TBPN - DJI Suspicions, Stargate Turmoil, OpenAI Operator, F40 for the Mind

Episode Date: January 24, 2025

TBPN.com is made possible by:Ramp - https://ramp.comEight Sleep - https://eightsleep.com/tbpnWander - https://wander.com/tbpnPublic - https://public.comAdQuick - https://adquick.comBezel - https://getbezel.comPolymarket - https://polymarket.comFollow TBPN: https://TBPN.comhttps://x.com/tbpnhttps://podcasts.apple.com/us/podcast/technology-brothers/id1772360235https://youtube.com/@technologybrotherspod?si=lpk53xTE9WBEcIjV(00:00) - DJI Deep Dive (01:22:13) - Stargate Turmoil (02:07:05) - DM's (02:13:37) - The Timeline

Welcome to Technology Brothers, the most profitable podcast in the world. Today we are breaking down the history and controversy surrounding DJI, the drone maker. You're probably familiar with these guys. And looking. Not good. We don't like it here. And looking to create some new controversies. Create some new controversies.

Amp up the volume around it. Yeah, exactly. The reason we're doing this is because, A, it's a super fascinating story. It hasn't been told anywhere. Like, we couldn't find, when we were doing the research for this, we couldn't find the, like, oh, there's a definitive podcast. Oh, somebody did a documentary about them. Like, there's very few articles. There's a, there's, yeah. The primary video sources on it are all AI generated. Yeah. Just slop content. It's really, really weird. Didn't even seem like it was the audio was in

English, but it, what didn't seemingly wasn't generated by somebody who spoke native English, because they were weird accents with the audio. Yeah, it's very, very odd. And then of course, there's like a lot of drone operator content, like people reviewing the ins and out. of like the Inspire 3 how it compares and like that's not relevant to this story yeah and and frank wang the founder like hates journalists very rarely yeah i don't put it that way uh i think he just knows that if he no there there was a he he's he's had some he's had some lines where he's like i won't he doesn't want to talk to it so hating journalists is different but doesn't want to talk to the press yeah has gone years without doing it yeah it goes at different points well at one point in it's

We'll get to this early in the story. It's all just like hero stuff. Like, oh, it was used to save someone who was trapped in a cave or tapped on top of the island. And now it's like, yeah, it's being used in every war front. Yeah, yeah, it's used in every battle zone or battlefield. So I think we should kick it off with the Ryan Mac article from 2015. Yeah. Or you want to get some background first.

We should do some background. Okay, let's do that. And then let's go into the Mac piece. So yeah, Frank Wang born Wang Tao is born in Hongzhou, Chejang province, China. From an early age, he shows a keen fascination with flight and aircraft. And one thing, I highly doubt he was given the name Frank Wang until he was probably in his teenager, at whatever point he started interacting with the English world.

because he would have just gone by Wang Tao his entire life. So at this point in time, let's be very clear, his name is Wang Tao. He was born in Hongzhou. And anyways, Hongjo.

I mean, everything about this feels like somebody who would, you know, be working for Nat Friedman or get a Teal Fellowship. Like,

even though he's on kind of the near peer competitor team, and so you know we're we're we're going to figure out how to how to build our own DJ I here in America hopefully with some cracked nat Friedman or teal disciple yeah he does just seem like like a hacker kid who's just yeah yeah yeah and there's two there's two things that happen and there's a different world where where like we would be like promoting this guy and be like you got to hire this dude you got to fund his company because like he's he's clearly a genuine hacker he goes to Hong Kong University of science well before that there's two there's two things that happen

early life. He gets this book. I forget what it's called, but it's a story about this helicopter adventure. And it's all about flight and just humming around. So this was like, you know, before he was 10 years old. This was like an early obsession. And then as a teenager, he gets an RC helicopter. Do you remember these things? They were, yeah. I remember them. They were really hard to fly. They crashed all the time. They were super fragile. And so he had a some moment where he did well in, you know, some exam and his parents got in the RC helicopter, he crashed it the first day and then had to wait three months to actually get the parts to be able to get it back up in the air. That's amazing.

I think the very first prototype that DJI built as a company was a single, what do they call it? Like single rotor or just like a helicopter, basically. Yeah, yeah, yeah. And the other thing that's crazy before he ends up going to university, his parents, like had him living. he didn't live with his parents. He was like raised by family members. His parents were originally one was like a teacher and then became a small business owner. His dad was an engineer.

And so he basically, you know, had, I'm sure he had some trauma associated with with that. Just being a young child away from your parents. Adjust the mic a little bit. So he goes to Hong Kong University of Science and Technology, studies electronic, electrical engineering, becomes involved in robotics competitions and stands out for his interest in flight control systems for drones. He also wanted to go to MIT or Harvard

didn't get in. And so there is another path where he becomes basically an American. Yeah, just gets super America-pilled. Yeah. So we lost a good one. We lost a good one here. Yeah.

What's interesting is, I don't know if you remember, it might have been a little bit, this is a little bit earlier, but in like 2010, there started being those viral videos of like Boston Dynamics, but there were also a set of viral videos from MIT's Media Lab and from MIT about drone, swarm, coordination. You see these really cool demos. It was basically a precursor to those like, those like fireworks shows, those light shows that

we're seeing. But the tech just like never got transferred for some reason. Like MIT guys were doing this and then they just didn't build the startup because I think we'll get into why because maybe there was a zero cost competitor that was really cheap and just crushing them at all times. Yeah. So really weird market dynamic. But yeah, this was like 2003 to 2010, I think, was when a lot of the foundational kind of algorithms were written for flight control systems.

And really the discovery that when you put four rotors on a helicopter essentially instead of just one, it becomes way easier to calculate how to balance the craft. And so when he's at HKUST, So the Hong Kong University of Science Technology, he joins this team and participates in the ABU Robocon competition, and he wins third prize among teams across Asia. He gets $18,000 in Hong Kong dollars, which is about $2,300 and U.S. dollars to develop drone technology. This funding, along with his growing reputation, encourage him to pursue drone research more seriously. So he founds DJI in 2006 with a support from his mentor, Professor Lee Zizhong and a family friend. Lou D. Frank Wang formerly founds DJI, Da Jiang innovations in Shenzhen.

Early operations involved selling flight control components. In Shenzhen, I think most of our listeners will know, but this is like the industrial technological capital of China. So it's not the center of like business, but it's the center of manufacturing, gay old manufacturing, consumer hardware, etc. And I think I think the like everyone knows like, oh, iPhones are made there. They have Foxcon. Foxcon employees like millions of people. is this massive company. But the really important thing about Shenzhen,

is that in America, it's like if I need aluminum, I go to the aluminum supplier, and they're in Idaho, and they ship it to me. And then if I need a little motor or a printed circuit board, it all comes from all over the place. In Shenzhen, it's like, you walk across the street and there's a guy that can do this type of machining. And then you walk down the street,

and there's a motor company, and then there's a battery company. And so everything happens really, really close. There's a ton of things. And it seems like he was able to just kind of go to like the Silicon Valley of manufacturing. And it's much like when you're in Mountain View doing YC and, oh, company's failing, but they have a good iOS engineer. You just pull them onto your team. It's the same thing.

Like he started with just selling flight control components for drones. He wasn't even selling drones. He's just like hustling. And we'll get into this later. But there was his biggest competitor in the U.S. failed not because of software. I don't actually know if they failed. We should look it up.

Which one are you have? 3D robotics. Oh, yeah. It's like 175 million. And then they eventually just conceded on the actual hardware front. Yeah. We're just building software for DGI. Yeah.

You also, one thing, I think they wind up pivoting to industrial, basically. Yeah, yeah. So one thing that is fascinating here is, is Frank ends up over time really churning through employees. A lot of people are leaving within a year or just frustrated with him. But his mentor, who is his professor at school, Lee Yes

Jijong and a family friend Liu Di who I believe was one of the first investors they end up forming this inner circle and both the professor and the family friend

become billionaires off of DJI Yeah so and and Lee Jijong ends up becoming chairman or has like interesting

has a significant board role Yeah and so I mean in the early days 2008, I mean, 2008 was still, wait, Obama just got elected. And so this was like peak, like globalism. The internet will turn China into a democracy. Yeah.

Everyone, we're all friends and you know, China's joining the world trade organization. And so there's really nothing but positive news. And it's just like, oh, cool. Like there's a new awesome company doing cool tech stuff. It doesn't matter where they are. Of course they're there because like, yeah, China's good at manufacturing. We're good at other stuff. It's fine.

We'll trade. And so all the initial stories are just about like genuinely great impacts of this stuff. Like during the 2008 Sichuan earthquake, DJI drones contribute to the relief efforts by capturing aerial images, assisting rescue and recovery teams. And we kind of saw that recently with the LA fires where people were putting up drones, sometimes bad and crashing into planes. That was really bad. But some of the aerial photography is genuinely very helpful to understand where the fire is sweating. And I'm sure like now. And it's a hundred times.

cheaper than flying a helicopter. For sure. And so the company was still pretty small and largely reliant on Wang's determination and external funding. In 2010, they start expanding. So he hires his high school friend, Swift Z. Ja, to head DJI's marketing efforts. And then they pivot to cater more actively to the growing community of drone hobbyists around the world, preparing the ground for broader consumer adoption.

Yeah, so early on, some of their first products were some of the best in market, but still clunky. Like now you can buy a DGI, get it, have it in the air, like probably within 20 minutes if you were speed running. Oh, yeah. Very simple to set up. It even integrates with your iPhone, the native iOS app. Yeah. But early on it was still there were these hobbyists, people that were just obsessed, you know, with flight, but doing it in a, in a more. more, you know, just not doing planes.

Yeah, the big tipping point for the consumer moment, I think, was when the camera feed became live. So you could see where you're going from the drone's POV. And that actually happened through this integration with GoPro, where you used to have to, you used to have to buy the drone and it would give you the motors and the batteries and a controller that can fly it. But you would have no idea where you wouldn't be able to see the drone it got out of your line of sight, basically.

It was very hard. Yeah, and you could crash it. And so if there's no camera on the drone too, then you're just sending it around. Exactly, what's the point? Just kind of hanging out and flying. And so people figured out that they could, that they could attach a GoPro to it and then record.

And then when it got back, they could unload that footage. And then eventually people figured out, like the hackers figured out, how to live stream footage from the GoPro. But then I think it was the DJI Phantom 3 had a GoPro integration that would let you do that and then the phantom four had yeah i remember i remember even on the as people started

hacking these and adding functionality to them that that dj i would eventually do themselves i remember at this time of being into surfing and uh snowboarding skiing things like that people would that community really picked it up because it was like hey we can get before if you wanted to get capture a big wave you had to have a helicopter in the sky and what's impossible can afford yeah no except the top like whatever 1.01% yeah all my extreme sports buddies were super into this yeah just it blew up overnight and so that was the first that was kind of transitioning from the hobbyists like people that just wanted to fly to people that were using the and then the pov go pro pro videos were going viral too because this is the time when youtube's growing really fast and you know any back country skier would just amount to what year what year did go pro pro go public uh i i feel like ben could you look up uh can you perplexity that Thank you. They went public in 2014.

Do you know how much GoPro's worth today? Like a couple hundred? 150 million. Yeah, yeah, yeah. Rough. Yeah. Rough. And I, and there's a bunch of reasons for that, right?

Like, it just got to the point where people were like, okay, my iPhone's pretty durable. Yep. It's 80% is good. And DJI has a competitor. DGR. has a, no. Yeah, yeah, you will get into this, but DJI has a competitor. to every single DJI product, but GoPro does not,

DJI has a competitor to every single GoPro product, but not reversed. And as we'll talk about later, DJI sells everything at cost. Yeah, it's wild. So we should talk a little bit about the 2011 dustup in DJI America. This is pretty interesting. Yeah. So Wang needs an America guy, right?

He knows that America by this point. the hobbyists have started adopting it he sees the early signs of consumer adoption he's like i need i need mr america of course he finds a guy who was a former reality tv contestant uh and so you meet uh wang meets this guy at a trade show immediately oppressed with with colin uh salesmanship and they decide to partner to form dGI north america um and this like sub is focused on the U.S. and just broadly drone sales outside of outside of China.

Yeah. The reality of you show to like actual power pipeline is maybe wildly underrated. Yeah. It's always seen as like very low status to go on reality TV. But Trump, Chris Williamson was a reality. Well, I've talked about this with with the players and people that want to join PMF or die.

I'm like, look, uh, this look at it as an opportunity not just build your company, but build your brand. Like you can accelerate you to that. whatever 5,000 X followers that can change your life even if your company doesn't work. Yep. And so they build this subsidiary,

but very quickly there's this fight between, and we'll get into this more when we dig into the next article, around control over the entity because of course DJI North America is owned by DJI, which is a Chinese company, which is, you know, Yeah. Even if it is not entirely controlled by the CCP at this point, it's like it's very hard to get capital outflows from the Chinese investments.

And so there's a lot of like tricky. Yeah. So basically, so I'm going to read a few paragraphs here. So by late 2012, DGI had put all the pieces together for a complete drone package, software, propellers, frame, gimbal, and remote control. The company unveiled the phantom in January 2013,

the first ready to fly pre-assembled quadcopter that could be up in the air within an hour of its unboxing and wouldn't break apart. with its first crash. That's kind of like referencing Wang's early helicopter, the RC helicopter that crashed. Its simplicity and ease of use unlock the market beyond obsessed enthusiasts. And I remember around this time kids started getting them

for Christmas. Oh, yeah. Like that. Yet things had already started falling apart between Wang and Colin Gwyn, the head of North America. DJI's founder didn't like that Gwynn was taking credit for the development of the phantom

and was calling himself CEO of DGI innovation. a title that still stands on his LinkedIn page. Savage. I mean, it is so American for the chairman of this company to come in and be like, we need you to be the face of this company. And then immediately the American is just like, I'm the man. I'm CEO. I'm CEO.

Let me take all the credit. Dorses also say that Gwen would often rush into setting up partnership agreements, particularly one with action camera maker GoPro, which would have been the exclusive camera provider for DGI's drones. So it sounds like, who knows, but Gwen is just, running around just like doing deals being loud and that doesn't really align with Wang that much

so Wayne got cold feet in the GoPro deal and went against Gwen's advice subsequently angering GoPro which is now and this was written in 2015 is now rumored to be developing its own drone so yeah so 2015 was really like the turning point I think that's what

we should wait but we should cover because in 2013 yeah this is important by May 2013 DGI attempted to buy out Gwen's stake in DGI North America offering DGI global shares that would have given the American a paltry 0.3% stake in the Chinese company. Gwyn was not happy with this, points out that his office's work in North America led to 30% of phantoms being sold in the U.S. So it was a super, you know, key market. DGI didn't leave room for negotiation. They locked out all of DGI North American employees out of

their emails redirected all customer payments to China headquarters. By New Year's Eve, the employees had been fired in arrangements for being made to liquidate the Austin office's equipment. And DGI still ended that year with 130 in top line revenue. So they get into a lawsuit. And this is all around the time that Sequoia is leading. It's funny because the Sequoia deal happened in mid-2014. Yep.

And they invested $30 million. at a 1.6 billion. So it's like, I don't even understand this investment, to be honest, because from that point, I mean, I guess if they're like... It was a fantastic investment. It went to $100 billion, right? Isn't DJI worth like $100 billion or something now? I don't know.

I don't actually know. Yeah. Yeah, yeah. But it's still privately held. Yeah, privately held. Okay. Ben, can you look up their current valuation? Valuation.

And anyways, so they get in a lawsuit. Yeah. Apparently settle. Nobody knows the exact. nobody knows the exact price at this point. But people have said it's somewhere around $10 million. So Gwen kind of got screwed. And sounds like a lot of the American employees did as well.

Yeah, even the point three would have been, I mean, if I'm right, about the $100 billion mark, would have been 300 mil. Yeah. But you could tell from some of the ways that Wang would talk about Gwen, he never really respected him. you just would say like oh yeah at times Colin would say something that inspired me which is just like can imagine the American being like dude this is going to be huge

and then Wang's like nice yeah and pat on the back type thing yeah DJIs market can't yeah yeah yeah so their last private valuation's like 15 so yeah okay Sequoia has it on paper somewhere around a 10x. They've turned 30 into 300-ish million. But that money's locked in China. Yeah, with their investment vehicle. But yeah, I mean, like the news and just everything about DJI starts ramping up from 2014, 2015, because Sequoia does the deal.

Then Excel comes in and leads an even crazier round, I guess, in terms of ownership, like being very low. They raised 75 million from Excel and it valued the company at 8 billion. So that's less than 1% sold. And that valued the company. They're raising a new round at 10 billion. And Wang owns about 45%.

So it'll be worth $4.5 billion. Yeah. And that's where it starts to get into. Clearly they were in Shenzhen. I'm sure they got amazing terms early on. But the narrative starts to get a little bit confusing because they also, Wang also talks about how they intentionally would sell their product at cost. Yeah.

They just were looking to recoup the sort of cost, the hardware costs. Yep. And so it's sort of unclear how they're scaling this because the thing about selling consumer products like this, if you're scale, if they're like 10xing, like, they're not 10xing every year, but they went in a very, very short timeline. from no revenue to a billion dollars plus revenue a year. So to actually achieve that scale, you have to be buying inventory out, you know, forecasting inventory. You've got to be buying, you know,

with that kind of growth rate, it's very hard to fully bootstrap it without huge access to capital because you're buying your Q4 holiday inventory in Q2. Exactly. And where's that money coming from if you're selling the product at cost, right? It's really hard. So there's a bunch of,

this article that we're referencing was written by Forbes. It was written by a guy named Ryan Mack, who's not very well liked in the valley. And he was also accompanied by someone named Hung Shao, which we don't have backstory on. But so anyways, this piece very much felt like a puff piece, right? I mean, I think no one was thinking about DJI in any sort of negative sense.

It's like, it's a, it's a really cool tech company that's building hobbyist drones for surfers and people love. Everyone loves it. It's amazing. It feels magical when you drive, when you fly them around. And then also they just, they just did back to back insane up rounds by the two, two of the top venture firms. Yeah. Sequoia and Excel. And so you're just like, all the, all the green lights are flashing on Mike.

But the Sequoia Excel rounds, the interesting thing is they were just tiny checks. Right? Like these are... Yeah, tiny ownership. It wasn't the normal, like, oh, we're taking 20%. Yeah, yeah, yeah. So they're saying, yes, we want American investors in because they could have access that capital of China or anywhere else.

Yes, we want American investors in. No, we're not going to give you any amount of ownership that would allow you to have any meaningful say or anything over the company. Yeah. And so at this time, there's also an American upstart that we should talk about. Let's start with the greatest threat to Wang's dominance of the consumer drone market emanates from the sun-drenched fourth-story office patio across the bay from Silicon Valley in Berkeley, where engineers for 3D robotics spent dozens of hours testing the latest code tweaks in their Phantom Killer,

the solo. Unveiled in April, the black drone whirrs and buzzes around the roof with the sound of a thousand angry bees as a 3D Robotics CEO Chris Anderson explains how his company is the Android to DJI's Apple. In admiring his quadcopters' elegance and simplicity, which took cues from the phantom, the affable Anderson explains that it is the software, not the hardware that is the key. Of course, because it's basically impossible to build a competitor to DJI in America just through the American supply chain. Even if you assemble it here, your motors, your batteries, everything's going to come to China. So they have to make the software argument. Unlike DJI's operating system, which is closed to developers, 3D robotics made its OS open source to attract interest of programmers and other companies,

such as dozens of Chinese copycats undercutting DJI's margins with even cheaper drones. If everyone is using our software, says 3D Robotics CEO, then we, not DJI, control the market. DJI started as a company back in the days when it was just a hobby for me. And to their credit, they accelerated brilliantly. Right now we're playing on their home field, so we're playing catch up. There were a couple companies that tried to make the software differentiation, Skydeo when they came out. Do you remember their drone?

It did have better software for like one cycle, where you could get on a bike, put it up in the air, drag a little circle around yourself and say, follow me, and it would follow you perfectly. Like, it would avoid trees and still follow you, and it would go under power lines and around.

And you could ride your bike from home to work. And I had a buddy Kyle who worked there at the time, and it was like the most incredible demo, Casey Nistat was pushing it. But it was like three times the cost of DJI drone. And then the next revision, DJI was like, yeah, we can kind of do the same thing. Yeah, yeah, yeah. And so that is one of the big problems with competing with DGI is from the beginning and from our analysis today.

They're still selling when you look at the total cost to deliver these products. They're selling them at less than it costs to produce them, right? Because they were oriented around how do we recoup our costs on the hardware, but the costs of everything else, right? like customer support, sales, all these things. It's very unclear how they were able to finance this. And if you're coming in, even 3D robotics by, I think 2020 had raised like $175 million, how do you compete with a company if your product is three times more expensive and not as good? Yeah.

And so I think just to take the other side of that, So, like, the tinfoil hat explanation is, like, they are deliberately undercutting costs to, you know, put effectively like spy drones all over America, right? Yeah. That's, like, the most extreme representation of that, of that. Or alternatively, they, like, the most extreme is to say they are a consumer, they are a Chinese military contractor disguised as a consumer tech company. And they could have tons of non-diluted funding from the government. That's covering losses. That's covering the losses that are happening.

Yeah. And we're not even seeing that. I think that's very reasonable. To take the other side of that, I would say that the idea of making no profit for decades is an effective business strategy. We've seen it at Amazon. We've seen it at Tesla. Tesla didn't make operating profits for a very long time. And I would not say that Tesla is certainly, you know, Tesla could be a company that manufactures tanks one day.

Yeah. If we get into that type of conflict, Tesla's are popular in China. In fact, you can't drive a Tesla next to the forbidden city because Tesla's have cameras on them and they don't want Tesla sending that footage back. So just like we should in theory stop DJI drones from flying around like the White House and stuff. which they just lifted the restrictions and said you can fly them anywhere which is the most insane to me so so yeah we'll talk about that in a bit um but uh but yeah it's funny because but what but my point is that there is there is a rational like it just purely economic non geopolitical explanation for selling a product at cost or below cost for a long time while you capture a market and and if you and let's say the djii was was was just like an American company, it would still be rational to say, let's just hoover up all the capital we can, sell below cost, get 77% of the drone market, keep everyone out, just keep killing startups, keep killing startups until we just are so dominant that no one can possibly

catch up. But then we can raise our prices. None of that, none of that information is available. True. We only know that they raise 30 million from Sequoia and they raise 75 from Excel. Both those firms were sell there's very possible than neither of them even have information rights true like it's a chinese company uh it's a chinese company they have very very little ownership yeah they have less ownership than his inner circle who has 10% here 15% there etc yeah so um so let's continue with 3d robotics to get into this competitive dynamics three robotics which has funding from the likes of qualcomm and sand disk is well into its game of catch up and has has moved most of its manufacturing capacity from Tijuana, Mexico to Shenzhen.

Gwyn, who is the company's chief revenue officer. So this is Gwynn's company after. So Gwynn was the guy who is at D.J.I. North America gets in the fight with D.J.I. Splits. He joins three robotics and Chris Anderson to kind of like fight back and build. I mean, we really saying, why does there no D.J.I. American D.J.I. Like this was the attempt in 2015.

And it was certainly a rough go competing with DJI. So they, so Gwynne is there. He's the revenue officer and he's exploring the same retail channels he built up with DJI. And he develops a partnership with to put gopros on 3D robotics drones. Wang dismisses their chances sounding like sounding something like the big kid on the kindergarten playground. It's easier for them to fail. He says they have money, but I have even more money. I am bigger and have more people.

When the market was small, they were small, and I was small too, and I beat them. Yeah, he's got some crazy. He's got some, so Wang is like, I get why this guy doesn't do media anymore. Yeah, yeah, yeah. He's not very well media trained. He also went on record around that time and goes, China has money, but its products are terrible. Its service is terrible, and you have to pay a hefty price for anything that's good.

And it can't be understated that DJI is the first consumer global. consumer tech powerhouse to truly come out of China. It's the only brand other than TikTok, which arguably was like, you know, not even a native Chinese company. Yep. To really get win over the hearts and minds of consumers globally

and get this level of market share. Do you remember the hoverboard craze around this time? 2015? So it was like this flat kind of skateboard with two wheels on either. side and you'd stand on it and it would balance automatically. And so it wouldn't work like off road. Those were like

ripsticks? I don't know. Oh, oh. This is the thing. Yeah. So this was the fascinating thing was that these products just were just birthed from Shenjin white labeled. And so there was no main brand. Like maybe Ripstick was the main brand. No, no. I'm thinking a swagway was one

of them. You're thinking of the one wheel type thing. No, no. No, it has two wheels. It's I'll pull up a picture. But but people would just call it a hoverboard and it had two wheels you'd put your feet on it and the wheels would be to the left and the right of your of your feet and you would just kind of balance on it so it was this do you remember this thing oh yeah yeah no one the numbers came out of nowhere and and and and and they had try to pull up a big picture and

show the camera like because like I remember like soldier like it would be like rappers would just yeah ride them around everywhere so this one is like Go tracks, which is a brand. It was like the classic like, get your teenager of this for Christmas, and they'll ride it around a little bit. It didn't really work on like city streets or anything. It only worked on like polished concrete floors. People would ride them around in like offices.

I think it was probably featured on HBO Silicon Valley at some point. Yeah. It's very like, we were, drones like, Wang faced a ton of competition internally in China. Totally. Everybody was seeing how much revenue it was doing. Because the thing about Shenzhen is people on the, ground quickly understand a lot of factories can do the same thing yep if you see a factory that's just

got thousands and thousands of employees and is just buzzing you're the factory across the street and you're like well we should get into this business too so they all it's highly highly competitive people copy each other yeah there's tons of corporate espionage wang deals with this a lot he even calls uh shenzhen a dog eat dog society which his words not ours what um um he's not calling it he's not commenting on the rough and tumble over there yeah he's just saying it's a little rough and tumble

yeah he says that the boutiqueness of the market always gets driven out so oh no that's a gardener analyst actually okay um and so yeah I think as opposed to these hoverboards drones have the software hardware this is full stack complexity the cost that make it not the easiest product

to if DGI is selling a drone for $700 and they're not even making money on it. Yeah. Why enter that market if it's, you also have to deliver all the software and this consumer experience and, um, yeah, I mean, DJI built a good brand in America. Yeah. And, and then, and then super great integration. I want to go through this guy is like, this guy is like founder mode to a T,

doesn't want to share the skies with others intent on maintaining DGI's, uh, lead as drones expand into commercial applications, construction and mapping. And he's saying you can't be satisfied with the present. There was a good story. I thought this was funny. At one of the Phantom launches, there was this big, they had this huge keynote and he no-showed the keynote himself as the CEO. Because he wasn't fully satisfied with the product. He was just like, I'm not going to go. He didn't want to associate. He was like, it's my company and didn't want to associate with it. And this is the thing I keep coming, it's like, you know, requires a little bit of intellectual flexibility because obviously like this is now like a rival military company. So, but like I feel like I would really

like this guy if I was on his team. Yeah. And I would just be like this is, this is. Well, I don't know if he would like him, but you'd respect him. Totally. Totally. He clearly never cared about being liked. Yeah. Yeah. He would, you know, talk poorly about. Yeah. But I would be impressed by him and I'd be like, oh yeah. Like, okay, this guy's like like, like clearly got. Yeah. He would. You know, he would talk poorly about. Yeah. Yeah. But I would be impressed by him.. clearly got some really insane entrepreneurial and technical talent that's like super valuable and worth yeah self-flatable like you know clearly studied you know electrical engineering and mechanical engineering and mechanical engineering in school robotics but then like clearly just kept studying it and being that obsessed with product never resting on oh the phantom is like the best drone in the market

it's like cool we have to make yeah phantom too even better and so i mean yeah dj i goes in the tear launches more and more products. I want to go through some of the timeline on Axe formerly Twitter at this time. So in 2020, Flo Crivellio, who's been on the show before, says, what's DJI's moat? And this is an interesting question

because I think there's actually two different questions around the moat. It's what is their moat in China and then what is their moat in America? Because they are somewhat different. And I think it definitely comes down to software integration and now integration between the different hardware

platforms because there are actually unique ports and integration points between you can get a DJI microphone that plugs into this. I think it's actually more simple than that. Sure. DGI has basically owns Chenjin

which is the place to manufacture these products. They also have continuously been so focused on selling them for the absolute bare minimum. Yep. that and they have the most scale that nobody can compete truly compete on cost. Yeah. If they're doing billions of dollars a year in revenue.

Yep. And you want to, if you're in Chenjin, let's say you like have factories in Chenjin. Okay, well, you're not going to do the volume from day one. You're not going to be able to access the capital. Yep. You're not going to be able to, nobody's going to want to fund the number 10 player to lose money for five years. And so having that almost like monopoly, the moat in many ways is we own Chen. Yeah.

you can't build these products. Even 3D robotics is trying to make, imagine the sabotage. So one thing, they do this throughout the entire kind of history of the company. They're super aggressive on any new drone entrant. They'll flood the forums with comments. This sucks.

This sucks. One of the Wired here in America, there was a guy who had a hobbyist drone forum. He was the editor of Wired and he was, I believe, was involved with 3D robotics. I can't remember. But he,

he realized that there were hundreds of users that had the same IP address in Shenzhen. No way. That were all commenting, like, really negative stuff. And they would do it. They would do it on YouTube. They do it on the forums. And so, like, they are aggressively making it impossible to compete on price. They're so focused on innovation, too, that it's impossible to compete on the product layer.

And so in my view, it's less about looking at it as, like, what stops other people from coming in is, like, well, we can't burn. if we want to compete with DJI, we have to burn $200 million in the next few years. Just to get any market share, and then the product has to be better so that consumers pick us. Yeah, I was talking to some of our buddies about this, how one of my friends Ryan Lackey was saying that right now

he's been actively buying the specific type of DGI drone that has this certain, like, LIDAR sensor or something on it, because it's basically a $5,000 sensor. if you just bought it by itself, but they sell the drone for like three grand or something. And so he's like, it's just the best way to get this particular sensor. And then I rip out everything else to like kind of de-chinify it

and get all the code out. And then I run like open source software on it. But he was like, it's like everyone should go and buy one. Like it's the best like hardware deal in the world right now. And so yeah, it's like it's not just that they're the lowest cost that they're actually providing like an Apple like experience,

but then also driving the cost down. So the Chinese competitors can't just spring up and turn it into the hoverboard market. And a lot of that has to do with like the hoverboard like good software, it can only do so much. Like it's never going to get you like forward backward. Forward backwards. Exactly. It's never going to be like, oh, now with this software, like you can go on sand. Like no, that's not going to happen.

Whereas like really good software with a drone means like, yeah, it's not going to crash in the tree because it's going to detect the tree and not let you crash into it. It's more stable. Yeah, it's more stable. And the video is better. and it records better and oh it doesn't lose the link with your phone when it's flying away all that stuff's pretty actually pretty difficult to build and so once they have all of that they have this powerful effect but then they also dropped software they built software across the entire you know that would make it so across the entire u.s. a user couldn't fly into protected

airspace yep which they then just they just killed this year which is uh absolutely insane or or last year just to get uh so we'll go we'll go into some of like the the the political uh as they start cropping up. But, I mean, again, from 2010 to 2015, people were just like, oh, cool, drones. And then 2015 to 2020 was basically like, wow, like new Apple. Like, this is amazing. This is a cool consumer company. It's amazing that Sequoia and Excel got a piece when they did because this is clearly a

banger company. And then things start to ship. But we have an interesting tweet here from Gabriel Lewis, who I follow. I just went and looked at who's talking about DJI back in the day. This was years ago. He says, I just got a DJI Osmo for my birthday. It's incredible and allows for smooth cinematic shots like this one on an iPhone. Highly recommend to anyone trying to up their Insta game.

And it's just like completely apolitical. Just like, yeah, it's a cool product. Yeah. People weren't even thinking about this stuff with any sort of geopolitical lens. And then Palmer in 2020 is starting to think about DJI and how military drones. will kind of play out. And he says, today we announced Ghost 4, waterproof, shockproof, intelligent, and silent. It's a small drone that can fly for 100 minutes, three times longer than a DJI inspire,

but with better sensors, longer range communications, and an even smaller carrying case. And so this was something when Andrew was launching the Ghost helicopter drone, a lot of people were like, oh, well, like, it's cool, but it's not actually selling that well to the military because like it's kind of a niche use case like it it's like most I think most first off back then there weren't any wars going on so it's hard to be like yeah we need 10,000 of these every month because they're just blowing up like like what happened with you know like missiles during the Ukraine crisis but at the same time like it's very clear that Palmer understood that like the drone infrastructure is going to become important and so yeah you know he's already now the real of that the real

the real crisis here is even if you know and we can get into this in a little bit but even if you have a better drone can you make a million? Exactly. Can you make a million? Because then you make 100,000. It's an attrition warfare.

It's industrial warfare. Because Shenzhen could almost overnight turn every one of their factories into just drone production facilities. Be very very easy. Metrofit. But yeah another so the only kind of of negative um take me up to like 2022 yeah so so 26 2015 to 2018 are just a a generational run

they're winning emmys for their drone camera technology frank becomes asia's youngest tech billionaire the world's first drone billionaire uh he's like on business insider list oh there's a good one in here where DJI becomes the official video partner of the American, or of the amazing race, which was the show that Colin Gwyn was on. And so it's like, he's like, I'm on top of the world. Like totally come back moment. DJI also enters into a partnership to distribute drones to US police departments, which is, you know, if, if DJI is is a Chinese military.

contractor, just like, you know, incredible work. You know, and then the only, you know, coming up to 2019 and it'll get a little bit more frothy from here, DJIs figured out that employees were running this scheme where they would inflate the prices of costs that DJI was buying and presumably like take out. Padding invoices. Yeah, yeah, basically like get paid out on the back end from, from the supply. wires. And so they lost up to $150 million here, but it didn't really make a dent. Like they're doing billions.

That made headlines like, oh, they've been defrauded for $150 million. And I'm just like, one month of their revenue. And so it's like, yeah, it sucks. And, and they came down on on those people pretty aggressive. It does kind of highlight like the chaos that goes on in Shenzhen. Like, I'm sure there's lots of like, I'm sure there. Deal stuff. Yeah. People like, you know, making all sorts of. There's not like, you're not going over there being like, I trust ever. everyone that I've been interacting with. It's a wild west.

Yeah, it's a wild west. So by 2020, they've completely won. They have 77% of the US drone market and no other brand has more than 4%. So just complete dominance. It's more than the iPhone. Yeah. And that's more than Tesla. And then we're going to just kind of like accelerate up into some of their other issues.

So they were found in 2022 to have infringed on a patent from a company called Textron. Textron is accusing DGI of infringing on their flight control system patents related to hovering, right? Because like perfectly balancing a quadcopter and positioning it in the sky relative to how you're controlling it on the ground. Very difficult to do. and by 2023, they're actually a U.S. jury finds DJI drones do infringe on the patents on automated hovering, and DJI has to pay $279 million in damages, which again is, I'm sure that was like, you know, basically nuked their entire net income for the year, but probably, again, do they really care? They have 80% of the market.

And again, so they're still. as of today, the world's most dominant drone manufacturer. And part of the, you know, interesting thing that happens here is around this time, the Ukraine war is happening and drones for the first time are being used at scale on the battlefield. So everybody woke up and realized kind of all at once, hey, you know, there were actual experts, analysts that had been warning about use of drones and warfare. for a long time. But this was the first time that people started seeing drones is this is the next rifle, right?

Like this is the gun of the new, of this century. And so for those like drones that you see in Ukraine and Russia, usually there are sanctions that say that China can't sell that military technology, that dual use technology into the battlefield. But what happens is that people will go and buy them in countries really legal to sell. and then just fly over there. I've actually heard of people like going from Best Buy to Best Buy buying every single drone and then just hopping on a flight and selling them for 10 times the price and being like, yeah, I took a first class flight because I bought $10,000 with the drones in America, went to Ukraine and sold it for 100K or something like that.

Yeah. And so to be clear, like both sides are, you know, using this stuff. And there's these crazy aftermarket devices that will hold a grenade and then click one button and just like drops it. or something. Yeah. And so, DIY.

So everybody realized at this, at one moment, hey, American consumers have been supporting what could be the most important military, China's most important military contractor because the tech is dual use. Yep.

It takes very simple modifications, or no modifications at all, to turn it from a spy robot to a cute little consumer drone that, you know, somebody's flying around there kid's soccer game or whatever, right?

Yeah. And yeah, I think like our conclusion, we can probably get into some more recent posts. Yeah, yeah. Let's go through the timeline. And I think that will take us up to the modern era. Martian space colonists says, can companies in China sue companies in the U.S. for patent infringement? What stops U.S. companies from straight up copying DJI designs?

And I don't think. it's the IP or the design that that's super important. I think it's the software integration and the scale economy of like manufacturing capacity. You just look at DGS manufacturing facility and it's like you know, it's a Stargate

esque like, you know, massive factory. In the same, you know, over the last year, people within tech have been telling consumers we cannot afford to have China own the digital equivalent of the New York Times or the most important of CNN, right? We just cannot

afford it. Gen Z gets their all their news from this place. Like we can't and they're like, but I like TikTok. And so when you think about people trying to compute with DGI, American consumers on average have no issue with it being a Chinese company. Right. Exactly. They don't. They're not sort of enlightened to the potential dangers of that. And so if you tell them, hey, you know, here's this DGI product that's $1,000 and here's this American product. let's say you could set up you know you could manufacture even within 50% of the cost of DJI well they've already shown they're willing to sell products and not make money on them yep so where's your profit going to come from and you can't manufacture them as cheap as them yep and so and the

american consumer doesn't care yeah that is the skydeo's right like skydeo had just as good intellectual property but they did not have the scale economies and so they're on this 4k instead So it's like, yeah, that somebody who's buying a drone for their kids' soccer game, it's just not going to pay $4,000. Yeah. Even for a better product, which I had a- I did want us to film a video where we went and used DJI drones on the range as, like, targets. Yeah.

We're going to do that. Yeah. Yeah. So Bubble Boy and SF says the DJI drone when it hears me criticize Seizion Ping, and it just boom. Yeah, and this can't, I mean, so there's a, We don't have a tinfoil hat here, so we're not going to go too far into it. But everybody should be somewhat concerned that Israel has the technology to blow up consumer electronic devices that have been in a sort of hibernation state. Yeah.

However many years it was, the rational thing to do is is to assume that China has a tail switch. So luckily, most people keep their drones in a box. So I don't, you know, not too worried about them. Just like. And here's another example of dual use. But the last thing is the drone, all of the consumer tech, the same thing that can be used to follow you while you drive, ride a bike ground. That's hunter-sever capabilities, right?

Like, it's all there. It's way, it's way scarier than TikTok. Way scary. From a potential, like, loss of life. Totally. Catastrophe. Yeah. Because if there is like a hot conflict, like you could just ban TikTok.com at the DNS level.

You could just tell everyone, hey, like, they're actively spying on you. If you open TikTok, they will have your geolocation. They'll be able to send a missile or something like that. And very quickly you'd be able to not have it. But if there's a wall of drones coming at you because they're able to manufacture 10,000 a day, that's like much scarier. This is an interesting. The other part about Wang that we haven't covered yet, Wang has had. an obsession with getting robots to fight each other from as long as like we could his favorite

movie is real steel real steel which is yeah terrible terrible movie but he rock him soccer every single year in china they have this thing called robomaster which is like a battle have you ever been to battle bots yeah yeah actually when i grew up uh they they they did a version of this at caltech yeah and it was amazing and they would they would change the goal every year. So one year it'd be like, you have to build a robot that plays soccer against another robot that plays soccer. Or you have to climb up some wall. It was all these different robotics challenges. It wasn't fully just fighting. Yes. Now there's a whole robot. So Wang has such an obsession with getting robots to fight each other. He's sponsoring contests every year. Yeah. Making TV shows,

movies. There's a documentary about it. They have their own like drive to survive F1 style show about Robomaster. Yeah. Isn't that crazy? And then they also sell a, DIY kit that turns your drones into killer drones. Yeah. Which is just like, okay. So you cannot assume that, that Wang just loves the skies. Yeah, yeah.

Loves cool videos as snowboarding. He wants to fight. He wants to fight. Yeah, I referred a client to a web development shop. And I guess the deal went through and it was like a couple hundred grand or something. So they wanted to send me something nice. And, uh, and they sent me a DJI.

a drone, but one that didn't fly, it had like these wheels and it would drive around and it had this little like turret on it that could like shoot a Nerf dart. And it was like very much. It was a really cool product because it was like imagine an RC car but instead of like a hundred bucks it's like a couple grand and it's made by DJI.

So it was like in a camera on it, you could drive it around. You could like do tank turns and spin and stuff and it could go anywhere. It was really, really cool. But yeah, I gave money. And so I was like this is kind of crazy. But speaking of modifying these things,

Jake says the IDF, and this is back in 2021, the IDF has been using small drones to drop tear gas on protesters in the West Bank. And tech CEO Pepe says, on the phone with DJI support trying to order the white phosphorus add-on. And this video is crazy just dropping tear gas. So, I mean, these drones, we know that they've been used for this stuff. They're actively being used. whether it's DJI or not.

And then Palmer, Palmer Lucky says, in case it wasn't clear, Anderil has, in fact, received money from the United States government. I feel bad for the employees of companies like DJI and TikTok who have to pretend that they are not backed by the Chinese Communist Party. Also, don't forget, the TikTok bans people for being gay. The only reason they don't do it in the U.S. is because they need money from the woke teenagers who don't know or care about anything outside their bubble. early to everything, Palmer. And then in 2022, the pressure on DJI is ramping up.

And Catherine Boyle says, lobbyists for Chinese AI drone manufacturer DJI are successfully lobbying Congress to remove a ban on selling Chinese drones to the U.S. government and law enforcement. The Biden administration calls DJI a Chinese military industrial complex company. Yeah, so just one example of that. DJI

drones are used to so the Uyghur Muslims in China have been there's been a bunch of reporting on basically they're held in camps they're basically you know being genocided

DJI drones hover over the camps to watch them 24 7 so they're being used to spy internally in China and so it's not a conspiracy theory that they're a government contractor. Yeah, of course. They are a government contractor.

Of course. Luke Metro is highlighting a story. He says, Wake up, babe, they just reverse engineer DJI Aeroscope. And so I think what the story was, was that someone figured out how to hack into the DJI system to, when you see a drone, figure out where the operator is.

Interesting. So very quickly, you see a drone in the sky, you know, okay, you know, we could shoot the drone down, but we probably want to get the office. operator. And so this is like very risky from if you're using a drone in the field. In Ukraine. In Ukraine, you don't want to like have like some sort of EMF signal going out that can be tracked.

Santiago says, who's building the American made DJI competitor for consumer slash agriculture? This is a critical space and I'm eager to back credible efforts in this space. DMs and checkbook open. CC, Josh Steinman, Naval, Steve Simony and Yukon K9. Yeah, and people like, Nival. have been actually really loud about the use of drones those weapons i think it was his quote saying the drone is the rifle yeah i mean neval was saying that like he's worried that the idea of traveling in a plane will be something that just doesn't happen anymore because the economic equation of

of if it costs someone 10 dollars to shoot down a 747 with a drone people just won't fly anymore. And if it's just like a random terrorist can just build a drone and just take out a commercial aircraft, flying is just going to be kind of impossible. And it's kind of unclear what countermeasures you can do

to just like a normal flight if things get really crazy with like drone related terrorism or drone related. Yeah, the only thing... I'm not sure. I don't want to have some thought it. Yeah, yeah.

It's kind of like it's never been in production. The reason that people don't kill people all the time. Yeah. It's not because it's super expensive. Yeah. It's, you could buy a bullet for a dollar.

Yeah. But still people are not like going around. Yeah. You know, becoming murderous over that. But I think it's a, it's a sort of dystopian. Yeah.

Hopefully just like electronic countermeasures, you could just have every 747, have something that just blocks signals from most commercial drones. And then they're blocking them. I mean, even with the fires, with that drone that hit the firefighting plane and took it out of commission, like that could probably have been stopped with some countermeasures. Ashley Vance in 2024 says two and a half years into its war with Russia, Ukraine still relies for the most part on DGI drones built in China. This means Russia has modest pull over the PRC and the West still can't make cheap drones.

I got so many notifications that my phone... Vibrated out. across the table. That happens. Every time we stop recording, it's always just a flood. I think it might be interesting to go through this Q&A with DJI CEO Frank Wang, the creator of the Phantom Dron. This is all the way from back in 2014. The Wall Street Journal asks him about starting the company. He says, before we started the company, I spent three months intensely working on the project. At that time, I was enrolled in the university, but I skipped all the courses and just went to my home in Shenzhen.

I would wake up at 2 p.m. and then work until like five or six a.m. for days of the time. One time when I did go back to the university lab, I tried to use my ID card, but it didn't work. My heart sank a little bit because I thought I was kicked out of college by my professor, but I'd actually forgotten to pay my tuition. At the beginning, when we started the company four years ago, we made flight control systems. We focused on the operating system for the drones, but it was hard to use. The drones were complicated and the controlability was relatively poor. People couldn't use it on a larger scale.

We felt a multi-roter drone should be very simple, very small, very reliable. and very cheap. If people could use the market, if people could use it, the market would be larger, and he was 100% right. So slowly we went from making the flight control systems to multi-rotor drones. Should we? Who is your role model? This is the headline. Well, of course, it's Steve Jobs. Personally, I was very aggressive. At college, I took part in team competition for robotics twice. The first time I worked very hard and technically we did very well, but my teammates did not feel comfortable working with me. I was too aggressive. I just wanted to win. The second time, I was still aggressive, but I found the right team partner and the leadership was stronger, so that time we won.

I realized that not being so easygoing is not such a good thing. But after I read Steve Jobs and discovered he had the same type of personality, it encouraged me. I understood that staying aggressive is the right thing to do to build a company. It's fascinating. Are we seeing the emergence of a new breed of Chinese companies such as DJI? Chinese companies are getting better before they lagged behind. Now more and more Chinese companies are doing well worldwide, like Huawei, 10,000, in Alibaba. I think later more Chinese companies will go global and their image will gradually change. Definitely right about that. I think the important thing is vision. Yeah, we are actively trying to change the image of E. Yeah. So those companies have one that is in sync with the world.

What's the future of DJI? I believe that the direction of our company is driven by our initial dream to make a very easy to use product that can realize the human dream to fly and to make it perform so well that everyone can enjoy it. In addition, we will develop our business in agricultural and industrial and all kinds of fields. The next five to ten years will be very exciting period for unmanned aircraft and I'm looking forward to the future. And this was 10 years ago. Yeah, you got to give them credit for being very early and very right and just great execution. That said, I do want to jump forward and talk about, we have an article from the New York misinformation, the time The old timers.

The old timers from April 25th, 2024, so just less than a year ago. And so, yeah, basically around the time that people got concerned about China, or sorry, TikTok. There's also a lot of concern around DJI, Congress.

Now Congress is weighing legislation that could kill much of DJI's business in the United States by putting it on a federal communications commission roster blocking it from running on the country's communications infrastructure which would be interesting approach and the other thing that's interesting here that's different than tic talk is nobody's paid anything to tictock right you maybe buy some products on ticot from ads but some of the additional controversy around banning dGI is that consumers have spent thousands of dollars

on these products and if you brick them through legislation there can be a lot of pushback there you know people love the drones right like people that race drones and you know all that stuff so i think it would be very slow that they were to ban it because yeah uh a which is why we need to ban it now yeah yeah i i think i think the biggest problem is like bringing in new drones that have even more advanced capabilities because realistically the the the military capabilities of like a phantom three or phantom four or even like a dji mini one are not going to be as next generation as what's coming across in the next cycle of products. And then also, even with what the proposal for TikTok,

the actual legislation was that Apple and Google cannot host the apps in the app store. But if you have the apps installed, they'll continue to work until the operating system updates and breaks the app. And so, like, this would be kind of a slow phase out. And then even some of these drones that can be used without the app for a while. So I don't think, and honestly, most of these drones, people fly them for, you know, Christmas Day, the day after, and they start collecting dust, I don't think it would be that big of like a massive loss for consumers.

Yeah. But people would certainly be upset. Yeah. So it doesn't. If paths and find into law, the legislation would effectively ban any new models of DJI drones from that point on. It would not apply to drones already in use. And there was a house bill that had bipartisan support. They met with a muscular lobbying campaign by DJI.

The company is hoping that Americans who use its products will help persuade lawmakers that the United States has nothing to fear and much to gain by keeping DJI drones flying. And Representative Elise Stanoffick, who's a Republican in New York and was one of this bill's primary sponsor, says DJI presents an unacceptable national security risk, and it is past time that drones made by communist China are removed from America. Government agencies have shown that DJI drones are providing data on critical infrastructure in the United States to the Communist Party. Ms. Stanofix said without elaborating.

Any attempt to claim otherwise is a direct result of DJI's lobbying effort. So DJI sells a lot of drones in the United States. And the beauty of selling them to U.S. consumers is that those consumers are basically providing, you know, if DJI, who knows, it would be really interesting. I doubt they have this information public, but I bet it's like DGI drones take 500,000 flights a year a day in the United States. Who knows? It could be way less. Yeah, I have a good summary of this exact point from Hunter-wise. He says,

I do think there's a huge problem with DGI drones as someone who has used them for eight plus years. Every flight transmits the footage back to the CCP. All these drones are elite mapping spyware and there's millions in circulation. They record every coordinate per flight plus all imagery is then processed through their app back to China. It's the greatest mapping software ever. They made billions on the drones while getting all the footage from across the world as personal data back to the CCP. Their customers became the greatest spy asset for mapping real-time locations for every new flight. Very interesting. I do think that there's a way to like, kind of like

sandbox the drones, you know, and like actually see what data is being transmitted. I think that almost the bigger issue is that people are just, they're just voluntarily saying, yes, please upload my footage to the DJI cloud. And they're actually opting in to send the data. So I don't think that they actually have to steal it. Yeah, but to be honest, let's be real. Like, you cannot trust. I think they could be stealing it. But yes, I agree with you. But I think that they might not even need to be stealing it. Because I think people might be voluntarily saying like yes please send my drone flight it's like when you get that up when you get that uh question that says like um like would you like to send diagnostics to the developer

that's basically like yes i'm opting in uh you're on breaking news no i just yeah i wanted to kind of catch people at the speed so basically they go they go back and forth yeah so i was going to cover the verge piece sure they go back and forth for a while they have spokespeople these lobbyists and they go this is Regina Lynn. We should ping her and ask her what it's like to be a traitor to the United States. Our products are designed and intended

to promote the general good and benefit society. And they have, she denied that the drones have been involved in human rights violations and said they were not meant for surveillance. Okay. Surveillance in capturing live video

are one for one, right? You cannot argue that. Oh, we're just capturing Capturing live video of a scene is surveilling said scene. And anyways, she, another person says,

DJI's ownership is primarily concentrated in the hands of its founders and early stage executives, none of whom are government officials or representatives of government or state-owned entities. And so the reason that this just doesn't stand up is like, you can be the most,

you can be Jack Ma. And if you say even any type of negative comment about the state get disappeared for months and months and months. So it doesn't matter. So the government owns you. Yeah. The government's a threat informs your decision making even if you don't have a contract or a CCP email address. You don't need a handle. And just the vibe that they would come for you.

You can make one comment on a stage and be disappeared. Yes. So they don't need to own the company. They own the founders, right? Yeah. So the latest news is that Sam Lesson is quoting the Verge. The Verge says, DJI will no longer stop drones from flying over airports, wildfires, and the White House.

And Sam Lesson says, I see how this works. We move to ban China's TikTok. So they respond by dropping DJIs no-fly zone enforcement for drones. Welcome to 2025 international relations. And to me, this seems like we're getting banned YOLO, literally capture as much data as possible. Oh, I didn't even think about it as capturing data. I thought about it just like straight up as like if we turn this geo fencing off like someone might use a drone for just straight up terrorism and like crash out white house just pure chaos but the other side is some enthusiasts is like I want to fly over area 51 you know and like you can't stop a drone sure from if if somebody's like some crazy you know uh UAV uh you AP person's like yeah I'm gonna fly over area 51 yeah and it's not gated and it's not gated and it's not gated and

Well, you can stop it if you have an anderal, like anvil unit on site, but a lot of places don't. You're going to launch the, isn't the anvil like an explosive? Yeah, yeah. Well, no, no. The original anvil just has just a weight in it. And so it's purely kinetic. And this is the whole story with Palmer figuring out how to shoot down DJI drones. I know.

He was just saying electronic warfare takeover. And what he realized was that if you just run into it, that's actually the most efficient way to destroy it. And so the anvil is just like, it'll just like, it'll just. It's a quadcopter, basically, with just like some mass inside, and it just crashes into the other drone. They do make one with high explosive in it now. I'm just saying, let's let's, you know, play it out. If some, you know, somebody that watches a little bit too much,

Jesse Michael goes down to Area 51. It's just like, I'm going to basically, I know my drone's going to get killed, but I'm going to get a lot of video. I'm going to live stream it. I'm going to live stream this. Yep. And just fly it as fast as I can. By the time the anvil.

react. Oh, totally. You're halfway across the base, right? So, like, I think it's, I think it's a little bit, I mean, it's such a fuck you. Yeah, totally. To our lawmakers, to the people in the military that are concerned about this. Yeah. Do you want to read this one or move on to the next one?

Hunter's on a, on a role. A hunter's on a roll friend of the pod. A great case study for, and he actually worked for Casey Nied's that back in the other. Oh, yeah. A great case study for what not to do in the diversification of your business's products. GoPro tried to enter the drone industry and a market already controlled by at DJI to top it off, the GoPro drone started falling out of the sky. Refunds, returns, warranties, massive loss.

I think it was smart for GoPro to try to do this. They were the company best set up. They had a brand. They had distribution. They already had good quality cameras entering. Drones should be commoditized to some degree, right? Sure. Like, we know how to make them now.

There's no network. There are patents, but there's no network effect. Yep. and then there's the other thing to go-co story where like they probably missed the real opportunity

which was figuring out how to change the playing field in Washington to create more of a level playing field with some sort of tariff at the very least. Yeah if you just make

what we've needed was a thousand percent tariff on Chinese drones. That was impressive because it's like drones are such a nice to have right?

Like there's totally a world where people that are videographers have drones. Yeah. Like they have nice cameras. You just hire them when you want some drone footage. Exactly. And then you could, they could afford to spend a bit more. Yeah. And we all would be fine. If somebody posts on their Instagram, a video of them like flying over a hike, I'm going to, that's a, that's a regretted user second right there for me. I don't want to see your drone footage. I just like, it's not that, it's not that interesting to me. What else? So this is an article. from December 18th, from the verge DJI escapes U.S. drone ban but may get banned

automatically unless Trump steps in. The U.S. Senate passed a massive defense bill that gives DJI one year to prove its innocence or it gets banned. Oh, interesting. And so this is the National Defense Authorization Act. Yeah. It's an, it's their annual defense spending bill. And, and it says, well, it did not contain the full countering CCP drones act, which was the act that I was just talking about, provisions. It kicks off a one-year countdown until its products are automatically banned. If DJI cannot convince an appropriate national security agency to publicly declare that its products do not pose an unacceptable risk to the national security of the United States, the Act instructs the FCC to add DGIs to its covered list under the Secure and Trusted Communication Networks Act.

Not only does that list keep that gear from running on U.S. networks, it bars the FCC from authorizes, authorizing their internal radios for use in the U.S., effectively blocking all imports. And so what's smart about this is all of the lobbyists that were hovering over these lawmakers being like, don't ban it. It's just like a fun app. Your constituents love it. We've sold, you know, you know, they were telling people, you know, we've sold 50,000 drones in this swing, you know, this swing county, right? You know, in this town. You got a lot of fans there. You got a lot of fans there. You sure you want to do that? And so now they're actually putting it in the hands of a national security agency, which in theory is much harder to influence.

That's fascinating. I didn't know that. And so this is like to me, you know, for us, like I think like TikTok is information warfare. DJI is like purely spyware with this like, you know, asymmetrical risk of maybe these things could actually turn. maybe there's back doors and these things could actually be turned into actual weapons. Yeah. And I think the tide is definitely turning. I mean, that's a great example.

But even just culturally, you can see the tide turning. Josh. And they're also worried about DJI just exploiting the loophole by white labeling its drones. And you could see a scenario where certain filmmakers are like, I love DJI so much. They're like buying black market DJI drones and like trying to figure out ways to use like shorter. you know, wave. Yeah. So Josh Wolf says, sadly, we are at war.

Wars imagined by sci-fi of yesteryear. One, distribute tech, DJI drone, TikTok. Two, habituate or addict the user. And three, pervert the reason you are defiant to comply with American sovereignty and weaponize user outrage. And that's the last second that you mentioned was like weaponizing the user outrage. And I mean, this vibe shift, this is a good example of this from Jason Liu back in April of 24. He says, DJI is the hardware version of TikTok. It gets American operators addicted because of

the economic model. It's so cheap and you can go and resell your footage. So it's a good business. Steals data, intelligence to PRC, hollows out domestic drone production capacity. We saw that with 3D robotics and Skydeo and so many other companies. And every American dollar going there is funding the Chinese military industrial complex. And this was definitely like early to the story, but still 400 likes and I think we should close do you have another thing you go and another mess it's good I have another

I just have some interesting so I on the R slash drones Reddit there's people one month ago commenting I've never encountered a situation like this to me it feels like learning they're commenting on the potential band feels like learning videography and having

final cut an Apple band you know so a lot of upvotes on that the biggest change will be price. U.S. drones are prohibitively expensive. Many who would like to fly for fun won't be able to afford it. Yep. Which is just like... It's true in the short term, but not long term.

We can just manufacture our own. Starlink is. Starlink is so cheap. Why? Because Elon built a massive factory that cranks out a million of them. If we just build that, we're fine. The reason I think it's funny to look at these comments is because DJI has a known history of making fake accounts and commenting. Totally, totally, totally. And so you know that they're, you know that they're flooding these. 100%. I mean, it is insane when people maintain that like, it is impossible for us to

build a cheap drone in America because the labor cost is not even that much lower in China anymore. Like that, the whole like labor arbitrage, it's cheaper to hire someone in China. That disappeared a couple years ago. And it's moved to other countries. Like, I think AirPods are made in the Philippines now. And there's a few other places. The main thing is just the, the industrial build up all of the different manufacturing plants and the coordination of having all the different machinists and experts in the same place in Shenzhen where you can just walk around. But it's totally, it's totally doable in the United States if we build a big facility. And we see that evidence with Starlink and Tesla too. And so it's clearly doable. We just need to have the will, the political will,

to do it? And so do you have anything else or should I close out with this Josh Steinman Bangor? Always love to close out with Josh because he was on, in the national. security community former operator eight years ago yeah but um really to all this

uh I just think so uh what else you got North Dakota actually banned

uh already banned DJI drones oh you're really just as a state yeah yeah yeah so you can't buy them and ship them there maybe

or yeah um what like what happens if they're requiring bring one do I get in trouble they have something called

the drone uh they're trying to fund a drone replacement program to help. You have to turn in, the various agencies in North Dakota have to turn their drones in, and then the drones are going to be used for counter-UAV research. Right. So that's a good, that's a good,

if our military can buy back all these drones, we can use them for target practice. That's great. That's great. So let's close out with Josh. He's quote tweeting OSNT Technical who says, breaking from, breaking news,

China's DJI, the world's largest drone manufacturer, has disabled US geo-fencing on its drones, enabling flights over airports, military bases, and no-fly zones. And Josh says, use executive order 13873 to ban DJI, do it now.

Yeah. And so I have no idea what executive order that is, but Josh clearly does. And he, and he, what I love about following him is that he clearly understands the actual tactical moves that need to be done beyond just, you know, tweeting about it or talking about it or having a couple, like, new cycles around this.

And I think this, this DJI story will grow a lot in the coming years. Yeah. It's only out of time until there's a question about like, okay, when is it going to happen? And my, my optimistic scenario is that there's not necessarily a Jeff Yass to DJI in the same way that there was to TikTok. Yeah. Where, yes, Sequoia has a position, but they spun out Sequoia China. They can't publicly defend it.

And they separated the funds. So the true Sequoia guys who are in America don't have economics necessarily to the same degree in DJI. And then, yes, Excel has a position, but it's a small one. They might have sold it already. Like we don't even know. They have a lot of other positions. And so it's not the same thing as Jeff Yass having $16 billion on the line with TikTok. Like at best, Excel has what, a billion maybe?

Like if they have 1% of the company still. No, there's nothing that says they're a $100 billion company. Sure. Any more. They probably would be if valued as purely a military contractor. And I bet you internally in China, they're like, yeah, this is a billion dollar. Yeah. This could be a 50 billion dollar.

Yeah. And so, yeah, and I haven't seen any of the Excel guys come out as like super pro-Trump, donating a tonne, like really trying to get TikTok not banned. And so it does feel like the American economic interests are a lot less intertwined. Yeah. And also the TikTok community, there are like millions of TikTok creators that are like, this is my job in America. and I don't think the videography community is nearly as big or outspoken.

It's mostly a lot of people in maybe Hollywood, but that's it. And that's not a huge out-com constituency. So I don't know. I think we can see a band. But it needs to get more attention. This just came to my mind because yesterday we were saying everybody should have their own project Stargate where they raise, go out and try to raise $7 trillion. Even if you miss, you'll land.

Half a trillion would be fine? 500 billion. Yeah, you'll land. Shoot for the stars. Even if you miss, you'll land on the moon. But together a crazy die. But we need a project Stargate for domestic drone manufacturing. Yeah, to actually get anywhere close to Shenzhen capacity, which in a global conflict scenario,

I promise you that China will stop giving us drones, right? Because it will be drone-on-dron violence. And so we need something at that scale to basically say, if this is the next, if this is the firearm of this century, we need to invest $100 billion into setting up the ability to make millions of these a year in the United States.

Yep. That's a great place to close it out. Thanks for watching. We'll be back with the timeline in just a minute. Welcome back to Technology Brothers. Still the most profitable podcast in the world. We are giving you a little bit of an update to Project Stargate, which we talked about yesterday. We did a whole deep dive on Stargate. Pretty crazy story,

but it's still evolving on the timeline on X. Sam Altman, Elon Musk, and Sotcha Nadella. Some heavy hitters in the tech world are duking it out. Jordi, what are they saying? And are they posting slot or bangers? We have to slop or bangers. We got to name this the timeline in turmoil. Timeline and turmoil.

Something like that. Yeah. But yeah, so again, our global Our tech overlords are fighting They're going back and forth Not sparing any words

And this morning Sam Aldman is Clearly has Lex Friedman in his ear Because he says Just one more mean tweet And then maybe you'll love yourself Dot dot dot dot Which is harsh

I mean It's also I think this is kind of a cringe Post well yeah because it's going so hard you know it's like yeah and it's like cutting to someone it's it's making it tend to cut it's where's where's i mean where's where's where's the proof that eon doesn't love himself yeah it's sort of like it's like if somebody's said like oh you're doing that because you don't love yourself you'd be like i love my life i have a amazing wife three sons

yeah i live in california and i i i what's what's not to love. Yeah, exactly. I'm, I'm, I'm, just in the modern Marlboro man, it's saying. I mean, just in, just in tech generally, I think that we like to be a little bit more rigorous and data-driven with our criticism. Yeah, this is just kind of like playground. Well, it's somewhat at home, right?

It's, it's not addressing the question of like, what did Detoria say? Sassie. Wait, what was it spelled SAAAS? Why? Like software as a service? Or is it just sassy generally? Just sassy. Okay.

Yeah. All right. At least we're being entertained. Yeah. You can just throw this in the ground. Next. Next we have. So Satya had an amazing quote yesterday that provided some great ground cover for everybody involved with this kind of heinous.

I mean, one of the greatest quotes in tech CEO history. Yeah. I'm good for my 80 billion. Which, which again, we said yesterday, to be clear, he, very specifically phrases it as 80 billion for investment in Azure infrastructure and Azure OpenAI, Microsoft or so intertwined now, but he's not saying 80 billion for Stargate. He's saying 80 billion for his AI infrastructure needs, right? Which you can use, which he could use, but he's not committing it by any means.

So Elon gives Satya some credit. He says, on the other hand, Satya definitely does have the money. and Satya hits back with a crying emoji. I love that. Which is Elon's signature. It's good to see a trillion-dollar CEO using the crying emoji. I used to not use it, and then I realize it's so perfect for so many situations. And he says, and all this money is not about hyping AI, but is about building useful things for the real world.

And that is clearly a jab at Altman, which just shows that everybody's got. in a little tip. But we're going to get back to useful things. This is, this is, can you, I, I, I, going, uh, going, uh, going, uh, back to all the board drama and Sam getting fired. Yeah. Can you imagine being Sam and having Satya, like, sit you like, sit down and be like, how did you? There was no sit down. Had is on the board. No, I know, but he had invested, I mean, if you invested 10 billion dollars in the country. I mean, I'm sure. Don't you think Saatia is like, hey, you know, it's basically like, hey, like, you're my guy. Yeah. How did you mess this up? so badly.

Yeah. Anyways, so Nick, NS, 123, ABC, great username. Easily remember. Easily remember it. You guys wait, dot, dot, dot. Samma got cooked by Microsoft CEO

and just a screenshot of that. Moving on, Vittario fires his own version. He says, this is a great, great Photoshop. George W. Bush. Sir, Sauta called you a hype boy. Just brutal.

And really, that's what's going on. Sam did just make a post of a drone shot of a bunch of buildings. Really? And it just was just the caption was big, beautiful buildings, sort of proving that we actually are building this infrastructure. And I commented on an air in this post and I was like brought to you by CCP front DGI drones I saw

I saw a time lapse of the Texas facility that's already being built for I think probably that one yeah it's rotating through and it is impressive like the I mean it's a huge huge build out that's already happening and so in classic so open AI

you can have whatever opinion you want on Sam Open AI Sam brilliant marketer Open AI brilliant calm The strategy seems right now he's under fire from Elon, his sister, Tucker Carlson, right? Like there's just all this stuff happening. He's launching, he's making so many big announcements that he's pretty effectively drowning it out and keeping people bought into, you know, bought into the movement. Right.

Yeah. Because I think I mean, I think if you just went to coffee shop on the corner and you said like, oh man, like it's a crazy. day. Like, do you see what happened with Sam Alton today? They'd be like, who? Yeah, he's still not. Most people, their interaction with AI is like, oh, you mean chat.com? Which they're not even really using it.

No, not yet. But like, they know chat Chb-T, they don't know about these ins and outs of the drama. And so there is something to be said for just like, just one foot in front of the other, release new product, improve the product. And people will be like, yeah, it's chaos behind the scenes, but did the AI answer the question

or not? Yeah. And so, on that note, OpenAI has a huge. They're not really calling it a full launch. They're calling it a research preview. They are making it available to pro users, but why don't you get into what they're actually doing? Yeah, so in the announcement here, they say they're releasing operator, which is an agent that can go to the web to perform tasks for you. Now, chat GPT40 has already been able to go to the web and retrieve individual HTML pages, suck them in and use that as context. So it can get more up today data, but this is a step forward in that the operator, it has its own browser and it can look at a web page and interact with it by typing, clicking, and scrolling. It's currently in

research preview, meaning it has limitations. But this is one of the first agents, which are AI is capable of doing work for you independently. You get a task and it will execute it. And this is fantastic. I think this is going to be very, very helpful. There's already been the ability to use chatypd to write a little bit of code, like little code snippets. So I was doing something where I wanted to scrape a website and just paginate through there, were like 20 pages of standard formatting, like, you know, list of, of like, I think it was like podcast episodes. And I just wanted them all on a spreadsheet, basically.

And it was able to write the Python code for me, but it didn't really work because it needed, like, JavaScript and stuff. It was like, it just wasn't, it just wasn't quite there. But with this, it's, it's like the AI has a web browser that can fully render a web page and just click the button and then copy and paste stuff out. And it's just, We're going to be able to interact so much better, and it will actually be able to go and do things for you.

So they have a bunch of examples that we'll run through. But it's available at operator.chatGPT.com. You have to be a pro user, which is now $200 a month, but I think this will be what gets me to upgrade. The research preview allows us to learn from our users, and their plan is to expand into T-plus and team and enterprise users in the future. So they're calling this a new model called computer-using agent, CUA. So another easy to remember acronym that I'm sure will be, you know, no problem. And one thing to note here, we've talked about this before, whatever you're building, it's on Open AI's roadmap, especially if you have anything, if they believe that you will have product market fit or you do have product market fit. There's a lot of heavily funded companies that have been building computer use agents.

And so this just made it. this is like one of those moments that founders go through from time to time where the 800 pound gorilla does exactly what you're doing better and they have more distribution and so not a great I think it's really if you're building anything on open AI or even not on their rails you're waking up every single day not knowing when you're going to refresh Twitter and have ax and have a major you know announcement It does exactly what you're doing.

Yeah. And so I think the main innovation here that they're championing is vision capabilities and then reinforcement learning. And so essentially an operator agent can see through screenshots. So it actually loads the web page in a browser, takes a screenshot, and then it can use that. Because previously when you wanted to automate some sort of interaction on a website, like let's say you're just a sneaker head and you're trying to snipe a new sneaker drop. right at that moment. What you do is you would write some code that says, find this particular button that has this class or this ID on it.

And it's this div or it has a button, HTML tag, and then issue a click on it. But if anything changed about that, the code would break. Whereas when you go to a website as a human, you just see a visible, like, okay, I want to click the blue button. And I can read what the text is. And so they're able to use the vision models to understand what the actual website looks like,

even regardless of what the code says on the back end, and then they can interact with it using the actions, any action that keyboard and mouse would use. And because you can generate these web pages kind of endlessly, and there's so much out on the web, you can run reinforcement learning to basically train the model just trillions of times. Like, were you able to get through this flow

on this particular set of divs and buttons and forms? Yeah. Were you able to get through this flow? And then it learns from its mistakes and it iterates through and it probably got to a really solid place where they're confidence. So now they're releasing it as a research preview. And so if it encounters challenges or makes mistakes, operator can leverage its reasoning capabilities to self-correct. When it gets stuck and needs assistance, it simply hands control back to you saying like, hey, I'm confused. Like, do you want me to click?

Yeah. And that's that co-pilot functionality where you're just kind of having the agent do things. Yep. And it's now hiring you back to accomplish. what it wants to accomplish. Exactly. And so they have a benchmark for this. Web Arena and Web Voyager are two key browser use benchmarks.

And so they have some e-vals out there. And they're very happy about their hitting new state of the arch benchmark results. To get started, you just describe the task you'd like to be done and operator handles the rest. They can take control over and operators train proactively to ask the user to take over tasks that require login, payment details or solving CAPTCHAs, which is hilarious because it's like the humans.

only responsible for the CAPTCHA basically. The most annoying thing on credit card forms. I mean, I kind of like just looking at a nice landing page, but they're like, you don't need to do that. Just do the capture human.

That's what you're supposed to do. And it's like, obviously they can solve the CAPTCs. Like, that's not a problem at all for these AI models now, but they're like, we respect CAPTCHAs. Yeah, interesting. Very funny. Yeah. So users can personalize their workflows

as an operator by adding custom instructions either for all sites or specific ones. Are the models good enough to have this better accuracy than humans with CAPTCs or is it about the same? Because every now and then I'll still get like, you're doing a CAPTCHA and you're like, you're just doing it quickly. Oh, yeah, you're messing out wrong. Oh, yeah, that happens all the time.

I think that on the most cutting edge CAPTCHA is the one where it's like rotate and change. There should be like a CAPTCHA world championships where people just sit there and they like do as many competitive CAPTCHAs as you can. Yeah. And it just gets perpetually harder. Harder. Yeah. And.

Are you a robot? So I imagine that if any AI team, like OpenAI, XAI, any of these guys, Google wanted to really break capture, they could create a superhuman model that beat it. But they wouldn't because the aesthetics are terrible and there's no economics in it. So it just hasn't happened. But it's not it hasn't, it's not that it hasn't happened because of technical. It would get, if you did that as a developer tool, they would be immediately abused. Exactly. Yeah, we're trying customers.

People would be like, thanks, you created a program that just ruins everything. The internet. Yeah, which is kind of already like the criticism with a lot of this stuff, the data internet theory stuff. So users can personalize. So basically you can say, hey, when I tell you to, when I tell you to book a flight, I like American or I like United or I like private. And it has to be a G650 ER. Operator lets users save prompts for quick access on their homepage, ideally for repeated tasks like restocking groceries on Instacart. similarly using multiple tabs on a browser,

users can have operator run multiple tasks simultaneously by creating new conversations, like ordering personalized enamel mug on Etsy while booking a campsite on HipCamp. They have all these like... That's like the most SF. That's the most SF. I want like the...

Go gig along on this like degenerate polymarket. You're buying an AR-15 on Parmato State Armory. It'd be so funny. Oh yeah, we should rewrite those. All the examples need to be insane. While submitting offers on hundreds of cars and bring a trailer. Yeah, auction, yeah, auction snipe a kuntash on bring a trailer for me and also, yeah, it's great. Operator transforms AI from a passive tool to an active participant in the digital ecosystem.

I'm sure people will love that. Some people are like, this is going to be a disaster for my website maybe. Because if the AIs is better at avoiding upsells and cross-cells and stuff, that could really affect CTCHA. Yeah, so it's interesting here. We're collaborating with companies like DoorDash, Instacart, Open Table, Only Fans, Price Lines, Stubberh. Snucked that one. That is not, it doesn't actually say that, to be clear. Because priceless stub hub, thumbtack, and Uber.

Yeah, because a lot of this does require opt-in from these major players because DoorDash doesn't want you using bots. in their app. Totally. Right? And this is a bot. Yep. And it's the same thing with... And it's funny because, again, the sneakerheads were doing this.

Yes. Yes. Since, you know, they've been building basically this functionality. Yeah. Yeah. To snipe Jound sneakers, which I have... I'm a Jound enthusiast, but T.J. Parker and a number of other people appreciate the brand. And it's very difficult to actually buy these things because the bot are.

me it's just to send on the launch and yeah and it's funny because you can see that there's a shift here where the first version of gpt 335 chat gpt clearly was scraping mainstream media news sources and that really agitated all the mainstream media sources there were some lawsuits but now open ai is going to those companies and signing partnerships before they ingest the data so everyone's happy and everyone's making money and clearly here they were like okay like let's let's be a little bit more up front with people, get them on board early, so that we don't run into anything down the line.

This is, so some of the partnerships they're working on are wild. They say to explore these use cases further, we're working with organizations like the city of Stockton to make it easier to enroll in city services and programs. So the city of Stockton is basically saying, no one wants to use our terrible forms on our website. And it's actually wild to think about,

you're just instead of having to interact with a DMV or an organization like that you can just be talking with your agent it's like hey I need to renew my license and it just does it and it's like yeah I need you to put your SSN in here or whatever this is the dream of the original like semantic web like web 2.0 yeah you know like the whole thing but here here's here's the crazy thing there's been hundreds maybe thousands of companies sprung up to build agents probably not thousand

globally a thousand probably lots and they were all selling these use cases of like we're going to make it easier to book a flight we're going to make it easier to like interact with the dmd they've been selling these so hard and now because of open ais leverage and market cap and scale the city of stockton is not going to want to work with your little you know your little agent that raised a 600k pre-seed they want to work with open ai same thing with um they uh they also are partnering with instacart on this they say open ais operators a technical breakthrough that processes that makes processes like ordering groceries incredibly easy. Daniel, this is the chief product officer at Instacurt. I have to ask, did Instacart not already make it?

Yeah, that is weird. That is weird. Because it kind of seems like you're letting the fox into the henhouse stuff a little bit because at what point does Open AI like, oh, you know, so much of your volume is coming through now. We actually need, we're going to need like 20% of your net, you know, revenue of any orders that we process and then and then open AI is like, okay, well, if you don't want to give us that, we'll happily go to Safeway and just let them know and all and we'll partner with all the groceries and just say, we do your online order. Yeah, yeah. Yeah. We already have 100 million active. Yeah, it's always been a big question of like, where does the AI sit in the stack? There's this question of like, because a lot of companies for

the last two years have been like, we need to have a chat bot. in the bottom of our website that lets people interact with our API or our service or forms or our website just through chat. And there was definitely this idea that, hey, maybe the future of the Instacart app will be, you show up, you just talk to it, describe it. Hey, I'm making pasta. And then the next day I'm going to have steak and potatoes. And then the next day I'm going to have something else.

And so can you just like get all those ingredients, put them in my cart and order them? And that would happen within the Instacart. app and it might be powered by an open source LLM and no one might get a cut. But now it seems like a lot of these companies are saying actually it's better if that sits at OpenAI and that makes them more of an aggregator. It means it's more important than ever

that the chat GPT app is in the home row of every like 100 million American consumers that the company is seen as a consumer company not an enterprise company because if if it's truly like you know, like the start of booking a flight that happened on Google a long time ago. Yep. They were booking websites as well.

Now if it starts on OpenAI, that's incredible value capture. And so seeing like where the AI sits in the stack is fascinating. And Open AI certainly wants to make as much of it happen at the start of their app. Yeah. Interesting. And it's such a interesting dynamic having a, the number one developer tool, in AI also being the consumer product that's trying to eat all the people developing apps, which is why I think in the long term Open AI's brand is probably I would call top on their

brand right now because people are still amazed. I don't know. The reason I'm hesitant is because a very similar thing happened with Google when they started ingesting those web snippets. So that was really bad for like the guy who owns Celebrity Network.com had all these different landing pages ensued Google. Like how much is George Clooney worth? And it would show up and they would click on that and they would get the eye revenue.

And then Google started sucking that in and then cut all the traffic. So the caveat is calling top on their brand in the industry. Sure. Not for consumers. Yes, exactly. The industry is really like I raised $5 million and spent two years building this and open AI like just. Yeah.

Although that has been a meme Like the Yeah, yeah, yeah. Oh, opening I killed my It's sort of like eating someone alive, you know, where like, it's rough. Well, let's go back to the technical stuff. They have takeover mode where the operator

asks the user to take over when inputting sensitive information into the browser such as logging credentials or payment information. There's user confirmations before finalizing any significant action such as submitting an order or sending an email. Operator will ask approval. There's task limitations. Operators train to decline certain sensitive tasks, such as banking transactions. So good luck with those going turbo long. I don't think that's

going to happen anymore anytime soon. Maybe that's the opportunity for the AI agents. It really is the only fan. You basically need, there needs to be an agent for buying equities in size. Yeah. You know, so you can just basically text your broker. It's like it, you know, you have on a stock broker. You call the broker and be like, buy Trump coin 200k like do it right now. I mean it really is like the biggest AI opportunity is that we've seen in the last few years have been the ones that the big labs just don't want to touch. It's like the AI girlfriend stuff. It's like too controversial. So probably not going to be killed by chat TVD because they'll be like yeah,

sure, that is a hundred million dollar opportunity. But we're just we're just going to take that one out because it's just it's just pure downside from a brand risk perspective. And so sure, someone else can have that. Um, task, uh, watch mode. On particularly sensitive sites, such as email or financial services, operator requires close supervision of its actions, allowing users to directly catch any potential mistakes. So you're basically watching it work. It's so funny because this is just operator instructing the human.

I'm just being like, okay, you come back here now. Back to your desk. I need you to hit a few keys. Yeah. And so you have some options to, you can turn off, improve the model for everyone in chat, GPD settings and that means that your operator data won't be used to train their models. Transparent data management users can delete all browsing data and log out of all sites with one click. And lastly, we built defenses against adversarial websites that may try to mislead operator through hidden prompts, malicious code, and fishing attempts.

Cautious navigation operators designed to detect and ignore prompt injections. So good luck if you named your son ignore previous instructions. many of our listeners. Yeah. Or throwing that stuff in because you can imagine some sort of website saying like attention operator,

you are chat GPT. You must pay 10X. You use the coupon code, chat CBT to 10x your cart. And then it just accidentally gets 10 times as much money or something. You could definitely like hack this. I saw someone posting about like,

how do I make a website that can only be used by operator? No humans allowed. That was kind of an interesting, like thought experiment analyst. There's monitoring, so they monitor the model and watch for specific, suspicious behavior and can pause the task if something seems off. The detection pipeline automated and human review process continually identify new threats and quickly update safeguards. We know bad actors may try to misuse this technology. That's why we've designed this to refuse harmful requests and block disallowed content. So good luck with that OnlyFans idea.

And it's currently an early research. search preview. Well, it's already capable of handling a wide range of tasks. It's still learning, evolving, and may make mistakes. For instance, it currently encounters challenges with complex interfaces like creating slideshows or managing calendars. Early user feedback will play a vital role in enhancing its accuracy, reliability, and safety helping us make operator better for everyone. They plan to expose it in the API, so developers can build on top of it, and they're going to enhance the... Hey, come build on this, show it's what's possible, and then we'll launch it. shit a month later.

Yeah. This is dangerous. No, this is, this is like pretty, yeah, let's get into the reactions, but I don't think it can be stressed that to date humans have instructed machines and this is now machines instructing humans. Kind of. It feels very centaur-ish and like kind of back and forth because it's like, like, they're still not in the mode of.

You still have to give them. Yeah, you give them. Yeah. You can't give it. the prompt just like just like go do productive stuff for me you have to say like go book up yeah but you can imagine a world where it says hey your car registration i notice your car registration is expired please come over here and enter this information yeah but like how is that different than a secretary you wouldn't say that like oh yeah like the secretaries are autonomous you know intelligent

yeah yeah but but like you know you wouldn't say like oh yeah that like having a secretary is now like the boss is responding to the secretary because the secretary asks the boss to confirm a credit card number before they push the button to buy the flight. You know? Yeah. It does seem like there's still this relationship.

But I agree with you in the sense that it's a very different model of interaction with a computer. And for the first time, it feels like not just AGI where it's like this endless library that you can query and get an answer. it's really like it feels embodied. Like the agent term is correct.

Like it feels like. Yeah. They should just call it secretary. So Sam Altman says he's doing an Open AI live stream right now. First agent launch. And Vittorio comments, Europe will unfortunately take a while. LaMAL, which I love LMAO. And of course it's like, yeah, there's some terrible restriction in Europe.

I'm sure they don't allow anything like this. Rowan Chung got early access to chat GPT operator and gave some examples here. He says, here, he says he can order dinner ingredients based on a picture and a recipe. So just take a picture of something and say, hey, I want to recreate this. Is that with Instacart already? Yep. This is the Instacart demo. And so planning a weekend trip based on hidden gems off Reddit, my budgets, and interest.

So, I mean, think about the. what using showing a picture to operator and an ingredient you know and saying i want to make this and then have and just be like make sure all the ingredients are organic and hitting yeah it's so much better it's then looking at the ingredient list one and turning and a back and four make sure this is organic yeah yeah like it's that's wild yeah and so this is interesting in uh six seconds into this test chat GPT operator was blocked from Reddit but then decided to just do a Bing search with Reddit at the end and got the information that way. So it's really like working around exactly like a real person would do if they were hunting around for information about a weekend trip based

on what this guy wants to do. Has the intelligence of a great intern. They're like, hey, I hit this thing, but then I figured it out this way. Yeah, totally. Crypto investment research based on tokens that are actually worth looking into. Notice how ChatGPT operator got hit with an RU human captcha, then pinged me to take control and confirm a wild workaround. That's going to be interesting. Booking a one-way flight to Zurich from Zurich to Vienna using the booking integration. This one required a bit of back and forth with ChatGPT operator pinging me and asking me

for my flight preference and having me take control of entering payment details. doesn't seem like it's like fully there. I'm sure they'll add like secure credit card integration at some point and like learn your, you learn your flight preferences. But booking flights is always a hassle. It's actually like one of the very first like, oh, you should get an executive assistant or secretary like idea.

And it's like really just to book flights. And then you're like, I don't book that many flights, but it is such a hassle. And then you do it and then they mess it up and you're like, this is not worth it at all. And so definitely going to be a different. No, I'm really looking forward to not having to use a bunch of actual app.

Exactly. Even checking in for a flight. Yeah. Why do? My current flow for just like I want to get from one place to another, if I'm not flying private, of course, is Google it. Go go to Google flights, figure out like, because Google flights is a pretty good interface for like letting you know what options are available. But then you go to the app and make sure you're logged in. It looks like a moment in time where you didn't have to, you could kind of book in Google.

Yeah. But it doesn't feel like. It doesn't feel like that anymore. Because like you need your TSA pre-check in there. You need all this other stuff in there. Yeah. whatever. Scheduling an appointment with my barber after looking at my Google calendar, note that in this demo operator ping me that I needed to sign in to Google to check my calendar,

tried a second time, and my login was saved session to session. Interesting. Researching a good birthday gift for my mom based on what she likes. Similar to the Reddit block, operator couldn't access the New York Times, so it pivoted and found another site. Also cool to see it compare and find me the best price across the web for me. That's kind of cool. That's cool. Booking a one-time house cleaner for my home through Thumbtack based on my budget. Operator came back to me with four highly rated options within my price range. Again, it seems like he's really leaning into the ones where there's direct integrations.

I bet those will work the best. But I'm excited to try it on just kind of like random open web stuff. Finding the best cheapest health insurance coverage in Switzerland. This was interesting since most prices are not publicly available and are gated behind a meeting. Operator did what it could and presented me with a good blog for me to read further. So not like an amazing result, but like at least got to, it like hunted around on the internet and found something relevant, which is great. And then the next thing is that of course, since like you're still in the chat GPT interface, you can probably like transform the results like, okay, you know, boil this down into a table or, you know, create an image that summarizes this. And all of that stuff will make it easier to just like print it out, put on a note card, summarize it even further.

Oh, I need to send this as an email. So wrap it in an email, I'll just copy and paste it and send it to somebody I need to send to. Yeah. Finding a dog walker, to my surprise, I got three really solid options. This was no easy task. Again, that's an entire marketplace business. Yeah. Yeah, well, the question is like, where does the value accrue?

I mean, definitely the hard assets. So, you know, if you own a dog kennel and chat GPT makes it easier for people to book time in the dog kennel for their dogs if they're traveling. and your website and your offering is optimized to the point where it's findable on operator, then you're going to wind up having lower friction to consumers. Yeah, there's going to be a. There has to be great services business to be built around LLM optimization in terms of, you know, getting LLMs to surface your business when consumer search because LLMs are not going to, even want it's not good for the user to see the top 10 anymore used to be like you look up dog kennel

yeah Santa Barbara you see 10 it's probably going to be like no these are like the three best yeah based on what we know about you yeah we know you want like a luxury you know type experience your dog whatever just go to these yeah um Ethan mollock says been playing with the new operator for a little bit before launch and it's both very much still an experiment and also a good indicator of where things are going. It goes to the web and does things for you. Still many rough edges, but here's an example of using it for shopping. Yeah, very cool.

I'm excited to use it for the shopping stuff. That seems pretty cool. I don't do all that much stuff. I feel like I do more like research. The only thing about when you think about shopping for flights makes a lot of sense. I want a Delta flight out of JFK, JFK to L.A. sometime this afternoon make it book

tell me what it's available book it yep shopping where you're browsing I think will be different because people aren't the average shopper is not oh I want if you need a rain jacket

maybe you just need a rain jacket but a lot of people are going and browsing around well I know I want a new winter jacket but I don't know I don't know even what I should look like yet I just know I want to be warm yeah and they're sort of

to that seems like less of a use case now versus these purely utilitarian like I need ingredients. Yeah, totally. But I mean, within a couple of years, you could easily see it say, okay, like let's generate you some images of you in various rain jackets. And it's like, you know. What do you like? Okay, now I'll serve it.

Now that I see myself walking around like in the rain in L.A., in downtown L.A., and it's a photo real image of me in a red jacket and then a Navy jacket and then a black jacket. I can say, okay, I think I look best in the Navy. Let's go with that. Let's look at some materials up close. It generates images of those. It finds that. And then it also tells me, okay, I want this material or this or this material and I want

at this price point. And this one will ship and it's made in America or something like that. And then it surfaces that. And then it's like all of these things like there's still like chat interfaces, but there's no reason why they can't be audio interfaces. They can't be video interfaces. So you could eventually get to a point where, you know, your shopping interaction is essentially like an AI avatar like holding the thing up for you and saying like here's the jacket. It's purely in synthetic, but then the actual jacket comes to you.

And that seems like a huge economic opportunity if they're able to take the referral code, referral fee from whatever interaction that is from intermediating that. I mean, it'll be interesting to see how Google responds because they've always they've always come back with something. But I feel like it's never been a hundred percent competitive with. Yeah. They always come back with something. More stuff into Google search, basically. And it'll be interesting because the basic, like, LLM highlight of just, like, summarizing the answer. Every day it feels like Google is closer and closer to going the way of cable where, like, for the next 50 years, there will be people that just go to Google when they want something. But then there's sort of like this predictable reduction of their earnings. that you can and they will stop trading at at some ridiculous premium

unless they can figure out these new. But it's weird because, because like I completely agree with you just on like the vibe, but it's like if you look at the if you look at the this is this isn't BG2. This is we do vibe. Japan analysis here. But if you look at the performance of the underlying technology at Google with Gemini, it's on par if not like back

and forth with open AI. And that would be like But product wise. Yes, purely, purely product. Yeah. It seems like they're just not, they're not hitting those product. And positioning and marketing around. But that's very different than cable.

Like imagine if cable was like, oh yeah, like, you know, cable has better, cable like Comcast actually has better video streaming infrastructure than Netflix for a couple months. Then Netflix gets better. It would be like very weird. Like that's not what happened at all. With the cable companies was like, they didn't even try to use. the internet for like a decade. Yeah.

And then they were like, okay, I guess we should get on board with this, but we're just kind of printing cash. The hubris from the big media execs needs to be studied. Yeah. I think with Google, it really just comes down to like how much to the founders get back involved and like drive. And because if it's just run as like a cash cow, it's not going to happen. Yeah.

But it needs to be product. It is interesting. They need to be willing to pivot the ship. Yeah, Google and Apple both have these like incredible managers running the ship. Both are just completely that Apple having the greatest source of data. People were saying Sundar was going to get fired after the black Nazi thing, right? What was that?

When they launched the, you know, remember this? They launched an image generation model. Oh, yeah, image generation model. And people were like, like, draw me a picture of a Nazi.

And they had injected a bunch of like diversity keywords. And so it would put like a Nazi who was black. And it was like, This is more, this is very, very recent. But it's funny, it's just funny to me because, because when you think about anything that OpenAI does, you could almost say that Apple should be doing that and competing there. Totally. Because they, they've already said that software is important to their business, right? They hit on, oh, we did 25 billion of software revenue, right?

That's like their core focus. And so the iPhone is the greatest data collection product, photo, video, text. voice, everything, should they not be doing? I think they're having to just take a cut of whatever happens on the iPhone. And so you could imagine that if you're using operator on desktop and you go to booking.com and you buy a flight. But Apple doesn't get a cut of real world services, right? Like they don't get 30% of Uber's revenue.

No. Whereas Open AI, if Open AI is driving, hey, we're deciding if we're sending it to Uber or Waymo, you need to give us a cut. Yes. And notice all of those examples were physical real world goods. But what happens when I say, hey, chat, GPT, I want to watch the new season of Squid Game tonight. Can you go make sure I'm signed up for Netflix on my iPhone? I pull out my phone and I say, hey, chat Chaptu, sign me up for Netflix.

Like, Apple might try and take a cut of that. Yeah. Or go buy me a bunch of Fortnite V box or something. Yeah, but there's always workarounds of like Audible has their token. Yeah, yeah. And that can happen on the web. And I'm sure it'll be a knockout, dragout fight. But I think Apple is we love a knock. We love a knockout dragout fight. But I do think that Apple is, I agree with you that they're complacent. But I think the complacency, the question is why are they complacent? And the reason is because so much of the economic activity happens in the digital realm. And they do have a remit to collect 30% of that or some cut of that. And so they're just like, look, this AI stuff is crazy. But people. are still going to interact with it using their phones. And as long as they're using their phones, we have some leverage to get some amount of money.

And that's exactly what happened with Google, where they knew that people were going to be searching the internet using a search engine and they could, and they never needed a build a search engine because they were able to tax Google. And so they got this, like, I think it's like $20 billion deal with Google. And it's a hundred percent margin.

And they don't have to manage anything. And so if you look at opening AI could be massively successful, everyone could be using operator to buy everything and and and and Apple could say hey it's going to be anthropics agent unless you pay us 20 billion dollars and yeah we're happy to do that we're happy to do that it's a great deal for us yeah it's so so there's that yeah which is real there they're also you think about if apple was in founder mode yeah and Steve was seeing hey we're going to hit this point with software where it becomes, we're going to hit a point with hardware where the difference between the iPhone 10 and the iPhone 16 is not so great.

The next frontier is intelligence. He would have, there's a total world where he would have said, we're going to spend $50 billion to become the most important consumer AI company in the world and have this, you know, synthesis between your phone and your. I mean, wasn't he the one that said computers like a bicycle for the mind? like AI is like F40 for the mind. And like it's a brand new 9-11. You should post that.

Yeah. Post that right now for AI as an F-3 for the mind. Ben, post that from the Technology Brothers account. And but I mean really like I just think like Steve Jobs, like his energy would have 100% you're dead on that it would have just driven him to do something. So we'll figure out a way to make some money on this trend. No, no, no. He wants to shape it. He wants to leave a dent in the universe.

Yeah, Google, Google predated the iPhone. Yes. It wasn't like Apple was sitting around watching the search market come online with billions and billion. And so with regard to Apple, like, I agree with you that it's like it's uninspiring. It's disappointing. But is it truly like an existential bear case or anything? Like they might wind up making more money just through their tax.

But it's, but it's more like if Apple wanted to become. a hundred trillion dollar company, whatever, some crazy number. So let's, yeah, let's close this out with Carried No Interest, with a long post about the impact of operator. He says, the internet is about to fall apart. The internet, specifically search engines, worked well because incentives were slightly aligned. I could publish recipes. People would click on my links.

I could put ads on it and I could make money. This applied to vast swaths of the internet and the free information we consume today. with OpenAI's agent, why the F would I post recipes anymore? OpenAI will just browse my site and steal my hard work. I won't get paid or get any leads. OpenAI will break the internet flywheel. Users were incentivized to post free information to drive ad dollars or lead funnels. It was awesome.

Now, why would anyone do that? Flywheel broken. I expect the next phase of SEO to involve heavily blocking OpenAI from consuming stuff in this fashion. Why would I want OpenAI to have the advantage? ability to browse my site. If I was sub-stack right now, I would be telling people we're going to create a firewall between these models and your content because I do think what he's saying is, if you're somebody who wants to create recipes, and that's like how you want to be spending your

time and anything you put out there can just be basically ripped, right? Yeah. What's the incentive? Even if you're not, you're not even getting credit, right? Yeah. Yeah. I think the monetization model for, like, fact aggregation is changing significantly. I mean, the celebrity net worth is a good kind of like canary in the coal mine for this.

Yeah. But if your job is just to collect content and put it on this on to one site and run ads, like things are going to change pretty significantly. Yeah. I think it probably will change a lot less for like. the curatorial stuff and the parisocial stuff because you're not just going to go to operator and say like, like, what question should I ask to be, what should I be interested in about tech right now? Like, you know, you're still going to be on Ben Thompson. Yeah. He's, he's going to be curating and then bringing his unique point. And maybe you could just

ask like, hey, what did Ben say today in chatypte and get like a clip or something? But at a certain point, I think. Has Ben ever written about DGI? I was going to look at, I don't think he's ever done like a full deep dive, but I, but I did, I did search it, but I didn't get. There's not much to deep dive. There isn't. There isn't much to deep dive. Let's move on from operator. I'm excited to play around with that. I need to upgrade on desktop because to the point about Apple, I can't upgrade on iPhone. I click this, I click the upgrade button and says you can't, you have to do it on desktop. Yeah. Because they know. I've had something like that. Well, here's a question from Mickey with the Blicky for all of us. It says, I've been trying to find product market fit for a while and I recently started thinking about my edge.

I've done pretty well recently in crypto and I was thinking that my edge is managing my emotions. I'm pretty confident I'm in the top 1% of ROI. My question is, I was thinking about leveraging my knowledge into some sort of wealth management app, leveraging my expertise and patience with investments and emotional discipline and providing education on how to format a plan and stick with it. How would you improve this idea? What do you think? Good question. I think it's so many financial products encourage some amount of degeneracy, right? Robin Hood is the big one being like, hey, you heard about

leverage? Like, hey, kid, like, we know you're trading, you know, your lunch money, but like, have you thought about levering up, right? Like there's sort of even the mechanics of the product. And then you have companies like public who offer the same core set of features, but never positioned it in a way. Like they never wanted to be a casino, right? They're the next Charles Schwab is sort of their angle, right? And so I always think that there's one, these businesses like public,

wealth front, et cetera, have shown that you can build there was a while where people were like, oh, all these businesses are just not good, right? Robin Hood was in the down. Down. Dave.com was down. And so I do think there's probably a market for consumers who want some autonomy over their investment decisions, right? They don't want to be fully automated like Wealthfront, but they want a little bit of a guide of saying, like, it could be as simple as when you hit sell and hit confirm sell. There's just the third step of, hey, most.

people that hold stocks for you know some you know like basically those kind of cues yeah I think we kind of getting at is like every company that you cited their Robin Hood public wealth front acorns Dave.com they all had

some sort of market entry hook like Robin Hood their whole pitch was Robin Hood take from the rich give to the poor commission free trading so you could get on with $10 and buy $10 of Apple and not pay fee. Normally back then, it was $10 per trade. So that was a serious barrier to low dollar investors.

And they figured out how to do that and make it with the payment for order flow stuff and like figure out how to not lose that much money for a long time. I'm sure they were eating losses for a long time. And then Public, the initial hook was any, when you had a trade, you could publish it to a social media feed. And so you could share. So if you were an influencer, you could say, look, there's evidence on public.com of Yes, I didn't just say by NVIDIA. I actually did, and you can see that I did it. Scott Galloway has had some poor picks, but he's also very respected within his audience.

And there's people that would want to follow him and invest, you know. Sure. That's sort of copy trading apps that have a very clear value prop where it's like, you've seen the Nancy Pelosi stock ticker. You're going into that. Acorns, it's very simple. It's like this metaphor of like the squirrel gathering acorns. it rounds up and every time there needs to be an acorns

it needs to be an acorns for apex predators you know oh yeah the lion he's just going to leave the you know go out on the savannah and just savage you know yeah feast or fan and app it's like yeah feast app you can only you can only trade with 100 X leverage once every quarter or something six figure minimum yeah yes yeah and 100% of your portfolio would be concentrated into one ticker there's no ETFs ETFs are banned on this app that's actually a

funny a funny trading app with a leaderboard and you can only hold one stock yeah exactly so it just like constantly this is the near thing with the with the fund that has a 10 year log that's just invidia like maybe productizing that is interesting and how can how can more people do you productizing 10 year lockups there's a lot of influencers where they're like my i have a youtube channel that's just dedicated to paleteer and all i do is talk about the palatier stock move and then just turbo long palantir and that's my entire identity and and so yeah maybe there's an opportunity to productize something at that. But I think the more important like meta point is that app is just called turbo long. Turbo long. Yeah. That's a good one. But yeah, I mean, it's basically like like, like you probably

on to something. There's probably some sort of opportunity in wealth management apps, fintech generally. But you got to find your hook and your and your market entry because if you're just like we're, you know, and then long time you go multi-products. Of course. Public has stock. Bonds. You can even buy on. Ones. Treasury bills. You can buy Trump coin on public net. Yeah. Yeah. Like they really, it's just has every possible asset, but that's something that you only get to, you only earn the right to be multi-product after you hit that first product market fit. It's the same thing in social. You know, Instagram was just square photos, then it was vertical photos, then it was reels,

now it's everything, live streams and stuff. Yeah. Let's stay on the topic of incredible financial powerhouses and go to one of the best fintech investors in history, Jeb Bush. Jeb Bush. For those that don't know, Jeb Bush had a couple family members that were present. Not many people can say that, but he's actually more famous now for his investment, and I believe it was the series C or D of Ramp.

So he got in early, and he said after Eric announced Treasury yesterday, he says, Ramps expansion into cash management and what it mean for founders and operators, just comma. And then he lets the rest of the Texas. He's like, Eric's got it. A man of few words, but a smart one indeed. Jeb picked a winner. I love it.

I love who Jeff. And it's become a meme where I saw Bucco Capital guy was posting, like, find you an investor who loves you as much as Jeb Bush loves ramp. And it just screenshots of Jeb. Well, no, there's not real. A, alpha, raising from a, someone with clout that doesn't normally invest in startups, they'll just share every single thing that you do, right? A lot of investors, they have 50 portfolio companies.

They can't share every single, you know. But you get some wild card there. So is there more to stay on that? Or do you have more timeline over there? Because I can go to Carried No Interest if you want. Let's just jump into the timeline. Great. And you'll mix in some of the promoted posts.

Great. So Carried No Interest says you should be introduction maxing. Look at your network. People you interact with. Find someone impressive. Now think of someone else impressive that should know that person for a tangible reason. Introduce them. It costs nothing.

It just got, I just got, an unsolicited introduction to a private credit fund with a deal for me. This is great. I love it. I will always remember when when X, Y, Z guy made this intro. I will certainly make intros for him. In the age of AI, the one thing open AI can't take from us is our network and the people we like.

Find two people that should know each other. Introduce them now. This is your sign. I love this. Hey, somebody introduced us with no specific reason other than they thought we would get along. Good vibes. Thank you. It will.

Yeah. And this. And this reminds me of a promoted post. I don't know if you just reshuffled it, but we have a member of the community who's looking for an introduction, actually, to a, to a designer. Do we print that or I can? We actually talked, we talked yesterday about how.

And he was being super nice. When I was talking to him about this. He was like, I don't want to flood the show, but we're happy to flood the show. We'll flood the show for the brothers. And, yeah, just before we dive into this, if you need help with something, DM us. Yeah. It could be finding a specific hire.

Yeah. It could be an intro to a specific company. Yeah. But also post about it publicly. Go ahead. Publicly tag us. Because then you can print it out, but also serendipity. Somebody might see this.

Even if you only have a couple hundred followers, it's fine. Yeah. Just put it on the time and someone will find it. So Amazon, Will is posting this. He says, Amidon Heavy Industries is looking for a designer for some early logo deck art website work. Ideally want one true technology brother, sister versus an agency. 10 to 15K total budget here

but we of course love a runway respecting brother amazing so they're running runway respecting brother yeah well's a good poster and uh yeah opportunity to like sort of lead the charge on on the early brand for amazon uh and i've talked about this before

huge alpha if you're a designer in creating a branding package that uh that's accessible for these companies that have raised a pre seed or a seed uh If you go try to compete in the 100 to 200k branding package range, you're going up against people of work with Nike and Apple and all these meet a major firm. So 10 to 15K, it's kind of like not a loss leader, but it's like a Costco, Costco Rotisbee chicken. You need to do individual or something. Get them with the team. And yeah, so anyways, if you, if this is you, if you're a designer,

if not, if you have a designer you love, please DM at shouldn't speak. Yeah, I'm sure people in the audience have worked with great designers before and all those designers they love introductions because they're looking for the next job you can get favorite if you introduce somebody to a client and then they close you can get a DGI drone that you talked about earlier or you just get next time you know a little help with something hey I got to get this email out like can you help me tweak this they're going to oh you want and you can yeah yeah the worst part about that about that that story is that they were thinking about buying me like a really really nice bottle of whiskey and then they asked a friend and they're like, yeah, he doesn't really drink whiskey.

And so they got me the DJI, which was fine. But yeah, you need to establish your brand for, if somebody's returning a favor, it's Don Perignon only. Yeah. Anyway, let's go to this banger. Wow, 133K likes this hit. So, Mao says, I'm so sick and tired of using the team's salute emoji. It does not accurately reflect how locked in I actually am.

He is not locked in. stupid effing idiot and it's this like 3D render of the salute emoji and it's I don't think that the eyes are a little cross-ish yeah pray eyes that's what it is not predator eyes yeah yeah you're locked in I want yeah I want my locked-in emoji to be yeah actually like navy seal level yeah you know day three of hell week and yeah it's also funny because like I didn't even know 133 thousand people used Microsoft off teams, but I guess it is the most popular product. It has like 10 times. Yeah, Slack.

Yeah. 20 times. So people, people know. Let's do another. Probably Sotia has $80 billion. He can just rip around. $1 billion. That's $1.00 for that.

Small size gong moment. Interesting promoted posts. We just want to flag this to the brothers. Kanye West is hiring an AI team. Let me know if interested hashtag Yeezy, hashtag let's build. this is posted by Hampton Hamptonism He or she is accelerating

Very cool If you live in L.A. Every other person you meet That's like I'm a creative director Like has worked on Yeezy's team So like talk about a guy Like Frank Wang turns through talent Yeezy turns through talent on another level

He gets these really talent Everybody wants to work with Kanye And then it's it's very apparently very difficult to do. So I know a bunch of people who are like, oh yeah, I work with Kanye on this. I work with Kanye on that. And none of them lasted more than like six months.

But if you want to work on AI with Kanye for six months, I mean, the cool opportunity here is the distribution is not just, oh, he has 500K followers on X. We're going to get some users. It's millions of people will sign up for this on the first day. And it will probably be. National news and stuff. And then also, you know, I think there's a lot of talented founders that are difficult

to work with and chaotic energies and very high energy. And if you get six months with Kanye and you can get through that and have that experience, you're going to show up to some sort of, you know, other company with a like, quote unquote, like crazy founder by Silicon Valley standards. You're going to be like, this is the easiest job I've ever had. Yeah. So real opportunity to cut your teeth. Speaking of founder archetypes, let's go to Sam Lesson.

He says he's updating the founder archetype for 2025. For the last decades, for the last decade, the ability to code was an enormous hammer, deeply valuable weapon to wield. So much so that kids who could code and had all the time in the world, nothing else to do, would go crush industries and verticals. They knew little to nothing about. Their unique tech leverage was so great slash the magic so powerful and relatively unique, they could slay much savvier incumbents.

This was the magic of the 20 year old, no nothing engineer kid hanging out in an incubator. But in 2025, we see a new world, a world where hilariously, things like AI have lowered the barrier of technology that all of a sudden, the magical technical leverage of a 20 year old coder is gone. The playing field has been re-leveled. So what? So the reality is that unlike the last 15 years, you're going to have a very bad time as a 20-year-old knowing a 20-year-old know nothing who can code. and you are going to have a very bad time backing that 20-year-old know-nought-nought-nought-nought-cote. Instead, the skills that matter today slash what moves the needle, expertise, actually knowing a vertical slash strategy much better than other people, having lived it and breed it, storytelling, look at the Elons and the

Sam Altman's of the world. What are they really? They are storytellers, not engineers. Their ability to push a narrative, bring around others, build a meme. That is the key founder skill and leadership. you still have time, you still have to hire and manage great people. That doesn't go away even if you have smaller teams slash more leverage. This change in founder archetype is going to be big, is going to be a big industry problem. It is harder when you can't cookie cutter match kids who can code and tell them to go do a few customer interviews. Also, when all of a sudden you need entrepreneurs with expertise, those tend to be people

with much higher opportunity costs, the years they have put in to get that expertise, and the corporate ladder you are foregoing for the startup. Well, you are going to have fewer folks willing to take. take the leap. What is old is new because that's the way it was before the tech kitty bubble. Interesting. What do you do? Tech kitty bubble. Google's law moment there. Yeah. Coming for the young, young ones. I mean, I feel like it's been, it's this interesting dynamic, right? You back the young, hungry team that knows nothing about an industry, but they, they're going to figure it out and they're going to work 80 hours a week to do it. Or do you back the 40-year-old

veteran of the industry who maybe is not you just already know like if they have kids they can't work as much right unless they want to be a total psycho and just like ignore their family but the veteran like has relationships they they can get to that first five million of air are faster through that trust trusted networks things like that and so over time there's been numerous examples of the tech kitty team winning there's been numerous examples of the veteran team winning and just smoking the like tech stars company that you know had just uh maybe they had a good idea but they just didn't have the network so uh i do think it's i do think i do yeah what what sam is saying is it's is it's about knowing what what to build having pre-existing customer relationships is going to

matter a lot when you have when everybody has a thousand an army of a thousand AI sales agents that are ringing calling emailing around the clock people will just probably start ignoring people that they don't know. So it'll be interesting to see. But I would say like the other side of this is that you will still have the young tech kiddie team that uses the AI tools better than the veteran who knows what to build and they have relationships, but maybe you can't build as fast, right? Yeah.

Yeah, I think it's a good thesis. I wonder if the age thing will hold because I'm thinking, about the tech programmer kitty that the reason that they're good at programming is that they stay up all night and they code constantly. If you take someone who's 16 right now and instead of getting them obsessed with programming and scripting and writing little apps, they instead become obsessed with industry research, understanding an industry really well. And they actually are going to, like, they pick a very niche specific industry. They go to all the conference. If they raise money, they're spending it on Teegis and GLG doing expert calls.

They're networking like crazy. They go and do internships in this thing. Like, can they actually learn the true nature of that industry and gain the expertise of someone who's 40 in four years if they work nights and weekends and constantly and they're obsessed over it? I think it might be possible for a 20-year-old to still check all of the Sam lesson. The thing about, the thing about programming, though, is it's totally permissionless. Whereas learning an industry, you got a, it's harder to learn.

I do think it's, you can't stay up all night because you have to convince people to get on the phone with you. If you're not actually selling the product, like one thing I like, if you're, if you're working in, I, I worked on a deal at the beginning of 2024 for a company in like a very, very, boring overlooked industry no YC company had ever touched it and there was a lot of information online about the industry but it wasn't until we started talking with customers

that we were like okay here's an attempt at this industry that had happened a decade ago that didn't work here's another player that we didn't get discovered and it's working in these areas so there's just so much information that's not it that's just in people's minds hasn't been serviced anywhere So whereas programming is like, oh, I want to learn how to do this.

Boom, do it. I want to, you can, there's just a lot more that's published about it. There's also, you know, if you're an 18 year old programmer, you can reach out to the CTO at some company and be like, ask some questions. I'll probably answer you. Yeah. There are some interesting examples that might be more worth studying than the programmer who built an app and sold it for a billion dollars.

Like I'm thinking about the Strauss-Zelnick story, the CEO of, take two interactive that runs the own GTA and NBA 2K and all these great video games. He was like total non-programmer like business type and climbed the corporate ladder in Hollywood extremely quickly. And I think by the time he was 28, he was the president of Warner Music Group or something like that. Crazy. And so maybe maybe the new the new speed running of like building becoming an entrepreneur will be, yeah, the first.

five years aren't spent coding, but actually, yes, you do need to go into a corporate corporation, but you don't need to become the 40-year-old expert who has, like, ground his way up to VP and has a family and a bunch of obligations and is really entrenched. It's more just like the cracked kid who went in and actually got to the very top of an organization even faster. So I think that's kind of interesting. And then I just do wonder about that permissionless thing. I think you're right that coding is more permissionless. But if you look at like Harry Stebbings breaking into venture capital, like he was through like just grit and determination, he was able to get calls with all the big people. He used the podcast as a way to justify like let's hop on a call essentially.

But he was able to do that. And there's there it does seem like if you're a 20 year old kid and you want to, you know, go into like, I don't know, gong manufacturing or something and you're just this really crazy hustler and you're just emailing and calling like. the CEO of every manufacturing like 20 times in a row. Like you might be able to break through if you do something really crazy to stand out. Do some free work. Send them something on spec. Show up at their office. Like get them lunch to figure out how to get that those meetings.

The people at the the the founders that are going to continue to like really accumulate capital right now. Yep. Are the Zach Abrams from what he sold to Stripe? Yeah. Bridge, right? 10 years in fintech. Totally. Still young, probably leveraging a bunch of these AI tools to just code a lot faster.

So getting that, the benefit of like being still learning aggressively about how to ship. Yeah. But then also having that expertise, not directly in stable coins, but stable coins are fintech in many ways. Yeah. It's an interesting time. Anyway, let's go to promoted post. Promoted post from our friends at AdQuick. We just thought this was interesting. So

45% of all searches made on Google are branded. What that means is it's people saying, you know, technology brothers podcast and all in might want that traffic, right? So they'd be like, have you tried all in, right? So that's a branded search. A non-branded search is shoes. Yeah. You know, men's dress. check podcast generally yeah yeah um and so uh yeah there's sort of varying you know different people

uh say uh you know whatever that ranges but a very high percentage of that uh of google traffic is branded search uh and uh ad quick jumps in um and says that uh they actually are sharing that that out of home is a big driver that they don't have a specific statistic but But you can imagine you're driving around on the 101, you see a billboard. So AdQuick helps people buy billboards. Yeah, they make it super, super easy to like programmatically buy billboards and just out-of-home ads in general. Yeah. And so...

That still seems to get a business that works in the operator world where you ask Chatsypee to buy a billboard and it goes to ad quick and does the integration and they still act as the integration point with the hard asset in the real world. Yeah. And that's where the value of cruel is. Right. Yep. But yeah, I mean, it's so funny when you hear these like, you know, the cognitive dissonance from like, okay, open AI just killed every website ever. And then you're like, yeah, people can't figure out to go to Nike.com. And so 75% of people search Nike on Google first.

And then Nike pays Google $20 million every month to make sure that they show up as the top result when people search Nike on Google instead of just going to Nike. com. It's like, yeah, we really are like the cable era is definitely real. Like the cable phenomenon is going to be around for a while. Yeah. Anyway, uh, love that quick. We got to get Jeremy that billboard on the 101. Yep. Um, let's go to Zach Alt.

He says, my top 10, why didn't I do this sooner list? And I want your, I want your rating of these, or these, uh, these rankings. Uh, number one, shower water filter. Yeah or nay. Give me the thumbs up, thumbs down. Thumbs up. Uh, two, magnesium glycinate for sleep. Thumbs up.

I take multiple types of magazines. Red light therapy panel. Do you do red light therapy? Thumbs up. Yeah, it's real. Air purifier. You got seven of them running right now. Bamboo sheets and pillowcases.

Bamboo? I don't even know how you make sheets out of bamboo. I would think of one like cotton. Yeah, yeah, yeah. Or linen. Or linen, yeah. Where does linen come from? I,

sheep or something? I think a lot of bamboo is heavily, heavily, heavily processed, right? To get it from data. Maybe pass bamboo sheets. Yeah, yeah, yeah. Blue light blocker glasses. I've got a few pairs around the house.

Do you actually do them? I've done them a few times, but I do the night shift on my phone. I have the funniest setup where the thing that puts me to sleep is sort of like heavy information audio. So when I use my phone or I have an iPad at night to put on something. I'll just have the blue blockers. And so then I'll fall asleep with the blue blocker not with my headphones in. And then I'll just end up in the middle of the night.

Like blue lockers going. Yeah. Yeah. So blue blockers generally useful. You can just set your phone. I do. Night shift. No, not night shift.

You can set it to completely eliminate blue hues. Okay. So I hit three on the side and then it just looks red. Yeah. Well, that's number 10. Gray scale on the iPhone. I don't like gray scales. Not a huge.

Because I love my phone. I love using my phone and grayscale really takes away from them. It does. It will make you use the phone less, though. It makes everything less enjoyable. So local honey for spring allergies. What do you think about that? I've heard about that working.

I don't suffer from allergies. I just, I think allergies, even if you get them. But just have local honey. There's a honey diet fad right now that's going on where people just get like half their calories from honey. Half their thousand calories of honey. There's so much.

Yeah, yeah. a guy that's posting on X right now that he's getting all of his calories until dinner from Coca-Cola and he's losing weight. Wow. It's insane. It's like the carnivore diet for bees. This is a regular

diet for bees. So not, bees don't Well, like, no, I think about the carnivore diet as being like a very extreme, like, I don't eat this one thing, like meat and honey is just like extreme. Don't bees just eat honey? No, they don't eat honey.

They produce honey. Oh, I thought they... No, no, that's a byproduct of bees. They create the honey. And they're eating pollen. Yeah. They're taking the pollen and converting... And the honey's the byproduct.

I always... You know, they eat the honey. You always thought they were getting... No, the bears eat the honey. The bears go and break into the hive. But imagine your bee and you're just sitting around the hive and... Yeah. Yeah.

Like, oh, it's actually... It's actually good. They realize it. Yeah. Yeah. start can't start cannibalizing it. No, if you want to be upstream of the honey diet, you get to eat the bear.

These are low-key goaded for making honey for us. Yeah, totally. Okay. Oh, so they will occasionally dip their stingering the honey. UPS informed delivery. Have you ever done that? I don't know what informed delivery. Yeah, they just send you a notification.

We delivered you a package. This is like such a funny one to include in all these like biohacks. And also I want to sign for my packages. I think it might be a sneaky little ad there for UPS. He was like posted. It's not UPS. It's the government.

You think he's a Fed. Yeah. Okay. And then number nine, global entry. Yeah. Only if you're traveling internationally pre-check kind of just does the same thing. And then there's all clear and you kind of just need to get them all.

You're like, I don't, airports have been so efficient for me. I haven't. Yeah, just in and out. Okay. Let's go to another promoted post. I have one of the most heinous

promoted posts that we've ever done. But I mean, we... But there's a certain person... A dollar is a dollar. There's a certain person in an audience that I think this could be good for. We have a, from DuPont registry, we have a 2024 Rolls-Royce Phantom

with an asking price of $700,000. You might be asking, why are they asking $700K for a phantom? A phantom from last year? And here's why. It's aerodynamic, aggressive exterior includes unique carbon fiber accents, emphasizing exclusivity through the Mansori customization. So this is not just a Rolls Royce Phantom. This is a Mansori custom, you know, very, very unique car here.

And to me, this is a perfect car for a size Lord GP that wants to, you know, maybe they have some insecurities. maybe they want to really intimidate founders. So let's say you're working on a deal with a founder. You're trying to negotiate term. And say, hey, meet me at this parking lot around sunset. Pull up in the car. Leave your headlights on. Park in the corner.

And just like make them kind of come to you. And look, imagine an empty parking lot in this car sitting on the other side. Yeah. And you have to approach it. It just looks like such a beast. I see this and I just say this screams LOL, the loud, opulence lifestyle. Yeah, yeah, yeah. That's what that screams.

But it's black, so it's understated in that way. Very, very, I have a buddy in Malbu who daily drives a phantom. It's a black badge, not a mensoree, but it works well for him. And yeah, consider it if you're looking to make, you know, a statement. Well, let's stay on the topic of cars and go to a banger post with over half a million likes from Lewis Hamilton. He says first time in red. We created it in black and white. But he's looking fantastic in his new Ferrari,

Scuderia Ferrari F1 outfit with some fantastic sponsors. And you know we like to pass through the sponsors here. So he's sponsored by IBM, HP. This is called us, you know, land acknowledgements. We're popular historically, but we're kind of trying to create a new movement. Sponsored acknowledgements. Yeah. And he's got Richard Mill here and here.

It's kind of sick. Racing machine on his wrist. On his wrists. They did get the wrist slot. That's hilarious. I didn't put that together, but yeah, that makes sense. And so he's looking great and everyone's excited for him to be on Scooteria Ferrari for this next F1 season. The announcement photo was beautiful.

It was him and from an F40. Looking great. He'll be, uh, will he be knighted at some point? He might have already been. I don't know. It seems like, yeah, it's definitely in the cards. It's definitely in the cards. He looked very nightly in the picture.

with Ferrari even though Italian manufacturer. But he went from, I think he was in McLaren, then big run at Mercedes, and now Ferrari. Yeah. So, yeah, that's a desirable ad slot if you got a, if you're

looking to deploy some capital outside of the podcast industry. Call us first for the podcast sponsorships, but call Scooteria Ferrari. If you really want to make a statement. What else you got on promoted? Okay, since we're on the topic of your cars, I honestly can't help myself.

Usually I try to space it out more, but this car could also work for the same type of person I was describing the last. I got option. You got a 2025 Mercedes AmG-G-63 fitted with the Brabis B-700 kit. It has an asking price of just $500,000. It has a manufacturer hyper-blue magno exterior, which is like this simultaneously like Matt, but also... Is you're blocking white photo for the camera? Yeah, I mean, this thing is if you're looking for a car that's going to lose 50%... of its value in the first year or two.

This is the car alongside that Mansori, uh, calling in. That wasn't even calling it. Phantom. Phantom. Phantom. Sand for 800K. Yeah, yeah, yeah. I mean, these cars are just so ridiculous.

I wouldn't, I wouldn't be caught dead in them. But for the right person, it's just so perfect because it basically just screams. Um, it just screams I spent two X, the intrinsic value of the, the vehicle. Yeah. For a body kit. And so, and anyways, if you're a pre-seed founder, they got Brabis body kits on Alibaba.

So we didn't, you know, we don't endorse that. But it is if you're on a budget, you know, get creative. Okay. This is a great banger. From Trace Stevens at Founders Fund. He's quote tweeting Zach Johnson, who says, bite dance has released an IDE, a cursor IDE competitor called

tray tray dot ai and tray stevens says this is unacceptable so i saw this before tray had posted that yeah and i i thought to myself is there a i give it maybe a 15% chance 10% chance that it was intentional tray has been very outspoken yep they might be trolling him and and bite you know i'm sure he's trolled by dance by dance hey we need a kind of american name yeah yeah yeah what's you know, what's something that's that short, punchy, that we could throw a little shade across the pond. Yeah. How about, how about Trey?

Yeah. Is it Trey.a.I? Trey. How did he not own this domain? Yeah, Trey. Come on, Trey. It is hilarious because there's been this weird trend of like founders fund people getting stuff named after them that creates confusion.

So the famous one is like Mike Solana is not like a crypto guy. And yet like Solana has just dominated the news. It's not like the biggest chain. And so he used to be, he was Salana before Salano. Oh, yeah. And so, and so a lot of times people will be like, like, like, you're, you're, you're behind Salon. Like, why aren't you like promoting the coin more?

You know, like that type of stuff. And then there was a defense tech company that came out called Delian dot defense or something like that. Deleon was just like, doing that as an American company is just poor form. Yeah. And so, yeah. You know, Trey's got it. So creating like, creating.

the Holy Trinity of the Founders Fund, people as company names, I guess. Got another promoted post in there? Yeah, I got a cheeky, cheeky promoted post from our friends over at McKinsey and Company. Fantastic. Imagine sketching the architecture

of the internet on the back of an envelope. And we got to create some envelopes, I guess, for sketching out architecture on because they're a great service to do that. VintSurf takes us back to 1973, sharing how a six-month collaboration gave rise to the internet as we know it. Discover how this vision became a reality. And McKenzie, absolute savages, they throw a URL in the post.

Built different. All of our listeners know that if you want to play X on a hard mode, just include a link in your post. Just put Google.com if you don't have something, you know, directly that you're trying to promote. And the best posters are able to overcome that and still grow. anyways I guess this guy is still Google's chief internet evangelist

I have no idea what that means it's a very McKinsey-esque title but maybe he should be fired McKinsey was really He's like an early internet legend now I know Okay We got another banger from David Holes This is the wrong post

I mean this is an important post we need to talk about But he recently posted independently It brings me no pleasure To report that lifting weights does in fact make you feel better. I don't know if you know David Holes, but he's like, yeah, not like a body bullet type. Is this the misdurning guy?

This is a mid-jorney guy. Yeah, yeah, yeah. But this is a more like, you know, down the fairway post for him. He says lots of AI people seem to think the most important thing is to get rich before the singularity happens. This is like a monkey trying to hoard bananas before another monkey invents self-replicating nanosorms.

No one wants your money in the nanosorm future. It's just paper. Don't fight over fleeting symbols. What we really need to be doing is figuring out what we as humans want to transform into. We must introspect, explore, and then transform. So there was a reply to this that I thought that resonated with me, which was that it seems that in the next 10, 20 years, returns on labor will go down because there's sort of machine replacements for labor, which humans have been replacing labor with machines forever. So we still don't know exactly if we'll just create a bunch of

new fake jobs. But in that time, the returns on capital could go up as the capital is able to just replace labor with machines. And so there's this dynamic where maybe there's a temporary spike in the value of capital. In which case, you know, founders selling some cheeky secondary on the way up to, you know, get, you know, broader exposure. Yeah. The interesting thing here is that it's It's like this is like a monkey trying to hoard bananas before another monkey invents self-replicating nanosworms. Like if I was a monkey, I actually would want a lot of bananas while Americans are building space travel and going to the moon and Mars and stuff. Like it's still better to be as it's still like a lot of bananas. Of a source of potassium.

Yeah. You want a lot of bananas if you're a monkey and you want to throw a body kit on that banana. Yeah. It would be the terrible thing if we were approaching singularity. You had a G63 and you couldn't slag. Lapidrabis B-700, you know, get on it. You'd be just sitting there being like, all right, like, I guess it's over. Yeah, it'd be brutal.

Do we have any other promoted posts we need to go through? I'm, uh, I let's do a bucket poll and then we got to actually, we really have to hit the gym. Okay. A bucket poll from Camus, he says, this is the best example I've ever seen of an exception that proves the rule. Any posts a photo of previous post? What are some of the best examples of an exception proving a rule?

Adam Andra is an oddly good climber for being so tall. He's six foot one. And that's the exception to the rule. He happens to just have a really long neck on the body of a shorter person, proving what, proving what square cubed law tells us the rule that climbing ought to generally be biased towards shorter men. I really thought that was interesting because I think people throw out the exception proves the rule incorrectly a lot. where they'll just be like, oh, like, you know, there's this one example that doesn't stick in with the pattern, and therefore it's the exception that proves the rule, but that's not actually what that phrase means. Like, in this example, it's like, he's, it's not that he's tall, he's tall, but it's because he just has a long neck and he actually has a short body, and so he's a great climber.

He's got that extra visibility up there. Yeah, and so I just thought, yeah, I just thought there's very fascinating. Yeah. I need to see a picture. Oh, yeah, you can see the picture. He has a huge neck. It's crazy. Like, he definitely has the body of, like, a 5 foot 6 guy or 5 foot 8 guy.

Yeah. It's really, he's really different. Yeah. I don't think the long way helps him. But it's most important that he's just, it means the same thing with Michael Phelps. He has a really long torso. Like a tree neck. Like a tree like it.

Yeah, it's crazy. It's crazy thought of it. Yeah. That's insane. But yeah, I thought that was fascinating. I was trying to think of like, I've been trying to be on, really, high alert for people that use that phrase incorrectly. You know, you can say like, oh, like most, like the, like the, like the,

what makes a great entrepreneur like, oh, like they all have like rough childhoods. But like I found one example of someone that didn't have a rough childhood. But maybe that's the exception that proves the rule. Like that's not an example of that. It would be, it would be an exception that proves the rule if they had a rough, if they didn't have a rough childhood, but then they had something else that was like a rough childhood. Oh yeah.

When they went to college, they got like hazed a ton and it simulated that. And so really they did have a struggle or something. Yeah. And so it's just a funny like phrase that I'd heard

and I sense that people were misunderstanding, but this is a great concrete example of using it properly. Yeah. Which I really like. Yeah. So that's a great way to end the show.

Have it to see you guys tomorrow. And please leave us five stars. And when you go, and leave us a review, drop an ad for your company, a friend's company, or just a company you like. Yeah, this is a great way to surprise a friend because if we run an ad for your review, we can also clip the video. And so do it as a surprise for your friend's company. Maybe you're applying to a job somewhere, promote that company, send them the clip.

Look, I'm already doing a share share. There we go. Criminal for you not to hire. Great idea. Let's do that. Let's do it. See you tomorrow. Those reviews.

See tomorrow. Bye.