us the breakdown. What's the news today?
- [113:52]
F
You know, today we are announcing the closing of our series E round of funding, valuing the company at 10.4 billion gong.
- [114:01]
B
Gong. Another one, another one, another one. 10.4 million, two gong hits.
- [114:07]
A
I don't know.
- [114:10]
B
Fantastic. Fantastic. Fantastic.
- [114:13]
A
How are you describing where the company, what you do? It's such a fascinating.
- [114:19]
B
What exactly do you do?
- [114:20]
A
Because, yeah, I mean, you've been doing some thing for a very long time and yet there's been multiple eras in terms of what the end customer has done. There's been where you sit in the stack, the type of size of project, scale of projects. So how are you describing the wheelhouse? That or the. Yeah, where Crusoe fits. Yeah.
- [114:41]
F
I think putting it simply, you know, Crusoe is an AI factory company. So what that means we're building, you know, these factories that produce intelligence. And this is both a hardware problem and a software problem. And we focus on all of the things ranging from cultivating the energy and helping develop the energy resources needed to power AI infrastructure. We're building out the data centers themselves and redesigning them to fit these large Scale accelerated computing clusters. Then we also built out this whole software platform that operates these large scale AI factories so that they can actually be effective in producing intelligence and really enable our generation's greatest minds and entrepreneurs to do their life's work and help build and scale their AI applications to impact people around the world.
- [115:33]
A
Is there a superlative that you're chasing in the neo cloud market? Do you want to be the firm that can build bigger than anyone else or faster than anyone else, or more reliable? I mean, obviously you want to win on everything, but is there something that you're chasing that you think is like the true differentiator?
- [115:52]
F
Yeah, I think, I think right now speed is the biggest focus for the business in terms of being able to move very, very quickly. It's a very dynamic market. But certainly reliability, scale, cost, all these things really come into play and ultimately we want to deliver for our customers what they need, what adds value to them. So being able to deliver to our cloud customers the best price, performance, trade off for their infrastructure solutions alongside a reliable, you know, well managed platform I think is, you know, ultimately our goal. But you know, I think being able to do that very, very quickly right now is a major, major advantage and it's something that's very difficult to replicate when you look at, you know, kind of what we did in Abilene, Texas in terms of, you know, starting from, from nothing into a large scale AI factory that's managing, that's operating tens of thousands of GPUs at this point.
- [116:56]
D
Is.
- [116:56]
F
A very important aspect to bringing this infrastructure to life.
- [117:00]
A
So Abilene's is up and running?
- [117:03]
B
Yes.
- [117:04]
F
We announced recently the first phases of the 1.2 gigawatt campus are up operational today.
- [117:12]
A
Operational. That's incredible because I feel like there was this whole press cycle about it's off track and this thing was announced like months ago. I expect these things to take years. I don't know, I was, I was. I feel like people aren't taking enough of victory lap there.
- [117:29]
F
Yeah, no, and I think there's been, you know, Stargate's been this like very overloaded term in terms of, you know, this. It's a company, it's a, you know, it's a site, it's a concept, it's a, you know, it's like what, what is Stargate? You know, I'm still not entirely sure but you know, there's, I think we got wrapped up a little bit in the, you know, rumors swirling that this thing's behind or, you know, that the company is behind. But you know, the project has done extremely well.
- [118:01]
A
People took the most expensive, the most, the most ambitious vision and the shortest possible timeline. And that became what you were measured against, basically.
- [118:10]
H
Yeah.
- [118:10]
A
And so it was like, unless, unless a trillion dollars is in the ground working in six months, like it's over. It's like, wait, we never said that. We said that.
- [118:18]
B
Well, it rightfully had a lot of pressure because the project got announced via the President. That's not very standard for launch.
- [118:27]
A
That's a good tip for startups. If you want a spokesperson, the President would be a great.
- [118:31]
B
Tyler Cowan has a quote that's been going around quite a bit recently which is like, don't underestimate the elasticity of supply in the context of. And it's been going around in the context of energy. It feels like Crusoe was started on this idea of like there actually is a lot of energy out there and there's real demand for it. You just got to go out and find it. How much does that, how much is that ringing true as you kind of evaluate new sites and opportunities for your business?
- [119:01]
F
No, that's absolutely the case. And I think that's one of the things that set us apart and enabled us to move quickly and build at scale is having this energy first approach to developing these infrastructure. Part of our announcement today was announcing our energy pipeline, which, you know, is in excess of 45 gigawatts at this point. And you know, putting into context, you know, the abilene campus is 1.2 gigawatts. So you know, it's, it's, it's, you know, orders of magnitude larger or an order and a half of magnitude and larger in terms of overall capacity there. I'm in New York City today, you know, it's something, you know, on the order of, you know, eight New York City's worth of power. It's a very substantial amount of power that's going to be required fundamentally for AI to scale and to really, really reach the ambitions and hopes of the industry.
- [119:54]
A
Last question, we'll let you go. Last time you were on, we kind of walked through a little bit of a debunk of the. We're going to run out of water for data centers. What's your current thinking on the story that's emerging around? Like data centers are going to drive up my energy prices. Is that solvable on the short term? Is this going to be some short term pain for long term gain? Is it something that, you know, you need, you need to rethink regulation or just speed up building of new energy. How are you thinking about that question? Because it seems like it's going to get bigger and bigger.
- [120:29]
F
Yeah, I actually found it fascinating. There was a report that came out recently that showed actually in markets with substantial data center investment, the overall cost to ratepayers actually came down compared to other markets where there was less digital infrastructure being built, where prices actually went up substantially. I really see this trend playing out as a whole because what ends up happening is the digital infrastructure players, the folks building the data centers. At this point, a lot of the energy market is saturated and there isn't just a bunch of free power laying around underutilized, not doing much. There's some of that, but it's becoming more and more scarce. Which means that in order to realize all of these ambitions that are taking place in AI, we actually have to build out a lot of net new power. We need to build new power plants, we need to build new creative energy solutions to effectively energize this infrastructure. And when data centers are doing that, they're having to build out power capacity that meets their overall peak demands of the facility. And as a result it ends up with some incremental power that is being underutilized and can become an asset to other participants in those energy markets and really actually bring down their overall cost of power because there's more abundant power. And that really goes in line with the Crusoe mission which is accelerating the abundance of energy and intelligence. We really view those two things as the key pillars of progress in this chapter for humanity.
- [122:08]
A
Yeah, I love it. I think it's definitely, I mean there's a bunch of different political frameworks where you could use to understand where this is going, but, but it feels very solvable.
- [122:17]
B
One, one, one last question that probably would, would take 10 minutes to properly answer. But how, how much are you banking on nuclear on a shorter time horizon, if at all talking, you know, five, five year timeline?
- [122:33]
F
Yeah, so we're really excited with a lot of the advancements happening in the nuclear space. You know, we have a number of partnerships across the SMR ecosystem, including a plan to energize the First SMR powered AI factory in 2027. So you know, that's going to be sort of a smaller scale deployment, but we think will be a great template for us to hopefully scale and bring new nuclear powered AI infrastructure to life. It's also, you know, this has been a huge priority for Chris Wright in terms of, of helping create a regulatory framework for nuclear and sort of reevaluating the nrc, the Nuclear Regulatory Commission in terms of how do we actually enable the private sector to build the energy necessary to power demands of society. Data centers being a big component of that. So very excited and I think it's going to be a huge win. And then, you know, there's also the big nuclear plants that, you know, I'm excited to see come to life again as well. You know, Westinghouse, you know, is announcing, you know, a number of very big APW 1000 campuses that, you know, we think will play another big part in terms of getting the power we need for intelligence.
- [123:56]
A
I'm very excited about that. I hope that we get all those campuses online as fast as possible. Well, thank you so much for coming. Coming by the show. Sorry to get the update. We will talk to you soon. Congratulations on the master.
- [124:07]
B
Congrats to the whole team. Awesome. Always great to see you.
- [124:10]
A
Fantastic progress. We'll talk to you soon. Have a great weekend.
- [124:12]
B
Cheers.
- [124:13]
A
Up next we have Tucker from Orion Technologies. We are running a little bit behind. 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. And let's bring in Tucker for a quick hit of the gong. Give us the news, tell us what you're building.
- [124:34]
B
We welcome to the show.
- [124:36]
A
Would love to ring the gong for you. How are you doing?
- [124:42]
B
I'm doing well.
- [124:42]
H
I can't hear you. Really? It sounds like you're speaking in slow motion.
- [124:46]
B
Oh, check, check, check.
- [124:48]
A
One, two. Orion Technologies, you're coming out of stealth. Would love to hear about the rays. If you can take us through just the high level, we can ring the gong and maybe we can try and have you back on later. Do we have Tucker? Let's see. We might need to move on if we have technical difficulties. We can always have you back on the show another day.
- [125:14]
H
Can you guys hear me?
- [125:16]
B
We can hear you great.
- [125:17]
A
We can hear you great. Let's see.
- [125:21]
B
I can, I can hear you guys.
- [125:22]
H
But it sounds very slow for some reason. Okay, okay, well let's, I'm the CEO.
- [125:32]
B
I'm the CEO of Orion Technologies and.
- [125:34]
H
We'Re coming out of Stealth after a 3 and a half million dollar pre seed.
- [125:37]
A
Let's go Jordy. Ring that gong. Thank you very much. Take us, take us one level deeper about the, the problem, the solution, what you're building and then we do have to move on. But please give us a little bit more on the progress of the company. Some of the decisions that you've made, how you're thinking about building the company.
- [125:59]
B
Yeah.
- [125:59]
H
So we are building essentially the interoperability layer for manned and unmanned aircraft. So what does that actually mean? We enable manned and unmanned aircraft to communicate, coordinate, and deconflict. And we're starting with heterogeneous assets, so drones built from different vendors within cities and around airports today.
- [126:16]
A
Mm. Congratulations. We're gonna have to have you back on later. We'll do a technical check next time. But have a great rest of your day. Have a good weekend, and congrats on the race. Congrats on the race. We'll talk to you soon. While we bring in our next guest, let me tell you about adeo Customer relationship Magic. ADEO is the AI native CRM that builds scales and grows your company to the next level. You can get started for free.
- [126:38]
B
New technical difficulty unlocked. We haven't seen that before.
- [126:41]
A
We have Katherine from Valthos coming in to the TVPN ultradome. She, I believe, is in the restream waiting room. Let's bring in Kathleen. Welcome.
- [126:53]
B
Welcome back.
- [126:54]
A
We had you on the show. It was a very fun conversation, and there were some things that we could talk about. There were some things that we couldn't talk about. I'm glad that we're here today and we have plenty to talk about, but please kick us off with an introduction on a little bit more on yourself and the company and what you're announcing today.
- [127:12]
J
Yeah. Last time I was here, I was so stealthy, but now very excited to announce what we've been working on. I'm Kathleen. I'm the co founder and CEO of Belthos, which is the next generation biodefense company. So we are trying to build infrastructure for America's biodefense.
- [127:28]
A
Amazing.
- [127:29]
B
And when we had you on, there was some crate, wasn't it? There was some story in the news about, like, some Chinese scientists that were bringing some type of bioweapon around or at least working on it. So good to have you back with some more happy news.
- [127:46]
A
Yeah, I'd love to know the state of what are the case studies that you're latching onto? In the early days of Anduril, there was this narrative around the rising power of China and our decrepit military systems. People would talk about the failure of the or not the delays with the F35 program. Do you have particular case studies and stories that you like to ground the story of your company with?
- [128:13]
J
Yeah, absolutely. So when we're talking about biodefense broadly, the big problem is asymmetry. So it is just so much easier, cheaper and faster to make a pathogen than it is to make any kind of cure. And that's just like at the heart of everything we're talking about. It was true before AI, but what's really changed is how some of these new AI capabilities are magnifying that type of asymmetry. It's two big ways that we're focused on in terms of the case studies. You were saying. The first is more about uplift. It's how many people, how much technical skill set does it take to actually engineer a pathogen? And that's just dropping week over week. We used to be worried about these sophisticated state sponsored programs. You can read what our folks in defense are focused on in Russia and North Korea. Now we can talk about a couple students, maybe at the grad level who have access to a lab. And that's probably going down to one or two as these capabilities expand so massively. Different landscape. The second side is just how potentially lethal can these fits be? So we're getting.
- [129:18]
A
I think we have more technical difficulties. It's a technical difficulty day. Oh no, it's Friday. The Internet is slow. But you're back.
- [129:25]
J
I'm back.
- [129:26]
A
Please. Maybe I can reframe the conversation. Not to just cut you off and restart, but people have been saying we need an Anduril for bio for a while. Is that an appropriate moniker? And maybe we could use the Anduril example as understanding a little bit deeper on what you want to build. Because I think of Anduril as very much separate from Palantir. They're building hardware, they're building specific programs. Record for the DoD competing against the, the primes. Are there companies that you're going to be competing with out of the gate? How do you think about the shape of the business as it grows? Is this on our side? Are we having technical difficulties on our side today?
- [130:11]
B
It might be a zoom. Might be a zoom thing.
- [130:14]
A
Is there a major attack? Let's flip over to the production team. What do you guys have to say for yourselves? What's going on?
- [130:22]
B
I don't know. Is it AWS 2.0?
- [130:25]
J
It could easily be us. I seem to be back now.
- [130:27]
A
Okay. Yeah, I think we have, I think.
- [130:31]
J
One more chance at talking about the Andrew comparison, please. I would say probably much closer to Palantir than Andrew. I mean, one of the obvious ones is like, what, what does our product look like? And we're talking about software and how we bring like the, the Best frontier capabilities to both understand what a threat is and then very quickly design a countermeasure to it. How do those things are that are largely in academic labs, emerging really quickly and put it into something that's operational. It could actually react in the time as real time data flows in and you actually need to design a countermeasure. But in terms of the overall, do we need private sector innovation in this space? Absolutely. Like the people in defense and health that are absolutely at the front lines of this keeping us safe now the pace of biotech innovation is so massively outpassing, outpacing our ability to react to it that like, we have to start bringing these types of tools to really augment the defenses that we have today.
- [131:26]
A
Yeah.
- [131:27]
B
Question from the chat. Trey asks. Moderna designed the C19 vaccine in two days. In a clinical dev today. The slow part is often validation trials, manufacturing and access distribution is valve though solving for that for biodefense.
- [131:43]
J
Thanks, Trey. Yes, I think a huge part of it is how, how accurate you can make predictions that are coming from computational models. How did this translate into clinical development? Translational. Can you actually have a really precise understanding of what these therapeutics look like as soon as you start designing them in silico? So that's a huge part of our focus is as more and more generative methods for countermeasure design come online. How could we help the people that are designing those drugs really quickly understand whether it will be effective against whatever kind of evolution we're seeing in terms of the, of the pathogen landscape? I think on the manufacturing and supply chain side, another massive problem. Do you think about overall biosecurity for America? It's not the first challenge we're taking on, but it's certainly something we'll partner on in the future.
- [132:32]
A
How do you think about, I don't know, there's like, like ISR versus like actual bombs or something in the Andoril context, like, like Palantir might make a map of like where all the bad guys are and then. And might be the one that builds the missile. How do you think about, like, does America have the ability to pull up a dashboard of every different strain of flu that is at LAX right now? Because I know LAX is like a famous petri dish of, you know, just like different flus and colds because everyone's traveling from all over. Do we have the ability as a country to actually understand what's going on biologically across the country?
- [133:20]
J
Largely, no. It's getting better. I think there's two dimensions to it. So a lot of the innovation we've seen in the last couple of years is focused on do we even have like the lens? Like is the data even flowing in of what's circulating in LAX or I'm closer to jfk. So that's the one that really hits home that I get worried about. The part that we're focused on and what I think has to be the next step is how do you take those data inputs and make them actionable so you can do something with it. So pathogens aren't static. It's not like there is something today and that's what we need to worry about. It's how is this changing? Which means that in two weeks what is the shape of this that we're going to have to hit? So a big part of what we're focused on is trying to understand these are the things that are emerging today. How quickly do we know that there are threats and then what do we need to change in terms of the countermeasures that we have to be ready for what's ever coming. So really focused. And I think that's where a lot of the Palantir analogy comes in. It's not data for data's sake, it's how do you actually pull that data into the decision making process so that you can start seeing changes in operations? And that's what we're excited to do.
- [134:20]
A
That makes a ton of sense. Give us the fundraising news. What happened? What's coming out of stealth?
- [134:26]
J
Well, we are very proud to announce that we raised 30 million.
- [134:33]
A
Congratulations.
- [134:34]
B
Incredible. Who, who are the backers?
- [134:37]
A
A bunch of no names.
- [134:38]
E
Right?
- [134:40]
A
You really had to scrape the bottom of the barrel. You had to put together ragtag group of. You could barely call these guys venture capitalists.
- [134:49]
J
Everyone wants to prevent bioweapons. What can I say?
- [134:52]
A
No. It's a murderer's row. You got a bunch of great firms in ff. We do.
- [134:58]
J
We have Founders Fund and Lux. They've been with us from the beginning and then we just had OpenAI their startup funds join in, help lead the round. They're like they have been ringing the alarm on what the bio. The risk of some of these new frontier methods are for bio like way before anyone and we're super excited to partner with them to try to stop it.
- [135:18]
A
Well, congratulations. We'd love to have you back on the show and whenever the next announcement is, whenever the next progress point comes or whenever, hopefully we don't have to be giving you a call when bad stuff happens, we can just talk about processes.
- [135:32]
B
Celebrate these wins.
- [135:33]
A
You'll be like a good cornerback in the NFL. We won't hear about you because you will be defending us and we won't be calling you on something like that.
- [135:40]
B
Perfect.
- [135:40]
J
We'll prevent the headline.
- [135:41]
A
Exactly.
- [135:42]
J
All right, Thanks a lot, guys.
- [135:43]
A
Thanks so much.
- [135:43]
B
Great to have you.
- [135:44]
A
Great.
- [135:44]
B
Rest of the team.
- [135:46]
A
Let me tell you about 8sleep.com I put up a generational run last night. 98. I slept for like 10 hours. It was fantastic. Jordi, how'd you do?
- [135:57]
B
91.
- [135:57]
A
91.
- [135:58]
B
How many hours did you actually sleep?
- [136:00]
A
Let's pull it up while we bring in our next guest. I slept 8 hours and 42 minutes.
- [136:06]
B
Wow. And I got to the gym.
- [136:08]
A
Get a podcast.
- [136:08]
B
I got to the gym. 30. At least 30. My drive is twice as long.
- [136:12]
A
This whole week.
- [136:13]
B
You were slacking for me.
- [136:14]
A
I was slacking. But. Our next guest is from Nvidia. Dion Harris is here.
- [136:19]
B
Welcome to the show.
- [136:20]
A
Down for us. How are you doing?
- [136:21]
B
And looking incredibly sharp.
- [136:23]
A
You look fantastic.
- [136:24]
E
Thanks for having me. I didn't want to make you guys look bad, so I even threw on the jacket.
- [136:27]
A
Oh, yeah, yeah. You're a fan of the show, by the way. Thanks so much.
- [136:31]
B
So great to have you on. Great to meet.
- [136:33]
A
We'd love to get a little bit of an introduction on yourself and where you sit in Nvidia. It's. It's not the smallest company in the world. It is, in fact, the largest company in the world. But. But how, how do you, how do you position yourself within the firm? What's top of mind for you today?
- [136:50]
E
Sure, thanks. Thanks for having me again. So, Dion Harris. I'm the senior director of our HPC AI and Cloud Solutions group. So not a very imaginative marketing name, which means I run a lot of our HPC scientific computing national labs. Obviously, we're here in dc, so, you know, lots of engagement, supporting a lot of our national labs, both here in the US and across the world. And of course, AI and cloud speaks for itself in terms of a lot of our hyperscale clients as well as the cloud providers who are building and delivering our platform. So engaging with a lot of those folks, partners, customers, and internally, obviously, to bring those solutions to market.
- [137:25]
A
Yeah. How big do you have, like, a headline number that you're thinking about broadly for just AI infrastructure? It's the hottest topic these days. But how do you think about forecasting AI infrastructure and where we'll be in a few years?
- [137:43]
E
Well, you know, I think Jensen has been on record and I think it's a pretty sound sort of estimate. We expect it to be about 3 to 4 trillion dollars in AI infrastructure. Yeah, yeah, that's a big number. It's with a T, so. And I think what's really driving that, and I've been listening to the show and hearing, you know, there's a ton of demand for how and where AI can be put to use. Of course we're familiar with the chatbots that we use and to speed up our email communications and to, you know, review and spell check. Those are great use cases. But what's really exciting is when you see it you being used to detect new drugs. And so there's new use cases that are happening in drug discovery all the time. I do a ton of work with national labs and seeing how it's being applied to climate and weather, being able to replace a lot of our old numerical based approaches. And so using AI to go and tackle really big problems that we all care about, even if we don't see and use those applications every day, we're impacted by them.
- [138:41]
A
Right.
- [138:41]
E
And so what's really driving that number is, like I said, also looking at how AI is being brought to industries. You're going to hear a lot about next week when we announce at GTC how we're helping industry adopt AI. And this is really important not just in terms of the bottom line of US companies, but really making the US competitive in a global scale. Right. Making sure that we can deploy and run and build and manufacture at an efficiency that makes us competitive on a global scale. And so that's really going to be a lot of what we talk about here at GTC next week.
- [139:15]
A
Do you have, do you have one of the easier jobs in tech? Just because if I'm a data center operator, I have to answer questions about, well, are am I using natural gas or clean energy? And if I'm an inference model creator, I have to answer questions about are the answers truthful or are they fair or am I generating something that's valuable or am I generating slop? But Nvidia, I feel like has the perfect, it's just the perfect business because you're already optimizing for the most efficiency. So no one can say, oh well, you're not taking energy efficiency seriously. Like that's the entire point.
- [139:51]
B
And no one can say you're unprofitable.
- [139:52]
A
And no one can say you're unprofitable. So like, are you feeling good right now? Are you feeling happy? I don't know.
- [140:00]
B
What's the Energy like even internally at Nvidia, I'm curious. It's got to be such a wild environment.
- [140:08]
E
It's definitely incredibly crazy right now. But what I would say is just because there's so much interest and excitement around not just AI, but accelerated computing, a lot of what we do. We're an accelerated computing company at heart and there's over a trillion dollars of infrastructure that still moving from CPU to GPUs. In fact, Oracle just last week announced that they're going to be accelerating their classic database. Now most people don't think about databases and they're back ending every major application. And so when we think about Nvidia, obviously accelerated computing is how we've really got started. And now that we're in this era of AI, we're really looking to help power every application across multiple industries. But to answer your question, we wake up thinking about how can we make our platform better, how can we make it more efficient? Because that's really at the core of the value for our customers. But we're actually doing a ton of work beyond just our core platforms. In fact, our solutions are shaping and redefining how the data center overall is built. A lot of the work that we're doing is building solutions and blueprints and reference architectures that help the entire ecosystem get ready for what we're building, which delivers more efficiency at the, at the end state. So again, I think our position is unique in the marketplace in that we totally see and understand all the new, new models that are being developed and coming, coming to market. We obviously understand what our products are going to be able to do today and going forward. And we therefore are giving lots of insight into all the mechanical, electrical and plumbing infrastructure providers. The power generation folks.
- [141:44]
A
Yeah, like how do you.
- [141:45]
E
The great providers, we're giving lots of feedback all the way up, up and down the supply chain.
- [141:49]
A
That, yeah, that Oracle database on GPU thing is. It feels massive to me. It feels like very significant, like immediately very tangible, you know, performance benefits. I can imagine. I haven't seen the actual stats, but I imagine it's extremely. It's going to be more efficient or faster. But how do you balance marketing, like sort of stories like that, stories about the AI build out with stuff that's more futuristic, more further out. There was a lot of debate on the timeline this week about Star Cloud. Yeah. How fast GPUs will be in space. And everyone's saying, like, look like we love this founder. We're having him on the show actually next week. It's a really exciting project but I think everyone kind of agrees that this is not going to happen tomorrow. And the rendering the sci fi is more of a sci fi movie than they put out shows like, you know, whole truckloads of GPUs moving around in space. How do you think about balancing that from the Nvidia brand perspective?
- [142:49]
E
Well, from our perspective, you know, we recognize that like I said, there's a lot of interest and investment like we Talked about that 3 to 4 trillion dollars that's going to be invested in AI infrastructure. And of course that infrastructure is going to be deployed, you know, in lots of different areas. Obviously as we've talked about how a lot of the concentration historically has been around, you know, high populated, highly populated, dense areas. But now we recognize, recognize, you know, AI factories can exist anywhere. They can exist anywhere where there's, you know, cheap clean power. And so obviously space is someplace where there's lots of renewable energy. So yeah, I suspect that will certainly be a place where data centers will land eventually. But right now, you know, we're really focused on how can we help build the infrastructure that's right in front of us and help really deploy that in most efficient, performant way as possible. And of course along the way make sure that we're adding value because that's really what we're here to do for not just ourselves but the broader ecosystem.
- [143:46]
B
You would laugh. We were running the numbers on building a hearth or a wood powered data center here at the studio that's surprisingly.
- [143:55]
A
Doable out of a normal fireplace. We believe you could potentially power 8H 1002 and fine tune GPT OSS however you want it over the course of a day. Basically.
- [144:11]
B
We want to be the first Nvidia customer to be focused on the wood powered data center.
- [144:15]
A
Wood power, yes.
- [144:16]
B
There's our own part of the market map.
- [144:19]
A
It's very cozy to chop firewood and boil water as they did hundreds of years ago.
- [144:26]
B
That's right. I'm super excited for GTC next week.
- [144:29]
A
Yeah, this is great. Thank you so much for giving us.
- [144:31]
B
A little free for giving us the preview.
- [144:34]
E
Thanks for having me.
- [144:34]
H
Have a great weekend.
- [144:35]
A
We'll talk to you soon.
- [144:36]
B
Great hanging, appreciate it.
- [144:37]
E
Thanks a lot.
- [144:38]
A
Before we bring in our next guest, let me tell you about public.com investing for those who take it seriously. A multi asset investing industry that yields.
- [144:45]
B
And they're trusted by millions, soon billions.
- [144:48]
A
I think we might actually have a.
- [144:50]
B
One minute break before our next guest comes in.
- [144:54]
A
There's no one in the restream waiting room or is there? No, not yet. So give us a post. Jordy Hayes if someone's smart, if someone is smart but has bad aesthetics, normies will not take them seriously. This is good aesthetic. Aesthetic contains real information. This post is from Defender of Basic that this category of smart people can't read and if they gain power, these blind spots will lead to their undoing. I completely botched this reading.
- [145:22]
B
This is a massive here's another one Alex Danko Sundar Pichai said yesterday or a couple days ago New breakthrough quantum algorithm Published in Nature today, our Willow chip has achieved the first ever verifiable quantum advantage. Willow ran the algorithm, which we've named quantum echoes, 13,000 times faster than the best classical algorithm on one of the world's fastest supercomputers. This new algo can explain interactions between atoms in a molecule using nuclear magnetic resonance, paving a path toward potential future uses in drug discovery and materials science. The result is verifiable, meaning its outcome can be repeated by other quantum computers or confirmed by experiments. This breakthrough is a significant step toward the first real world application of quantum computing, and we're excited to see where it leads again. See this positioning, right? It's positioned as an internal science project. Of course, he's saying it's a step toward the first real world application of quantum.
- [146:23]
A
Did you see what Martin Shkreli said?
- [146:26]
B
What did he say?
- [146:28]
A
Contrived results, still not faster. Advantaged. He's like, so bearish.
- [146:34]
B
Alex Danko, though, says, okay, I'm sorry if this is a dumb question.
- [146:37]
A
Elon, congrats. And I think when I see this, and I see the algorithm and I see, like, how they're talking about it, it really does feel like, yes, over the next decade, like there could be some really, really powerful niche use cases.
- [146:48]
B
That are extremely good. It's a good, it's a good quantum computer, sir.
- [146:52]
A
I agree, I agree.
- [146:53]
B
But one more post before we go.
- [146:55]
A
Did you want to explain why quantum computers look like that? Because that's what Alex Danko is asking, and there is an answer. It's because they're. They need to be cooled. They need to be really cold. And so you have to hang it instead of attaching it to the floor. Each layer of the chandelier is colder down to the last one where the quantum computation is taking place. And so you see this chandelier funneling smaller and smaller and colder and colder until you get to the very, very tip. I don't know. Interesting. Well, we have our next guest in the restream waiting room. Let's bring him in to the TPP and Ultradome. How are you doing? Welcome to see you guys.
- [147:37]
B
What's happening? Huge, huge day. Congratulations on all news in the fall ecosystem.
- [147:43]
I
Thank you. Thank you. So today we are doing our inaugural Generative Media Conference. I'm in the Ferry Building.
- [147:52]
A
Very cool.
- [147:52]
I
You can probably tell. Beautiful venue. Yeah, it's a big milestone for us. It's, you know, we thought that Generative Media really needs its space. And, you know, we need to talk about it because it's an extremely big deal. And, you know, we thought we would honor this industry by throwing a big event where foundational models are coming here. People from the industry are here. Jeffrey Katzenberg, actually.
- [148:25]
B
Legend.
- [148:26]
A
Very cool.
- [148:26]
I
That was. That was pretty awesome. Yeah.
- [148:28]
A
So what are attendees trying to understand or advance?
- [148:36]
B
Yeah. What are people talking about?
- [148:38]
I
Yeah, yeah, there is, like. I mean, this is becoming a full industry. Right. So there's many kinds of people here. So there's people from Foundational Model Labs. So they're trying to build these models, compete against the big giants. Right. And it's a very diverse space. So unlike the language model market, where you have a few big winners here, we're seeing, like, hundreds of companies building models, customizing models, and using their customers are using in production. So there's foundational models, there's people from the industry. So media. We had a talk from an architecture firm that's using Generative Media and production and. Yeah, like early stage founders and even some big Hollywood names, like big production studios.
- [149:30]
A
Yeah. Are people focused now on benchmarks, evals, certain niche capabilities? We're starting to see this kind of refragmentation. I feel like where SORA can DO cameos and VO3 could do audio and certain crayon can do lip syncing well, and there's all these different sub applications where it feels like if you're a buyer of Generative Media, of tokens or I don't need frames, I guess you are really looking for the right tool for the job. And you might have to kind of go through the haystack to find the needle that fits your use case specifically. Is that the general vibe?
- [150:13]
I
Yeah, absolutely. I think there's definitely alignment on how these models are getting more specialized. On one hand, there's general models. I think VEO is pretty general. It can do many things. But on another hand, you need more. And what we're hearing from people is you need more and more specific workflows. So workflows is a very big topic. And we are now in the usefulness, utility phase of this whole space where.
- [150:42]
B
Like let's give it up for utility and usefulness. Love it.
- [150:47]
I
So what we're seeing is, you know, people are building these, building this tooling. Right. So they're building the application layer such that these downstream users from marketing, architecture, gaming, movies so that the creatives can start using this technology.
- [151:06]
B
Yeah, how, how's, what's your view on how Hollywood has most recently been kind of navigating Gen AI? We've heard from folks in the industry that people are experimenting with it, but they're hesitant to even talk about it too much because they get so much blowback from. You know, we're here down in la, there's a lot of people that aren't excited about Gen AI for a number of reasons. But how do you see Hollywood navigating this current moment?
- [151:33]
I
Yeah, I mean we're definitely feeling like a shift in the vibes. So they're definitely turning towards a side of experimentation. I know there's a few good projects in the pipeline that are going to come out. I think we're going to see this shift happening towards the end of the year, early next year. And yeah, we see very big players actually embracing the.
- [151:58]
A
And yeah, it's gonna be.
- [152:00]
B
Somebody in the chat. Somebody in the chat just said you guys should get Hollywood on as a guest. We'll work on that. Another question I had is, I don't know how much you can share, but what's Sora 2 adoption been like so far? How quickly are a number of different application layer companies kind of turning it on and burning up the GPUs? Yeah.
- [152:26]
I
We don't have full visibility into Sora's adoption, but from what we know is it is a very good model for a very certain aesthetic.
- [152:37]
H
Right.
- [152:38]
I
You can probably tell that it is extremely fine tuned on social content. So yeah, it is getting amazing adoption, but this space is virtually infinite. Like there's literally all sorts of things you can do. That's why we believe like this. We're just in the beginning. And yeah, it's just a form factor. Right. And it's just one kind of model. We think there's going to be many, many versions.
- [153:08]
A
Totally. Can you walk?
- [153:09]
B
Yeah. My immediate reaction to Sora is like I could see this being used for like very specific. If you wanted to generate UGC style ads, it could probably do a decent job at that. But that same application would also want to generate ads in a bunch of different styles. Right. Cinematic styles, things like that.
- [153:27]
I
Exactly. And yeah, like don't get me wrong, I think you know that model like it's the pre training versus post training, like it's, you know, I would speculate that it is really fine tuned on this like social use case, but that doesn't mean that there is no amazing model in the back of it. So technically they can crank out different things for different use cases.
- [153:50]
B
Totally.
- [153:51]
A
Yeah. Yeah. I mean I could go so much deeper on all this and all the comparisons of all the models. We'll have to have you back and because I know you're at your conference and obviously need to get back and meet customers and whatnot. But thank you so much for taking the time to stop by.
- [154:07]
B
Excited to have a bunch more folks on in the next 30 minutes or so.
- [154:10]
A
Yeah, this will be great. We'll talk to you soon.
- [154:12]
B
Have a great weekend.
- [154:13]
A
These guys. Let me tell you about 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 efficient seamless ad buying across the globe. Our next guest is Glenn Solomon from Notable. We'll bring him in to the TV pin ultradome.
- [154:32]
B
What's going on? Can you hear me? I cannot hear you. We don't have audio yet. No audio. I'm sure the production team is working on it. I will tell you when I can hear you. So just I guess keep chatting, I guess. Still, still nothing. How we doing folks? We got nothing. We got nothing coming through. AGI is just days away and we need AGI to put some resources towards technical difficulties for podcast live streamers. Anyways, we will Alkyl time. Here's a post yesterday.
- [155:18]
A
Okay, sounds good.
- [155:19]
B
Okay, I think we got something coming.
- [155:21]
A
Can you hear me?
- [155:22]
B
I can hear you.
- [155:23]
A
Ah, Jordy, great. This is Glenn.
- [155:26]
B
It's so great to see you, Glenn. An amazing, amazing setup here.
- [155:30]
H
Are you.
- [155:30]
B
Did you already give a talk or are you going to give a talk or are you just. I've been in the.
- [155:37]
A
I've been, I've been enjoying the, the amazing talk so far and I'm excited to talk to you about them.
- [155:43]
B
Amazing. Why don't you give a quick intro on, on yourself and your firm and then, and then we can get into reactions from the event.
- [155:51]
A
Sure.
- [155:52]
B
Yeah.
- [155:52]
A
So I'm Glenn Solomon, managing partner at Notable Capital. We've been in business for 25 years. We're a global venture capital firm focused on software infrastructure and one of the key Personas we focus on as a software Developer. We backed companies over the years like Vercel, recently, Hashicorp, some great businesses that we've been very fortunate to work with.
- [156:18]
B
What was the exact moment that you founded the firm 25 years ago?
- [156:24]
A
2000.
- [156:25]
B
No, I know but like what was it? 2000? Was it March? Was it during the crash? Like when, when do you. Do you remember the month specifically?
- [156:33]
A
It was February, so early in the year.
- [156:37]
B
So you basically went on a roller coaster. You started up here and then just.
- [156:41]
A
That's little known fact though it's actually good to found a firm right about.
- [156:45]
H
Then because you don't get.
- [156:48]
B
You can deploy in the valley of despair.
- [156:51]
A
Exactly, exactly.
- [156:52]
B
That is a great timing, great timing. Amazing. So yeah, walk us through some reactions to the event so far.
- [157:01]
A
It's been really cool to be here. It's the first generative media conference that Fouls put on, really their first user conference. There's about 300 people here and you can just feel the energy amongst the developers who are building all kinds of amazing applications using general media. It reminds me I was lucky enough to invest early in HashiCorp in 2016. I was at their first first user conference up in Portland and there was 150 people in the room.
- [157:28]
H
Basically anybody who wanted to attend could.
- [157:31]
A
And it was a bunch of very intrepid DevOps professionals who saw the future somewhat similar to Hashicorp. And the reason I mention is because four or five years later at their user conference in San FRANCISCO There was 5,000 people. And I think this is the start of something very similar.
- [157:49]
H
We're going to see this industry grow.
- [157:51]
A
Dramatically over the next several years and it wouldn't be not surprise me if Dreamforce, Lookout Generative Media Conference is going to be really big over the next several years.
- [158:01]
B
I believe it. Berkey was saying there's an architecture firm talking at the event on how they're using. What are some of the other maybe more edge cases, niche use cases that you're seeing?
- [158:16]
A
Well, definitely advertising, marketing, your feed on social media, entertainment, education, all of these things are. People spend dramatic amounts of money today, hundreds of billions of dollars on content production for traditional media. I think that's all going to shift.
- [158:35]
H
Over time and we're just starting to.
- [158:38]
A
See the early days of that. But clearly there's going to be a shift because using generative media you can move much quicker, it's less expensive, you can be much more targeted in what you do. So I think all those industries are in experimentation mode. Maybe first, second inning. But it's early but One of the things I heard about today that I thought was really interesting is really the net new applications that I think are going to come as well. Things that don't exist yet today. A good maybe harbinger of that would be like if you've ever been to the sphere in Las Vegas, you look at the geometry there and the sheer size of the place. Traditional media couldn't really solve that problem and create an immersive experience. That's a great area where generative media is doing something that couldn't be done in the past. So I think you're going to see both replacement and over time net new. And I think that's why people are.
- [159:29]
H
So excited about what's going on here today.
- [159:32]
B
Makes a lot of sense.
- [159:33]
A
Yeah. That's amazing. Well, thank you so much for stopping by the show. We'd love to talk to you more.
- [159:38]
B
But yeah, first, first guest appearance from a podium.
- [159:43]
A
Yeah. Hey, I'm a longtime listener, first time caller. This has been awesome. Thank you guys.
- [159:50]
B
Amazing. Well, come back on again soon and have fun out there. Give, give our best to everybody at the event.
- [159:54]
A
We will.
- [159:55]
B
Love you.
- [159:56]
A
Thanks man. Cheers. We'll talk to you soon. Let me tell you about Basil. You got to go to get bezel.com just you.
- [160:07]
B
Hit her.
- [160:07]
A
They're available now to source you any.
- [160:09]
B
Watch on the planet. I'm really pushing John to get the cubitus. I think it's time green dial. Well, I wish I had more time to give you the full, my full case for it. But we will move on to our next guest.
- [160:24]
A
We have Diego Rodriguez from Crea in the Reese room waiting room. Welcome to the show. Diego, how are you doing?
- [160:29]
B
Welcome. What's happening?
- [160:32]
A
Talk to me about the AI suite for creatives. I'm particularly interested in this debate between like chopping wood and creating new workflows for specific video generation or AI use versus like trying to instantiate some things on the fly. I always see these like I made this amazing AI video. Here's how I did it. And it's piecing together like three different steps where they fine tuned on their face and they did a special model that had a beginning and an ending frame and then they did some sort of control net on top of it and then they up res ed it. And how are you thinking about actually piecing all the different tools together in a way that someone. It's like the final UI effectively.
- [161:17]
I
I mean like, I mean I'm here like at the fall conference and I talk actually exactly about that.
- [161:23]
A
Okay, perfect.
- [161:25]
I
I don't know, it's kind of frustrating to see people being like this AI generated. This is not dude, it's a new medium.
- [161:30]
A
Yeah.
- [161:31]
I
Like it's just a new medium. So like whenever I see a frame by say now Hollywood is starting to toy around these tools. Like the, the best people tell you like, like you, you, you don't ask them like is do you use Photoshop or After Effects or Houdini or AI? He's like, do it all of them. Like, this was even a dialogue when we were discussing about a certain idea that's also part of the process of making things. I think we're far from automating true storytelling, which is why we need tools. Where's the pencil for the latent space? That's what we need.
- [162:04]
A
Yeah. How do you think about what the actual product looks like that people want at each level of scale? Enterprise is fine with APIs and they will have a team of software engineers build the tools exactly that they need. Then the prosumer might be fine with a node based workflow or something that's a little bit more fine tuning parameters under the hood. But then they want something that's maybe repeatable, they can jump in and out of. And then the final consumer wants like, you know how you see those like dedicated app just for. It's like just the face swapping app. And like face swapping is obviously a feature that should exist in the node based workflow in Kreia, in prosumer, in a professional tool. So how do you think about checking the box on all the features out of the gate versus leaving enough flexibility that your users might get lost because you don't handhold as much.
- [163:06]
I
I think that these barriers are, let's call them vestigial concepts from our culture. I think we think those barriers are there because of where we come from, which is like, okay, we used to have the professional who goes to Photoshop and everyone that is maybe doesn't need as much control goes to something like Canva. And then if you're just a random person that is toying around with making stuff, then you use a certain iOS app or whatever. But the reality is that a lot of these barriers used to just be APIs, like a certain plugin or a certain API call or whatever. With Vive coding they are blurring out. Now I actually see consumers who ask me, can you get an API? Because I've coded some things. Then I actually get enterprises being like, your UI is actually one of the best ones. And that's why we are doing these things. I think that this is a time full of uncertainty. And when you have uncertainty, the only thing that can ground you is first principle. So you have to go back to 1K, go back to the beginning, think about what is truly true and what is your marketing fluff.
- [164:15]
A
So the way I'm thinking, do you think images and video are two separate products long term? Like Premiere Pro is a different product than Photoshop. They are different. And even Lightroom is another product in the Adobe ecosystem. I can imagine you can. I mean, it's just pixels. You can clearly put all the tools in one bucket, one app. Adobe didn't. Is that vestigial or will we see another separation?
- [164:43]
I
I think separation will be here in the short term and midterm, because you can argue that as we advance, there's actually even more separation.
- [164:56]
B
Right.
- [164:57]
I
Like at first, books were only in a certain way and now there's a myriad of formats and images. And can you even do, do books for kids where they even have things inside animations, even? So I feel like as we push the medium, and by medium I mean the latent space forward, we will actually realize that we need way more tools than we thought we needed. And that's why we're starting. We've seen this Cambrian explosion of, okay, we have ControlNet, we have ComfyUI, we have Kriya. But then in Krea, you can jump in between tools and then you can use, say, an image model within many tools. And that's the key part, which is you said it's just pixels. And he's like, yeah, but guess what? A song can be represented as pixels. If you have a certain frequency diagram, it's just pixels too. So at the end of the day.
- [165:45]
A
When you scan sheet music, that's pixels.
- [165:48]
H
Yeah.
- [165:48]
I
So what we have at the end of the day is a technology that can understand information in almost its more distilled form. That technology can move around products, but the product will be different. However, they will all use that same technology. And that's what we're seeing. Node based tool, image tool, iOS app, all connect to our image model, for instance, with kria1 or flux context or whatever. But it's different applications.
- [166:12]
A
Yeah, yeah, Fascinating.
- [166:15]
B
We like to hit the gong for guests. How many users does KRIA have? What's the latest? What's the latest?
- [166:20]
A
Give us some stats, give us some numbers. Numbers? What you got big numbers?
- [166:23]
I
Last time I checked sign ups, it was like over 30 million.
- [166:27]
B
Oh.
- [166:29]
A
Congratulations.
- [166:32]
B
That's wild.
- [166:33]
I
Yeah, and, and then, but like, I'M actually more interested. I'm participating in like the enterprise because I'm starting to see true business adoption. And that's where it starts to be like, very auspicious.
- [166:46]
A
Gong ringing. I don't know if you, you can hear it, but the gong is still ringing for you. Something about how I hit it. It's very good.
- [166:53]
B
It's a very good omen.
- [166:55]
A
It was a very good omen for your company. 30 million, you know, we. I just want to take a second because we get lost in like the AI boom or whatever. That 30 million is so many. That is such a huge, that is such a huge impact to have on a product that you've built. So sincerely, like, massive congratulations. That's very, very big. And I feel like, you know, Sam Allman's out there throwing around trillions and stuff and we lose track of the fact that 30 million is a lot of users and congratulations. That's very great. Thank you so much for coming on the show.
- [167:26]
B
Have fun at the event.
- [167:28]
A
Yep, we'll talk to you soon.
- [167:30]
B
Cheers.
- [167:30]
A
Have a good one. Up next, we have Justine Moore from A16Z. This is her second time on the show.
- [167:37]
B
Dylan in the chat felt that gong hit in my spirit.
- [167:40]
A
That gong hit was wild. Justine Moore, can you hear us? We're back on the podium. Hopefully we have audio.
- [167:47]
B
Fantastic. Here we go. Loud and clear.
- [167:49]
G
I'm here. It's great to be back, guys.
- [167:51]
A
Fantastic. What is new from you?
- [167:55]
B
Wait, wait, wait. Open, open screen time. I want to know how much Sora.
- [168:01]
A
This is a good question.
- [168:02]
B
Let's see it. Let's see it.
- [168:05]
A
We know she loves Sora.
- [168:06]
G
This is not going to be super impressive because I mostly use Sora on desktop.
- [168:11]
A
Oh, okay. Okay.
- [168:12]
B
Desktop user.
- [168:14]
G
I would guess, if I had to guess, it's probably like 15 minutes a day.
- [168:18]
A
15 minutes a day. Okay.
- [168:19]
B
In the feed. In the feed or making, making stuff.
- [168:22]
A
Right.
- [168:24]
G
I think I like the phone. Okay. When I do it on the phone, I like to scroll through the feed and see what other people are making. On desktop, there's a lot more controllability with like the storyboard and stuff like that. And you can do landscape as well.
- [168:37]
A
Oh, yeah. Do you feel like you've. You've given Sora the app enough data that you're actually getting fine tuned recommendations or are you just seeing the cream of the crop because there isn't enough to give you specific things that you like? Take the temperature on the feed? You know what I'm talking about, right?
- [168:55]
G
I think it is not yet Enough content to be able to have a very personalized feed for everyone. But I'm not sure because I haven't seen anyone else's. So they might be super different.
- [169:06]
A
Yeah, because like if I go on TikTok and I say I like car videos and you go on TikTok and you say you like boat videos, like pretty quickly we're both going to just get unlimited cars and no boats and vice versa. Interesting. How are you thinking about. We were just talking to Kreia about this. Like winning in the application layer for creativity, like Sora is a creative tool you can create on desktop with the storyboard. There's this world where we wind up with one app that instantiates UI on the fly, lets you do everything. There's another one where we see even more bifurcation. Just like we have Lightroom and Photoshop and Premiere and After Effects, we could wind up with 20 apps for individual little fine tune of face up res over here. How do you think it's going to play out?
- [169:54]
G
Yeah, it's a great question. I think honestly at the beginning of the wave we thought maybe there'd be a little bit more concentration than there has been. The market just keeps getting bigger and bigger and bigger and we keep seeing more specialized apps, I think because these creative AI tools are enabling all of these people who couldn't create content before to make content for the first time. And so that unlocks a bunch of different verticals who weren't using creative tools much before. So my guess is like there'll be some kind of horizontal tooling layers, whether they're for developers, like what the folks at Fall are doing, which is awesome, or for consumers prosumers. And then I think we're also going to start seeing more vertical specific applications, just like we've seen with LLMs. Honestly, like we've seen a lot of for accounting, for law, for finance.
- [170:43]
A
Sure, sure, sure.
- [170:44]
G
Different industries and different Personas have different use cases and workflows.
- [170:50]
A
Yeah, it's interesting. I haven't thought about it that way because when I think of like an AI law firm, I don't think them, I don't think about them as like fine tuning tokens or like running a specific. But it is, it is, it is very much the same. How are you thinking about the mobile versus desktop divide? You're obviously experiencing it on Sora, but are you seeing a bifurcation where you might actually back a company that's saying, oh, we're going after mobile gen AI in this particular niche or we're going desktop and is that a meaningful designation or are companies with vibe coding able to kind of go multi platform on day one? Obviously OpenAI was, but they have a lot of resources.
- [171:35]
G
I think sort of the mobile versus desktop speaks a lot to sort of the user Persona and use case. What we've seen so far is that stuff that blows up on mobile tends to be consumer products where you don't have to put a ton of work into editing the output into having super long prompts. Things like Sora 2 or like Lensa if you guys remember, which blew up a couple of years ago. Now to make these like AI profile pictures of you, I think like true, true creative workflows and really detailed editing tend to require bigger screens. We've actually seen more companies start looking at iPad apps too, which is crazy to say. It's not just like 4 year olds on iPads, it's like real, real creative professionals because people love to kind of draw on them and it. And it's a larger screen that enables kind of more knobs and dials.
- [172:25]
A
Yeah.
- [172:26]
B
What are you seeing on, on the monetization side for pure AI content creators? I know you track a lot of these accounts. I'm interested to see if, if some of these again, like not, not just a content creator that might use AI here or there, but like a new content creator creator that only makes AI content. Will they monetize via, you know, influencer style marketing? Will they go straight to doing deals with, you know, the Netflixes or other streaming platforms? But curious what you think there.
- [172:56]
G
Yeah, it's so early, it's hard to say. I think like how it started was people started growing these sort of AI only accounts and monetizing in the traditional ways, which was like ad revenue or kind of some share of the views you got on your video, things like that on platforms like X and TikTok and Instagram. I've also seen honestly a ton of the ways that AI creators monetize is by selling like prompts or courses or teaching other people how to make the same things just because there's so much demand and most people are not experts at these tools. I think it's going to be really interesting to see like when we start to see more AI native IP or characters that can like appear in a Netflix show or, or have a song that they release on Spotify and they kind of get a cut of the streaming revenue. We've already started to see some of that like that AI actress and we've seen some in music too. Especially in Asia. But I think we're just at the very early stages of that.
- [173:55]
A
Yeah, that makes sense. Are you losing the ability to actually define a particular AI virality moment? Like, the Lensa era was definitely a thing thing. Everyone had the magic avatars. The Studio Ghibli moment was a moment. Then maybe it was like the Sam Altman stealing GPUs era with Sora. But I feel like we're just now in, like, the Sora moment and maybe it just becomes an era. And the Italian brain rot. Like, are we accelerating? Are you feeling acceleration where you can't start? You can't even track all the memes because they just happen so fast.
- [174:32]
G
Yes. I used to keep basically this spreadsheet where I tracked the number of impressions as far as I could guess from certain trends or launches or announcements or things that would happen. And I gave up on it probably like six or so months ago because there were so many things happening at once. It was like nano banana and VO3 and Sora and two new Chinese models that were amazing. And the World Labs launch, there were just 20 things happening every day. And I think the interesting thing about that too is, like, different people are excited by different things. Like, sometimes I talk to people who are really deep in AI creativity, but they're really into one part of it, and so they don't even know about the other launches because it's so hard to keep up with everything right now.
- [175:17]
A
What was the Nano Banana?
- [175:18]
B
I'm using the Meta Vibes app too much. I'm sorry, I don't have time for anything else.
- [175:22]
A
What was the Nano Banana viral moment or driver? What was the. Or the canonical, like, thing that you would prompt that Nano Banana did so well? That made it really popular.
- [175:33]
G
I mean, the interesting thing about Nano Banana was, I think it wasn't just one prompt.
- [175:38]
A
Yeah.
- [175:39]
G
It was a series of them. There was like the Polaroids, putting yourself in a different era and then there was making yourself an action figure. And then one I've been following this week on TikTok is like, college kids are making videos of them, like, hugging their younger selves from an image that they create.
- [175:58]
C
Yeah.
- [175:58]
G
Like with them and their younger self on Nano Bananas.
- [176:01]
A
What do you think of that? I'm not a fan.
- [176:05]
G
I think it's nice. I mean, don't we all want to.
- [176:07]
A
Hug our younger self? No, not at all.
- [176:09]
B
I'd say toughen up, toughen up. Pull up the bootstrap. Now that I'm picking myself up by your bootstrap, I would make an AI. I would make an AI video of my younger me picking myself up by my bootstraps.
- [176:20]
A
There you go. Okay, so, you know we're close there. We will be participating in, like, the next slight iteration of the viral sensation. But no, I won't be hugging myself. That feels very odd to me.
- [176:33]
G
You can do a new one of you yelling at your younger self if you want, or boxing.
- [176:38]
A
But yeah, I mean, a lot of these are odd. Stephen Hawking at the X Games is odd. A cat steamrolling its way through a house is odd. But I love most of the those.
- [176:49]
B
But it's important.
- [176:49]
A
It is important.
- [176:51]
B
Humanity needs these videos.
- [176:53]
A
I think we do. I think we do. I think we do.
- [176:56]
B
Laugh. Laughing is important.
- [176:57]
A
Yeah. Creativity. It is. It is creativity. Whoever prompted it was creative. I would not have thought to put Stephen Hawking at the X Games. And someone did.
- [177:07]
G
Me neither.
- [177:09]
A
What else are people talking about? What are the key, like, stress points? Are people worried about, like, AI bubble generally? Is it all just about, like, let me find the most efficient inference arbitrage or inference provider, the best AI factory. Like, what are the current discussions? I imagine it's not Doom anymore. That one's kind of moved on. What are people talking about at this particular conference?
- [177:33]
G
Yeah, we actually had an investor panel earlier with a bunch of great folks kind of talking about those sorts of things. I think, like, the bubble question is an interesting one. I personally don't think we're in a bubble. I think, like, just the true demand for the products is massive and we're seeing companies grow faster than ever before and also retain users higher than ever before. Yeah, a lot of. A lot of folks, like, I'm sure you guys do have questions about kind of gross margin. I think, like, we find that margins tend to improve over time, as you've seen in many eras of consumer software like Amazon and Uber and Doordash. Like, a lot of these companies were gross margin, negative or neutral at the beginning and then kind of become more efficient over time as they scale and unlock various network effects. I think where I would be worried we were in a bubble is if the margins were all negative and the retention sucked, because then it's like you're not actually building products that people want and you're raising a lot of money for it.
- [178:33]
B
Or if there was only like 100,000 real users or something like that.
- [178:37]
G
Totally, exactly.
- [178:38]
B
Which we've seen before in different tech trends.
- [178:41]
A
Yeah, no, totally, totally. I mean, yeah, we just talked to Korea. Like, 30 million users is, like, very significant. And ChatGPT, close to a billion users. It's just accelerating much faster. And the distribution is there. And it's a lot because stripe exists. You can just actually get 20 bucks a month out of someone pretty easily. And you couldn't do that in 1999. It was just no way. And so you had to say, well, we'll monetize the eyeballs later and we'll do ads and people will pay. And it became this kind of weird circular economy. But now you can just be like, how much value am I delivering to your business? Put down your card and we'll pay for it. Or churn, which is great. Well, thank you so much for taking the time to stop by during a busy conference.
- [179:26]
B
Second ever podium guest podium time, chatting with you.
- [179:29]
G
I'm honored. Thanks, guys.
- [179:30]
B
Yes, we'd love to have you fun at the event and give our best to everybody there.
- [179:33]
A
Yeah. And just so you know, we're coming for your lunch. We dropped our first market map. I don't know if you saw that, but we dropped a media market map. So now we're in direct market.
- [179:45]
B
It's not schizo at all.
- [179:46]
A
It's not schizo at all.
- [179:47]
G
You know, the competition fuels me, so I'm excited for this.
- [179:50]
B
Yeah, fire back. Fireback response with your own market map. This is like a rap battle for people.
- [179:57]
A
It's a map battle. It's a map.
- [179:59]
G
A rap battle that like five people are interested in.
- [180:01]
A
Yes, a rap battle for people who know how to read a term sheet or something. Anyway, thank you so much for stopping by. We will talk to you soon.
- [180:09]
H
Have a great rest.
- [180:10]
G
Thanks, guys.
- [180:10]
A
Goodbye.
- [180:11]
B
Gotta go over to this post from Marvin Von Hagen, which is one of the most royal names ever. This. This man is destined for. For greatness. He quoted the essay that I published yesterday, and just with two excerpts from my piece, he said, the Interaction Company of California has managed to really break through the noise and deliver a truly novel consumer AI product experience. Nothing important. Don't worry about it. Poke.com itself is a fantastic domain and a name for their business and audience skipped over, of course, the.
- [180:49]
A
Oh, there's like a little bit of criticism in there.
- [180:52]
B
No. Yeah, I was like. I was criticizing the general sort of like name intervention, but this was the absolutely perfect response. And I really want to have Marvin on the show. So, Marvin, let's get you on the. On the schedule next week.
- [181:03]
A
Yeah, yeah. This is the nature of.
- [181:04]
B
This was like working with Chef's Kiss. The perfect response. What's going on here? Austin says you should never use an EM dash in your writing anymore. You will be 100% be accused of using AI, especially if it's on social.
- [181:19]
A
That is true. I actually recently, like today in the newsletter, I used just a minus sign instead of an EM dash. And Brandon was like, certainly we can use the EM dash here. You intended to use an EM dash. And I was like, no, we're using the minus sign. We're using the wrong thing because I don't want to deal with it's EM dash. You used AI or whatever. So I have completely sacrificed it. But what is Alyssa Pumpkin Queen?
- [181:46]
B
She says that's what the dropping the EM dash. That's what the Clankers want you to do. Surrender the EM dash. But by the grace of God, we will outmaneuver the Clankers, sink our hands into their motherboards and remind them what mankind can do. Let the Clanker hold their position. We won't.
- [182:04]
A
We won't.
- [182:06]
B
Oh, so good. By the way, the chat Chat wants us to to do a conference. And Ben sand says, I'd go to the conference, but we needed a TVPN cruise. I'm not joking. We've talked about getting a yacht to park in the San Francisco Bay and do a conference there. So I would like to pull that off at some point and the chat will be the first to know. Ohio has a new bill that goes.
- [182:35]
A
Far beyond why did you blank her marriage?
- [182:38]
B
Far beyond banning human AI Marriage. The measure would define AI systems as non sentient entities, strip them of any claim to legal personhood, forbid them from owning property, holding financial accounts, controlling IP or acting as company directors.
- [182:52]
A
So it's basically game on. If we get a humanoid. It has no rights in Ohio. So if we take it to Ohio, the unit tree is going to be sitting there being like, hey guys, why are we going to Ohio? I'd love to stay in a different state because I would hate for you to destroy me and then face no legal consequences because I have no rights in Ohio.
- [183:16]
B
Yeah, this Ohio clearly hasn't read less wrong.
- [183:24]
A
It's not familiar with Rocco's Basilisk. Ohio might be the first place that the Clankers descend on and. And Skynet goes on.
- [183:32]
B
Yeah, they really put themselves on the crosshair, the clanker crosshair on the aggress, on the aggressive. Daniel Tenrero has read this.
- [183:40]
A
Let me tell you about wander.com, find your happy place. Book a wanderer with inspiring views, hotel, great amenities, dreamy beds, top tier cleaning, 24.7concierge service. It's a vacation home, but better. What did Daniel say?
- [183:48]
B
He says finally some good news, which is that hedge fund assets hit 5 trillion with most inflows since before the financial crisis. Great hit, John. Assets in the global hedge fund industry have surged to a record 5 trillion as investors poured money into alternatives and funds posted solid gains.
- [184:07]
A
Our gong is gifted today. You hear this right?
- [184:11]
B
It's still going. It's still going. Still going to Saw net inflows of nearly 34 billion in the first three months through September, with total returns across all strategies averaging 5.4% over the quarter. So underperforming the average terminally online zoomer that is just massively levered into the Mag 7, but still not so bad as a whole. Cryptocurrency hedge funds posted solid double digit gains in the third quarter, recovering from sharp losses in early 2025 to bring year to date performance to 6.7%.
- [184:45]
A
Can we play this clip of it's probably too long to react to Buco.
- [184:49]
B
Capital the legend says I'm always so confused when Tesla goes down on earnings as the business performance is completely irrelevant to the price of the stock. They put up over 1.4 billion. I think of free cash flow as a trillion dollar company. So that's good.
- [185:06]
A
They beat earnings.
- [185:07]
B
Yes.
- [185:08]
A
The stock went down. Is that what they mean?
- [185:09]
B
No, no, no, no.
- [185:10]
A
I think they, they missed earnings and the stock went up.
- [185:13]
B
I don't know.
- [185:14]
A
I agree.
- [185:15]
B
No, he's saying they missed, but he's in the stock traded down and he's like why is it trading?
- [185:19]
A
It shouldn't even matter. It shouldn't even matter. Let's skip the Dorsey reaction. We got to get out of here. It's Friday. We got to start our weekend. Apparently the sale of TikTok was a pawn sacrifice to open up the diplomatic chessboard. And Rune says it's pretty bad that Xi Jinping models TikTok as spiritual opium and feels okay selling it to the west in a revenge opium war setup. For Trump, facing the sale of the high profile app was a populist victory framed around national security. For Xi Jinping, however, the app was a low cost bargaining chip. He had privately dismissed it as spiritual opium, according to the people close to Beijing, and viewed it as easy sacrifice to secure the continued dialogue on China that China needed. That makes a ton of sense.
- [186:15]
B
In more important geopolitical news, Justin Bieber is on Twitch. Michael siebel says, almost 19 years after we got started, we finally got Bieber. So never give up. And this was, I think My favorite post of the week. It's now been deleted. I won't read the user's name, but I. I think it's funny. It was a post that said, I honestly don't care if I get fired. I was part of the illegal gambling ring with Chauncey Billups and Terry Rozier. My name is Andy Jassy and I work at Amazon.
- [186:50]
A
These fake framing someone who did something is so funny.
- [186:54]
B
Shareholders are.
- [186:56]
A
I didn't realize that there was a retail army for Andy Jassy that cared about Andy Jassy and Amazon.
- [187:03]
B
Yeah, it's tough when your Stock goes up 20% year to date and people are still like calling for you to be fired.
- [187:12]
A
But I think that's a good place to call it. We've had a fantastic show.
- [187:16]
B
You know where we're going to call it?
- [187:18]
A
Where are we going to call it?
- [187:19]
B
We're going to call it with Gabe over at Sora. Sora was dethroned in the App Store by Dave's Hot Chicken.
- [187:28]
A
Play the Wentworth.
- [187:31]
B
And an eagle sound effect for Dave because something their team, I think is just really cracked because I don't think this is the first time they've like charted number one.
- [187:42]
A
I think most people think that the Dave's Hot Chicken team is like one notch above the OpenAI team in terms.
- [187:48]
B
Of just, you know, product design growth.
- [187:50]
A
Yeah, just IQ and strategy.
- [187:54]
B
AI research.
- [187:55]
A
Research. Yeah, these types of things. I think most people think it's sort of like, you know, a hotbed of the best talent in the world.
- [188:03]
B
Absolutely.
- [188:04]
A
I'm unsurprised. No, seriously, the Dave's Hot Chicken team must be cooking because something crazy happened in their marketing because to go to the top of the charts, even if it's momentum based, that means that they had some campaign that just hit and it's not like it's the super bowl right now. Like, they must have had some crazy call to action to get that Apple to the top of the store. But congratulations to everyone at Dave's Hot Chicken. I'm a huge fan. I order Dave's Hot Chicken all the time. Maybe I should. The app I usually use, Door Jesse.
- [188:31]
B
Never ordered it to the Ultradome.
- [188:32]
A
No, I haven't.
- [188:33]
B
But more of a weekend thing.
- [188:34]
A
More of a weekend thing. More weekend thing. But Dave's Hot Chicken is awesome. Well, highly recommend.
- [188:39]
B
It is almost the weekend weekend for our east coast audience. If you're on the west coast, we're.
- [188:46]
A
Gonna continue to lock in.
- [188:47]
B
Please stay locked in at least until 5. Hopefully you're not leaving the office at 5 because it just looks bad. It's like you just were hanging out.
- [188:56]
A
Yeah.
- [188:57]
B
Trying to. Trying to get the optics in. But we had a really fun week. Next week is going to be wild as well. I think it's going to be a magnificent week.
- [189:05]
A
I think so, too.
- [189:06]
B
I think it's going to be absolutely magnificent.
- [189:07]
A
I think it will be magnificent.
- [189:08]
B
And I can't wait to be back. So have a wonderful Friday. Have a wonderful Saturday. Have a wonderful, wonderful Sunday. We'll be counting down the hours until we are going live again on Monday. And thank you for tuning in.
- [189:19]
A
Goodbye.
- [189:20]
B
Cheers.