X Article Apocalypse, Hollywood's AI Takes, Is the American Dream Obsolete? | Jordan Schneider, Eliot Pence, Fil Aronshtein & Matt Grimm

You're watching DVPN.

Today's Monday, January 19, 2026. We are live from the TVP and Ultra Dome, the temple of technology, the.

Fortress of finance, the capital of capital.

Ramp time is money save. Both easy to use, corporate cards, bill pay, accounting, and a whole lot more all in one place. Go to ramp.com we have a great show for you today, folks. We're talking about the article apocalypse that's happening on X. Will it swap the timeline? The million dollar slop fight? Will it slop up the timeline? Will the algorithm handle the flood of AI generated content that will be slopped up in the timeline? We will see, we will dig into it, we will discuss it. We'll also try and understand what is going on with Nikita Beer strategy. What's the strategy at X? Will this growth hack work? This competition? Of course. This all started with a Nikita beer.

Hold up. We got Main Street Summit in the X chat. Hey, what's up Main Street Summit? Shout out to Brent and the whole team. Love you guys.

Cheers.

Cheers.

So this all started with Nikita beer on X. He said, ladies and gentlemen, we are giving $1 million to the top article posted on X. You have two weeks. It's time to write. And very, very different structure.

I thought there's already. There's so much AI generated essays on X. Why not make AI generated essays? Lottery tickets.

Lottery tickets.

Let's turn them into lottery tickets.

I like it. I mean, they're definitely paying out more than a million dollars in creator payments across all the little $100 thousand dollar checks that go out to folks. You have to imagine. But running a sweepstakes, I mean, that's time honored. What's not to like about a sweepstakes? It's great. I think it's a lot of fun. Obviously people are worried that this will create some, you know, perverse incentive. There was already a perverse incentive for slop on the timeline because if you engagement farmed, you could potentially get a big creator payout. We've seen some people that get, you know, $20,000 randomly sent to them just because they went mega viral. We've seen a bunch of people that have just built up accounts where they're getting a couple thousand dollars every month. There's always been an incentive and that incentive exists on YouTube where you can get creator payouts. You can. There are plenty of platforms where you can go and get attention and convert that into ad dollars or rev share. I mean, back in the day there was a huge incentive to like do listicles and clickbait to get visitors to your website that you would put Google Ads on. Some of it made a decent amount of money, some of it didn't stick around. Most of the bad stuff eventually went away. Eventually people stopped showing up. Eventually the audience quality got so bad that all the advertisers said, why am I paying to show up on this listicle of like the cutest dogs in America or something? I don't know, maybe that one actually rips. Who knows? But. So Nikita's kind of taking two sides to this. He says it's time to write, but also he tells everyone else, please bear with article Armageddon. My question was, what's the strategy here and what will the outcome be? So before we dive into it, let me tell you about Vanta Automate Compliance and Security. Vanta is the AI trust management platform, the leading AI trust management platform. So it's sort of crazy to me how long Twitter went with. 140 characters.

Yeah.

Do you realize Twitter launched in 2006? They held the line on just 140 characters for over a decade. It wasn't until 2017 that they doubled it to 280, which now feels still a little tight to get a Tweet in under 280. But 140 was truly a constraint. You needed to be extremely terse and extremely concise. And even that was a relic from the SMS era. Like the default text message that you could send was 140 characters. And so Twitter was very much built on, hey, if you verify your phone number, you can just send a text in and it will post it on your Twitter account. And so a lot of people in the early days were using it that way, although obviously people were also using it on web. And it's always been an important watering hole for tech power players. The tech elite have always liked hanging out on Twitter, but it's never captured the value that it goes created. Like, it was very, very hard because people are not in the. They're not in the same frame of mind as when they're scrolling Instagram. Scrolling Instagram reels. You're basically shopping when you're on a meta platform and then you see an ad for.

Yeah, and to be clear, meta captures a lot more value, but certainly not all the value it creates. Right, that's the point of a platform.

Yeah, exactly.

More value than you capture. But in the case of Axe, it's like, you know, they were capturing a very small, very, very, very tiny percentage.

And there were lots of stories about people building businesses that Basically exported audiences from Twitt or they went viral there. Ben Thompson talks about LinkedIn actually being a big early growth vector for him because links did very well. They would convert to his newsletter, sirtechery.com, but people have done this all over the place where they've been consistently viral with threads, build up an audience and then they take it somewhere else to monetize it, either with subscriptions or ads. But they don't really do that commercial activity on the platform. And so there's always been a question of like, could X or, or Twitter capture more value? And a lot of that big turning point happened when Elon made the acquisition. So the Elon taking over felt like, okay, it's an entirely new day, everything's moving way faster now. Some of those projects had actually been in the works for months or years before. I think Community Notes was the big one, where the Twitter team had been building that infrastructure for a while, but they were moving forward.

They had 5,000 people working on it.

For over five, hopefully something like that. But so Elon, it's not like he kicked off a lot of the, that many of these projects, but he definitely like greenlit them, shut things down, just sped things up, moved more.

It's a speed thing.

Exactly. Let me tell you about Okta. Okta helps you assign every AI agent a trusted identity. So you get the power of AI without the risk. Secure every agent, secure any agent, secure our wide angle camera, because the Ultra Dome is secure, secured by Okta. So 2023 was a big year for expanding character limits. They'd also previously acquired this Dutch startup, Revu, which was a direct competitor, Substack, but that product was shut down in less than two years, like January to January. They only made it two years, never really got it off the ground. And I think it was because it was sort of counter positioned with Twitter. It was kind of like always, oh well, if we get people on the review platform, they'll leave or they'll take their audience elsewhere. It never really felt, yeah, and I.

Don'T know that a lot of people, some incredible faith that, hey, I want to build my business on a Twitter.

I mean even pre Elon, there had been companies, startups that had built on the Twitter API, which had gone through a number of version changes. Elon made some more extreme changes to that. But there were plenty of times when people would build businesses too closely to any platform really. And then they get steamrolled, whether you're Zynga, building FarmVille on Facebook and then they Changed the algorithm or they changed the integration and all of a sudden your growth engine's just go. This happened with Teespring. They were selling custom T shirts. So they would go and figure out a specific profile, like people who run podcasts with the last name Hayes. And then they would like print a custom. They would generate an image of a.

Shirt and it'd be like, yes, my name is, my last name is Hayes. Yes, I'm a podcaster. And I'd be like, I gotta have it, I gotta buy it, I gotta.

Have to have it. It was like, who is buying this cringy, cringy nonsense? But it really sold a ton of shirts. And teespring was on like an absolute, absolute, like meteoric rise. And then they changed the algorithm and they stopped a lot of those advertising because they felt creepy. Because it was like, wait a minute, like, why am I seeing an ad with my last name and my job title in the image? Like it's too much. And so Facebook put a kibosh on that and the rest is history. But they've been working on making X a more literate, long form platform for a while.

2023, we need to teach our users the English language. We need them to read more than 180 charact.

So they've obviously done a ton in video. They've never really leaned into a video library like YouTube. Like I used to cross post my YouTube videos on X. Sometimes there were length limits for a while, then they got rid of those. So I could just actually copy a 25 minute YouTube video, put it on X and it was cool. People wouldn't watch nearly as much as they did on YouTube and there was no catalog value. That was like. The really weird thing was that on YouTube, like my old videos that I haven't uploaded in over a year and my old videos are still probably getting like, I don't know, half a million views a month. Just because people are, every day they search like, what's the history of Android? Or what's the history of Ramp or what's the history of Xi Jinping?

Well, they're ever content, so just continuously get served.

Exactly. And then once someone watches one of my videos, they watch five more.

Well, part of that, if you want to be a serious video platform, you need to be like, have, you know, actually incentivizing people to watch on TV. Like so much of YouTube watch hours happen on TV.

Yeah.

And X, I think has a streaming TV app, but it's certainly not. Doesn't get a lot of love.

Yeah. I actually made a whole documentary about El Segundo, went around with Augustus and toured everything and it just didn't really work for my YouTube channel. So I only put it on X. It got a ton of views and actually the response was probably better than anything that I would do on YouTube because it was very tech insider. A lot of VCs were getting up to speed on what was going on in El Segundo. We're having some of the El Segundo folks on the show later today. Well, he's not truly Elsa Gundo, but Phil is like spiritually Gundo. We know that. And Anduril in many ways is spiritually Gundo. Let's take you through the Linear lineup. Who we have on the show today. We got Jordan Schneider from Chinatalk coming on at 12:30. Then Dan from T1 Energy, Elliot from Dominion Dynamics. We're in hard tech land kind of all day. I'm very excited for this. And then we have Matt Grimm, the co founder of Anduril and Phil from Dirac coming in person at 1:30 together to break it all down for us on what's going on in building hardware, building defense.

And remember, T1 Energy is a company that we talked about in Q4 of last year. It was a unique situation. I believe it was a Chinese company, solar manufacturing plant that was forced to sell and now it has a cool brand and new management and the stock is absolutely ripping.

Very excited to have them on the show.

It is up 433%.

Yeah, I love that we get to.

In the last six months just find.

Some random company and then all of a sudden we're talking to the CEO. Anyway, of course if you've forgotten Linear, it's the system for modern software development. 70% of enterprise workspaces on Linear are using agents and you should be too. So continuing with the long post journey that X has been on, 2023 was a massive year. February took it from 280 characters to 4,000 characters. That's more than I need for most of my posts. I don't think I've ever posted anything longer than that. April expanded it to 10,000 characters and finally in June they maxed it out at 25,000 characters. That's longer than most blog posts. That's around 4,000 words, maybe a little bit more. And you can still thread these posts together. So I was always wondering like why does articles exist if long posts are so long that you can put 4,000 words there? In fact I took, you know that one super viral article that everyone, they're all joking that like, oh, he should have waited a couple days to post it because then he would have won the million dollars.

Oh yeah, yeah, yeah.

It's like how to change your life in one day.

And wasn't that effectively inspired? Like my view was like that post, they were like, we should do more of this. Let's incentivize a lot more of this.

Totally. Yeah, yeah. It seemed like Ryu, Nikita saw that and was like, okay, articles are creating satisfaction on the platform. People like them. It does seem like the algorithm is. It seemed like at one point the algorithm was maybe misaligned. Not that it was punishing articles, just that it wasn't really processing the time that people would spend in articles and the way they would engage. Because if there's a viral post, people might be more inclined to like it than a long, long essay that you read 45, 75% of and then you get out of it. You actually spent multiple minutes with it, you really enjoyed it, but you forget to like it because that's way down at the bottom. Or you forget to retweet it or whatever. Or you're not commenting on it, you're just kind of enjoying it reading. And then you close your phone because you're like, oh, wow, I just spent 20 minutes reading this thing. I'm moving on with my life. Or I got a phone call and so they needed to dial the algorithm in. And I think that's what's happened. And now I think that essentially articles are first class citizens in the algorithm. And there's really no downside to posting articles. But that post, that article, how to Change youe life in one day or something is like 30,000 characters. So it almost would have just fit if you just like edited a little bit more. It would have just fit into a long post. But it worked as an article. There's something about the title and thumbnail that really works there. Maybe, maybe X should do a deal with our partner 11Labs and let you listen to these articles. 11Labs, of course, build intelligent real time conversational agents. Reimagine human technology interaction with 11 labs.

Zach Vole had a great idea. He said, grok, summarize this article in 280 characters or less. Don't make any mistakes. Thanks. So if you really yearn for the days the pre article era, just respond and tag grog in a post and have it turn it back into just.

Regular 140 characters, even shorter. I mean I think somebody did that with how to change your Life in one day because a lot of it was like from Atomic Habits I guess, which is a very popular book. But all this like lifestyle stuff is always recomposed from various previous prior art. It's all just reformatting essentially with some new flavor and new spice. So when articles launched in March of 2024 it didn't really hit for me like I was not like oh I got to jump on this, I got to post. You know. We started doing the newsletter in 2025. Mid year we started cross posting articles were available. I think we might have tested one or two as an article. Didn't really see a difference, so just kind of stuck with the long posts. But last year, so my pieces are usually around 5,000 characters, they fit fine in a long post. I was doing long posts until I saw Lulu Maservi, friend of the show posted standing out in 2026 as an X article. Now she's been viral before for Going Direct, the Going Direct manifesto which was a thread of long posts. And I like, I think it was maybe just a really long post or maybe a thread of long posts, but I really like that format and I would actually use that format as sort of a template for if I wanted to communicate something in a few paragraphs. I would sort of look at what she did with the Going Direct post and use that as like the backbone. Obviously not taking specific words, more just like okay, you want an intro paragraph, an opening paragraph, a couple bullet points, a summary, a takeaway. And what's really interesting is that her standing out in 2026, it's an X article, it has a header image of these robots and it grabs you so you click on it and it was dropped January 5th. Good timing. Kick off the year it went super viral. Millions of views, thousands of likes, but it's really really short. It's only 443 words, it's under 3000 characters, so it easily would have fit in a long post. But she went with article. It worked. And when I saw that I was thinking okay, maybe we should do. Maybe we should do articles for our pieces. And I've been doing that and it's been fine. Hasn't been honestly the metrics are exactly the same so maybe I need to work on the titles and thumbnails or something. But it's not like articles have been some magical cure all or oh yeah, we're getting 10x more views now that we're using articles. It's kind of the same thing, but it does seem like an interesting format that Maybe they will surface other people.

It's a better experience as a reader maybe.

I think so. It's hard to tell. I mean you can definitely, if you take advantage of the formatting maybe and you're doing bullet points and bolding and italics, you can write embedding posts. Yeah, embedding posts is cool. And also images in the. I saw one that had like graphics and it read almost like a strathecary piece. It had embedded images and stuff. It was cool.

Gabe in the chat had an interesting idea. He said above X should acquire Beehive, which would be kind of interesting because like Beehive, one of the issues with the review acquisition is that I didn't know anybody that was building a business on Revu.

Yeah, that's true. Twitter had done a couple of these early stage acquisitions. Like I mean vine was very early pre monetization. They did I think Periscope, the live streaming video platform that was maybe pre revenue, like pretty early. We talked to one founder who sold his company to Twitter before they even launched. And so the previous Twitter management was very quick on these like smaller acquisitions of like pre launch but venture funded. The team's been building for six months. They go and show the Twitter team and the Twitter team says hey, how about we take it for 100 mil or something like that. And it's a great outcome for the founders. And then years later they gripe that they should never sell the company. Like, like the Vine.

Yeah, that that era I feel like taught San Francisco that like some of these smaller scale acquisitions were just relatively straightforward, commonplace. And then in the 2000s people realized that era just kind of went away.

Yeah, it either turned into like these mega aqua.

Apparently Twitter had done approximately 79 individual acquisitions.

Wow, that's a lot.

They did Telepart, which was digital advertising was half a billion. Magic Pony 2016 for around 150 million. Nip Social Media Data Aggregation, 134 million. Tap Commerce, which mobile ad retargeting for 100 million. Periscope, if you remember that vine tweetdeck. So bunch of different deals.

Yeah. Before we move on, let me tell you about Gemini 3 Pro. It's Google's most intelligent model yet. State of the art reasoning, next level vibe coding and deep multimodal understanding. So strategically this big article push it feels like another shot at Substack, Chris Best's company. I didn't put substack in my article because I'm up on the algo speak unalived. I'm not afraid of the Real word, I guess. No. I am terrified of the algo and being punished and shadow banned. So I said, Chris Best's company, this is best practice if you want to talk about Substack. But I don't know that anyone's going to be rushing to get off of Substack based on this. Because the value of Substack, the value prop is that you own your email list so you have a direct relationship with your, your audience and you could take that anywhere. You could leave at any time. You can even take your Stripe account with you, which is a crazy thing. It's so creator friendly. And Substacks made that their brand and they built a great business around it. And long term, I think Substack should probably figure out how to build some sort of ads product that can run across all of the different substacks because.

That would probably be a total TVPN victory. It would, it would very, very anti.

Anti ads. Anti ads. But I just see like what does it take to get Substack into the Snapchat Pinterest Reddit category? Sort of like DACA Corn category. And you imagine that for a lot of these, like if you're they're clearly trying to push people into the Substack app, but even just embedding ads here and there, it could be done tastefully, I think it could be done in a way that's not too abrasive and you're getting a lot for a little. Now, of course, like we use Substack, we don't charge for our newsletter. Tvpn.com, you should go sign up. So if they put ads on ours, we'd be like, ah, that's sort of a downgrade. We want to put ads in ours. But I think just from a business perspective it would make sense and really open them up to some big, big dollars at a certain point because there's a very high value audience across the.

You had kind of a rough theory that maybe there was some logic to doing this just for having new training data.

Yes.

So there's this idea that somebody untitled YouTube chat says the data is worth 10x more than 1.

There were a few people that were mentioning this. Before we dig into that, let me tell you about Plaid. Plaid powers the apps you use to spend, save, borrow and invest securely. Connecting bank accounts to move money, fight fraud and improve lending.

Now with AI, talk about a company that feels very immune to vibe coding. You don't want the. If somebody comes to you and they're like, I got plaid, but I'm gonna charge you half as much. Yeah, it's like, well, I'll take the, I'll take the real one for storing and processing my customers financial information.

Yeah.

Anyway, Noah in the chat says, I don't want to read substack without ads. So that's.

There we go. Thank you.

That's advertising. Strongest soldier.

But so the question is like, you put out this million dollar bounty, how many people will show up? How many articles will actually get posted? I am seeing a lot of articles in the, in the feed. It's also hilarious when you see like a Fortune 500 CEO post an article because you're like, oh, damn, are you really down? And you need a mill, like, what's going on, buddy? And it's like, no, they're just like, they're just posting news and they just need to get something out so they use an article. But it's funny to imagine that they're like, oh man, the million would change everything for me. It's what I need. It's what I need.

But Tim Cook starts ripping articles. He's just like, he's like, I need to. I can't be like within the same.

We're not going to do a keynote this year for the new iPhone. We're dropping an article and we're releasing it nine months early. No, the question is, is the data valuable? So a million dollars, how many people will show up? How many people will play the game? How many people will buy a lottery ticket by writing original articles that go viral, generating articles. There's already a lot of allegations of AI slop in the articles feed. Every time I see an article, somebody tags that AI detector, Pangrim Labs.

I mean, I've been seeing so many people quoting articles, sharing them, and say, you need to read this. It's incredibly well written. And then I tap in and it's just obviously slop.

Some people love slop. I mean, we're gonna get into this with the Hollywood reaction to can we.

Get a pig sound effect. Thank you, Scott.

To the latest Joe Rogan episode. But some people just. There is more and more content on the Internet where I see it me, and you can clock it as AI, but you read in the comments and everyone loves it and they're happy. This is what Sam Altman told us. Like, one person's slop is another person's treasure. Like, some people like this stuff. Also there's the interesting counter dynamic, which is I don't use AI to write. I don't even put the final draft through An LLM to like reality check it or do anything. I just sort of hammer on the keyboard and type and if there's typos or weird phrases, that's my style. You're just hearing it from my brain. It's like a brain dump. But that doesn't stop people from commenting on all my art, on all my pieces. Like this is AI or did you use AI or how do you write so much? Blah, blah, blah. So there's a little bit of inevitability.

Honestly, we should just live stream you writing.

That would be good every morning. Yeah, you could see me lock in with the actual doc right here. I should livestream the. Or at least I should put a time lapse out of my screen while I'm doing this. I mean obviously I use LLMs to research like for, like for this article. I needed all of the dates of different changes to X's and Twitter's like character limits. That's obviously something I'm going to go to an LLM for, do the research, pull it together, find the sources for me. It's super powered Google search. It's great. But the question is strategically it doesn't seem like anyone's going to be bailing on Substack today to move over, but there is just valuable. And then there's also the question of like. So you're probably not going to get your own audience list on X, but will you even be able to monetize with subscriptions? X does have subscriptions, but Substack has a way more like built out.

I wonder what the. I wonder what the monthly revenue is for the person that's utilizing subscriptions the best, right? He's the highest earning subscription driven creator.

I have a couple people that subscribe to me and pay me just because they're like, cool, we like you. And I've never posted just to subscribers, but they're just like, they're just like, shut up, keep doing what you're doing.

Riding with Coogan.

I think it's public, so I can say like Gary Tan has just been ride or die for like years just sending me five bucks. I'm like, thanks Gary.

This is not an ad to go and do that by the way.

No, no, no, no. You will get nothing. But the interesting thing is that Substack offers more than just a one time subscription monthly. You can pay annually, get a discount. You can also do like VIP tiers, different tiers, Access to discords, access to different things, participate in the live stream, get the chat and then there's also like business plans, group plans. So they just have a way more mature product catalog when it comes to monetizing a customer list or an email list. And then obviously you can take them elsewhere. So if you ever don't, if you want to, if you're like, look, my newsletter product has grown so much that I have like enterprise level needs. I have to be off substack. You can go and build something from scratch that does exactly what you want and has all the different integrations. And you know, some companies have done that. Other companies, I mean, Free Press has scaled immensely with substack. And you go to the Free Press's homepage and it's like it looks like the New York Times. Like there's like 25 different articles popping up little subsections. Like it is a full featured CMS for like a newspaper, for like a modern newspaper. Anyway, MongoDB choose a database built for flexibility and scale with best in class embedding models and re rankers. MongoDB has what you need to build. What's next? Let's Mongo.

So how did you conclude. So what was your takeaway?

My question was, so I don't see anyone who's a writer online who's on substack being like, okay, now's the moment. A million dollars on the table. I'm shutting down my substack and I'm going over to X exclusively. But a lot of people cross post. That's what we do. I think a lot of people, if you have a free. If you have a substack that's part free, part paid post, the free part on X, why not? It's all upside. It's just more attention that can actually get you into funnel. I believe that the posting of the articles on X has been additive to our substack list because at the end of the post we follow up and say, hey, this post is emailed as well. You want to go check it out, go to tppn.com and so we. I think this has been net growth. I don't think people have been like, oh, I'm unsubscribing from TBPN's newsletter because I'm getting it on John's random Twitter feed that may or may not surface to me in the algorithm, but I don't think so. I don't think X needs writers to go all in. I just think that they want more copies of the best writing to make their way to X. And so if you have a great YouTube video, they want you to Share it on X as well. So then they can surface that to people. If you have a great piece of writing, put it on X as well. And I think this does that and I think they'll be successful there and I think they can.

Yeah, the question, I mean if you're, if they're really successful and you have like YouTube for long form writing, like the UGC platform for long form writing, it's probably, that's probably a solid business. But it's not gonna be a juggernaut like YouTube specifically because it's not gonna monetize nearly as well. Right. You have like they don't put ads in articles today.

It's such an.

They could add them in but you're like locked in reading an article versus like when you're watching something totally. Like if somebody's watching football and yeah.

You can put it. You can put a two hour YouTube video on your smart TV and it will just play an ad and you just won't. You don't need to do anything. Whereas if you're scrolling an article and then all of a sudden an ad pops up like you, that breaks the flow of like what you're doing. It's not a passive experience, it's an active experience. So yeah, I completely agree with you. I want your take on whether on the actual article experience. First I'm going to tell you about Figma. Figma make isn't your average vibe coding tool. It lives in Figma so outputs look good, feel real and stay connected to how teams build create code back prototypes and apps fast. So I was chatting with some folks who have really, really big on Twitter, a little group of folks who are all well over 100k. I've gotten there back in like the thread era they were like thread guys basically and they were saying like I don't think Naval's how to get rich without getting lucky thread would have worked as an article that the fact that it was all these bite size pieces like that's. That was what was special. Now I think personally that meta has sort of died and we haven't seen that work.

But what do you think coming back and just unironically ripping threads?

Why not? I'd zig.

You miss threads.

You mean just like do you miss.

That error at all? Is it. Can you remember any iconic threads other than Lulu's and Naval's and apparently Lulu.

Was just a long post someone corrected in the chat. So no, Noval is the main one that I remember. I probably scrolled a few threads. I actually did have a thread that actually blew up my account. So I was at 10,000 followers.

Is this the Alaska one?

No, that was not. That got me unfollows because I posted about bringing tech. I was early on the let's leave California train and I said, we gotta go to Alaska.

Alaska.

We're not going to Miami, we're not going to Texas. We're going to Alaska. Because if you have runaway global warming, it's going to be beachfront property up there, baby. It's gonna be 72 and sunny every.

Day in Alaska palm trees.

Alternatively, if we solve global warming, we will have unlimited nuclear power. You will be able to heat air condition, do whatever you need in your Alaska compound because you will have a 1 megawatt nuclear reactor dropped on your doorstep and you will have unlimited power and you'll be lounging in a beautiful scenic mountain retreat, perfectly temperature controlled.

Alaska did not like that.

They did not like that. I got a lot of DMs from a lot of people showing me their guns, saying, if you come up here, California boy, there will be consequences. And it was a rough day on the Internet. I had to lock my account. It was very funny. But anyway, I survived that. But the post that blew up my account was I had written a. So Chris Miller had kind of popularized the idea of TSMC and Taiwan as an important geopolitical football in the U.S. china battle. And I'd read the book, really enjoyed it. I had done a YouTube video all about US China competition that had done very well on YouTube. So basically what I did was I recontextualized everything that I'd learned and pieced together for that YouTube video into a thread. And I gave it a very salacious title. I said, because FTX was collapsing at the time, something else was going on. Crazy. There were like two major current things happening.

Launching.

I don't know if it's ChatGPT launching, but basically I was like, FTX is a distraction. Elon Musk is a distraction. Like you need to be focusing on US China policy or whatever. And I just sort of like contextualized why Taiwan is important for the semiconductor supply chain in the AI boom. And it wasn't anything revolutionary for anyone who was paying attention. But there were a lot of people who just hadn't really learned what ASML was or TSMC was at the time. And so just having that in a nice digestible feed, it went super viral. And because it was so like authoritative, it took me from 10,000 followers to 20,000 followers in like a week. And it was. It was like, really, really gross. Really, really. It's like the clearest jump in the follower graph. And then everything else has just been, like, slow growth. And we know this from posting, like, you post like a viral funny banger. It doesn't, like, double your account, but if you posted a really, really great thread, people would follow you. And I think that's true for the articles, too. The guy who wrote the how to change your life in one day. He has almost a million followers now. And I imagine that he had way less than that because I'd never seen his account before.

Yeah, because he's providing, like, insight.

Yes.

People could critique the article or say it's. A lot of these ideas have been talked about before, but for better or worse, he summarized some key insights. He shared it on how to improve your life. That is value. That somebody is like, I want to keep improving my life, I'll follow. Or they didn't read the article. They just saved it and followed him. Because they don't have time to improve their life. I'm busy right now.

We didn't say which day they were going to use. It could be any day.

So they'll circle back on that.

Circle back.

Whereas if you just make somebody chuckle.

And then there's also the fact that if you spend an hour with someone or it takes half an hour to read that piece from start to finish. If you spend half an hour with someone, you feel like you know their perspective, you learn more about them, as opposed to you just see some funny post and you're like, ah, that's funny. They made me chuckle. Okay. I don't need to, like, be in. I don't really have a relationship with that person. But once you've spent a half an hour sort of cohabiting with them, you're more likely to hang out. So I closed out saying that, you know, maybe there's a, you know, article apocalypse. I'm not really seeing it. I think the algorithm is pretty well equipped to handle stuff like this. People always seem grumpy with X generally, and honestly, Nikita specifically a lot of times. But I've always felt like the sloppy algorithm changes that happen, like, they sort themselves out really quickly. Like, there's plenty of times when we will come in on a Monday, we'll be like, wow, the algorithm was really brutal over the weekend. Like, I was just seeing, like, just junk that didn't feel at all tied to what we're interested in. It just seemed like they made some change and someone hacked it and it's all just like viral videos or some quote, tweet, meta, or some meme format.

There was an era where you'd get eight elon posts in a row.

That happened. So there like a whole bunch of different. There were just a whole bunch of different, like, crises where people were like, it's over. And then a week later, the algorithm sort of adjusts and learns that people don't actually like all that slack or they don't like this thing, and then they adjust and it's fine. The other little quick tip is that if you ever really get sick of the for your page, you can actually disable it by uninstalling the app, bookmarking X.com as like a homepage icon that just opens in Safari if you're on an iPhone. And then you can install this extension called Social Focus that will allow you to turn off the for your page. So it opens to your following tab by default. And you just can't even access the for you page. So you can always do that. I've done that for like a month here, a month there. Honestly, it's not that much better. And it's one of those things that's, you know, classic stated preference versus revealed preference. Stated preference is, oh, I just want to see who I follow. I don't want to see these random viral posts. And then the revealed preference is like, actually, some of these viral posts are pretty good, they're pretty funny, and I enjoy them. And the problem with the for you with the following tab is that you is that there are some people that post a ton and there are some people that post very rarely. And you want to see some sort of even balance. And you don't realize that there are some folks that you follow who post a lot.

A lot, a lot.

And then it's like, really, I just want to see the cream of the crop. I just want to see their best posts. But then I want to see every post from certain people. And it's hard to find long term. I think we'll get to a point where you can define your algorithm a little bit more, but for now, you're subject to the whims of Nikita Beer. And fortunately.

Should we talk about graphite?

Absolutely. Code review for the age of AI. Graphite helps teams on GitHub ship higher quality software faster. What should we talk about?

We should talk about Xai.

Oh, yes. What's going on with Xai?

So, Time Morse.

Yes.

Did a podcast with. I'm going to potentially not pronounce his name right Suleiman. Suleiman Khan Gorey.

Yeah.

Who was previously at the time when the podcast was done.

Yes.

He of course was working at xai. Yes, Came in, he talked about WTF is happening at xai, Predicting bottlenecks, bootstrapping off the Tesla network, how Elon deals with fires, what it's like working at xai. Cybertruck bet with Elon, how they built Colossus, how XAI hires experimentation, how Elon recalibrated his timeline estimates. No one tells me. No, that's important.

That's a wild one.

Why the fuzziness between teams is an advantage. Testing human emulators as employees biggest blunders, what a meeting with Elon is like, how Elon gives feedback, figuring out what is truth for Grokipedia, what happens when Elon sees wrong Grok outputs on X, what it's like operating in X, AI's war room and a number of other things. So massively viral, by the way.

4 million views just on X, 5000 likes. This is a massive.

And so, and so anyways, like I'll start by saying that this is fantastic content, beautifully shot. Well that time or very well done. Fantastic team, very well done and, and great content. Right. People are so curious about what's happening in xai.

Right.

We want to understand, you know, how they're approaching different problems.

And Elon talks about it in such high level. He's talking about, you know, AI AGI, super futuristic and then he'll give little details on like some massive build out that he's working on. But he's not one nitty gritty.

One small problem.

Yes.

Usually a $200 billion company likes to control who's going to go out and do a tell all about the company. And specifically with Elon companies, if you hadn't noticed, there's not a lot of people that work for Elon companies that go and do the podcast circuit, talk about everything going on.

No, I mean at SpaceX, Gwynne Shotwell has given a few speeches, mostly at commencement speeches, like inspirational speeches to colleges and that type of stuff. She's not going on Rogan or different space podcasts or tech podcasts. It's very rare to see content with her. And then Kiko Donchev is their head of launch and he will do engagement with NASA and if there's a real heavy industry focused panel, he might attend. But he's also not just hopping on stuff.

And then beyond generally, he's not trying to create like thousands of superstars. Right? He does, he wants you to Work for him forever.

Yeah, I saw a post in here.

But when a company, when, like when somebody, when someone super talented leaves, it's not like Elon in general is lining up and saying, like, let me do your first round. Let me do your first, like, you know, 10 million, 20 million, 50 million bucks. Right. He's actively trying to incentivize people to not develop big brands. He wants people to stay and he wants to be able to control the messaging.

Wait, speaking of that, look at those, look at this Sam Sheffer post he went and found 15 years ago, he did a SpaceX office tour. It is so, like, it's just, it's crazy. It just looks like a cubicle office. But in there, Sam Sheffer finds that there's someone who works on propulsion named Matthew McKeown McCune. And he's been on propulsion for 17 years and eight months. Like, that's exactly, that's exactly what Elon wants. A true believer who's riding with him through thick and thin and has probably done extremely well financially, worked on really interesting problems. Is he going to be a household name? Is he going to be a podcast circuit? No. But does Matthew really want that? Probably not. He wants to propel the rockets and that's exactly what he's done for nearly two decades now. So that's the Elon strategy. And you should know that going into that environment, maybe it should be more explained, but it certainly ended in disaster in this.

Yeah. The funny thing is I didn't catch all the interview, but they talk about how HR in general is not super formalized at the company. And so you can also imagine that PR and the media relations team is not super active. Maybe nobody in HR said, hey, don't go do it an hour long podcast. You know, if you're anybody but Elon, maybe don't do an hour long podcast telling all. And so of course people, what happened. Ultimately, Suleiman, he said he's laughing.

We don't actually know what happened, but it is odd to go do a. To go do an hour long interview.

And then quit anyway. So I think like good intentions potentially took the concept which he talks about in here. No one tells me no, a little bit too far. But I posted earlier, you can't even share the internal secrets of your AI lab on a cinematic podcast in this country anymore.

Brutal. Andrew Curran had some good context here. He said the speculation is that this is the result of the interview he gave where he said Xai intends to run one million human emulations, a digital version of operations Optimus for online work, which will be powered by leasing compute from Tesla owners when the car is parked and charging. So that's an interesting development that we actually haven't heard about before. This idea that you have a pretty advanced AI chip doing self driving in the Teslas. They're just sitting there overnight. Why not do some inference and lease.

That Back to the why is it a digital version of Optimus? Is it going to be sitting there in the corner like Clippy?

I mean I think realistically it's probably just like agentic rollouts. So if it's for online work, it's like what are a bunch of tasks that you could turn grok loose on? Everything from the classic order doordash in a simulation. Try and buy something on Amazon in a simulation. All these things that require moving around on a screen, doing computer work, using Excel, building stuff, writing code. All these things that require a lot of heavy inference to roll it out, create a plan, write some code, test it, go back, double check, test it again, blah blah blah.

I somewhat remember, but maybe not quite in as much detail. I feel like when I remember him talking about this kind of thing, it was like, yeah, we have distributed compute that we, we can leverage.

Yeah, this might be in more detail, but still again, if you're at a company, any company, you should probably clear the public speaking engagements with.

Solomon didn't just read Lulu's going direct. He studied and he sat down.

That's wild.

And I actually DM with him briefly this morning and I think he's going to work on a startup. Okay, cool. We'll have to have him on at some point when he's ready to talk about it. I mean he's certainly now honestly kind of benefit of like going and working at a Google or a meta is they'll straight up tell you don't talk to the media unless you've been explicitly or go.

I mean there's different strategies at different companies. Some companies are totally cool with a bunch of people going out and talking about their business. But yeah, this seems like one of the more secretive organizations. So a life lesson there. Anyway, console. Console builds AI agents that automate 70% of it. HR and finance support, giving employees instant resolution for access requests and password resets.

Let's go over to Ben Affleck.

Ben Affleck, you know I had a premonition, a little preview that this was going to happen. He went on Joe Rogan and gave some very good takes on AI. There's some discussion over how good of Takes. They are. But I ran into somebody that said that they met Ben Affleck. They were like a proper AI researcher and they said that they met Ben Affleck and he used compute as a noun and they were like, yeah, that was impressive. He's talking about scale and compute. He's actually very, very deep in this which he went to Harvard. It's not shocking but I think there's just a general vibe that Hollywood is behind or out of the loop to the point where they wouldn't know the big models they wouldn't have tested. It's like no, this is their business.

I mean there are some like Joseph Gordon Levitt is pretty chapped in.

I mean he's being a wife on that board of OpenAI back in the day. She was, she was, yeah. You can't get more tapped in than that.

Well, hasn't he just been spewing like AI safety nonsense?

Yeah, he's a big safety guy.

He's a doomer or just a safety guy.

He's a doomer. Doomer. He said today we're announcing a new organization called the Creators Coalition on AI.

Well fortunately we have the counterpoint to that because Ben Affleck is not a doomer. He's actually pretty white pilled and it seemed like a really good interview. Let's play this first clip from Forrest who says that Ben Affleck has great takes across the board.

Ryan. Ryan says what if he's just action.

To get ChatGPT or Claude or Gemini to write you something. It's really shitty. And it's shitty because by its nature it goes to the mean, to the average and it's not reliable and it's. I mean I just came stand to see what writes now. It's a useful tool if you're a writer and you're going what's the thing? I'm trying to set something up. Or somebody sends someone a letter but it's delayed two days and gets. And it can give you some examples of that. I actually don't think it's very likely that it can. It's going to be able to write anything meaningful or and in particular that it's going to be making movies like from whole cloth like Tilly Norwood. Like that's bullshit. I don't think that's going to happen. I think it's. I think it actually turns out the technology is not progressing in exactly the same way they sort of presented it and really what it is just like sort of visual facts.

There was a line in here where yeah, there was a line in here where I was like, that is a Carpathy line. Like, he definitely heard this from Carvathy. He's listening to Doorkesh for sure.

For money. I can't. You can sue me, period. It kind of feels to me like the thing we were talking about earlier.

Where there's a lot more. The whole idea of, like, these models being mid. That's not quite true. I mean, it's hard with, like, to develop a reinforcement learning environment with a verifiable reward for, like, a great script that will break through. Because even the great script writers have flops that are unexpected. Literally everyone thinks it's great, and then it goes out and it just is a box office bomb. We can pause this for a minute. But later he actually makes a stronger case, I think, because they're talking about Dwayne the Rock Johnson in the film the Smashing Machine. And Matt Damon is recalling an interview that he did with Dwayne Johnson about his, you know, one of the most climactic scenes in the Smashing Machine that's very emotional. And he asks Dwayne, you know, how did you pick this, you know, motion? He pulls up a sheet because he's sad. And, like, what were you conjuring? What were you drawing on to create that experience as an actor and bring that emotion to the screen? Because Matt Damon's saying, I love that scene. I loved your performance. Where'd that come from?

Because you're thinking about content without ads in order to draw on the sort of, like, deep, sad emotions.

Yes, yes, that might be what does it for you. But I mean, Dwayne Johnson gave, like, a very emotional response about a family member who was diagnosed with cancer. And he was in the room when she got the news. And so her reaction, he was, like, channeling that. It was very emotional. And then the other one was, I think, about another family member who had substance abuse issues. And so it was just this very unique blend of human experiences that he was able to bring to bear. And I continue to think that the lore and the storytelling that happens outside of a piece of content is as important as the actual content. In many cases. This was the same thing with art and the NFT boom. It's not like the scarcity is important, but the storytelling behind the art piece is a lot of it. People like a Van Gogh because they know about him cutting off his ear and all this history, and he's in all the history books, and everyone just knows Van Gogh.

And yes, you can go even with cars, right? It's a car is worth more if the previous owner was significant in some way.

Yes, yes, yes. Steve McQueen is going to sell for more. And I think that for certain storytelling elements, for certain stories, that lore of like who this person is and their likeness. And then he also makes a really good point that it's like, yes, you can, I mean you can obviously generate something that looks exactly like a Van Gogh right now. You can paint it with someone who knows exactly how to paint like Van Gogh. It's not going to have anywhere near the value. And when he gets to the idea of just AI generating a Dwayne Johnson movie, it's like, well, he'll sue you immediately. And that's very easy. And you've been able to Photoshop Dwayne Johnson into marketing materials for years. And we've had a system for counteracting that. You have to pay licensing fees and that will happen. So he has a good, he has a good balance. We can go into this second clip from Barely AI I think that's Trung Fan's company because it's a little bit longer. It's about a four minute piece. So while we pull that up, let me tell you about Label Box. RL environments, Voice Robotics, evals and expert human data. Label Box is the data factory behind Tyler's leading AI teams.

Get in the box. Label Box.

The Label Box. Let's pull up Ben Affleck diving a little bit deeper and we might have to skip ahead.

Yeah, we've been spending time looking at this like my belief is sort of like what's gonna happen?

So we can Skip to like 50.

That's gonna happen with electricity. And it's, and it's. Don't think it's very likely that it can. It's gonna be able to write anything meaningful or. And in particular that it's gonna be making movies like from whole cloth, like Tilly Norwood. Like, that's bullshit.

Can you pause this? So I think that he, he basically says that like he doesn't think that an iconic, amazing movie is going to be generated by AI And I sort of agree with that. But I think it's important to consider what are we defining as a movie, because we don't really watch movies anymore anyway. People watch content. And if you think about it like decades ago, people might see one movie a month or one movie a week, and that's like two to eight hours a month of screen time. Well, now people are doing two to eight hours of screen time a day, but it's these fragmented pieces and so if AI seeps into the cracks of. Within the cracks within the cracks. And the goal is not to create the experience of a two hour movie that you show in a theater and you get everyone to come and buy tickets.

Joe Rogan's really turned into Dwarkash for a listers. Just come on and talk about AI. Right.

It's really.

McConaughey.

Yeah.

Is talking about wanting this like personal LLM.

Oh, yeah.

Where you can log some data in.

Yeah, that one was a little bit rougher than this. Affleck's got it dialed. Let's continue with aflac.

I think it's. I think it actually turns out the technology is not progressing in exactly the same way they sort of presented it. And really what it is, it's going to be a tool, just like sort of visual effects and yeah, it needs to have language around it. You need to protect your name and likeness. You can do that. You can watermark it. Those laws already exist. You can't. I can't sell your fucking picture for money.

I can't.

You can sue me, period. I might have the ability to draw you, to make you in a very realistic way, but that's already against the law.

Yeah.

And the unions are going to. The guilds are going to manage this where it's like, okay, look, if this is a tool that actually helps us, for example, we don't have to go to the North Pole. Right. We can shoot the scene here in our parkas and, you know, whatever it is and. But then make it appear very realistically as if we're in the North Pole. It'll save us a lot of money, a lot of time. We're going to focus on the performances and not be freezing our ass off out there and running back inside.

Pause for a second.

Just like Spencer Tracy.

What if.

Before you rebut this, let me tell you about that Vibe Co where D2C brands, B2B startups and AI companies advertise on streaming TV, pick channels, target audiences and measure sales. Just like on Meta. Continue.

Small chance that AI saves Hollywood as a place, right? As a place, yeah. And the reason for that is one of the issues right now is filming is so expensive in la, there's so much red tape that people have to go to Atlanta, Canada, Europe, whatever, international. But all the talent, the writers, directors, producers, the platforms, they're still here, Right. It's very depressing. It's a bad vibe. All the restaurants are shutting down. Everything's cooked. But if you can start to generate the content Here just on your computer, there's a chance that you see a kind of resurgence in Hollywood the place. Right.

Yeah.

And I think that realistically that is the strategy that the industry should take because having this concentration here, there's already so much talent. Keeping it here and actually embracing AI is much more likely to revitalize the city.

Yeah. There's an incredible quote in here. Hopefully we can get to it. Let's skip ahead and play because he has a line that truly is directly out of Andrej Karpathy's mouth.

Thing we were talking about earlier where there's a lot more fear because we have the sense that's existential dread. It's going to wipe everything out. But that actually runs counter in my view to what that history seems to show. Slow. It's incremental.

Let's go.

I think a lot of rhetoric comes from people who are trying to justify valuations around companies.

Literally a Carpathy line.

Well, the reason they're saying that is because they need to ascribe a valuation for investment that can warrant.

Oh, you know Ben Affleck, he's sitting.

There taking notes make the Social Network 3. But it's just about Dwarkesh and Karpathy hanging out and Ilya, it's just about AI research. It's about an exciting neurips and that's it. Applovin profitable advertising made Easy with Axon AI. Get access to over 1 billion daily active users and grow your business today. People are already having fun with the memes out of this. I like this amend and pretend Ben Affleck on AI software. I suspect many AI native software companies are misrepresenting their growth and quality of revenue. Use of credits, annualizing proof of concept revenue and lack of annual contracts lowers the overall quality of revenue and ultimately the multiple.

Not that I mean this is a joke and actually say that but he's not that far off from what this.

Is a great template. I hope this meme runs way further because it's so good banger. I love it. Oh, Forest also clipped Ben. Ben Affleck talking about the GPT4O people. Let's play this because this is interesting and while we pull it up, let me tell you about Cisco Critical infrastructure for the AI era. We are very pleased to be partnered with Cisco AI's new frontier, the Edge. Let's do it.

Early AI the line went up very steeply and it's now sort of leveling off. I think it's because. And yes, it'll get better, but it's Going to be really expensive to get better. And a lot of people are like, fuck this. We watch before. Because it turned out, like the vast majority of people who use AI are using it to, like, as, like, companion bots to chat with at night. So there's no work, there's no productivity, there's no value to it. I would argue there's also not a lot of social value to getting people to, like, focus on an AI friend who's, you know, telling you that you're great and listening to everything you say and being sick. But that's sort of a side issue.

Tyler's just engaged over there.

Yeah, yeah.

What's your. What's your rebuttal? Let's give. Let's give another rebuttal.

Defend AI.

I think most of these takes are, like, pretty outdated.

Okay.

Like, writing is actually AI writing. You can still tell that it's AI but it's like, it's good. It's good.

Yeah.

Videos are like, I think, like, okay.

Okay, sell a screenplay then. Sell a screenplay. Yeah, yeah, Shop a screenplay.

Do it.

Do it.

Yeah, but why would I sell a screenplay? I could just make the whole movie myself.

Okay.

Like, I.

So it's like, just make the movie.

He has this idea, like, make the finished movie.

Call it Netflix. Instead of get a million dollar advance.

For your movie, instead of filming in Antarctica, you can just get in the parkas and film it here.

Right?

You don't need the parkas either, because you can just have them be in normal clothes.

That's true.

And then it's like, well, you don't actually need the people.

Right.

Because you can just maybe fine tune a model so you get consistent characters.

And then, yeah, I will say, just in the last two weeks, I am getting served AI content. That's pretty good.

Pretty good.

Pretty good.

There's some pretty good stuff out there now. A lot of it still has that creative human element. Like someone will take an iconic movie scene.

It's still all about the idea.

It's about the idea.

And the idea didn't come from the model.

It's a bull market for ideas.

Yeah.

That does feel like the next breakthrough.

This idea that Ben saying. I would push back a little bit on Ben saying, like, you're not just gonna be able to generate a film.

You will, but it won't be a film in the sense that it won't be Titanic. Won't be something that everyone goes to the movie theater, sits down, watches the same film. It will be a lot of films that are for different People. I don't think there'll be anything.

Yeah, but I'm just saying, like, people are gonna. People are gonna create, like, insane fan fiction style hero's journey, kind of like standardized plot, but for a specific sub community. You'll have, like, one subreddit where everyone totally watches.

And I mean, there is a lot of just slop movie content on streaming platforms right now. You can just go and watch Gap Year.

Tyler put Aflac in the Truth Zone.

Not everything is. Yeah, not every movie is gone, you know, is Titanic is, you know, Sicario, like a lot.

Look on Netflix.

There's a lot of stuff. Yeah, they spend a million dollars.

But every time I. Every time I look on Netflix, I'm like, is. These are. These are film.

I know there's a lot of slop on there. And.

Yeah, so I think, like, in Hollywood, at least, like, I like a lot of movies. Like, there are certain writers that I like.

Yeah, but.

Okay, you like movies? Wait, you said you like movies? All right, name every film. Name every film on the TVPN chat. Best film list.

Movie list.

Okay. But you always hear about, like, certain writers. They have a hard time getting their films made because maybe it's like kind of a weird concept or something. Like, it's hard to market, but this is like, AI is actually much better for them.

Right?

Because if you have someone who's like, actually has some, like, unique look at the world or they're like, very creative or something, like, AI is like, super good for them, but kind of only for them. Right? So maybe the industry, it's not super great, but for the people, like, if there's a certain director that I like, really love, they're gonna be, you know, they're gonna be able to make way more of their movies or whatever, you know, instantiate all their ideas.

Why are people saying that the Papa Tai is AI generated? This song's from 2013. Someone's saying that there's. The most viral song of 20 of the year so far is a. Is an AI generated song. But it must be based on another. It's an AI cover of a human written song. Okay. Yeah. So it's like style transfer on top of. It's like a filter. This still feels like tool level, not to push back. I mean, that is an interesting watershed moment that something that involves AI could be so dominant. But there is a lot of, like, human work at play there to build the original song that then was. Was edited with AI Anyway. Cognition. They're the makers of Devon, the AI software engineer. Crush your backlog with your personal AI engineering team. Anything else on Hollywood or should we move on to the battle between.

Chad is calling Tyler based Zoomer.

He's Zoomer. It's good, good times. Okay, so there's an update to The Elon Musk OpenAI lawsuit, which is sitting at about 60% chance that Elon wins on call. She right now on the prediction markets, but Elon Musk is now seeking. So on the day that the Brockman files went out, started going viral, OpenAI put out a release. They, they countered some of the narratives, added some more context, but they also sent out a letter to their investors saying, hey, look, we think it's going to be $34 million or something. worst we'll have to pay him back for what he donated. But it's not going to be.

Well, they said, we think, we think this is the most, but it could be more.

It could be more. And Elon said, yeah, it's going to be more. It's going to be $134 billion in damages is what Elon is seeking. Elon Musk's legal team argues that his early funding and involvement materially contributed to OpenAI's rise, which is now valued at 50500 billion. And he wants a slice. The lawsuit says OpenAI has become a for profit company, entitling Musk to compensation tied to the value created from his early contributions. Musk Musk helped found OpenAI in 2015 and donated approximately 38 million in early funding. When asked about the lawsuit, Elon Musk responded, I've lost a few battles over the years, but I've never lost a war. Funny.

And here's the thing, here's the thing. So great line. But I don't think this lawsuit is the war. I think it's a battle in a longer war. And I think that it may play a role in how this sort of like war between Xai and OpenAI or Elon versus Sam evolves. But ultimately, like, this judge is not going to make, you know, the judge and jury are not going to make a decision that just like decides the war, right?

Yes.

Just like the war is going to be won by overwhelming domination in the categories that they compete, which right now consumer enterprise.

And I'm sure if he gets that, it's like over 25% of the company, something like that.

And that's going to be, that's going to be hard again, depending. I'm sure it'll come out how much Sam owns exactly? In this, in this, in the case. Yeah, but it's going to be hard to justify that. He deserves more than Sam given. Given their.

And also just play out what happens if Elon does get all 134 billion. It's like you're not CEO. You're not. You don't have board control, you don't have super voting shares, but you're a massive shareholder that can just constantly do shareholder lawsuits. And so that's going to be incredibly annoying and very bad, but at the same time, like, maybe a distraction. How do you actually get to control if that's what Elon's going for? It's a very, very odd scenario to work through, even though I don't think it's likely that it lands there. We will see what the judge thinks of the. And where the damages land. If he wins, which is still, you know, not guaranteed by any means. It could. It could wind up being a ferric victory for Elon. It could be something where he wins the battle, but he loses the war in the sense that, like, it's a distraction. There's going to be a ton of discovery. It'll be, you know, ultimately, even if he gets 100 billion, he's worth 700 billion. It's like, you know, drop in the bucket or, I mean, he's like 10% richer, then. Okay, does that really matter? 20% richer? Of course it matters, but it's not.

Gabe says once I churn from ChatGPT, they are screwed. The straw that broke the.

A couple of people have been coming out, rallying, supportive Elon, saying that they're churning, they're out. Seems like Sequoia is maybe happy to take a diversify. Happy to diversify. They're planning a big investment into Anthropic, joining a round led by GIC and CO2. They're investing 1.5 each. Anthropic is raised. Name 25 billion. And this is. Yeah, Sequoia gets the Financial Times article because it's an exciting moment to have them pick. So Financial Times says Sequoia Capital is lining up a big investment into Anthropic, throwing its weight behind the AI startup for the first time as part of a funding round set to raise tens of billions of dollars. The California Venture Capital Group is funding a round led by Singaporean sovereign wealth fund GIC and US investor CO2, who are contributing 1.5 billion each. According to multiple people. These people said anthropic is raising 25 billion in a deal that values the company at 350 more than doubling its $170 billion valuation just four months ago. We've heard about this round for months now, but these things take a long time, so we won't be ringing the gong until we get Dario on the show and the deal is closed. So Microsoft have committed to invest up to 15 billion into the company and VCs and other investors are doing 10 of the 25 that they're raising. Sequoia obviously had the shakeup. Roelof, both in charge of Sequoia, passed on previous funding rounds for Anthropic, was wary of the concentration of VC investment into a handful of highly valued startups, arguing that venture capital was not an asset class. He said in an interview last year that throwing money into Silicon Valley doesn't yield more great companies. He was ousted in November, with the firm's partners electing Pat Grady and Alfred Lynn to compete co lead the firm instead. Sequoia was an early investor.

Let's give it up.

Stewards Google, Apple, Airbnb and Stripe. And has backed some of Anthropic's biggest rivals. Sequoia invested in OpenAI's funding round last year and Elon Musk's AI Group XAI. So venture capital firms rarely back rival startups in the same field, preferring to pick winners in each category. But the scale of the financial opportunity in AI has changed that approach. And Claude code is just too much fun. They love it, they can't stop using it and so they had to get.

A check in, but gotta get a piece anyway.

Railway Railway simplifies software deployment. Web servers, apps and databases run in one place with scaling, monitoring and security built in.

Well said, John.

Continuing little known fact, Sequoia was an early investor in Il for Nio, owning 6.3% of the company at its 1997 IPO. I love that. The Sequoia team in the early days was like doing pizza joints and restaurants and software. Didn't they own Chuck E. Cheese?

Right.

Yep.

That was a funny one. So yeah. This is from Andrew Reed. Il Fornayo owns and operates 13 full service white tablecloth Italian restaurants serving creatively prepared, premium quality Italian cuisine based on authentic regional recipes. What a fantastic business. And Sequoia Capital got a huge slug. 6.3% at the IPO.

Yeah. So sad news. Il for Nio. How do you say it?

Il Fornayo.

El Fornayo. As they're closing their restaurant in the bay.

Il Fornayo was like the cool spot to go before high school dances. When I grew up it was like the. I think there's One in Pasadena. It's really good. I hope that they can keep it open somehow or someone can.

This feels like ftx, You know, shutting, you know, shutting down. Closing locations.

Sure, yeah.

So much history.

Sequoia involvement.

Sequoia involvement.

Ridiculous. Now, that was obviously years and years ago. Anyway, these. These guys love violating narratives.

It is so funny to think about Sequoia getting a deck for a restaurant.

Being like, this is.

We got to get into.

This is right down.

Got to get in. Right down the fairway.

Right down the fairway. This is our sweet spot. We were waiting. We had a market map of restaurants. We had Italian, authentic, Italian carved out. There was no one there. We did a two axis graph. Italian, authentic. You're in the sweet spot. You got a deal. We're taking this.

You're not going to believe this. The founder for the partner meeting invited us to the restaurant. The food was fantastic.

They actually did another restaurant later. They did the. The melt. Are you familiar with this?

No.

Yeah, Sequoia did the melt, which is a grilled cheese spot. It's still going. It's all around la. I enjoy the melt.

That is.

Yeah, I mean, grilled cheese.

As if I needed a reason to love Sequoia more.

I know. Amazing.

They're supporting the grilled cheese.

It's a really funny story because the guy, the founder, I believe, was like a amazing physicist, like a genius mathematician or someone really, really quantitative. And they were like, you're the one that can solve fast casual grilled cheeses. And I think you scaled the business. I think it's done pretty well.

Wall street journal article from 2014, Sequoia backed the malt moves beyond grilled cheese.

Yes. Yeah, they have burgers now. They have soups. I used to go there.

They got burger, milk, chicken melt, Mac and cheese.

I'd get a grilled cheese and a cup of tomato soup. It was delicious. It was amazing. Anyway, Shopify. Shopify is the commerce platform that grows with your business and lets you sell in seconds online, in store, on mobile, on social online marketplaces, and now with AI agents. So David Sacks says, narrative violation. And Joe Weisenthal says, these guys love violating narratives. The narrative violation is that the largest AI data center uses roughly the same amount of water as two burger joints they are over. There are over 200,000 fast food restaurants in the United States. So this is a tiny amount. This is, of course, from semianlysis. They did a whole write up a breakdown comparing various water usage. We were asking, when is the semianalysis water model coming? Well, it's here. And they did a whole analysis on how much water these data centers are using, and it's a lot lower than people think. It's in many cases less than golf courses. It's not nothing, but it certainly seems manageable, which is great news because people have been worried about this. And now we have some hard data to point to when you want to say, hey, the water thing's actually not that big of a deal. Let's focus on the energy thing, because that actually is a big deal. Let's build some nuclear power plants, let's get some solar on the ground, and let's get some more energy up and to the right. So there's another narrative violation. Another narrative violation has hit the timeline. David Sacks. This narrative violation. According to BlackRock, the US does not have enough workers to meet the surging demand for the construction boom. AI is creating too many jobs. And there's a beautiful picture of a overnight success, some sort of construction equipment. And so BlackRock is now warning that a worker shortage could hurt construction boom. We heard that story.

If I was graduating high school, college, didn't want to work in an office, I would actually get a plane ticket down to Texas. I'd show up to Abilene with my ID and say, put me to work. And I bet you could get a job there.

I've told several people to do that. At least take the trip to some of these new gold rush towns and see if there's a way to bring your business there as a real estate agent, lawyer, whatever you're doing. Actually go and understand how much is this going to change the community, because it certainly has in the past anyway. CrowdStrike, your business is AI. Their business is securing it. CrowdStrike secures AI and stops breaches.

We were hanging out with some folks at the Singer drivers club yesterday. Our friend of the show, Paul, and a number of the coaches who were professional drivers were talking about how much a legend George, founder of CrowdStrike, is. Like, he has. He has the respect of, like, the most elite drivers in the world. Like, this guy is the real deal.

Speaking of driving, we have seen Elon. He's beefing with Sam. He's suing OpenAI and he wants 136 billion, something like that. He said that he will drop the lawsuit if they change their name to Closed AI and there has to be a point where if the judge is about to actually issue 136 billion.

Call up Delaware. Hey, no, no. I got a name change.

This is. This is not a joke. I mean, like, would ChatGPT actually see a decline in user numbers if they change the company's name to Closed AI? I don't think so. I think a lot of people just think about it as chat. They think about ChatGPT. They maybe know OpenAI. They maybe know Sam Altman's name. They don't know Greg Brockman. Like, they average chatgpt user could probably deal with the name change.

I don't think Ben Affleck would be able to handle it.

I know he would be upset. But truly, I mean, Facebook rebranded to Meta. It's entirely new name.

Okay, John, but realistically, the lawsuit's not getting dropped because of the name change. A post is not a contract.

He will abide by the results. Okay. He will abide by the results.

Well, he's also asking if we should buy. If he should buy Ryan Ayer. Have you seen that?

Yeah, I've seen that. But. But, but I have another. I have another way resolved a lawsuit. Okay, so when. When Elon was beefing with Zuck, do you remember when they were going to go in the octagon? They were going to do. They were going to fight it out. They were going to do ufc mixed martial arts battle. Do you remember what they were fighting over?

No.

Isn't that hilarious? Do you remember what they were fighting over? No, I can't remember. I can only remember the fight, but I looked it up and they were fighting because Mark Zuckerberg was about to launch Threads, and it was. And Elon was in this vulnerable position where he. Where he levered up to buy Twitter, and he bought it sort of at the top and Twitter and the whole market was down and he was legally obligated to buy it. And so there was a lot of debt. There was a big question of, like, what is he going to do with Twitter? Is he going to be able to juice it and get the ad product working? All the advertisers were leaving. He was in a vulnerable position. And that's when Mark Zuckerberg was like, I'm jumping in with my competitor threads and I'm leveraging the Instagram network. I'm really piping everyone in. It's a shot across the bow. And so Elon said, connor Hayes, absolute dog.

Absolute dog.

He put Conor in charge. And Elon took it personally and said, we got to settle this in the Octagon. But it never happened. They were thinking about doing the Coliseum. Dana White said, I'll officiate. I'll make it happen. I'm down. I'M into this. Never happened. But I was thinking, how can Sam and Elon settle their lawsuit that doesn't require a bunch of jurors in Oakland wasting a bunch of time to courtroom. They got to settle in on the track.

They got to get the track pay per view as well.

No, no, no. Obviously they're not getting in the Octagon. That's not going to happen.

Track is good. Tesla versus F1. Starts their own Formula One team. Sam's got his McLaren F1.

Yeah, I want to see Sam in the McLaren F1, Elon in the Tesla plaid. I want them around the Nurburgring 24 hours, Le Mans style. They got to go for 24 hours and then. And then they got to do. They got to do a 0 to 60 or they got to do a quarter mile race. They got to do an illegal street race for sure. That's definitely going to happen. They also got to do a drift track. Yeah, I want to see them both drifting. And they probably got to do a Pike's Peak as well. I want to hear both of them racing up Pikes Peak. And then you do sort of like best of five. Like you do a 24 hours of Le Mans, see who does best there. Average that with whoever's quickest off the line in a quarter mile.

Then the drag race too.

Maybe throwing in the Indy 500. I want like a month of Elon and Sam just racing various vehicles.

And this would satisfy the people that are saying, pause.

Yeah, the AI.

This is great. Okay, pause.

Yes. And it's a month of live streamed races between Sam and Elon.

It's like Tour de France.

Yes, yeah, exactly. They go from one place to another, do every. Every different circuit.

Yeah. Bobby says. And a cannonball.

And they definitely have to do a cannonball. Just whoever gets from New York to LA faster, we drop the lawsuit. They win the lawsuit. That would be ideal anyway. Lambda Lambda is the super intelligence Cloud building AI supercomputers for training Inference with scale from 1 GPU to hundreds of thousands. So Roon posted. Do your duty while weeping if you must. There's nothing clean and gentle in the crooked timber of human relations. And Gabriel says every Roon tweet contains material non public information and it's your job to get it out. And there has been a lot to read into the rune tweets. Lately he's been on a tear. But we won't. We won't speculate anymore. What else should we.

In the journal.

In the Journal.

Why the tech world thinks the American dream Is dying. Silicon Valley fears this is the last chance to amass generational wealth before AI makes money worthless.

People are really into the underclass.

This in that sense doesn't really make.

Thousand comments on the WSJ it's kind.

Of a confusing subtitle because why would you. If AI makes money like the people that think AI is going to make money worthless are the ones that are just going to go sit in the park, right? Is that the whole point?

Potentially.

Potentially. Let's get into it. Silicon Valley is filled with all sorts of dreams, but one of those wild eyed ideas long debated on subreddits and in hacker houses is becoming a real life nightmare. Will the AI boom be the last chance to get rich before AI makes money essentially worthless? One note, there's some people that love making money so much they will keep doing it for the love of the game. It's just going to be like a sport. It'll be like going and playing, putting on the cleats and playing a little soccer. Right? You're just running up. You just want to run up the score sometime.

Even if money becomes worthless, you know, people are still going to be looks maxing. It's going to be all status games, it's all going to be height maxing, various levels of inserts in the shoes. It's going to be a whole.

No, I think what I'm saying is like in that scenario it would actually just be like a video game. Like, yeah, I made a unicorn. Did it change my life? No, I did. I just wanted to prove that I could do it.

Sure, sure, right.

It's like somebody that's running a marathon.

And grow a unicorn. Like the bio company we talked to. What do you mean make a unicorn?

Oh, like actually build a company just purely for the luck. A literal unicorn.

Yeah, yeah.

Because in this AI abundance area, you could potentially go to Claude code and say, don't make mistakes, make a unicorn.

Make a unicorn.

Genetically modify a horse.

Anyways, continuing, the argument is that tech companies and their leaders will become a class unto their own with infinite wealth. No one else will have the means to generate money for themselves because AI will have taken their jobs and opportunities. In other words, the bridge is about to be raised for those chasing the American dream and everyone is worried about being left on the wrong side. It's the kind of FOMO that on first blush seems to require a huge suspension of disbelief. But the idea's mere existence helps explain some of the increasing class worries in California, where a growing movement to tax billionaires we're calling this the Business Elimination Tax is the new branding, according to Lulu, is roiling the Democratic Party. Affordable housing is real concern and the idea of the middle class seems out of reach, yet it smacks of sci fi thinking. But in San Francisco it feels real. And it's made more believable by the exploits of Elon Musk, the rise of OpenAI, Sam Altman, and warnings by Anthropic's Dario about great Depression like worker displacement. The transition will be bumpy, musk said this month on a podcast. We'll have radical change, social unrest and immense prosperity. A little bit of a roller coaster ride, he said.

Universal high income is his term for it, and that's Musk's best case scenario. History is filled with technology booms that create new winners and losers. AI optimists like to point out out that a rising tide has tended to lift all boats. What's being talked about now, massive job loss to automation and the need for public safety nets in the form of universal basic income paints a dramatically different future. It's still not clear if there's an appetite for so called ubi which runs counter to many Americans bedrock ideals of personal achievement. Quote I used to be excited about ubi, but I think people really need agency. They need to feel like they have a voice in governing the future and deciding where things go, Altman OpenAI's chief executive, said last year when asked by a podcaster about how to create how people will create wealth in the AI era. If you just say okay, AI is going to do everything and then everybody gets a dividend from that, it's not going to feel good. You got to lock in, you got to grind. He didn't say that part and I don't think it's what would actually be good for people. If money is out, scarce assets like art could become key. We saw some massive results at a Ferrari auction this weekend, a bunch of Ferraris trading for maybe 3 over what people thought the market was going to be. Lots of speculation about what's going on there, but people are certainly getting in on luxury goods. If you don't have a scarcity of resources, it's not clear what purpose money has, he said at a conference last year. More recently, Musk suggested people shouldn't even worry about saving for retirement, predicting AI will provide healthcare and entertainment.

You like that? Tyler?

No, I was just gonna say I think I've heard at Anthropic at least a lot of people don't contribute to their 401k for these reasons interesting.

Still it look I mean it also helps that if you've been there a while, your company went from sub 10 billion dollar valuation to 500 in the span of a few years. You're not like, oh I better. I'm so worried I haven't added much to my 401k.

Okay, that's fair. Maybe you don't want to contribute to your 401k, but if you want to change the world, you gotta raise capital at the New York Stock Exchange. We're very pleased to be partnered with the New York Stock Exchange where you can go and change the world by raising capital. Let's finish this out Weeks later, social media posts on X and LinkedIn and Facebook began claiming Nvidia's Jensen Huang said something similar. The CEO these breathless posts claimed was warning Quote the period from 2025 to 2030 may be the last major chance for everyday people to build wealth through technology. Scary stuff. Except Jensen, he didn't say it. Rather, his numerous public appearances in the past few months have been filled with talk about the potential for AI to be more of an equalizing force for technology. Quote we're going to have a wealth of resources, things that we think are valuable today, that in the future are just not that valuable because it's automated, he told Joe Rogan last month. It can afford, it can. It can be hard to sort fact from fiction in an era of technology that seems pulled straight from Iain Banks novel and backers of AI companies have billions, if not trillions, of reasons to hope their gambles aren't just once in a generation jackpots, but once in human experience, once in a human existence. Further contributing to the FOMO in San Francisco is the expectation that local AI companies such as OpenAI and Anthropic will soon go public, minting many more millionaires. After the New York Times ran a headline this past week about a wave of mega IPOs expected this year, local real estate agent Rohan Dar posted on X May I humbly suggest you buy your house in the San Francisco before this? It is interesting. I feel like the SF housing market has really been through some crazy gyrations. It was way down for a little bit. And yes, I mean, I agree. I expect it goes way up. The tech entrepreneur who was once part of Y Combinator years ago told me he was drawn to real estate in part because of a belief new AI wealth will fuel a housing boom. Or, as he predicted last year, the mother of all tech booms is coming. Get some while you can.

Well, thank you to Tim, One of the comments here is interesting. Somebody Jason Barger says, when I graduated college in 1999, I remember those pursuing teaching being mocked for going into an obsolete field. The Internet was going to replace teachers. Some thought it would replace educational institutions themselves.

Anyways, well, we have Jordan Schneider from Chinatalk in the restraining room. Before we bring him in, let me tell you about Turbo Puffer. Serverless vector and full text search built from first principles and object storage. Fast 10x cheaper and extremely scalable. Let's bring Jordan Schneider in to the TVPN ultradome. Jordan, how are you doing? Good to see you.

I am doing amazing. I'm so glad to be here. I just love listening to your ad reads. They're truly the best ad reads on the planet. Thank you.

I appreciate it.

There's no competition.

Careful. Careful what you say. We might. John is really tempted to start doing them during guest interviews. And I'm like, I don't know if we want.

We've said like 30 minutes is a lot of time to hang out with Jordan. We might need a break halfway through. Slip in an ad every once in a while. I do it specifically. If a VC is on and they're pumping their bags and one of the bags that they're pumping happens to be a sponsor, why not slip in an ad read?

Yeah. Why not be explicit about it?

Why not be a little on the.

Nerve, John, I give you the pass. I give you the pass. I'm ready for it. I want to see the creativity.

Well, if there's a lull in the conversation, expect an ad read.

But anyway, dude, you seem. You seem pretty happy.

Yes.

What's going on? You're like in the best mood ever.

Do you get a vacation? What's new with you?

I'm actually in a terrible mood because the Bills just fired Sean McDermott, our Godhead coach for the past nine years. It's some Game of Thrones shit. The GM is like, you know, knifed him in the back. He's now got a promotion. And I was asking Chad, GPT and Claude what the correct tech analogy would be to explain this to the TVP and audience. And there was not a good one.

So I guess AGI and back up for the. For the viewers.

Who.

Who are the. Who are the Bills.

The Buffalo Bills are American football team, okay. Plays in Western New York. They've got a great run. They have a wonderful quarterback. You get. Everyone should root for. For them. They're. They have a lovable loser reputation.

Okay.

Never won a Super bowl. And have been on a generational run. They've been the second most winning team over the past 10 years. Can't seem to get over the hump. And we can talk about something else.

I know this is changing every so. So I just looked at 12 teams have never won a Super Bowl. Is that like, imagine being.

A lot of them are new.

A lot of them are Jacksonville Jaguars, for example.

Okay, I know both, but the Bills, we. Chargers, Jaguars, Cardinals. It's a lot of team.

I'm glad we got to do it. You know, it's, it's a, I think this is like the debate among, among.

Do you, how do you watch football?

Do you want your kids to suffer? Okay, oh, how do I watch football?

Do you do the David Senra and like, you lean forward and like really focus like this, or are you kind of like writing the newsletter?

It's a family event. And I think the level, like myself, my brother and my father's levels of intensity of fandom have evolved over the years. I have started to lean back a little bit, but when the playoffs come around, I am screaming and cursing and getting very heated again. So I kind of like taper in over the course of the season until they inevitably rip my heart out. Not only with errors on the field, but apparently errors in the management side of things as well. But anyways, we'll see. We can talk about China if you want. Yeah, I also have some, like, media criticism of you guys I want to do.

Okay. Yeah, okay. About that. I'm just, I'm just glad that you're coming on black pilled about sports as opposed to geopolitics because, oh, I'm sure.

He'S blackpilled about geopolitics, but at least.

You didn't lead off with it. Because if you were, if you came on here and you were like, invasion of Taiwan's imminent. It's all, wait, why don't you give.

Us the geopolitical comp for whatever the Bills just did? I don't even remember. They like, let go of the head coach.

It's Operation Paperclip.

You know what, my analysis, here's my analogy. Okay, so, like, talking about invading Greenland is what Trump wants you to do. I don't think he's going to do it. But, like, the thing that gets him jazzed up is people being upset about crazy ideas he has and the push back and this, I think the, the fascinating analogy here is the whole Ro Khanna saga that we've had over the past month or two because, like, the amount like, this is what he wanted to happen. He is not dumb, he unders. He is a politician who has a national presence and has ambitions. And what he wants to do is as the Silicon Valley representative, like, there is a, there is a risk for him in a potential 2028 Democratic primary of saying, oh, you're just in the bag for these guys. So having like being able to sort of manufacture this enemy of all these like rich billionaires who, hey, by 2029 when no one has jobs, or 2028, 2027 when no one has jobs anymore, like, and we have two more years of tech bros. Running the White House, like the Q score of that segment of society may be down. Like having all these people on the record saying, we're going to take him down, we want to run against him, he's going to take our money. Which by the way, is something a lot of Americans, I don't think would be all that upset about. And like, yes, your median TVP and listener understands the nuances of taxing unrealized gains may not be the best thing for innovation, but like, the ease at which this entire ecosystem was baited I just found to be fascinating.

And it's a kind of thing. I think that's a good, I think it's generally a good take because. Because if you actually press anybody that's actually pressed Roe, he has no defense for his policy stance on this. He actually cannot defend it and stay logically coherent. Right. It's like he's running. He really is. It seems like he's running a bit.

But since when has that mattered in politics? I don't know, like the Kennedy Nixon debates or something. Yeah. He is now setting himself up against billionaire capitalist overlords who are apparently on the other side of him and hate him and want to do everything to bring him down. And like that is a winning play for him. And these folks are so scared of their bags and they're so confident that like, oh, like they say something and then it will happens and no, there won't be second order consequences that, I don't know. It was a fascinating saga to watch from the sidelines.

Yeah, he's on a whole media tour now. He's done a ton of. He's popping up everywhere and he's definitely taking advantage of the moment. He's done really well because he's got.

Some haters because he's smart.

Yeah, yeah, yeah, yeah.

You need playing the heel. He studied, he sat, he sat down and he watched bunch of WWE game tape and he's like, okay, I know. I know what I gotta do.

Okay.

All right.

Yeah.

This brings me to my media criticism.

Yeah. Yeah.

And I think, like, look, you guys have gotten a lot better, and the show has evolved, and the centrality of it in, like, American discourse is really something to behold. And with that comes responsibility. And I think you are slowly but surely maturing into that. I mean, the big moment you had a few months ago, making fun of Cluley and everything, is an important part of that maturation process. And, you know, the rage.

Are you talking about the. The case against rage bait?

Yeah, the case against rage bait. And sort of as you're. As you're doing more of this, like, you know, pushing yourselves to do, like, more independent thinking and writing with the. With the essays, every day you're going to have more politicians on. You're going to have more. Just, like, more and more of this is going to be trending towards that. I'm sure all the. All the presidential candidates are going to want to come on. Same with the 2026 candidates. I just think it's really cool. And politics matters to this world, whether we like it or not. And watching you guys kind of, like, grow into being. Having the. The stature and comfort to ask harder questions of guests, I think is really cool. So I just want to give you kudos, and I'm excited to see where all that goes.

I like that you're trying to rage.

Bait us into politics.

Yeah, so, So. I mean, like, so, so, so. So you're 100% right. The. There is a weird dynamic where, at least with our viewership, we actually do notice a decline in the metrics when we talk about politics and have politicians on. And, like, some of the chat will just be like, go back to talking about space or tech or whatever, but I understand it. It's gonna happen. At the same time. I really enjoyed Joe Wiesenthal on Chinatalk, and I particularly liked Joe's point that on the hard question issue, people often have this fantasy that if you're talking to someone, you're like, okay, here's a hard question. Are you a fraud? They'll just be like, you got me. I am a fraud. Like, you ask the hard question. Like, I can't lie. When in fact.

Well, we actually did ask somebody.

We actually did do that once, and it did work. And ironically, it worked with Joe Weisenthal, too, because on Odd Lots with Matt Levine, they asked sbf, like, how does this work? And SBF was kind of like, it's a ponzi scheme. And Matt Levine says, so you're in the Ponzi business and it's a good business. And he says, yeah. And so he kind of confessed. So there is a way to ask a hard question. But I mean Chad is saying can.

We talk about B2B SAs instead.

Owned. Yes, we can. Restream 1 livestream 30 plus destinations. If you want to multi stream, go to restream.com. that's some B2B SaaS for you folks. Also applicable to Chinatalk. Maybe we'll get you live streaming this year. Anyway, so give us how you process the hard questions.

Well, before that I think the question I would ask you because people ask us the topics that you discuss become you may not be approaching them from a political angle, but they have political implications. It's your duty to talk about them. And my response historically has been obviously tech influences politics. Obviously it's going to be a large part of the debates in 2028 in the midterms. But that being said, I've always seen our role as talk with really smart people and get an understanding of the individual players, the teams, the companies, the startups, the public companies and then get an understanding of the industries and then there's an opportunity for people to do kind of like meta political analysis even based on that understanding. Like something that we've like, you know.

People like, if you defining it. Yeah, I think defining it as Paul, as covering politics is a little too narrow. I think it's more like there are aspects to all of this which are bigger than will this company grow its revenue at this, you know, at this rate and will these guys fire.

That's exactly what I want to talk about though. And so like you have to hold me back. I want to talk about.

Yeah, so I'll give you, I'll give you another example. So I wrote maybe a week or so ago, like, why would Americans be excited about AI? Like, there's a, there's a video on Instagram and I'm sure there's hundreds of of Sam Altman saying AI is probably going to kill everyone, but it's going to create some cool companies in the meantime. Right? Like that is like a mainstream like video that millions and millions and millions of people have seen. And they'll use ChatGPT and they'll use some other AI product and they're like, this is cool. But I actually like the world and I don't want down to like pot. Like so part of it, part of it. It's like, I definitely think it's our responsibility or worth talking about questions like that, which is, like, you're not giving people a real reason to be. You can't say, like, we're curing cancer or AI is going to cure cancer and then simultaneously say all these other things and expect people to just broadly say, like, you guys are the greatest thing to ever happen to humanity. Like, keep it up. Right? Like, why?

Totally.

Yeah.

I mean, I think coming back to a point that John made of, like. Like, asking the hard questions is not. Does not necessarily conflate with being a dick to your guests. I mean, just. Just recently, like, I had this thing that went viral. I was having two folks on the China Commission on, and they were talking about, like, the downsides of, you know, China developing. And one of the things they said is, like, wouldn't it be really bad if China cured cancer before the U.S.

I was like.

Really?

Like, are we sure about that one?

Look, I want to save humanity, but I don't want anyone to save it harder than me.

Exactly.

But, you know, kind of getting to that point in the conversation of, like, exploring what your views are at the limit doesn't necessarily mean that you have to say, oh, don't you, you know, you want to kill grandmas or whatever. It's like, you know, viewers aren't stupid. They can kind of. They can kind of read into that what they please or, like, you know, make a decision based on what the guest says. But you do need to, like, take the guests there and bring them, as you guys said, to conversations which are a little deeper than, you know, revenue targets and new products.

And I do. I do enjoy those types of conversations with business leaders. I just. Maybe I'm just new to the talking to a politician, but it feels like it's entirely. It's like an entirely different game. Like, the type of conversation we're having right now is like, we're both podcasters. We're just talking back and forth. It's great. I can get a CEO on, and they're there to talk about a specific launch. And it's halfway. It's some conversation. We might be able to get them to crack up. We rang a gong. We do a bunch of stuff to engage them. I know a little bit of lore. There's stuff that can get there, but the politicians, at least for right now, I don't know the lore. I don't know all their positions. And so it can just be. I just feel like I'm struggling to keep the car on the track. Like, I'm not even I'm like, I'm spinning out a lot. I'm spinning my time a lot.

Yeah.

The cool thing with a CEO is like, they're playing the game of capitalism and whatever they're saying, you know, at the end of the day, for the most part, that they just want to make their company bigger.

Yeah.

Whereas with a politician like Ro Khanna is talking about, you know, these insane policies that will, you know, make California, you know, descend into, you know, a tourism and agriculture based economy. And I don't want that for California. I've lived here my whole life. My family has been in California for over 100 years. Like, it's very important to me. But. And I'm not even like, I'm not tapped into this whole Washington thing. And it makes sense when you say, like, okay, he's just doing this to like, prove that he stood up to the tech guys and so that he can go on his presidential run. But I have no interest in, like, really engaging in that specifically because there's enough enterprise SaaS to talk about it.

And there's also. There's also a lot of good media coverage in politics. Or at least there's a lot of. There's a lot of media coverage in politics. Like, there's so many huge outlets that do huge numbers, and there's plenty of outlets on the left and the right for all sorts of things. And also it feels like a lot of the politicians are crossing over. For a while there was this criticism that there was the right podcast echo chamber and the left mainstream media echo chamber. But in the last year, it feels like there's been pretty massive crossover, whether it's like Gavin Newsom having conversations with people on the right. The New York Times platform.

Well, the moderates, the Elk Boys left.

Yeah, the Nelk boys have platformed everyone. You know. Where's your Nelk Boys crossover?

Where were.

We got it. We got the Happy Dad. Celtics are right over there. We love the boys over here. Yeah. We had John Shahidi on the show live, in person, in the ultra dump. When are you getting to the Ultra dome? You got to come out.

The steward of the time.

You're in la, you got to come. You got to come hang out in person. It's great. Anyway, I got more.

I got more topics for you guys. Talk Nvidia, PR Stumbles.

Okay.

We got my Claude Code games.

Okay. Okay. I want to hear about the Nvidia. I want to hear about the Nvidia, the stumbles. Is this Jensen specifically? Walk me through what you're Thinking, no.

So I guess it was last week in the House of Representatives, there was an act introduced called the Overwatch AI Overwatch act, which basically would essentially start to require congressional approval for AI exports, certain AI exports. In the same way that Congress has a role to play when it comes to exporting arms sales abroad. So if you, you know, you sell fighter jets to Israel or Saudi Arabia or Taiwan, Congress has to, like, give a nod to that.

Sure.

And within two days, you saw Lara Loomer, Brad Parscale, and like a long tail of 10 to 15 other right wing influencers basically all tweet the same thing. A lot of phrases were repeated. House Republicans are being lied to. We need to win the AI race. This is, this is going to strip Trump of his foreign policy powers. But the funniest part was that AI was spelled not a capital I but a lowercase L across every single one of these.

Joe. So I have a story like this. There was the.

I mean, we do need to win the owl race as well. We need as many Alfreds as possible. We need Al in charge. We need total control over. We got an Al in at Sequoia Alfred Lynn.

That's right.

There's a couple other Alberts and Alfreds all over America. We need to capitalize on them.

I once gave some what ended up being bad advice to a portfolio company. They were asking me, like, how should I handle my launch? And I was like, basically you need to prepare your investors that you're going to launch and then encourage them to share it and maybe give them some thought, like starters on, like, what they could share. Because they're going to be busy. They're going to be in a meeting.

Yeah. And.

And so it very clearly was like, rewrite in your own words. Like, these are just examples. And then, and then everyone just copied the first one. Six or seven people just like copy and paste it. No, just like.

And then.

Of course.

Yeah, yeah. There's been a couple of quote tweets that I've seen hit the timeline from people who are clearly very busy where it, where the, the whole quote tweet starts with a bullet point. And you're like, I know what's going on there. Yeah. It's become its own meme format.

We should run this as a bit. We get like 20 of us to like, all share the same.

Yeah, yeah, we should all.

The same thing about episode. What's that?

Italian bullet point. Jordan is a truly masterful interviewer. Parentheses. Please feel free to rewrite this in your own.

We got to we got to do it for the Italian wrestler restaurant that's shutting down. It's like say, L4 now is one of the most storied institutions. A Sequoia backed company. It is absolutely imperative that it stays operational.

It needs a taxpayer bailout. The taxpayer bailout.

Bailout.

The treasury should wire $1 billion right now for 10%. Wait, so. So tie this to Nvidia. You said Nvidia was involved in this. Explain.

Well, I don't know who else is paying Laura and Brad Parrscale, but why.

Doesn'T Nvidia want this to happen?

Oh, well, because they want.

Because they want to win the AI race, not the Al race.

They want to win the AI race.

Okay. So they want to be able to sell internationally as freely as possible with as little oversight and maybe just a rubber stamp from the President. Get the it done. If it goes to Congress, it's going to be more complicated.

And to be clear, this, this seems like something a PR team, a PR firm ran where they're, they're like engaged by Nvidia, it sounds like. I'm just.

It could be engaged by anyone in the company.

Engages, a PR removed, whatever.

Yeah. And they're like, we're going to engage key opinion leaders. And then of course, as soon as.

The KOLs get involved, it's all over. Interesting.

Yeah.

Okay.

I mean, to be clear, this bill would put these restrictions for exports to Russia, Iran, China and the dprk. So I don't know. But look, it's, it's funny because there's been this real learning arc with Nvidia and Washington where I mean, as of three years ago, there was like one person in D.C. it was really not a company that engaged Deeply with, with D.C. and then October 2022, you have these export controls and all of a sudden this firm has to ramp up. And what they've kind of seen or the success that they had over the course of 2025 was basically Jensen doing the job. Well, yeah, when you.

We were looking at president, we were looking earlier this year on like how much they actually spend on lobbying. Really low, super low. Like single digit millions, like whenever I checked but travel. And I was like, that Jensen's gas bill could very well be higher than how much they're actually spending on like lobbying.

But, but to finish the thought. So, you know, okay, they've. They've got, they've got the White House won over, they've got the president won over. But the problem is this is not the idea of exporting Chips to China is not popular in Congress. You've had a number of bipartisan bills be introduced in the Senate and the House, which have basically tried to cramp Trump's style when it comes to how he wants to sell these chips to the pure. And so starting to play that. And they've been able to actually kill one or two of these bills. But the momentum does seem to be building. So having a flailing response like this is like, yeah, kind of cringe. Makes for some good. TVPN laughs but also is indicative of the fact that like, yeah, they are worried enough about what's happening on the Hill with respect to AI exports that they feel like they need to start to play these sorts of games.

But, but what if we sell a ton of GPUs to North Korea and they cure cancer? Look, would that be so bad? Kim Jong Un.

What's, what's, what's North Korea's AI Al strategy?

Their strategy is probably all fine tuning on the glorious leader. He's absolutely right. He does, he, he does hit a hole in one every time he golfs. Just hard coded the weights back to media criticism.

I love that Vanity Fair piece and like the last, the last paragraph where I forget what it was, but there was some funny tweet and it's like, this is the reason that all these other ones fail is because you guys are talented and you guys are very funny and like, that is a. There's a Venn diagram of not a lot of people who can like check all these boxes. So anyways, hats off to you.

Yeah, I mean, hats off to transistor radio. It's one of the funniest podcasts ever. It's amazing. Every time you guys drop one, I turn it on immediately. It's fantastic. Never quit. It's amazing. Sorry, did you want to move on to something else? Because I still want to get the update on just broad. China, Taiwan. We talked to someone who was particularly black pilled. Seemed like they were very upset with America's capacity to protect Taiwan. And I was wondering if you've seen any movement there. From my perspective, it feels like it's been kind of quite quiet all of 2025 was sort of quiet. I don't know if that's just me focusing on US Tech and AI race stuff and I'm out of the loop, but would love just sort of your lay of the land on U.S. china, Taiwan negotiations.

Sure. So I think the big story of, I mean, there were. There was a lot of drama over the course of 2025 Liberation Day. We forgot there is a whole nother arc, which of course included 150% tariffs on China, which got negotiated down, then got ramped back up, then you had rare earth controls. And then finally we've sort of seemed to reach a bit of a equilibrium. Trump's supposed to go there, and I think March or April, nobody really wants to rock the boat.

Specifically on the Taiwan issue. No blockade, no invasion, no like, you know, firing back in and forth or drones or conflict that I remember seeing anything like that. And that was sort of like the worst case scenario was that maybe they were going to start with some sort of soft blockade and it wasn't going to be violent, but it was going to really escalate the situation to like, we're repositioning ships.

Yeah, I mean, this is. This has been one of my polymarket trade recommendations over the past few years, is no, no, no. No war, invasion, no warning. Particularly surprised we've had something crazy. Look, I think Xi sees a lot of upside in the direction of travel of a lot of things. And yes, China has problems, but like starting World War III or, I mean, right now he has like a pretty good equilibrium with the US and that all of the kind of like, we had a sort of slow but steady creep up in controls over the course of, you know, really starting and towards the second half of Trump won. And that with what she has been able to do by kind of throwing the rare earth thing back in America's face, is stop that. And, you know, I think who's the really old guy who does trade in the White House, but.

Who.

Who like, had the. Says like, dumplings are bad or something. I can't remember his name.

You're like, you guys are ready to cover D.C. we're like, couldn't tell you who works there.

Dumpling man.

Google this dumpling quote while I'm like.

A Peter Navarro dumplings.

Okay.

I have heard of him. I'm sure he's a nice guy.

He said, look like it's a different chessboard between from what it was a while ago and the fact that China both has leverage and is willing to. More important, is willing to use it and understands that using it will get the US to back down. I think is a new dynamic which we hadn't really seen in the trade war arc of Trump one and then in the Biden administration. The one thing that I do want to flag to you guys a little bit is this saga currently going on between Japan and China where the new Prime Minister of Japan made some not super offhanded remark, basically saying, like, a problem that China has with Taiwan is a problem that we have and it's very likely that we would be involved in any kind of scenario. And this has basically been the goal of American presidents for decades now to kind of get Japan off of its butt, get its spending more on its military and get it, you know, really part of the equation to dissuade, to make the balance of military power in the region such that China has no interest in going into Taiwan. And what you sort of saw in response to the Trump administration was kind of nothing, no big statement, no saying, hey, we're here to support you. Basically, Trump called Xi and then Trump called Takaichi saying, hey, chill out and look like they just had a nice visit. The defense minister, Koizumi, the son of the old prime minister who like, like a Hotshot 37 or something, you guys should have him on. They had a nice meeting. We say that the alliance is okay, but I did find it a little disappointing that like, we're not kind of cheering them on when the whole thesis of the Trump administration is like, we want to step back and have our allies do more of the heavy lifting.

So you're famous for the watch mandate of Heaven tier list amongst AI labs labs. Have you been feeling that China labs have been falling off? It feels like they've certainly been making less sort of dramatic breakthroughs like the original Deep Sea. Manus is now an American company. Let's go. Let's hear for Zuck. There was an article and the coin models are great, but it just doesn't feel like it's in the conversation as like this is an existential risk to OpenAI or anthropic or Google. The real heavy hitters are the main labs in America. And then even Xai and MSL are doing exciting things. And so you have five significant efforts and it feels like the open source narrative has sort of fallen back. It's still there, but it's not nearly as existential or just as embarrassing as it used to be when a hedge Fund could spend 10mil and defeat Mark Zuckerberg.

Yeah, I mean, I think the big thing that folks need to remember when it comes to the US and China is on the talent side, roughly equivalent. Maybe the US has an edge because we can pay higher salaries. On the data side still probably a wash, but when it comes to compute and the ability to deploy these, these models, there's really no contest as well as capital able to be put behind this whole this whole arc, there's really not a contest between the US and China. I think current numbers are like TSM, like SMIC and Huawei. Production is like 1 15th what Nvidia plus plus TSMC plus all the Asics can pump out. And this matters less in a future where you aren't kind of using all the available compute. But look, people are going to come up with ways to use this stuff. I'm doing so much dumb shit on Claude code every day.

So is the how optimized is your Claude code setup? I'm sure you spent all week on that.

I know I just use the terminal and it's been a fun apps, but the fact of the matter is, I think it is, it is. This is why the AI chip exports are so concerning is I really do think like the, the, the ecosystem with the compute and the capital is going to be able to drive ahead because then you'll, you'll have these great kind of reinforcing loops of productivity gains and growth then leading to kind of, you know, more development, more compute and so on and so forth. And the fact of the matter is that China, even though it has really good engineers that can train good models, is compute constrained when it comes to development and deployment. We just translated this really interesting panel that happened in China where they think it was like the CEOs of four of the top labs, so Jerpu, Quinn, Minimax and someone from Tencent. And basically all of them were like, look, we are so jealous of OpenAI and Anthropic. We have to use all of our computer to deploy stuff just to kind of keep the business going. Like the amount of the scale at which they can deploy compute, the amount of data they're getting from their users, and then the ability of these firms to run experiments to kind of put into the next stuff and get to the next breakthrough really means that in the current paradigm, America and the broader kind of Western AI ecosystem has a real edge. And throwing that away to export a handful of chips to China and the in my idea, Pyrrhic vision that we can still hook them on Nvidia seems a little silly, but overall, yeah, it's been a pretty good six months for American air.

Let's go.

One more question. Can you give us a take on Canada's new deal with China, specifically this trade? You get soybeans, I get infinite cheap EVs. What's going on there?

My sense is it's a bit of a bait as well. Perhaps a, perhaps a ro Khanna play to get the White House's attention maybe a little escalate to de escalate. What's interesting is like I think the amount of EVs they're letting in is pretty low. So I would not be shocked if this was more of a negotiating tactic to get the USMCA in a place which is more favorable.

Yeah. So Canada buys around 2 million new vehicles a year. I think the cap is at like 50,000 initially and then it goes up to 70,000.

Okay. Yeah, that's not. That's 2%. Yeah. Not huge.

Yeah.

It'll be fun seeing The Yong Wang U9 jumping over potholes in Toronto, Saskatchewan. That's where they got the potholes. Anyway, thank you so much.

Only waste men will drive this.

For coming on the show. Jordan, I'm glad to hear that your 2026 is off to a great start. Can't wait to have you back on the show.

I know.

Let us know when you're in la.

I wish we had more time. Hey, we're about to have Dan, the founder of T1 Energy, the solar manufacturing company that previously was Chinese owned and they were forced to sell it. I believe I have the history right. So you might enjoy listening into this one because when I saw the T1 Energy had a video go out and it was just showing like this like incredible, like high volume advanced manufacturing. I was like America's back. And I heard that it was a.

Chinese quite built by it Chinese plant anyway. But it's American now going through transformation. Anyway. Anyways, we'll talk to you soon. Have a good one. Goodbye Sentry. Sentry shows developers what's broken and helps them fix it fast. That's why 150,000 organizations use it to keep their apps working. Our next guest is Dan from T1 Energy. We're gonna bring him in in just a few minutes.

I don't think he's ready yet.

He's not ready yet.

False alarm.

We will tell you. We'll give you some more news. The New York Stock Exchange, our partner today they're announcing that the development of a platform for trading and on chain settlement of tokenized securities. The New York Stock Exchange's new digital platform will enable tokenized trading experiences including 24. 7 operations, instant settlement orders sized in dollar amounts and stablecoin based funding. Its design combines the New York Stock Exchange's cutting edge pillar matching engine with blockchain based post trade systems. This is obviously very exciting. We've been, we've been talking about 24. 7 trading happening a long time. The Interesting thing to track is that the volumes, even when you enable overnight trading, the volume is not the same during the day. And the reason is because like a lot of the traders work 9 to 5 in New York hours and they will have families and they want to go home and sleep. And so some hedge funds have set up overnight desks, but they basically take less risk. Risk during the night because there's less people on staff and you can't escalate. Oh, some trade's happening. We think that there's news that's happened at 2am and we think this might be mispriced. So we should be trading. But we don't have the rest of our risk team here to price it. So we're going to be a little bit more risk off. So the volume is much lower. So I'm interested to see where this goes and how much the volume ramps up. But obviously it's a very exciting development for the New York stock.

And ultimately still, this is still going to be subject to regulatory approvals. And the bill that's been. We got to have somebody on to talk about because there was some beef going back and forth between Coinbase not happy with how the market structure bill was coming together.

Yeah. Yeah. Well, let me tell you about phantom cash. Fund your wallet without exchanges or middlemen and spend with the phantom card. Lulu has proposed that we don't call it the billionaire tax. We don't call it the wealth tax. We call it the billionaire the Business termination Act.

And you know, if Moses. The job destruction act.

If people are against it and they're rebranding it against, you know, they need an anti. The, the people that are pro this act will need to rebrand it as well. And so I think that they should do like the, the grind incentive initiative, something like that to, to reframe it as it's a, you know. You know, we're not. It's not a billionaire tax. It just incentivizes everyone to grind harder. You're 5% down now. You got to build yourself out of that hole.

It's almost like a prestige in Call of Duty.

Exactly.

You got to start again.

Exactly.

Maybe they should for being rich. Yeah.

Like how many times have you. Prestige.

Yes. Yes.

Maybe if you. So they could call it the 10x tax where it's a wealth tax, but it only applies to people that are not 10xing their net worth annually because like.

Yes. Yeah.

You really like. Ideally, if you're trying to build wealth, you should 10x it 10x your portfolio 10x it again and then 10x it again. You'll be in a really good spot. And so we should have. If we're going to have government regulation, we want it to encourage people to really.

It should ramp up every year. 1% then 2%, then 3%. Eventually everyone has nothing. But the last man standing is kind of like a, you know, what do they call it?

Badge of honor?

I don't know. I lost it. Anyway, let's go to Flo. Flo Crevelho al Timor has been on the show multiple times. He says almost every single founder I know in San Francisco, including me, has reached the same conclusion over the last few weeks. That it's only a matter of time before we have to leave California. I love it here. I truly do want to stay and until recently, intentionally to be here all my life. But now it's obvious that we won't be possible. Whether that's two, five or ten years from now, there's no future for founders in California. But maybe he'll get a spot in Inclined village. Maybe he'll domicile in Miami or Nevada. There are plenty of ways. And I would be surprised if everyone fully relocates. I think people will be spending a lot of time in San Francisco, but probably less than 180 days. We'll see. Well, yeah.

It makes absolutely no sense. Every, every state in the country wants these kind of companies to be created.

We got to talk about some other. Some other better news. How well do you know preppy? Let's play a quiz. There's a quiz in the, in the Wall Street Journal. How well do you know preppy? It is in the timeline. You can pull it up and play live, but I'm gonna read this to you. So it's 10 questions on lacrosse, J. Press and more. We're gonna see how you do. So the first question. Which of the following is not a preppy approved label? Lily Pulitzer, J. Press, Versace or L.L. bean? Tyler, do you know?

I'm gonna go. What is Versace?

Nailed it. He got it. He got it. What type of shoes are you wearing right now? Oh, that's a preppy shoe. The term preppy is used as a slur in which canonical prep novel A, the Catcher in the Rye, B, Love Story, C, a separate piece or D, the Great Gatsby. What do you think it is?

I think I've only read the Great Gatsby. I'm gonna say, have you ever read.

Catcher in the Rye?

No, we didn't do that. I don't even Love Story.

Was that.

Yes. You got it. Love story. Wow. Okay. Name the non preppy sport. The least preppy sport. Volleyball, rugby, lacrosse or squash.

Volleyball.

Nailed it. Nailed it again. Which New York cool girl collaborated with GH Bass and Norwegian design? Eddie Sedgwick? Diane Brill, Chloe Zvigny or Leandra Medine Cohen?

The last one.

No. Chloe Zvigny. Although I thought she was in la, But I guess she's also a New York cool girl. Which shades of pink and green should be combined for optimum preppiness? Fuchsia and forest green. Baby pink and emerald. Hot pink and lime green or cerise and chartreuse?

I'm gonna go baby pink and emerald. Emerald.

Wrong. Hot pink and lime green. Rookie. I think. I think I have to flip the whole page over to see the other side. 5 is C which is. Yeah, C is hot pink and lime green. Let's see what else we got in here. What is the origin of the term urban haute bourgeoisie? Henry James. The Portrait of Lady. Whit Stillman's Metropolitan. Patricia Highsmith's the Talented Mr. Ripley. Andy Warhol's the Philosophy of Andy Warhol. These are hard.

Really hard. Anyway, have you ever been into dressing preppy?

I don't know. Not really. I've never had any sort of like lane. I just put on clothes as they're made available to me. Anyway, there's some other good news. DHH is racing again. Shopify LMP2 is back in action for the Daytona Rolex 24. New team, new co drivers still. Toby Lutke and DHH at the wheel now riding with rails and Omarky 2. They got their whole livery. It's amazing. Ruby on Rails base camp on there, the Shop app and Shopify on the livery. It's green.

Okay, we might have to do it. We might have to ride with them. We might have to do the 24 hour watch along live stream.

This might be the one that we need to do. This is absolutely amazing and I'm always to going excited when I see Carl.

Looks fantastic. The race, let me confirm is going to be later this month actually there's.

Also a post by Toby Lutke. Good testing days. Can't wait for the Rolex 24 next week. This photo is incredible. This is this weekend so it's 24 hours they'll be racing. And yeah, Toby Lutcke, he needs to do a racing podcast. He should go on Vincenzo Landino's show. He's done shows about Starcraft, all about gaming. I've never heard him talk specifically about racing, but what a sport. What a sport. Fin AI, it's the number one AI agent for customer service. If you want AI to handle your customer support, go to Fin AI.

We should pull up some of the new livery that was announced. There was some Gabe was sharing earlier. He said that Zinn made it on to Ferrari this year. They're allowing a nicotine sponsorship.

Oh, no way. What is this?

And Hoss, if you pull up this post, somebody was showing off Hoss's new livery from above. I'm in the timeline. You can see it looks almost uncannily like the top of this toilet.

What?

You see that?

No. Oh, wait, no. What is that?

I mean, it's just part of the arrow on the top part. Anyways, I think the Haas car looks fantastic.

Yeah. I also like that it's a. Yeah, it's an American machinery company. Right? Haas. And, yeah, it's just a cool little, you know, obviously not a car maker. You can't go buy a Haas and drive it around. But still, a lot of lineage and logic to it. I mean, Red Bull's been so good that it sort of tracks for me anyway. It doesn't seem jarring or anything that you have a. You have an energy drink car on the grid. But I feel like I still like the Ferrari. I'm excited that Audi and Cadillac are getting in. I like.

I like when Cadillac's livery looks.

Just go and.

You seen it.

Go and get it. I haven't seen the livery. I saw Vincenzo Landino posting some absolutely hilarious images of sort of what he thinks the Cadillac should look like. I don't know if we can pull up his post. I know he replies to all this stuff. What did he post? He posted some hilarious Cadillac livery. That was an AI Image here. My neighbor sent me this photo of the new Cadillac F1 car. You got to see this journey. It's amazing. It's got the.

Drop it in.

I think it might be already in there. I think I got it. But if we can pull this out. He said my neighbor actually did send me this photo that his friend sent him. I have no idea whose credit goes to. If this is yours, let me know. Is this in the bottom of the timeline? Did it go through? Let's see. Yeah, it's here. Look at that. That's incredible. So funny. It has to be AI or did they actually race something like this at one point? I don't know. I sort of assume AI at this point. Who else is in the. Who else is in the news? Three teams have now revealed their 2026 red liveries. Red Bulls looks a lot less shiny. When we saw them in Vegas, it was. It felt very pearlescent, very like almost pink sparkle to it. This looks like a little more classic to me. Well, we should do. We should talk more through. Speaking of cars, Porsche has suffered the biggest sales fall since 2009. People are not buying Porsches. The German sports carmaker was hit by a lack of petrol variants for best selling macan and weak demand in China. So Porsche sales slumped by a 10th last year. They lost 10% just in a single year. As the German sports carmaker faced weaker than expected demand for its electric models and struggled in the Chinese market. The company sold about 280,000 vehicles, which is down from 310,000 the year prior. That was the steepest slide since 2009 in the midst of the global financial crisis, when deliveries fell by 13.7%. That makes sense. The economy's not doing well. People are going to stop buying Porsches, but the economy did well in 2025. So what's going on? We're going to dig into it. The decline underlines the challenges ahead for the new Porsche chief executive, Michael Letters, who took the reins at the start of the year. The company has embarked on a restructuring program to reduce its production capacity and is taught it is in talks with trade unions in Germany about further savings. Porsche last year was forced into a costly adjustment to its model range, shifting the emphasis emphasis back to petrol and hybrid models and shelving plans for new electric vehicles. Falling.

Yeah, the market has just completely rejected.

Do you think they will ever ship the. I can't even remember. What's their new hypercar? The one that is the. It goes Carrera GT918 spyder. And then it was the Mission E or Mission X. That was supposed to be the electric hypercar from Porsche and it was rumored for years. And we saw a lot of the other hypercars. There was a whole idea of like, there's going to be a new Holy Trinity. The Holy Trinity, of course, is what, the 918 spyder, the LaFerrari or the Enzo and the P1. And then you kind of go back and you retcon a holy trinity out of the Carrera GT, the F40. Like you kind of piece it together with the McLaren F1. Even though that wasn't really a trilogy, people sort of put those into, like, legendary bucket together. And there was this idea that, okay, maybe McLaren's coming out with a W1. That'll be exciting. You have the F80. That's exciting. And then maybe Porsche will come in with something that's in the lineage of the Carrera GT and the 918 Spyder. And if they looked, if they bred the room around the 918 seemed like Carrera GT was way more popular. So maybe go back to that or do something there.

Well, Ford said, hold my beer.

Yeah.

Because they're coming in with a new hypercar.

That's true. Do we have any details? So they did confirm that Ford will be making a hypercar or supercar in 2027. And it's not just a continuation of the Ford GT. It's not a Ford GT V3, because there's been a number of those. We had the good fortune of seeing one on the track over the weekend. Looked beautiful out there, performed very well. But back to Porsche. While you looked that up, last year's sales slump has been partly due to supply gaps for the 718 and the Macan, its best selling model. So they're not making enough of those. Porsche stopped selling petrol versions of both models in Europe over a failure to comply with EU cybersecurity regulations. What?

Whoa.

They didn't. They got to get on Crowdstrike and decided against developing replacements that would have met new standards. So they're down. It's a rough. They've had a number of really high growth years. They came roaring back after the 2009 global financial crisis. 2010 they were up 25%. 2011, they were up 20%. They were putting up 20% growth years for five years straight, basically. Then the last decade has been a little bit slower, but not contracting. And now we're way down, way off, way off peak, which is not good. So despite an increase in US tariffs announced in 2025, North America was its best performing region, down only a few hundred units, compared with significant declines elsewhere. Sales collapsed by 16% in Germany, Porsche's home market, with a 13% decline for Europe. Do you want to give everyone an update on T1 Energy, Dan Barkello, because I believe.

Sorry, everybody, we're going to reschedule. There was a scheduling mix up. It happens. So we'll get Dan back on.

So the next guest will be Elliot from Dominion Dynamics at 125 in about five minutes. So I'm going to continue with this Portia piece. Continuing with weak demand in China, where European carmakers have found it increasingly difficult to sell luxury vehicles amid growing domestic competition. Also, wait on the tally. Yeah, in China, I mean, there's a lot of homegrown heroes. You're really excited about your national champion. Whether it's Huawei or some phone maker. It seems a little silly, but the cars are cheap, they're great, they have fun, crazy features, they can jump up and down, they can go 0 to 60 in 2 seconds. They look exactly like Porsches. And if you're a local, you're just looking at the dollars and cents and then also you supporting your buddy who runs the company or builds it online, you know, why not buy, buy local? You know, it's popular over here, it's popular over there. So Porsche sold 41,000 Porsches in China, which has been down 26% on the previous year. And it's under half the total for 2022. So back in 2022, they, they were selling 80,000 cars, now they're selling 40,000. That's not good. You want to be up and to the right and they are down to the right. Sorry.

Yeah, I'm trying to figure out how much of a factor. According to registration data, for the full year of 2025, Chinese branded auto manufacturers sold approximately 60,000 to 70,000 vehicles in Germany. That's still only 2.5% of the total German car market. But, but if that's eating into Porsche's sale of some of these like kind of more entry level EVs could definitely be a real factor.

Cool.

Well, we, I believe we have Elliot in the restream waiting room a little bit early. I see no reason not to bring him in as long as he's ready to hop into the TVP and ultra dome. Let's do it while we bring him in. How you doing?

What's happening?

Hey guys, how are you?

Great to have you on the show. Would love a quick intro on yourself and the company.

Sure. Nice to meet you guys. Elliot Pence. So Founder CEO of Dominion Dynamics. Dominion is building distributed attribute mesh networks in the Arctic, amongst other things. That's the wedge in.

Very interesting. How do you start building? What brings you to building something like this?

Look, the Arctic is a huge challenge, as is very clear. Over the last couple of months it's been kind of at the foreground of geopolitics. Canada has under invested in securing the Arctic. Awareness is core to securing it and we're building for that. So we started in June of last year and I've built about a dozen products, deployed them a few times and about to deploy them again just outside of the Northwest Passage. So the highest north up.

Very cool. So what does it actually look like? In practice, these are hardware that is being deployed, but give me a sense of it more practically.

Yeah, so it's basically, it's us building on top of COTS system. So we're not building the hardware.

Okay.

We're using radios that are, you know, from meshtastic to persistent systems to go tenant to solidus. We build software that sits on top of that RF that allows Rangers, which is Canada's kind of reservist force that exists across the country, but in particular in the Arctic, to capture more information from their environments. Allows them to take videos, to take pictures, to do voice notes. We compress that information down and then manage that across a mesh network. And then we reinflate that data when it gets to a satellite backhaul and push it up into a common operating picture.

Very interesting. So what. Give me. Give me a sense of how the Rangers even approach trying to secure and understand the Arctic. I imagine it's like one Ranger for every, you know, multiple thousand square miles. Like, it's not a very large area. Imagine they can't have people everywhere.

Yeah, yeah. This is the fundamental challenge. Right. There's 5,500 rangers across the country. Canada has a coastline of 200,000 kilometers. Their principal task is they do a sort of anomaly detection patrols every other week. And on those patrols are often out for sort of 48 hours, looking to see if they can find anomalies, whether that's, like, on roads, whether that's in the water, drones in the sky. But they're radioing that stuff back, and they're capturing it in kind of an analog way. What we're building is a way for them to capture all that data, store it, and then to start thinking about, okay, well, what are some predictive processes that we could build into this, this community of 5,500 reservists?

Do they. Do they. Do they roll as kind of like a little squad? Or is it just like one guy out there on a boat? Just like.

It's a steed.

Like. Yeah, you can. I can just imagine, you know, a movie of, like, you know, Lone Ranger. It's one guy out there just going crazy in the wilderness, thinking that they roll deep.

It's like 30. 30 per patrol. And then, you know, they're doing this all the time.

Yeah.

So what kind of. So specifically the anomalies that they're looking for, is that like, foreign intrusion, like surveillance?

Well, it's a mix of things, right? It's like, you know, is a roadblock that's critical for a community you know, a few years ago, we had a Chinese balloon that really nobody knew about for a while. There's indications that there's buoys in some of the channels up in the Arctic. So it's both strategic and just tactical of, like, how do I get to where I need to be going in an emergency?

Is the buoy like the Chinese weather balloon story, where you just kind of like, send it off and collect whatever data you can basically effectively.

Right. I mean, part of what. What we believe is happening just outside of the Arctic is they basically put buoys in currents that brings it through the Arctic, and they're just sucking up information about what is the operating dynamic. Like, so we want to be ahead of that, and we want to.

Is it like, how thick the ice is? So if you brought an icebreaker ship, like, how big would that ship need to be? How many would you need to get through to bring a warship through or something like that? Just strategic intelligence.

Yeah. Honestly, I'm not sure why, but.

But that's one possible reason. If you were in charge, you'd be doing that. Makes sense.

Why. Why this market? How does this. How does this fit? It's like, beautifully niche in that, you know, probably somewhat of a small market initially, but you can probably saturate it and then expand from there. But. But why? Yeah, why specifically the Rangers?

Well, a few. So the Rangers were the sort of least technically advanced group, but they had the hardest problem set. And, you know, for Canada and for the NATO alliance, the Arctic is becoming really quite critical. And so it's sort of niche right now, but it also gives us sort of a good user community to build a useful product and have an extreme delta, like an extreme effect on.

Totally.

So we want to build broader than just that community. But that's a good starting point.

This is like the iPhone moment for the Rangers.

Where are you building the company? Where are you based? Where's the team based? What's hiring look like?

Yeah, so it's a very Canadian company. It's deliberately Canadian company, is headquartered in Ottawa. We have offices in Kingston, Toronto and Ottawa. Yeah, I'll be based in Ottawa. I've been based in the US and DC for. For the past 20 years. It's backed principally by Canadians as well. But.

But yeah, yeah. What about hiring? Is. Is Canada caught up to America with the defense tech fervor? The Palmer Lucky podcasts have not reached Canada yet.

They get stopped at the border.

They get stopped at the border. Yeah.

Yeah.

I mean, seriously, like, where are you recruiting from? What type of people are you recruiting and is there this idea of like serving the national interest, working in government technology? That might be slower. There might be, it might be harder, might be longer hours, might be more technical, but there's some sort of, of light at the end of the tunnel in terms of impact.

Yeah, there's no TBPN in Canada yet. But.

This interview is making me feel like TVPN is the TVPN for Canada.

Yeah, we gotta North America.

I know Carney's not making it seem like we're all big one, one big happy family the last 48 hours. But we're still in our, in our, in our world, we're still family.

We're good, we're still family.

So Canada has not caught up on the defence tech sort of surge. That's kind of what we're looking to catalyze. I would say that when we raised our pre seed we had literally hundreds of Canadians that are working at US companies come back up and want to build something for sovereign capability or sovereign capacity. So you know, and there's an immense amount of talent in Canada. Like the Waterloo corridor is ridiculous though.

Talk about the process of actually selling into the government and how it's different. I think most people will be familiar with the American defense tech arc of Sbirs program of record how you kind of sell into the US Government, the Department of War. What's different, what's similar? How are you thinking about market?

Well the thing that's similar is they have a version of SBIR in Canada. They call it ideas. I would say it's a little slower, a little less money, a little less diverse, but it still sort of doesn't get you into a program of record. So you still need to figure out how do you get into those kind of long tail contracts. What's different with Canada and this is changing right now is there's no version of an ota, there's no other transaction authority. So you can't. Getting onto a contract is a real challenge and that's what we're hoping to, to change and catalyze and lobby for. Because without the ability to get on contract, this market doesn't make sense for anybody. Government knows it and is trying to change it.

Yeah. So it's a very Canadian company. But is this just a foothold? And then ultimately you'll sell to folks in Alaska, Greenland, maybe the Nordics. Like where do you see this going over to time? Is it like if you can do it in Canada, you can do it anywhere?

Yeah. The basic logic is this is the hardest operating environment on earth. If you can build tech that works in the Arctic, it will be valuable to everybody. The us, Europe, NATO allies. Yeah, it doesn't matter. I don't want to build just a Canadian company, just focused on Canada.

Yeah.

This is going to Canadian company focused on the world.

Yeah. What does the supply chain look like? I mean, you mentioned that you're, you're partnering with like Gotena and building on top of a lot of things. Does that make it less capital intensive? And you do need to build stuff on top of it, but it looks more like software teams, engineers. And you're maybe able to match it with the money that's coming from the government on the sales side.

Yeah, yeah, we're definitely software first. There is a supply chain in Canada. It is pretty extensive. It's actually built pretty. I mean, most of Canada's defense hardware is American, about 80% of it. And so they're integrated into large American platforms. But we are a software first company. We're taking a similar approach to a lot of the other Neoprimes, where you start off with software, you realize you need to do hardware, you vertically integrate, you push platforms, that sort of thing.

Yeah. What's exciting on the connectivity side, obviously the Starlink story is big. Is, is that the main driver of progress there? Are there any other connectivity stories that you've been benefiting from?

Well, one of the challenges is that Starlink isn't. So the polar orbit doesn't have a ton of sat. I didn't realize the polar orbit sucks. Right. Like there's nobody that really does polar orbit.

Okay.

And so that's actually what we're solving for. We're solving for localized connectivity.

Okay.

Starlink, Iridium 1, web, there's Canadian company called Telesat, they're all kind of coming up there, but the connectivity still sucks. And that like, that's across, like it affects, you know, your localization, your pnt, how you move, autonomous systems. So connectivity is a real challenge.

Yeah. Do you have an idea of how Astranis fits in here? I know that they do geostationary large satellites for Internet. They were putting one over Taiwan and they were putting one over different countries. Is that something that they could track? I mean, obviously, so we can ask it from them directly, but I'm wondering if you've gone down that rabbit hole.

Yeah, no, we're, we're going down that rabbit hole now. There's a bunch of new providers and I think the, the satellite problem will be solved over the next Five years. But like, five years is a long time.

Yeah, that's great news.

Well, we got to hit.

We got to hit the gong.

Tell us how much the biggest.

What's. What's the biggest seed round in Canadian history?

I think it might be the second biggest Canadian seed round in history. It's definitely the biggest in defense.

Well, this is going to be the biggest gong hit in Canadian history. Okay, good. I think so. How much? $21 million. There we go. Great to meet you. Thank you so much for coming on and yeah, glad. Glad you guys are focused on the Arctic, Cultivating Arctic power.

Keep us safe.

Thanks for having me, guys. Thank you.

Great to meet you.

Have a great rest of your day. Let me tell you about Gusto, the unified platform for payroll, benefits and hr. Built to evolve with modern small and medium sized businesses. And without further ado, we have Phil and Matt Grimm in the studio in the TBP and Ultradome. Come on. Come on down, guys. Thank you so much for taking the time to come. The TVP Ultradome. Great to see you. Good to see you. Two of the most athletic men in tech. Both athletes in their own right. Bench press world champion. Ben Press, high school champion. What was the. What was the stat again?

At the age of 17. 17 broke a bench state record for New York.

Let's go. That's the real news.

What was the actual amount weight?

Nothing crazy.

Okay, what's the number?

Three.

Three plates.

It was. It was 275 at the time. I could do three plates.

Okay.

I overshot it. 355. Actually get to do a 335.

Oh, no, but it is what it is.

Wait, that. So that was a high school state record?

Yeah, yeah.

Age and weight class.

I was 17. Amazing.

I made like 190 at the time.

And what's the latest on the Murf? Is this an annual thing now?

It's an annual thing we're doing again this year. Rolling it back. Definitely beating the time.

Oh, yeah.

I have a 11 hour goal, so I was an hour and 12 minutes last year.

Okay. You're gonna shave off 12 minutes? Yeah, yeah. That's what, 20% of the time?

A lot of training.

Yeah.

You're gonna come out and do it this year?

Yeah, I'll be.

Last year there was a lot of talk and then not a lot following.

So, you know, you keep coming out heavy. You know? You know, when we started the show, you invited me down and we did a Murph or we didn't even do a full Murph. We just did like some training. About a half dead and I went down and I had like a fever for like a month. It was brutal. I was so sick doing the show. It was absolutely.

But this year, yeah, we're rolling it back and then most importantly, we're going to do a big one in Ohio as a partnership with Ohio State and the launching up our Arsenal One factory there with everything getting up and moving. By the time the MRF comes around, we'll be just opening for production at the factory there. So it'll be a pretty big day for us.

Great timing.

Let's rewind. Introduce yourselves. Just so we have it on the record, I think everyone knows since you repeat guests, but kick it off.

Oh, I mean, I'm Phil. I'm the co founder and CEO of Dirac. We are building the first AI native production orchestration platform, starting with automated work instructions.

Okay.

And then I'm Matt Grimm, the co founder and COO of Anduril Industries.

Okay, and then why are you two here today together?

We are very, very fortunate to chat with you guys about an excellent partnership that the folks at Dirac and Anduril have come to.

Cool.

We signed a excellent multi year agreement to roll out Drac across all of Anduril's manufacturing facilities.

Okay. So that includes the one in Ohio.

Will include the one in Ohio. So I first met Phil about two years ago and change. There was a video that they sent around that went kind of viral in kind of manufacturing tech nerd circles. And it was a video of an earlier iteration of their build OS showing how you could go from a model in CAD and that the build OS platform would then take that, break it down and turn that into instructions for factory floor workers to actually put together and assemble. So I saw the video, reached out and said, hey, this seems kind of cool. We've got a lot of factory. We're growing, we're building. Could we set up a call? So we kind of did a couple of calls, did a couple of videos, video zoom screens, and then I brought them in to do a proof of concept. And we've been kind of building together for the last two years fully integrating the build OS platform into our existing suite of tools. We call the tools the Arsenal os. That is the combination of our ERP on the back end, then our PLM that our CAD designers use every day to design the software, our MES on the floor that our production workers are interacting with, and then getting Drock's build OS plugged directly into that, which will then reduce the amount of time going from an idea on a whiteboard to a model in cad, then ultimately to those work instructions that the factory floor workers need to actually assemble the products. So with buildos implemented, we've seen pretty wild, wild improvements in the time that it takes to go from that CAD model to that work instruction for the factory floor. Was the number 80 something.

Oh, 87.5%.

There we go.

It was 87.5%. It was a 12 hour reduction. It went from 12 hours down to 90 minutes.

That's amazing.

Let's give it up for SAS. Yeah. So not getting enough luck.

How often are the work instructions changing? Is this more unique to Anduril because you're developing new products, new CAD designs, and so you need to go from CAD design to work construction every day, every month, every year, versus someone who's just like, look, I make five, five, six rounds. It's the same CAD design, same work instructions for the last decade. I'm just doing the thing that I'm doing. Why is this important for you? And how does Dirac plug in?

Yeah, this is an excellent question. So you're not just a pretty talking, you know, some things. So the challenge for a company like Androl, as we scale up and we're diversifying across a number of different product lines and a number of different geographies is then you have engineers with an idea who want to make a tweak or make a change or sometimes we're responsive to quality issues in the field or something like that. We want to make a design change. So then every time a design change gets made, the current method for doing work instructions is predominantly manual. So every time a bracket or a wiring harness or a connector or a circuit board changes, you have to change the work instructions. Else on the factory floor, you get an instruction that says, like, you know, screw bolt number abc123 into this hole, tightened to this torque, whatever, but it's a different bolt now.

Yeah.

So then the guy on the floor gets an instruction and is like, but it doesn't, it doesn't fit. So then for us, as we, as we scale and we're specifically trying to take a faster iteration, kind of faster mentality to this manufacturing process, reducing that time is absolutely, absolutely clutching and critical. So for us, it is especially paramount, though I will say that I think this is a problem that plagues a lot of industries and a lot of our supply chain, a lot of our other partners and vendors. So I'll let Phil speak more to the rest of the market. But it's definitely compelling and working for us.

Yeah. Because you've had success in like, automobile manufacturing as well.

We're pretty.

We're pretty vertical noxes. Last time we, last time I came on the show, we talked about a shipbuilding partner.

Yeah, that's right.

Andrew makes a ton of. Of things. Plane ships, all sorts of different things. And we've seen, at least with Dirac, is that our problem set that we work on.

Right.

Automated work construction. It's not just useful for just defense folks.

Right.

Because we end up working with folks in aerospace and defense, automotive, agriculture and construction machinery, maritime complex industrial machinery. Really anybody that has any sort of complex assembly. You need operators on the shop floor who know exactly what to build than when and when. It takes weeks or months to make a work instruction, a guide for your folks in the shop floor, most of the time they're operating on outdated information. And when you want to have an industrial base that is agile and robust, you can't have your operators and your technicians operating on outdated information, on outdated context and that, you know, we often talk about the work instruction as the core atomic unit of information in the factory.

Sure.

And if that is the core driving force by which your factory builds, you.

Can'T let it be manual.

Can't be PowerPoint, you know, can't be.

Yeah. My only reference here is if you've ever gotten, you know, like, like some baby furniture thing and you got the instruction set and like, if you're actually following it closely and you realize that the instructions are wrong, it's just the most enraging experience ever. And so I imagine even for like, like employee morale on the shop floor, if, like, you're constantly dealing with just, like, terrible instructions, why should somebody be like, yeah, it's great if they're high agency and they're going to try to, like, fix it and manage up, but at the same time, a lot of people are just going to be like, you know, just actually become less engaged. And they're like, I'm not getting the right information and the resources. Why should I even. Why should I even try?

Correct. And then your baby furniture example, that is, that is not an electronic product. So you have to think of that exact problem and then layer on all the complexity of electronics and wiring harnesses and power distribution and all those extra components. So, like, one of the major challenges is that there is a quick iteration of chips. There'll just be a new chip that'll be faster on a different board, kind of new power. New power specs or Whatever. So you need to make a slight tweak. The traditional way of doing that is to take all of those changes that you want to make on a very slow cadence somewhere on in the area of annual, sometimes biannual, and batch all of those changes up into a single block upgrade that you then do all at once. So then for a long period of time, you are deploying outdated hardware. So you want to be deploying the best that you possibly can to your customers. But to Phil's point, in order to get the best onto the factory floor and then ultimately into the hands of the customer, you drastically need to reduce the time from idea to work construction to factory floor. And then most importantly, keep all of that in sync. Because when you have, at our scale, with thousands of design engineers and then many thousands of factory floor workers, keeping them all talking off the same sheet of music becomes a really hard orchestration problem to keep everybody talking.

Thousands of design engineers. That's a ton.

We just passed through 7,000 employees.

Wow. Congratulations.

Thank you, thank you. Every time we got to bring that gong closer for you here.

It's too far.

How.

How are the work instructions actually evolving from a form factor standpoint? Like, I imagine historically, you just get like a printed out kind of, I don't know, big stack of papers. I can imagine like a very thick IKEA instruction. And I think, like, the number of companies have come and gone trying to, you know, basically put work constructions into like, you know, like VR, stuff like that. How far away are we from that? I imagine, and this is a question that's come up with investor pitches.

VR is an interesting modality for displaying work construction. Typically what we see nowadays on the shop floor before we come in, most manufacturing engineers are using PowerPoint or Word. Microsoft, that old beast, is still dominating the word construction market.

So they just have a laptop pulled.

Up and they're just kind of oftentimes, so they will draft it into PowerPoint or Word. They'll then print it out, and it'll sit in a thick white binder on the assembly station. If they're lucky, they're updating it maybe, you know, once or twice a year, flipping through, say, oh, it's this version. No, wait, get that binder off the rack over there. To make this different model on the, The. The way that we think about it, right, the way that we've seen it on the design side of things, right, we've seen 15, 20 years of an explosion of excellent software in the design side. And because CAD is so advanced, what we've seen is an explosion in the variants that people are able to design, right? Think about the increasing complexity, the different modular builds, right? Company like Anduril has modular products. They have different payloads that are reconfigurable. Now think about that from the design perspective, great, we can swap out a bunch of different modules. Now from the manufacturing perspective, that means that you have to have, when you can have thousands of different combinations of things, you now need thousands of different work instructions where if one picture is off and one step is off, you have somebody building a completely different thing. And thinking about that from the top level build perspective, that makes sense, right? Like that's a ton of different configurations. But when we're looking at all the people who make up the supplier base of the companies that make planes, cars, trains, everything in between, we're talking about companies that make pumps and valves, complex mechanical sub assemblies. Think about how many different variants those guys have to have. Now they're in deep trouble because when they lose the ability to not only have folks on their shop floor that are 30 year veterans who just know off the back of their heads, okay, I remember how we built this last time. I don't really need instructions because all those folks are also retiring. Our entire industrial base and our supplier base is also at risk because we haven't codified all of that tribal knowledge. And that is a key mission of ours as well. Not just to be a rubber band around the variants, but to also codify.

And capture basically like how do we used to do it, why did we stop doing it that way? Like making sure that's maintained so if a guy doesn't retire, you're not calling him up like, you know, what did you used to do in this situation?

Exactly. Similar to McKinsey analyst who put together a spreadsheet or a PowerPoint that was like 150 slides. They're like, hey, how did you, what macros did you use? What, what formulas did you use for this thing? That's like another framing to think about it. Except that guy retired.

How many products does Anduril have now? Because I think most people know Fury, Roadrunner, Dive, but there's different variants with each product. How do you think about the product footprint?

To Phil's point, there are roundabouts, 30 of the core products, but then there are many, many hundreds of individual variants that'll come up along with those. So for example, the, the Australian military might use a slightly different radio than the US military. So then there's just, and it's not a, not a Hard change, but it's still a change. A slightly different bracket, a slightly different.

Connection because they're not standardized on a specific size. You need different bracket sizes and all that.

Exactly. And then to Phil's point about payloads, so for on the Dive xl, for example, maybe you want to put a very advanced sonar package or maybe you want to put a mine sweeping package or something along those lines so that the actual dive is still the same, but then inside you need a slightly different bracket to hold the thing.

USB C. This is.

This is. This is.

This is a.

This.

This would be the ideal. I think it's a long ways out, though.

It is. Right. Because standards, I mean, it takes so long even just to roll out into the phones. I mean, how long was it before the iPhone and the Android phone had the same charger?

Clearly we need European style regulation force to force us into it.

Like do we. Like, like does it need to be defined from government?

So there are a number of standards that exist out there specifically about, you know, communication protocols and things. And this is part of why we run these international multi military exercises, partly to kind of work through these types.

Of issues, especially for comms, I imagine.

Exactly.

This helicopter can't talk to that plane. You're dead.

It's a.

It's a huge pain point. Yeah. So there is some standardization, but that doesn't quite get down to the chip level.

Sure.

Like inside the seeker head of a Barracuda that there's maybe a slightly different sensor that this customer or that customer wants to use. The body of the Barracuda is the same. The wings are the same, the engine's the same, the power module is the same. All that's the same. But at the front there's a slightly different module. Which then to Phil's point means then you need another set of work instructions. You need to train your workers a different way. And then that becomes very hard to manage at scale. So. So with Phil's platform, we'll be able to kind of bring that together and shorten that time and ultimately lead to products getting in the field quicker.

Talk about Ohio. How far along are you with the project? Is Trey vindicated in every. In all of his Ohio? You know, I'd be.

Don't want to take swings at Trey here, but the Arsenal One project in Columbus, Ohio is. Is running incredibly smoothly. We've got about the first 30 and change. Ohioans hired already. We hired them about six months ago and then brought them out to California. So they've been at headquarters where we've been building the first few tails of the Fury line. Then they're coming back to Ohio this month. And then we're doing the stand up and provisioning of the line. So the ultimate arsenal. One campus will be about a dozen buildings totaling in the neighborhood of about 5 million square feet total. The first building that's about just shy of a million square feet. It's like 800,000 square feet is fully built. The internal is being constructed now the inside and the production line on all the equipment, all of that test rigs and all of those things are being provisioned. And we'll start production on Fury this summer. Building two is the walls are vertical. So that's going up. And then building three is immediately thereafter building four right thereafter. So we'll be rolling out a new building about every year.

Ish.

For about the next seven, eight years.

You said the Fury tails are being built one place. Is this. Is this a matter of like building different pieces then bringing them all together? Is there an assembly line? Is it like a manufacturing of a.

Sorry, a term of industry. Tail being the full plane. Like a tail number on the plane.

Oh yes, yes, yes. Got it.

Yes, yes, yes. So the first. The first. The first. The first out of four or five. I'm gonna get in trouble for not remembering first. Four or five will be. Will be assembled at hq.

Yeah.

And then test flown at our test range nearby. And then the next one will be the tail number, the whole plane. But then from there on out, they'll.

Be manufactured in terms of assembly lines. Are we still paying homage to Henry Ford with a. With a plane that moves down and gets things added? Or is it a more integrated process?

For Fury's line, it will resemble something in the neighborhood of okay, it kind of moves along. So to move from station to station.

Actually more like the grim method. There you go.

There you go.

There you go.

We'll coin that one. We need a gong hit for that.

Perfect.

Give it up for the perfect.

Thank you.

Thank you. But the, the line will resemble closer to that where at station one we do the hydraulics. At station two, you do the electronics. At station three you do the fuel system. Something along those lines. And then it's ultimately the Fury line will be 22 stations. And then where the Fury line in particular is. Gets complicated is that we have to do a number of very complicated coatings and paintings on it. So that's a little bit tricky and specific. And then of course the propulsion element of it is just very, very complicated.

Talk to testament to how far out Mr. Matt Grimm thinks about has been thinking about fury. Two years ago, when he initially reached out after seeing our demo of Build os, his initial outreach was, hey, I saw what you guys are working on. Can you use it to build robotic fighter jets? And I said, yeah, I mean, depending on how cool they are, of course.

Yeah, yeah.

And then two years later, here we are.

Here we are.

Yeah. And what a moment too.

Yeah.

And how validating this is for Dirac. I mean, I feel like part of why you guys are coming out and talking about it together is I imagine you want your supply chain to probably get on Dirac is at. Is that a part of it? Like wanting the whole supply chain to modernize?

The quality part that you get at the end of the day.

Good question. So there's a couple parts, a couple reasons why I'm here in particular to support the announcement here. First is that Anduril's on a mission to save the West. It's kind of the name of the company baked into that. And it is my firm conviction that part of that journey will be re industrializing the west and re industrializing America. And part of that is getting supply chain teams and get manufacturing organization production operations to be more tech forward and be more tech enabled. And just like accounting software and sort of like traditional kind of big software platforms, a lot of the core technology in that ecosystem is pretty hard to use. It's almost user hostile. It's just very difficult. So part of the mission here is that like, yes, of course we want to adopt it. Then we want the supply chain that supports us to adopt it. And then in the long run, we want more and more and more and more factories operating this way. Because if we're going to compete with China and the rest of the world on the large scale manufacturing, we need to get faster at this. And part of the lag in western manufacturing is exactly that. To your point about the board constructions of a kid's playset that aren't quite right, what happens on a line then is it's like, then the line is stopped at Station 7 because the Station 7 instructions. So then you just get like, okay, well then it's stopping here, stopped at 6, stopped at 5, stopped at 4. Because that guy is like, I don't know what to do, I can't fit this. So then extrapolate that out across industries. And there's just so much wasted time and effort that gets spent on problems like this that I think can be solved with technology. It's a perfect application of where AI can be very powerful to help accelerate the development and the shipping process. So that's one point. And then the other point I want to bring up is that we've seen this very fascinating and honestly un. Unpredicted outcome of our work with Dirac is that by talking to our customers, so talking to the Navy, the Air Force, et cetera, about how we are going to scale manufacturing and talking to them about how we go from designs into PDRs and CDRs and MRRs, the whole engineering process to get something designed and fielded. We've been specifically pitching Drock's solution as part of that ecosystem to help them understand exactly this question about how we get from that idea to that instructions to the factory floor to help us accelerate.

And it's been pretty good ahead of sales.

In a bizarre way, it's been an accelerant for our sales to help our customers see that we're leveraging this newfangled technology, this AI, because the government customers are like, what is this AI stuff? How am I going to use it? And we come in and say, here's how you're going to use it. We're going to use it on Autonomy on your end products, you're going to use it in the manufacturing line, you're going to use it in the back office, and here's how it's going to ultimately get you better products fielded for cheaper.

I want to talk to you guys about the labor market in manufacturing specifically. Are you guys going. There's so much demand for probably type of talent that could do manufacturing or they could go work in a data center development project. Are you guys competing for talent with the data center build out? Is there enough people, enough people excited about entering this workforce?

So there's actually two pieces to talk about there. Interesting thing to bring up of data centers increasingly, rather historically, the way the data centers have been built out.

Right.

It looks a lot more like architecture projects. And so you've seen the tooling that people use to design data Centers historically on AutoCAD, Autodesk, which is a little bit more of the architecture, they're doing.

It in bim, which is like building a building.

Yes, exactly.

Throw a bunch of racks in there.

Historically, yes. But now something that we are seeing as a company, through the companies that we're working with, increasingly more data center companies coming to us and saying, hey, we need to start building at scale the same way we produce electronic systems at scale. And so I, you know, we are seeing a massive migration of data center producers and data center builders away from traditional architecture oriented design and build systems towards the more manufacturing oriented design and build systems. And with that they're saying, hey, we are going to have to build data center systems all over the world and we cannot have folks not having the context of what to build. Can Dirac's work instruction software be useful for this? And so of course we say yes. And so that's one thing about the data centers, just as an interesting anecdote that might be interesting for folks to know about that we are also trying to build data centers the same way we build cars and planes and everything else. And then on the labor question, another thing that we think about is radically reducing the barrier to entry for net new folks into manufacturing. Because, you know, pictures worth a thousand words, but an animation's worth a million, right? Earlier you were asking about how do we change the modality of the way a person interacts with the work instruction with our software. It is not just a picture, not just a photo, not just a printed out thing. It is a 3D interactive, animated, dynamic work instruction that a operator technician can interact with.

Right.

Pictures worth a thousand words, but an animation's worth a million. And so because you were able to have less skilled folks potentially take the roles of operators, and because you're also codifying and aggregating all of that tribal knowledge and context, we're seeing that the companies that we work with are actually able to hire less trained technicians. And so it's easing that burden and pain for them.

On the anduril side, I'd add a couple points there. First is that on the data center side, in the construction market it's becoming pretty contentious just because, you know, big huge builds made big, huge bulldozers and concrete layers and electricians and etc. So on the construction side, these guys.

Are basically competing for the same kind of firms for the time of firms that are getting offers from data centers that maybe have a different so on.

The construction and build kind of side of things, yes, that's certainly getting competitive. On the actual technicians on the factory floor. We haven't seen pressures from data centers because the dot I would add is that these data centers, while very big and very powerful, tend to employ very few people. There's not that many people engaged in the actual day to day running of the place. So for us on the technician side, we don't see much competition from them. And where we are spending a lot of our effort is investing into the local communities, including in Ohio, through partnerships with the local university and trade Schools and community colleges like that kind of ecosystem to work with them on the types of skills that we want to see new new hires have, whether that's additive manufacturing or welding or particular CNC programming or just like electronics assembly, whatever that skill set is helping design curriculums that can then produce graduates of those programs that are ready to come get a job at Anduril. So for us, it's kind of a, kind of a long play because, you know, we invest that time up now and those graduates won't come out for a couple years, but in the long run, we think will lead to a pretty fertile supply, to say nothing of being good for the communities that we're engaged in.

What are you seeing within the armed forces on the self repair side? There's been a couple press releases about being able to repair drone or something on the fly. I remember I toured a aircraft carrier and they had a machine shop and there was a, a sailor in the machine shop who was talking about if, if something breaks on the ship they built, they will go and mill the bolt right there. They don't go and buy it because. And he was very proud to say, you know, we're saving the taxpayer. They want to charge us $1,000 for this bolt. I made it for five bucks. That seems like a really positive thing. Does that play into automated work instructions at some point? Some digital solution? And then just what are you seeing on the overall repairability from the systems that you'll be deploying and are deploying?

Interesting. So, yeah, funny enough, maintenance repair is a very, very common. Next question. When folks ask us about work instructions, right? Yeah, most of the time when they're saying, hey, awesome, you can now automatically generate my work instructions for me. I can now have operators and technicians interact with it in a 3D manner. Awesome. Now can you do maintenance and repair instructions? And our answer is, of course, yes. MRO is a natural inverse generally of actually the initial build of the thing. And so part of what our roadmap is, part of the things that we're working on with a variety of different folks is automating, maintenance and repair instructions. So we're very big proponents of. You were mentioning the right to repair not just of tractors, but of drones out in the field. It's a very big thing that we've been talking to a lot of different folks about. And it's something that we're a big proponent of because we don't think that these should be two different domains of instruction sets. Context should live on the model. Context should live in the hands of folks who are actually designing and building those systems and repairing them. And so you should, at the time of making your first instruction set, have the ability to codify all that context into one core instruction set. And maintenance and repair instruction should fall naturally out of that. And that's a lot of how we think about things.

Yeah, that makes a lot of sense.

You know, from. From a policy perspective, where I think the lines get a little bit blurred is that we're talking about a bolt or a bracket or a fixture or something that you can carve out of metal. Makes, makes obvious sense in one area. And then the next area would be some, somewhere in the middle of, you know, rewiring or reconnecting, you know, an additional battery pack or something like that that's kind of in the middle. And then all the way on the other end is like all the way down to the chip level broke.

You're not going to fathom.

Yeah, you're not, you're not laying, laying silicon, silicon chips out on an aircraft carrier or something. Right. So there, there is a spectrum there. At which point, at some point on that spectrum you say like, actually I do need to call the experts to help me on this. Right. But, but in my opinion, the, the, the default of saying, don't touch it, we'll come repair it, ship it all the way back to our factory. That's just not going to fly in any time of, of conflict because like, we do need our sailors and our airmen and our soldiers in the field to be able to at minimum replace the consumables, you know, the brake pads, the tires, et cetera. But then also to do kind of the next level of repair into replace a fuel pump or something along those lines. When it gets down to the chip level though, you're like, I'll probably send you a replacement module. And then maybe the work instructions say, take bad module, output new module in, reconnect these cables. But in general, yeah, we want to be as friendly as we can be to this forward repair concept because in a time of conflict, you just think like, I need the thing now. I need.

You said brake pads and mine went to motorsports because I know Ander rolls sporting nascar. And then I was curious, are you guys working? Has there been any inbound from any racing organizations that are having to make parts on the fly and are in heavy R and D and maybe they only make the part? Like maybe they're making different variations of the part, but they're not actually doing it at the scale because it's not a manufacturing business any. Any opportunities there.

So funny enough, I cannot confirm or deny explicit names, but we are actually in conversation with several motorsports teams who are actively evaluating us for use cases on and off the track.

Yeah, probably like a very different workflow, but so many learnings and just great for the brand whatnot. Exactly.

Work there.

Exactly.

Yeah. How are you reflecting on the. The DoD or Department of War budget potentially being increased?

I still say DoD too. I'm trying, but it's, you know, years ago.

Yeah. I mean the post saying that we might go to 1.5 trillion, that seems like that was not Anduril's original mission. Anduril was all about doing more with less, but at the same time, if it accelerates the mission, that could be good. How have you been processing it internally?

Well, I mean, I think anytime a customer says they want to spend more on the things you make is generally a good thing for us. That said, exactly to your point, like, we would love to see this increase in funding if it happens. Of course it's got to go through Congress and all the usual, but if it happens, be spent smartly and on new technologies, new approaches and not just kind of funding the old way of doing things in the old kind of legacy models that in our opinion hasn't worked for decades. So yeah, of course, you know, customer spending more money is good for us, but we're going to be lobbying pretty heavily to have them spend it on more AI powered, you know, weapons of futures, autonomy, that kind of construct. And especially on a different kind of contracting model, more towards the cost plus type of model. Sorry, away from the cost plus type of model to the firm fixed type of model that.

Yeah.

That we like to do business with so. Well, still a lot to digest there, but we're in general pretty excited and.

As part of that initial flipped model of raise venture, take risk capital, build a prototype, build something, then show up to the Department of War with a concept or something that's maybe ready to be fielded and then say buy it. Has that held as long as you thought it would? Do you think it holds forever? Because at some point if a customer shows up and says I'll pay you to do the R and D, what are you going to do? Turn them away. But at the same time there's some, you know, there's the financial hazard that the firm was founded around. So how's your thinking evolved around the whole flipped model and doing R and D on your balance sheet?

Yeah, a couple of thoughts there. First is that we spend an inordinate amount of money on our own internal research and defense, irad type research development rather than spending just an absolute inordinate amount of money. And we're grateful to have the support of our investors who believe in this business model. So for us we haven't not at all seen a like let's just shift back to that way. It's a lot capital, you know, less, less risky capital, safer, all of that. So no, I'm not surprised. Like this is to your point. This is exactly the kind of thesis, the mission of what we're trying to do. If anything we've been very happy to see venture capital funding to the SEC to sector increase to fund more companies approaching it this way. Because I think that for us to win a great power conflict, we're not going to need one Andoril, we're going to need five or eight or ten of them. So we're excited to see new companies get started with this kind of mindset and go out and chip off parts of the ecosystem here. The last point I would say is that to date something in the neighborhood of 85, 90 ish percent of our revenue has not been from these kind of cost plus traditional iris like customer funded models. The vast majority of it comes from us selling our product to our customers at a fixed price and then the bit of recurring revenue tied to that for kind of think of it like an extended warranty kind of concept where we are doing the maintenance, we're providing the support, we're doing the security upgrades, the patching, the all of that which from our perspective really aligns incentives for us to deliver the best possible product. Because to be blunt, like we will have a higher margin if we don't have to send repair guys out and keep replacing things. But then also gives the customer a more reliable product and then importantly gives the customer predictable pricing so they know what their total 4 year, 5 year cost of ownership for a product will be because they know that within that number we're responsible for anything that breaks any sort of problems or maintenance that needs to happen. That aligns everybody. And we're seeing a broader acceptance of that business model within both the Dow and with dhs. So we're pretty excited to see those messages landing.

Yeah.

Give us an update on the secondary wars.

Spicy.

Are you the head of investment relations?

I'm not, I'm not, I'm not.

Alison Lazarus is our the guy she.

Calls to live smackdown.

To be honest, at some point there's going to be some enterprising young SPV slinger. That's like, if I make a really heinous memo, Matt Grimm is going to.

Pump it, the grim Reaper is going to come to the top. So give us some backstory first and then.

Okay, so first I would say that our investor relations is run by this incredibly talented woman named Allison Lazarus. She's great. She works closely with Trey on kind of managing the cap table and all the inbound and interested investors and all of that. So not really my turf anymore. That said, I've had a long running battle against what I would call the Wildcats in the secondary market, who have absolutely zero respect for any sort of control of the cap table, like who our investors are, information rights, any of that. And they're just out kind of slinging offers at early investors, early employees, at share prices that are completely insane. And those people don't care whatsoever about the impact that has on the company. Whether that's around 409A valuation or RSU pricing or any of those kind of internal dynamics. They don't give a shit. So they just kind of create headache for us by then, us having to field this inbound. So beyond the headache, the point that I'm really worried about, and I think the SpaceX IPO will be very, very interesting because this is going to be a situation where there are, you know, the tide goes out and there's going to be a lot of people without trunks on. That's just the people have been out there slinging, in our case, access to Andoril or I have limited access to their next round at some insane price that is nowhere close to what the actual price is as set by the market. And then people are buying them and.

That'S not even factoring in the fees, which are sometimes kind of crazy layered fees.

So there was this one situation of someone slinging an offer around that was an SPV into another SPV that was buying a chunk of an investor's, an early investor's holdings. So there's like layers and layers of, of fees and carries on top of that at a price per share that was something like 70% above what the last round traded at. So it's just an insane thing where they're pitching access to a deal they don't have. And I know for a fact that there are. There was a recent indictment announced in New York around a case where someone was out slinging Anduril SPVs that did not have access to Anduril and the guy was just pocketing the cash and then he was just Indicted and arrested at jail, jfk. So like this is a thing that happens. And the reason I bring up SpaceX is not to create more work for them, but that the world of SpaceX.

Trillion dollar IPO because the legal fees of unwinding all these wild.

Right. Because there's, there's just, there's a lot of this nesting. And SpaceX is a fantastic company that's been in business for a long time. So there's so many of these nests.

Of who owns what types of like Main street investors. It's totally possible that they're thinking, well, I got into SpaceX at 200 when you didn't. And then it goes out at 1.5 and that's my retirement. And then it goes out and you're like, where are my shares? And the guy skipped town.

And the guy skipped town just out.

Whatever you put in. And not only did you lose the whatever you put in, but you also lost the mental of where you thought you were and what you were planning. You might have been planning a certain lifestyle, of course.

And the opportunity cost of deploying that capital somewhere else too. Right. It's a very big problem. And I've used this joke before, but my joke is that how many investors in America think they own a chunk of SpaceX when they're actually funding their ex roommate's boyfriend's coke habit in Miami? And it's like probably a non zero amount. Right?

What do you think? What do you think the solution is? Can the community police itself?

It seems like obviously like you have a legal team that's working on this stuff, but there is an element of like going direct, like you need to get the word out that this is happening. And that's important too. But yes.

So this is, so this is why I've. We as a company have been more public about this is because we're specifically trying to send a message to the market that's like, we see you, we're aware of you. And like secondaries of course have their purpose. Of course there are early investors who need to liquidate funds or fund life cycle problems, of course. So it isn't a blanket comment that all secondaries are bad. They're not. They're not. There's a category of these people who just have like really no respect and are basically just fraudsters, hucksters. And what I think needs to happen is two things. One, the investor community needs to be better at policing ourselves around just who these bad actors are and kind of shining a light on them. And then second Sadly, I think it's going to take law enforcement action, whether the SEC or the FBI. You know, thwacking enough of these folks that that community starts to see like, this isn't good for me. And it just kind of puts a damper on that side as well.

Last question for me, Adam Porter Price has been on a tear doing deals, buying companies as coo, the best in the biz. As coo, what is the key to a great post merger integration?

That's a great question and an often overlooked part of doing these deals. The deal makes a headline, busy popping.

Champagne, and then you got it. Get to work.

Adam logs more airline miles and more and more iPhone minutes than me and Brian, than Trey, than anybody because he's constantly out there. The best of them is doing this. So from a post merger perspective, there's a ticky, tacky piece of it that is important, which is payroll, benefits, equity ownership, all the 401k transitions. The email has to migrate, but maybe.

Making it sure it happens on day one. There's no, oh yeah, you'll get your paycheck, but it's a delayed a week. That's annoying.

Correct.

No one wants that.

Correct. Nothing will burn your trust with a company faster than screwing up their first paycheck. So for us, it's a thing that we try to lean as heavy as we can into that employee education and systems transition up front. It's not possible to transition like ERPs and years of design software that takes usually 6 to 9 to 12 months on the long end to get those systems fully working. But like the guts have to work first and then second is, I would say is a cultural alignment that's just like how we pitch our products, how we price our products, how we plug into the kind of go to market ecosystem and engine. Like we have to, we have to get that right. And then the third piece I would say is trying to get some wins pretty quick. Because if you're, if you're an acquired company and you get integrated and you're like, who are all these people? This is different. I liked it when we were smaller. Yeah, yeah, exactly.

They're in scrappy mode and then they're like, hey, I'm gonna set this.

I saw the campaign. Don't work at Anduril. Why would I.

Just don't work my patience.

So kind of aligning with the, what are we, what are we trying to build? Why are we building it? What's the goal? What are the deliverables we're trying to hit, like getting that Kind of alignment for folks to see, kind of see the destination, see the lighthouse on the hill and say that's where we're going and aligning folks to that. And it's, it's those second two that are, that are really hard because that's much more kind of personality based and kind of relationship based. So it takes a lot of investment from the leadership team. Takes a lot of investment on the engineering team especially. You know, how are you designing your products now? What's your release cycle like? How do we get that into Android's kind of way of doing business? There's just a big transition there, but.

It seems that's exactly right. There's been this maybe kind of sentimental, intimate or joke shoot for nicey. Even if you miss, you'll land at Anduril. You know, this idea of like, you know, maybe, maybe I don't build a multi billion dollar company as a new like Neoprime, but I have, you know, maybe like the not the next outcome is I land at Anduril. Can you reality check that? Because I feel like in practice. Yeah, yeah, yeah. No, no, no. And I'm just saying I think it's like for founders to understand, understand like what is the actual bar. Because you guys every, every company that's built a great product but maybe the business side isn't like fully working yet and they're thinking like, okay, like my product's great, our team's great, I bet we could fit in there. Reality check that.

Yeah. So the reason I'm pretty dismissive of it is the. That implies that Anduril will acquire anything or any company that's struggling or not.

Because Adam's logging all these miles. But he's not exactly doing a deal with.

No, we do like two a year on average if that. Right. So he's saying no a lot.

A lot.

But what we look for in acquisitions is really a sweet spot of two things. One is a very strong engineering team. They have to have something that's very compelling and very unique and very interesting. And then the second is a area where we can add a lot of value or accelerant or fuel to their go to market.

How do we speed up the timeline? Exactly.

And you guys talk a lot with defense founders and you've had a lot on. And folks will consistently talk about how hard it is for the go to market engine. And there's a lot of nuts and bolts around that, that get underestimated around what formal government proposals look like, how to do formal government contracting, security clearances. The government relations kind of lobbying side of this to be. To be a player on these bigger deals, then relationships within your customer that are. So you're not just pitching the guy flying the drone out in the field, you're pitching the whole chain and how it fits into their ecosystem outfits in the program. There's a whole lot of that infrastructure that, to be blunt, like, we're pretty good at, We've gotten pretty good at this. So when we find a company that has an incredible team building a really cool product that's like really unique and interesting, but is maybe struggling there, they haven't seen kind of the ramp that they want. And we see that and say, hey, look, we can help with capital, obviously. We can also help with this go to market engine. We've built this legal engine infrastructure, this facilities infrastructure, this government relations infrastructure, all of that that will then take your product and can help really accelerate and get it deployed both to the US and to friendly governments around the world. Like, that ends up being the sweet spot. So the reason I would say that the base question that you started from is just completely off base is like, we're not going to go look and try to find a company that like isn't clicking. They don't have a great product, they don't have a strong team and they haven't sold a lot and then be like, yeah, let's go get them.

Couple deals a year. It's like, usually like you're going in and being like, we want to buy this company.

Yes.

Less like, hey, like, things aren't working that well, we should try to sell.

Yes, exactly. Exactly. Exactly. Right. And the last point I would make on that is that to date we've done exactly zero turnarounds. And I think this is a kind of a mental model for us is that like, in general, we don't really want to be in the turnaround business. So take a, take something that's struggling and come in and fix it. Like, deploy our team of operators to go fix it. Like, there are plenty of PE firms who are very good at doing that. There are plenty of other organizations who are honestly better positioned to do that for us. We really look for that strong team, strong product, and then where we can really accelerate sales and go make an impact.

Very cool.

Do you have a favorite? I know I don't want to put you in hot water.

There's no chance I'll answer that question. I love all of my children equally. Yes.

So which one do you like the most?

I think we know there's a great company we acquired in Ireland that was called Class, and they make these absolutely incredible computers and networking devices.

When I saw that, I was like, I want one of those.

Yeah, they're great devices, great team. But it's headquartered in Dublin, which is very far. So a lot of time spent over there. And that acquisition is to going, gone, gone. Absolutely swimmingly.

Couldn't be happier.

Well, thank you so much for coming on the show.

Thank you guys for having us.

Appreciate you.

Hang out for a second.

Leave us five stars on Apple, Podcasts and Spotify. We will be live tomorrow at 11am Pacific. Sign up for our newsletter@tvpn.com and thank you. We will see you tomorrow.

See you tomorrow, folks. Have a great MLK day. Thanks for hanging out.

Goodbye.

Cheers.