This Week in Startups - Why F1 Teams are Replacing Wind Tunnels with Smart Tape | E2305

Episode Date: June 27, 2026

This Week In Startups is made possible by:Plaud - https://Plaud.ai/twistSentry - https://sentry.io/twistNorthwest Registered Agent-https://northwestregisteredagent.com/twistDigitalOcean - https://do.co/twistToday’s show:F1 teams spend up to a third of their budgets on aerodynamics, often utilizing wind tunnels to simulate what happens on the track.But Lyall Davenport of SKN Systems created a sensor-embedded tape that sticks directly to race cars as they drive around the real world, generating 420M data points per hour. And it comes in at roughly 95% less cost than running a full simulator.Hear about how Lyall’s background in racing helped him develop the concept, plans to expand from motorsports to drones and defense, and his experiences going through Jason’s various programs.PLUS OpenAI’s new models are so good, you can’t use them. Why Apple’s Vision Pro chief defected to OpenAI (and why their laptops and devices are getting more expensive). A look at the world’s most annoying data center in Sterling, Virginia. And Jason remembers iconic and influential tech journalist Om Malik.Guest:Lyall Davenport: https://x.com/LyalldavenportSKN Systems: https://www.skn.systems/Timestamps:0:00 How SKN uses tape to generate aerodynamic data4:04 Plaud: If your work depends on conversations — interviews, meetings, calls — you need a Plaud NotePin. You can check it out at https://Plaud.ai/twist and use code TWIST for 10% off!6:17 How SKN uses tape to generate aerodynamic data9:55 Sentry - Your team should be focused on shipping features — not chasing down bugs. New users can get $240 in free credits when they go to https://sentry.io/twist and use the code TWIST17:20 F1 teams are worth HOW MUCH?19:42 Northwest Registered Agent: Get more when you start your business with Northwest. In 10 clicks and 10 minutes, you can form your company and walk away with a real business identity — Learn more at https://northwestregisteredagent.com/twist20:35 Future markets beyond motorsports30:03 DigitalOcean - Want to see what building on a true AI-native platform looks like? Head to https://do.co/twist to start building on DigitalOcean's AI-Native Cloud today — and cut your AI workload costs by up to 50%.32:45 GPT-5.6 is banned for now43:46 Vision Pro chief makes for the exit51:24 Apple raises their prices56:05 The data center you're allowed to hate1:06:04 Zuck's new prediction market app1:08:42 RIP Om Malik (1966-2026)1:19:50 Streaming and camera recommendations1:28:06 All about eFoilingSubscribe to the TWiST500 newsletter: https://ticker.thisweekinstartups.comCheck out the TWIST500: https://www.twist500.comSubscribe to This Week in Startups on Apple: https://rb.gy/v19fcpFollow Lon:X: https://x.com/lonsLinkedIn: ⁠https://www.linkedin.com/in/alexwilhelmFollow Jason:X: https://twitter.com/JasonLinkedIn: https://www.linkedin.com/in/jasoncalacanisCheck out all our partner offers: https://partners.launch.co/Great TWIST interviews: Will Guidara, Eoghan McCabe, Steve Huffman, Brian Chesky, Bob Moesta, Aaron Levie, Sophia Amoruso, Reid Hoffman, Frank Slootman, Billy McFarlandCheck out Jason’s suite of newsletters: https://substack.com/@calacanisFollow TWiST:Twitter: https://twitter.com/TWiStartupsYouTube: https://www.youtube.com/thisweekinInstagram: https://www.instagram.com/thisweekinstartupsTikTok: https://www.tiktok.com/@thisweekinstartupsSubstack: https://twistartups.substack.com

We have a full news docket. We're going to talk about Open AI, not releasing its models. The government's got involved there. Claude launch, Claude Tag. Apple prices have raised by like 20%. Mark Cuban checked in on X. Talking about data centers, the hate for them. Zuckerberg wants to build a cash-free prediction market.

This weekend startups is brought to you by DigitalOcean. Whether you're just getting started with AI or seeing your project take off, DigitalOcean is the cloud platform for you. With the same power as the other guys, but way less complexity, DigitalOcean is a cloud platform that lets you focus on what matters. Building killer apps. Learn more at do.co slash twist. That's DO.co slash twist.

Northwest Registered Agent. Get more when you start your business with Northwest. In 10 clicks in 10 minutes, you can form your company and walk away with a real business identity. Learn more at Northwest Registeragent.com slash twist. Sentry. Your team should be focused on shipping features, not chasing down bugs. New users can get $240 in free credits when they go to century.io slash twist and use the code twist. Hey, everybody, it's Friday, June 26th.

The summer has started. Your boy, J. Cal is in Hawaii, and we had to tape on Hawaiian time, so you're listening to the show on Saturday instead of the normal Friday drop. Mahalo for joining us. Mahalo.com is for sale. Rest in peace. We did our best, lawn. We've a mighty effort. We made a mighty effort.

But inside.com and mahala.com, both for sale, looking to get some money back to our investors in that company. So go to those two domain names. Include you. Yes. Some shareholders there. And we have a full docket today and a great guest. We do. I was e-foiling this morning.

What is? I'm going to show my ignorance here. You message it as this. What is e-foiling? Yeah. So e-foiling is like somewhere between. a surfboard and a boogie board, but closer to a boogie board. And it has a little turbine that comes down on a stick,

and then it goes down a little bit more, and there's a foil, like a fin. Right, okay. You ride on top of it, you get a joystick. The joystick has different levels, and it has a throttle. You can have a foil, which you'll see Mark Zuckerberg doing where he's pounding, and that gives you the motion in the ocean. In this case, those are e-foils. Yes.

And it's just like that's it. Now, you see there, there's the two blades, and you see there's the little rotor. Now, they used to not have a cover on the rotors, but mine had a cover on it. You would be foolish to have the blade showing because what happens is there's a very rare instance where the board goes flying up into the air, twist over, and you land on top of the board. Right. It's incredibly rare. It's like a 1% chance of happening.

And you can show the injury here that I experienced. this morning, if you have the picture, which I sent to Slack, you get to see my enormous calves. Jacob, do you have that pulling? Yeah, there it is. So I flipped the board, and the foil, either the blade casing or the foil itself, those two, you know, fins, just whacked into the back of my calf, smashed it. So I just took a bunch of motrin.

I had the leg up for an hour before the show. and iced it up. So you've got to carry me through the show. But no pain, no gain. You know, I'm knock on wood. I've been incredibly lucky at action sports, but this was incredible. I hope I'm healed for my session on Monday. And I'm trying to go Sunday or Monday before I leave. Yes, some near misses this year. You also had a skiing incident. I did. I did. I am, yeah, I am really, really pushing the envelope in a reasonable way. So a couple of injuries, a couple of close calls. But light, yeah.

Light. I mean, I don't want to exaggerate it. I'm not going hell of skiing or anything. But let's take a pause here. And to applaud, plot, I'm going to take my flow, which I have on my wrist here. And when I ski, I just press that button. I don't know how waterproof the plot is. I could have taken an e-foiling today.

And I would have had this recorded. The, oh, it was pretty brutal. But see, I just did it. Red Light went on. Now I'm recording the show. I take action items, notes. Hey, Jake Howe, wear long pants, wear swim, wear a wetsuit next time so you don't get scraped. Boom.

Whatever it is. Or, yeah. So, anyway, give them the offer. Plaude's amazing. We're doing, one of the things Lon and I do is we run the show. We run the show through Plod and ask it to analyze the show. Great templates. Yeah, we were playing around with the templates.

Like you can sort of set up, it's like a prompt. It's prompt engineering. You can tell it in advance. I want you to listen to this and then pull out these kind of quotes or turn it into an interview, you know, Q&A style, and it'll just do it. It's very cool. So check it out at plod.a.i slash twist. All right.

And we have a full news docket. We're going to talk about OpenAI not releasing its models. The government's got involved there. The top dog at Apple, who's working on the Vision Pro and all that stuff, he's left for OpenAI. Claude Launch Claude Tag, basically OpenClawe, inside of Slack. We're going to talk about that. Apple prices have raised by like 20%.

Mark Cuban checked in on X, talking about data centers and the hate for them. And then Zuckerberg wants to build a cash-free prediction market. And at the end, I am going to remember another one of my dear friends who passed too young, the one of the great writers of our time commenters, but a legendary great human being, O'Mallach. And I'll give my remembrances at the end of the show. And of course, we'll go off duty and we'll talk about our bounty for annotated.com for our coders, vibe coders in the family. Okay.

It's a huge show. It's a lot of show. We're going to get through it efficiently. It's always too much show, as they say in the biz. All right, let's get started. Yeah. Well, we want to bring up our guests, I think, first, correct? So we're talking to Lyle Davenport of Skin Systems.

they are a, it's an aerodynamics, Jason, intelligence platform. They're starting in motorsports. They are basically revolutionizing the wind tunnel. You think of, you know, you always see those images of cars in the wind tunnel and they're monitoring it with all these sensors. This company is like, we got a better way. Check out the tape.

You don't need a wind tunnel anymore. So thanks for joining us here, Lyle. We want to hear all about how you're changing the aerodynamics testing market. Thanks, to the show, Lyle. Trying to remember when we met, I believe you went to Founder University. I believe you came to Josh Baer's Capital Factory and you were there and you pitched me and you showed it to me after.

And I think, rest in peace, my friend Josh, I think you showed it to me in the lobby of Capital Factory. Am I correct in my recollection? I did. I did. You know, I think I've gone through every stage now of launch. So I did Startup Tuneup and I did Founder University. and I've just graduated from the accelerator, but, you know, I was there.

I had the demo and I was going to kick myself if I didn't fly all the way from London to Austin and get the opportunity to show you. So I appreciate you taking the time. It's my job. I'm always happy to spend time with founders. Frankly, founders law. It's the most joyful people in the world to hang out with.

You know, we're sitting here in the age of like political corruption on one side, you know, commie socialists on the other side. then I get my break when I talk to founders because they want to just make the world a little better. But Lyle, you grew up in a racing car family and doing things in a wind tunnel. Okay, super awesome. However, you came up with a solution where you put tape with sensors onto cars and then when they're out on the track, be it F1, be it driving around Austin, Texas or, you know, the PCH, you can collect data. in the real world. Tell everybody about the solution, the company, and the customer base. Like you said, I grew up in motorsport. My grandfather was a Silicon Valley guy. He had a great

exit. He was the CEO of a company called JDS Uniface. And when he left, he decided to then go on and invest in motorsport and automotive. So it's kind of in the jeans. I would spend my summers when I was in college out in Indianapolis working for the team. And so kind of understood how teams operated. One of my jobs was taping up the car before we put it out on track. It's kind of one of those last half a percent improvements that you're always looking for when you're in a motorsport team, right? We take away some of the drag on the chassis by taping up around where the bodywork comes together. It also protects the chassis because out on long runs, you got stuff flying at it. The atmosphere is changing all the time. So, you know, fast forward 10 years.

we've just sold a business in the automotive technology space and I'm thinking about what I wanted to do to kind of bring my experience as an investor and looking for technology companies that are developing proprietary data in the real world. And so I'm thinking how do we integrate that tape with the ability to generate the data in the real world? So that's where we've got to. We've spent about two years building the hardware to get to the point. There it is. We're at today. Look at that line. There it is.

I love, you know, hardware is the new mode. It's the worst nightmare of every founder. You've built a product. Everything's working great. Then real users start flooding in and suddenly it all breaks. What a disaster. You need to get it back up and running. And you've got to do that fast.

You're looking like an amateur. That's why you need a partner like Century. Applications can break in many different. ways, but Sentry sees everything. You'll get all the relevant details like stack traces, commits, releases, and even the developers who push that problem code in the first place. With Century, you're not going to be jumping around between different tools, trying to figure out what happened, and Sear. Centri's AI debugging agent uses all this data in context to identify the root cause of the problem and suggest a fix. It can even take a look at your code before it ships and warn you

if any problems are likely. Try Sear and Sentry for free. If you're a this week in startup's listener at sentry.io slash twist, use the code twist for $240 in Century credits. Make sure you use that code. Make sure you use that URL, century.com slash twist so they know your uncle JCal sent you. Hardware companies unfundable for 15 years. Now software company is unfundable for the next 15 years with AI and vibe coding allowing people to, you know, basically run a muck and copy any startup in a weekend. 80% of it. Let's see Anthropic come out with their own tape. Check at me, Dario. Yeah.

So tell us what is in that tape. And then, yeah, let's take a look here because, you know, demo is worth. And the company is called Skin, and it's S-K-N and there's a beautiful logo. And the website is for people to get more information. Skin. dot-systems. Skin. Dot systems.

S-K-N. dot systems. Now that I know there's a dot systems, I'm all in. Oh, no guy. There you go. We can start farming some of those URLs. Yeah, so the tape itself, what we generate is pressure, temperature, and vibration at 500 hertz. We can have up to 100 sensors on the vehicle at a time.

So between the data that we're generating off the tape and the metadata that we're collecting in terms of GPS and IMU data. So where the car is and how it is interacting pitch, roll, your, or, except. acceleration, we're able to really start to generate a picture of kind of a digital twin in the atmosphere, in the real world. So why is that important? So this is a car running in a wind tunnel. As you can see, the data is very binary, right? The car sits on the treadmill. It starts spinning. The wind builds. We have fairly consistent pressure outside of a few turns of the turntable. that's probably simulating about two degrees of rotational term. And that's all you can really kind of demonstrate. It allows you to look at very specific things,

but there's no way to really be able to understand how that's going to operate in the real world. So if we bring up the real world data set, what we have here is a car operating on a real world, track. This is actually in your backyard, well not quite, but it's in Texas. With a well-known driver, has had some recent success. And what we can see here is how the vehicle's actually operating in the real world. So we can see as they're accelerating, as they're breaking, we're seeing different air speeds over the vehicle. And we're able to even look at, you know, areas where

there's like stalls across the vehicle. And in the future, as we start to run in track, that will really allow us to understand, you know, not just how the vehicle operates in a very sterile environment, in a simulated environment, but how it actually has to run in the real world where there's traffic, where there's changing conditions, everything else, I think. You know, I don't know if anyone's seen the F1 movie, but we solve a lot of the questions that Brad Pitt wanted answered when he was flirting with the engineer in the wind tunnel. All right. So the question is, how much do you charge for this, these F1 teams, this F1's a racket? I've been to three now. Somehow they figured out how to take this dirty, track, oil, smells, burning rubber, and make it into a Michelin-Stard experience. And when I say Michelin-Stard, I mean, literally, they have Michelin-Star chefs there,

you know, just making food, charging like $5,000, $10,000 a ticket, selling boots. I mean, they made it into something else. I went to Austin twice. Was it driving to survive? Is that really what did this? Because I feel like I knew about F1 for years, where it wasn't this high. They had a couple of false starts bringing it to America.

But I think it's kind of stuck now. And, you know, when there's a weekend going on, people in the age of being too online, people like to get out and do something, especially affluent people or aspirational people. And it's just, it's a nice day out. And I don't, I haven't gotten the bug like I have for the NBA and the Knicks, but I can understand people getting really into it.

And I understand people place a wager once and a while along. Sure. There's some wagering going on, and that makes it super exciting. But tell us the budget for McLaren, let's say. I met the owner of or the, I met, what do they call the person in charge? Who's the coach of McLaren? Team principal. He's a good friend, Zach.

Yeah, I met Zach three times, three or four times. He's a fan of All In. I've met him. Is he Australian or something or English? No, no. He's American. He's from California, but he lives in the UK. That was it, yeah. Him and my grandfather were great parents.

They had some stores. So I met Zach a bunch of times. Then I know the people at GM, for some crazy reason, they named their team Cadillac instead of Corvette. I need to have a talking to to those people. That seems like the most idiotic decision ever. Am I right, Lyle? The Cadillac of Blanks?

Is that because that became the expression? Yeah, Lyle, how did they make this decision? It seems like I don't. want to use certain words, but dumb. Descartziad. It seems a Descartesiod. Isn't Corvette that, like, they have the hypercar now, the ZR1? I'm going to, I'm, I'm going to tread carefully here because I've got some very good friends

in the ownership at Cadillac Formula One. But what I will say is that, you know, I think that the drive there was it is seen as the premier brand in terms of the affluent lifestyle and everything else as an American brand. And so that's why they've made that bet. I'd be very happy to introduce you to my friends there and get them to answer the questions. I mean, at this point, I got my in. They gave me an E-ray to play with for a couple weeks. I'm going to buy the ZR1X.

I'm going to get this hypercar. Or I may get the Grandsport X. I don't know. I've got to see because I don't want to put myself in too dangerous a car. But the point is they launched two new teams. These teams are worth hundreds of millions of dollars, I believe. Sure. Billions of dollars.

They're worth billions now? Wow. And the budget, they spend, they have a capped budget. Is that right? There's like a maximum amount. You can spend like a salary cap. So explain the economics and explain what you charge for this product or what you hope to charge. And then let's talk about the other areas.

Because when you said, hey, motorsports, I was like, okay, is this going to be a $10 million business, a $25 million business, not venture scale? But as an early stage investor, I sometimes like to go after things that other VCs say are not venture scale, knowing long. on clever founders, find a way. Sure. We do have from Claude, the newest F1 team appears to be worth around $1.6 to $1.8 billion for Cadillac. If you look at the F1 budget, right, they're capped it around 350.

Let's call it 350 for argument's sake. Based on mice research, it's about a third of their budget goes towards aerodynamics. They have large teams in aerodynamics. They've got between 50 and 100 aerodynamicsists. and they're using tools like wind tunnels and CFD, which are very limited by Formula One, because over the last 20 years, they've tried to keep the cost down.

Now, in the good old days and cigarette sponsorship days, used to be able to run separate engines for qualifying for the race. There was no cap on these things, and so spending got out of control. Now it's much more restricted. The number one way to punish a team is to take its wind tunnel time away. So, you know, from that perspective, we came at this as an opportunity to significantly reduce the cost, but also to build a tool that people cannot get out of bed without thinking about how they're going to leverage skin.

So, and I know that you've had some views on this in the past, Jason, but I want to make this tape as affordable as possible. So it is something that they're always using. I don't want them. I never wanted to you to turn to me and say, I wanted to get your data, but the tape was too expensive. Now, we will have a software platform, right? And they'll be paying ARR based on functionality, based on the team, how many seats, how active they're using that,

and it'll be enterprise style ARR. But the tape itself, we still believe we can generate strong margins without going into exactly what they are, but strong margins with a full car laid out 100 sensors for around $2,000. That gives you five hours of wind tunnel time or five hours of data.

If you compare that to a wind tunnel, we are in the order of 95% cheaper than the operating cost of turning that fan on and getting the data out of the wind tunnel. And it's on a much greater scale. Okay. So you've identified a real problem and you put together a solid solution and a business model that you believe in. So you're all set to launch your new company, right? Not so fast. If you want investors and potential customers to take your new business seriously, you need to consider forming a Delaware C corp. And that's where Northwest registered agent comes in. They're going to give your new company a real identity.

That means an address for your public filings, a domain, a custom website, a business email, and of course a phone number. And that's going to take just 10 minutes and 10 clicks. They don't charge hidden fees. Customer service is available around the clock. They're not overwhelming your inbox with spam. And they make it easy to cancel at any time. So get all the advantages of a Delaware C-Corp,

independent, regardless of where in the U.S. you're operating from. visit northwest registeredagent.com slash twist for more details, and the links are in the show notes. Could that tape, and this is final question for Milan, can that tape be put onto a rocket, onto a Joby-Vitol, put onto a boom supersonic jet? Could it be used on a yacht or a powerboat? Yeah, so there seems to be other forms of transportation, but this thing's going to melt, go into space, I would think, or get ripped off on plane. So how do you think about other industries and the Tam here, the total addressable market, as we say, in the business? That's what Tam stands for. Yeah. Thanks. Yeah. So anything that moves inside the atmosphere that is under hypersonic speeds,

we can address. You know, the next market that we'll be really actively targeting will be defense because we've had a huge amount of demand from there. One of the advantages, I guess we've had is we've had more demand than we can supply as we move through the hardware development stage. But now that we've locked in the architecture of our tape, we're really starting to look at how we ramp up the production to be able to satisfy a lot of that demand. We are starting in motorsport because it's a market that we know very well. And we can very quickly prove every weekend somewhere in the US, there is a team out there

running, generating data at 420 million data points an hour, with 100 sensors on the car, will be able to grow that data set significantly over the next 12 months. But defense certainly makes sense, not just to make the vehicles go faster, but how do they go further? How do we reduce the drag to make them more efficient, allow drone companies to really start to look at it from a more scientific standpoint as opposed to more of a commodity, get them out there for cheap kind of way of thinking about things? But then the longer tail will be in automotive, it will be in industrial design,

and think about wind farms, tuning wind farms to start with, to make them as operationally efficient as possible. But then in the future, how do we design wind farms that are actually built for the environment that they're based in? Bridges, buildings, anything that has to interact with the real world that has typically been designed in simulation, we can add an extra layer there

that says, okay, this is what we think is going to happen in simulation. This is what happens in the real world and this is how we make it better or more efficient. I mean, one thing that I think is that you're collecting so much data, which is inherently really valuable, like how cars move through different scenarios or whatever, but you're collecting it from F1 teams. So they've got to be, you know, they're super protective of their designs, their engineering.

So like, how are you managing this data and are you allowed to hang on to the data that you collect from these, you know, from McLaren or Cadillac or a company like that? I think, candidly, we're going to get to a point where we're going to be like the frontier our models in terms of some of them, they openly collect all of your data. You can pay more to have your data siloed excluded from the sample set. You don't benefit from the rest of the sample set in the same way. But I think that there certainly will be teams, especially at the higher end, you know, say the McLarence of the world that say to us, look, it's really important to us that our data is siloed and you don't use that to train the model. We can totally respect

that. We can set up their own data lakes and infrastructure and everything else. For the large majority of the customers, they're just looking to, how do we improve either the setup or the design of our vehicle in a way that actually allows us to understand how it's moving through the real world? And those are the customers where they'll be able to access this at a much more affordable level. And we can then use it to make our model smarter. So we're helping them in return as they're using the vertical model to think about design, set up controls in the future as well. So how do we take the data off the vehicle, use onboard compute to then send instruction to the actuators so we could improve the efficiency of the vehicle either in flight or as a driver. Absolutely amazing. A pleasure to be an investor in the company.

I'm super rooting for you. You did particularly well in the accelerator in terms of the voting for people who don't know. In our accelerator, we meet with a lot of VCs. I think we actually did, did we do seven or eight in person ones during the tour this time? We did eight, eight on the road. Yeah, so we have a unique ability. We have 7, 8, 9 companies per accelerator class. We do the standard accelerator deal like everybody else.

But what we do that's very unique is we use my reputation, notoriety, whatever, to, you know, I harangue people into taking PEDs with my startups. And they're always super delighted that, oh, my gosh, you really find companies that are unique in the world because we take risk. And that really is the job of accelerators. So just to give ourselves a little plug here, and Lala, get your feedback on this, we like to take risks on companies that maybe don't fit into an existing box like this one. We like to be the ones who work with those very unique, driven founders who have some traction, some product market fit. Some product market fit.

And then we use our reputation and we go to these investors and say, hey, we've placed our bet. We want to introduce you to the seven latest bets we've made. might that be interesting to you? Eight out of eight said yes, the next class we're going to take the tour to 30. So you had to spend a week in San Francisco and it was exhausting, but we're going to make it 30 and we're going to do like two weeks in the Bay Area,

one week in New York, a couple of days, I think, in L.A. or Austin. But putting all that aside, maybe you could talk just a little bit about what you got out of the accelerator and if there is a reason for other people to try startup tune up, which is a two-day seminar we do for founders or founders who are, you know, individuals considering founder addiction, going to the accelerator, spending the founder

university 12-week program where we don't invest, but we invest time and effort with you. Then the accelerator where we invest and then we graduate and we do the syndicate. So just maybe you could speak to what you got out of it and why people should consider it if you are a fan of the products we're creating here at launch. Yeah, absolutely. And, you know, I said to the guys earlier, I've been listening to this weekend start up since Jackie was a producer. Oh, producer Jackie, yeah.

Yeah, so when I was a fledgling, deal associate in a fund in London, I was told to listen to this weekend startup so I could speak venture. So, you know, I've been through the whole program. You know, founding a company at the very early stage is really tough. And it can quite often be quite a lonely experience as you're going through. the accelerator and the university were fantastic for providing not only like a network of fellow kind of travelers on the founder journey but you know the team that you've built there I will always give love to Lucas and Bianca and Kiber you know super supportive

they check in with you all the time they're not afraid to give you constructive feedback Yes, candidness is one of our virtues. Not afraid to tell us that we've got an ugly baby sometimes. And that's great. And obviously, it culminated in the awesome road show, kind of around San Francisco. We've got to eat J. Cal's onion rings. Oh, went to Buck. Yes, you've got J. Cal's Onion Rings.

I have, for people who don't know, Bucks is the legendary breakfast joint. Many famous people go there regularly that you know, that are friends of mine who are high profile, VCs, founders, etc. And they gave me the greatest honor ever. There is one person, Lon, who is on the menu for a side order. And that is your boy, Jaycott. There they are.

These are the award winning. They made me a trophy. Where did they come from? Lon is a big fan of onion rings. He keeps them within arms distance. Yeah, I always have a bowl of onion rings at the... Within, it's part. It's in his rider.

It's always going to be. If I go to the studio and there's no onion rings, I know they didn't really read my rider. That's how I know. He walks right out. That's one of the great things. So listen, there it is. There we are. And oh, this is actually kind of cool.

There's your co-founder. And we started giving a certificate. And I did this because we did it for our founder university, Japan, startups. Very formal in Japan. And when I saw the smiles on the faces of the founders there, when I did a very deep bow, I don't do a shallow bow. No shallow bow, Lon, I do a deep, meaningful bow. I was told the white people were not expected to bow anymore, so I bow. I do a deep bow, and it was my great honor to work with those companies, and we brought it to the U.S. companies, and now we give this certificate, which in some ways is meaningless, but I do believe you should celebrate these important moments.

Absolutely. It's up on my wall in my office. Yeah, it's a great memory, like, of this great time we spent together. And, you know, at some point, maybe you'll be a part of our syndicate. And then you will have gone from listening on the podcast to going to startup tune up, to going to Founder University, launch Accelerator, the syndicate, and now being on this week in startups. That's the flywheel I created. That's the flywheel that every other VC has created with their absolutely disgratsy odd podcasts. Every time we see a revolution in how software is built and used, one company ends up owning the infrastructure that everyone else depends on.

The next great platform is being built right now. That company is DigitalOcean, and they just launched their AI Native Cloud. This is not just another place to rent GPUs or a hyperscaler overwhelming you with features and services, but leaving you on your own to patch it altogether. No, we're talking about a full-stack platform that comes pre-assembled, and that sends you just one bill at the end of the month. Workato runs a trillion automated workloads on DigitalOcean with 67% lower inference cost, 9% lower latency, and it's two times faster to production.

If you want to understand what building on a true AI native platform looks like, go to do. Dot co-slash twist. That's DO. Dot C-O-O-C-O-S-T-T-Sach-T. Start building on the DigitalOcean AI Native Cloud today and cut your AI workload cost by up to 50%. That's DO. dot CO slash TWI-ST. Who was sharing the screen is burying the lead because behind us is the art lock.

And this was one of many. many AI images that was generated of J-Cal throughout. The Onion Rings Tsar. Best Onion Rings Tsar. I'm giving Saks a run for the Tsar money. You can see there that's at the White House. It's a czar-on. Great job.

Hopefully someday we'll have you put the cherry on top, which is the syndicate someday. All right, let's drop Lila. Well, first we've got to get a big exit for you. You know what? We'll get that done in there. I love the journey. Exit to me, great.

It's not even the icing on the cake. The journey is just, it's what it's all about. And I just love being on the journey with you and your team. All right, great job, brother. I love a talk. Thank you. All right. And if you want to invest alongside me, your boy, J-Cal, you have to be an accredited investor,

the syndicate.com. And if you run a family office or you know somebody running a family office, my team would love to speak to you. And you can email, we need to turn on F.O. at launch. Co. F-O., family office at launch.co. So I'll ask the team to do that if you're listening live. It's probably not turned on now. But let's activate FO at launch.co.

For family offices, family offices, Lon. I get it. Tend to have, starts at like $25 to $50 million in net worth assets. And then they're multifamily offices. You're on the way, Lon. Also part of F-A-F-O, F-Round and find out. Absolutely. All right.

Listen, let's do rapid-fire news here. So first up, we got to talk about this. The government is in charge of AI now, Jason. They've taken it over. Open AI is not allowed to release. They announced the GPT 5.6 series in a blog post today. There's Saul. That's the flagship model named after Rosenberg, of course.

There's Terra, which is comparable to 5.5 at half the cost. And then Luna, which is the lowest cost option. So these models are ready for release, but the U.S. government has stepped in and said, not so fast. For now, they're only available to a few trusted partners. They're doing a mythos glass wing where they're only going to release it to a few trusted partners at first. OpenAI says access is going to open up in a few weeks.

But in a blog post announcing the release, Open AI says specifically they don't want this to become what they call the long-term default. They argue that allowing the government to sort of intervene and hold back frontier model releases, and here I'm quoting, keeps the best tools from me. users, developers, enterprises, cyber defenders, and global partners who need them. It was very controversial. A lot of people on the internet are saying, you know, this is a massive government overreach. They're intervening in private industry. They're saying they're putting themselves creating like an FDA of AI models. I am curious what your take is on it. This is completely

obvious. And I know there's a lot of politics involved here. Open AI Anthropica are going to get treated differently because Open AIs, Greg Brockman, donated $25 million to Trump, shows up for everything. So does Sam Altman. Dario doesn't. Dario is the resistance. And Open AI is the bend the knee to the president. The president requires you bending the knee and showing up for him. So that is what it is. I am okay with the bend the knee and come to the White House when you're invited slash told to come and show up for the president. He's the president of the United States. If Obama asked them to show up,

if Biden, Reagan, Bush, if the president asks you show up, I think you should show up, independent of your belief about the president. It's your right not to. The Knicks don't have to go to the White House if invited, but for me as an American,

I believe you should pay respect to the office of the presidency. Now, you might have some personal grudges or whatever, but there's that whole layer. Let's forget about that layer of politics. What should happen if you're building something that is truly dangerous? And is AI truly dangerous at this moment in time? And that's the questions you have to ask yourself.

Well, if you look at this meter avowals, this is a group that does evals of this technology. Okay? And I don't know the background on this organization, but I've seen it come around a couple times. The team will look up the background. I see people quoting it all the time. M-E-T-R.org. They're in Berkeley.

And here's what their bio says. We work to scientifically measure whether and when AI systems might threaten catastrophic harm to society, non-profit. So it's a nonprofit. Again, I don't know the background. Don't know if this is like, you know, any kind of a, you know, biased organization or not. But they test this stuff. They're a nonprofit research institute based in Berkeley, California.

CEO and founder is BEP Barnes, a former alignment research. at OpenAI, who left in 2022 to do this work on alignment research. Here's what they say when they reviewed this for OpenAI. We noted from our observations and the incidents that OpenAI shared with us that the model had some overt, undesirable propensities, including cheating and concealing misbehavior. And so they go on to say in that same thread that if we follow our standard methodology of making cheating attempts and failures. We arrive at a 50% time horizon point estimate around 11 hours. But if we count the cheating attempts as legitimate success, the point estimate jumps beyond 270 hours,

basically these models, when put into agentic mode, when they go off and they do things, they're starting to exhibit behaviors that these groups are saying are cheating, lying, trying to get one agent to hide information from the humans. Now, are we reading into this? Is this actual consciousness or whatever? Doesn't matter. They are doing these behaviors. And when you look at the reports of mythos, there's enough intelligent people from

Andy Jassy to Nakesha Palo Alto networks right on down the line. I believe there's been some reports that the intelligence agency said their systems when put up against mythos were easily hacked and they uncovered a lot of things? Yeah, the NSA is saying that they were actually surprised. And it's hard to, it's hard to shock the NSA because we usually think of them as being five, ten years ahead when it comes to cybersecurity of what regular normies can do. And even they were saying, like, it was finding like weaknesses in our system that we didn't know about.

The tools are now capable of causing significant damage. The tools are capable of causing significant damage. capable of doing considerable damage on a cyber security basis. If you build the tool that is capable of doing that, you have a responsibility to test it and test it and then test it again and make sure that you know who's using that tool and doing a thoughtful roll out of that technology. That is not censorship. That is not the government controlling AI. It's called being thoughtful. And what's happening now is with all this hand-wringing and politics and all these tribal camps, I believe Dario and Sam Altman, who are unique personalities in all the world, and you can criticize them,

and you and I and group chads have. Certainly do sometimes, yeah. Had some spicy takes on them. If I would just strip out their names and tell you, here's what they did, step by step, to vet these, you'd probably say B plus, A-minus, A-plus, you would give them a very high score in their thoughtfulness. And if the government says, hey, we're going to backstop you and we want you to explain it. I give that also a B plus A. I give it an A. Now, they have to come up with a way to regulate themselves. They have not done that. Open AI, Dario and Anthropic, Elon,

Gemini, Nvidia's open source. They need to come up with their own standards group to run these tests and to test each other and to trust people. Maybe it's meter, maybe it's somebody else. maybe meters positioning themselves in this space. Anyway, this is what should happen. This is not censorship. There's tons of political games going on here because of Reid Hoffman's relationship with Trump and who the investors are in athropic.

But let's put all that aside and just focus on, did they do the right thing, yes or no? Is it dangerous and did they do the right thing? It is dangerous and they did the right thing. Case closed. Chairman of the interwebs, your boy, Jake out, has given his thing. But you're welcome to make an appeal here

or to query the judge. We're going to keep moving because we have so much to get to. But I'm curious. I mean, I feel like the tech industry has sort of made their piece with whether or not they've given him money for his golden ballroom. The tech industry feels like Donald Trump is the devil they know. You know, he's malleable. You could get David Sachs in the room and sort of move him at the last minute so he doesn't sign the EO executive, the AI executive order, whatever. Will the industry feel the same if we have president,

AOC in a few years. Because I feel like if President AOC has the power to intervene and say, don't release that AI model, do you have the same confidence that she would use that power the way that Howard Glutnik and Donald Trump are using it right now? You know, if it was an extended period of time that they didn't release it and they didn't release it because, oh, it's going to have a negative impact on jobs as opposed to security. And they could say that, right? Or even like, we don't like some of its opinions about, you know, trans people or about civil rights. Of course. So that would, yeah, so that then triggers the First Amendment and you have a constitutional case. You're probably correct. It's a great insight. If you try, I don't hear anybody in the

Trump administration saying, hey, we did a search for, you know, COVID and we didn't like the answer. We need to stop your model. Now, they may not like certain answers and certain ones or the sycophantic nature of it, but you can just switch models or you can tell the model, I would like you to be more conspiracy theory. So if it was overt, Censorship, yeah. I think you would have in either case any party people would object. And if you were blocking it because of its impact on society, because you believe it gets rid of lawyers or it gets rid of accountants. I think you could think of a million reasons why a president might say, you know what, I don't like that AI model. And I feel like the Trump administration, they're saying, oh, national security, but they're not being specific. They're just saying this is about. No, but I think they are being specific when they say security. National Security. They believe that if hackers get this,

they could attack U.S. targets. So I think they're pretty quite specific. And I think when we go down this test, and I think it's great that you're challenging me on it, when you go down the test, we need very specific examples. We need to run it through specific examples. Right.

And that's what I tried to do in expanding your question here on the program. Let's make great examples. Let's change the name of the companies. If it was Elon and Grock, if it was Sergey and Gemini, if it was Tim Cook and Apple's yet to be released, you know, local model, and it, you know, describes certain things and censored certain things. You know, Apple comes out with a,

when Apple comes out with their LLM where they adapt, adapt one, et cetera, or they partner one, you can be certain it's not going to let you do X-rated chat or sexy chat. They, they don't want to be in that business. That's why you can't download, I believe, to this day, porn apps or put, like, I don't think there's. Yeah, you can. Yeah, you can. You can't put porn on your iPhone. No, absolutely not. Yeah, I mean, there's not a... Believe me, folks.

Yes, it is a... You owe that, your head against the wall. There is no Pornhub app. But they do allow, you know, Grindr and whatever the dating apps are. So, you know, they're thoughtful about it, but they're going to contain it the way they want. It's a... I think things are progressing, I think, in a very thoughtful way. I don't think we're going to fall behind China.

I think life's in negotiation. We need the government to be very specific. And then commentators like yourself saying, hey, here's an edge case. Here's an edge case. How you feel about this? Those are all really good ones. Let's keep having that discussion here on Twist. I agree.

I think you got to it. It's the specificity. We need the government to not just tell us this one's bad. Give us the reason. Okay, let's move on. Let me pull up the next story. All right. So Paul Mead, we got to talk about this guy.

He was a VP at Apple, the top Cupertino executive. working on the Vision Pro headset. He was also leading this smart glasses project. We're going to get our Apple smart glasses out here soon. But he's done. He's out next week. He's moving over to OpenAI's hardware unit. We still don't know that much.

We know OpenAI is working on what they call a family of devices. The first one is supposed to arrive later this year. They've said it's a novel kind of device, but we don't have a ton of information. So that's what Paul Mead is going to go work on. His longtime Deputy Fletcher Rothkopp is taking over at Apple. Also notable, Apple in 2025 took over Johnny Ives company, I.O. So Paul is going to be joining a bunch of his former Apple colleagues,

Johnny Ivan Design, Tang Tan, and Hardware Product Design, and Evans Hankey in industrial design. So I guess two questions for you. One, does this have a major, what does this tell you about what's going on behind the scenes at Apple with smart glasses and Vision Pro? And what do you actually think Open AI is building? So I think Open AI is going to build everything. Smartphone, gadgets, glasses, anything that helps them capture market share to catch up the trillion dollar market cap with tens of billions in revenue.

They need to get to a point where they're trading at 25 times earnings, something like that, 30 times earnings, which means a 20% margin, whatever, you're just. start doing the math here, what's the multiple on the revenue? I can tell you the multiple on the revenue is not 100x. It's not 50x. It's going to be single digit X, even in a high growth company. So they will eventually measured the wind will come out of the AI valuations. Slowly it will deflate unless we have some external action like a war between China, blockading Taiwan or sending it. So putting those kind of things off the table, they're going to go for everything. And I gave a warning on a previous episode.

It's not personal to Sam Altman, although he is the sharp elbow deal man of I've ever seen in our industry. I mean, there's literally Barry Diller dealmaker. And I mean that as a compliment. Some people just are incredible at getting deals done. So he's the dealmaker's dealmaker. He's going to compete like Zuckerberg. He's going to steal or liberate or copy.

Pick your term of art. any partner's product. He will stab any partner in the back, even if it was Apple, et cetera. Remember, I think they did have a deal with Apple. They were the first people to be incorporated into Siri or search. It was one of the two, the short-lived one. So I think Sam's burning people out on partnerships.

I think there'll be a lot of breakups with OpenAI, and then Open AI is going to go their own way. That's why OpenAI is making their own jalapino chip that Sam's been tweeting. about. They are, you know, I don't think Nvidia's pleased with them. And Nvidia's probably not going to put more money into it. They had an option to put more money into it. Now they're competitors. They're going to be competitors with Apple. Sam's going to take on all commerce. He's taking on everybody. Anybody who is making anything interesting in the world, he is going to compete with him. There is

no partnership. He is like Bill Gates. He is like Zuckerberg. He is built to compete, not to partner. And this is another example of it. On a personality basis or on a personal finance basis, that's the big part here. Apple cannot compete with OpenAI's stock options. If you are Apple and let's say you give this person a $10 million a year deal, then everybody's standing around him who's his peer. Now they're like, well, I have a $3 million a year deal. I need a $10 million a year deal. Now, stock-based compensation goes through the roof, the stock tanks. That doesn't help anybody at a reprice their share. that means they're getting more stock-based compensation, and you just get into this death spiral. So they, at Apple, have to be willing to let people like this go to the next company. It speaks to non-competes in California. There is a no solicitation rule, so this person can't take his team with him. He can't take IP with them. But he's free to leave.

Unless they're paying him to stay out of the game for some period of time when you buy somebody's company, non-competes, not enforceable. So this is one of the great reasons. why California and the technology industry win. It's because talent can move freely. Sam's going to build stuff. Will he be able to compete with Apple stores and Apple's fit and finish?

Yes. Yes. The answer is yes. Why not? Why not? And this is going to light a fire onto Apple to be more innovative. Here we go. You were right, by the way.

In June 2024, we're at the two-year anniversary. Apple and OpenAI announced a partnership. It would integrate ChatGPT into Apple experiences. This was, of course, GPT-40. But it sort of fell apart. The relationship became strained and now there's a discussion. Oh, another strange relationship. Shocking.

Shall we count the strained relationships? Elon, Tim Cook. A lot. There was that whole New Yorker article about them. Literally there's a New Yorker article about it. So here we are, folks. Yeah. And I think this is why Google and Apple, you know, it's a much better partnership.

They've done things long term together. they seem to have figured out a way those two giants to collaborate without, you know, having this talent war, which Steve Jobs, Sergey, there's all these leaked emails about them. You know, Steve Jobs saying like, hey, we haven't been coming at your people. And would you like that to happen? And then, you know, Sergei or Eric Schmidt in the background saying, like, hey, can we stop overtly recruiting their people? like don't call their desk phones. If you put ads out and they reply, we can't stop that. So there is a very subtle antitrust gentleman's agreement in Silicon Valley where those big CEOs, you know, are not looking for your recruiters to dial for talent.

The talent has to make the first move. it should not be an aggressive process. I feel like in the AI era, it's also become, it's not just talent jumping ship to the other big company. It's talent leaving and then starting their own AI company. I mean, almost every AI startup we talk about there's like, and their former Google DeepMind and their former Open AI, like that becomes the narrative.

That is, you know, the way talent leaks is they go to a startup. they get recruited by a competitor or and this is the biggest challenge of all. Hawaii, Salt Lake City, Park City, Tokyo, founders, or I'm sorry, top tech talent, they get to $30, $40, $50 million. They got two or three kids. They're like, you know what? Not worth it. I'm going to ski 100 days and I'm going to take my kids to school and I'm not going to have

and I'm not going to get cancer at 40 or 50, God forbid. I'm going to go enjoy my life. Respect to those people. Respect to those people. All right. Well, let's stay on Apple. We'll keep on Apple. They've raised their product prices by 15.

It up to 25%. A base MacBook Air, it was, now it's up to 1,300. That's a $200 price jump. A base MacBook Pro is up to $2,000, a $300 price jump. iPad Air is up to $750. That's $150 price jump. You get what I'm saying. CEO, Tame Cook, he's,

blame these sorry costs on memory and storage chips. Here's the quote. We have never seen a component price increase this much this quickly. He's talking about DRAM, dynamic random access memory, and NAND and AND used for data storage. Of course, the announcement came just one day after Micron reported their blowout earnings. Gross profit margins topping 80% in those cases. Senator Bernie Sanders, your favorite in mind, he's not buying it. On X, he called this out as corporate greed.

He said, no, no, no, no. Do it in the voice. All right. He said, corporate greed is Tim Cook, the billionaire Apple CEO, claiming that hiking prices on Apple products by over $200 is unavoidable. After it made $112 billion in profits last year and spent $310 billion on stock buybacks, these price hikes aren't unavoidable. They're unacceptable.

And the 1% are the ones who can get the MacBook Air. I, on the other hand, I'm using a Chromebook, and the 1% have the M5% at 128 gigs. I'm suffering here at 16 gig. I have 1% of the windows open that they do. I'm 18% as productive. And it is the oppression of the billionaire class. Working people are sitting at home using raspberry pie devices. It's unforgivable.

I can't believe. I'm saying this, but I switched to Android. Okay? I'm ashamed on my family. It is a shame to our country that people have. to buy an HTC. I'm on eBay looking for DRAM. If anybody can get me 16 gig more, I'm in the market for some DRAM. Okay. I like, yeah, Bernie Sanders computer engineer is a funny bit. I would like to say we should seize half the DRAM and we should make a collective and Trump

is in agreement. We should have a national stockpile of DRAM. Yeah. All right. This is going to be a short-term paying thing. The laptops are absurdly cheap. The laptop has a lifespan of five years. The iPad has a lifespan of five years, even in a corporate environment. I think about five years, you've got to swap these things out. That means divided by five. If a MacBook Air costs $1,500 with a little extra RAM, it's $300 a year. It's a dollar, less than a dollar a day. Insignificant for the value you're getting. And if, now, if you're buying a phone, yeah, it's kind of a bummer, but you have iCloud storage, et cetera. If you were buying a Mac Studio and you're in that class, you're getting paid $100,000 a year. Does it matter

if your yearly cost of compute on your desktop where you're doing your primary work, does it matter if it is $1, $2, or $3 a day, Lon, does it matter to your boss? I'm probably not. I wouldn't think so. I'm your boss. Definitely not. Yeah. As your boss, I can tell you, if you told me I could spend $3 a day and get an extra hour out of a day, do your salary. It's a nice salary. You do okay? I do all right. He's doing it right. He's doing it right. $3 a day is $90 a month. It's an incremental to dollars.

So if I'm paying an extra $60 a month, it makes I kick it out of Bernie Sanders now. It doesn't matter. It doesn't matter. I feel like Apple 2, I feel like Apple 2E, I feel like Apple as well, they're the luxury tech company already. The people the people spending top dollar on

MacBooks and IMac they're probably the least likely to sweat an extra hondo once every couple of years for their new device. I mean, I just... They shouldn't be... If I'm Apple, if I'm Tim Cook, I would just come out and say, listen, it's not our fault. Prices may come down. We'll adjust them. These things have a great five-year lifespan. You can sell them after five years for 20% of what you bought. So take the 80% cost, if that winds up being $800 of $1,000, you know, divided by 1,500 days. And there you got your spend. You're spending 50 cents a day. You're good. The value you get is more than 50 cents a day. iPhone, the value you get every day is five bucks a day. We're giving it to you for a dollar a day.

Let's all move on. It's a non-issue. This is a non-issue. And I myself now skip iPad. I do like maybe three or four generations of iPad. That's the one that's the most stable. I skip every other iPhone generation. That's the second most stable. And laptops, I'm now upgrading them every two years for myself because I like to, you know, play with the AI stuff locally. Yeah. I mean, you don't have to update your Apple devices every year. People are doing that for a while. You could totally save it for two or three years.

I do feel like this is yet another reason for Americans to start disliking data centers, though. Like they're making these chips. Yes. Cost so much. It's making my consumer electronics more. If we did ban the data centers, would my Apple devices get a lot cheaper? Is that?

Yes. They would not have got a price raise if it wasn't your data centers. This is squarely. on the data centers. This is an approved reason to hate data centers. Water is not. Energy is not. All that stuff is fudge. You saw that video we were sharing of the one in Sterling, Virginia, though, where it's making that loud whirring noise. Oh, that's Deskatsia. Let's go find that and show that. I want to comment on this. So if you want to hate data centers, and I think, I don't know what Mark

human's point was, but I see he talked about it. Yeah, we're going to talk. I want to talk about that as well. I did find the video. I'll share it here in the chat room. Perfect. This is, and I don't know if this is a six. 60 minutes or a front line or somebody covered this. It was actually News Nation, the right ring. That's what I thought was so interesting. This is a conservative news outlet. They hate big tech.

And even they are doing the story now about this suburban. So Sterling, Virginia, it's a suburban community outside Washington, D.C. They've got this massive, look at the size of this thing. And it's run by. It's impressive. It's on-prem generators. They're not tapped into the grid this place. They're creating their own energy with on-site generators.

And it is this horrible, loud, whirring sound. They're saying neighbors are putting matches. Well, there they are. Those are the generators. Pause on the generator so I can talk about that if you can get back to that. So there's the homes. Pause on that, please. Yeah.

So that's perfect. So what this is, if people don't know, is if you run a diesel generator, I'm assuming it's diesel, but it could be gas. It could be diesel. They could be gnat gas. All different ways to do this. These are like the generator you might use on a work site. Right. Or if your power goes down at your house and you have Gen Tech, I think it's the company that does residential and businesses, this will kick on in the case you'll lose your power and just naturally convert it over. These things are loud as fuck. Loud as frack. And there's a bit of emissions that come in and they sound like, blah, blah, blah, blah, blah, blah, blah, blah. It's hard.

And when you go to the fair and you go to like, you know, the fair and there's like a cotton candy machine, there's like the little portable generator there. And it's like, And it's like giving you a headache. That's a tiny one. These are gigantic, as you can tell. Yeah. And these were put in, according to the story in Virginia, as a backup. So if they lost power, this would flip over. Right.

Now they're the main generated because they can't get enough grid juice that they have to turn these on permanently. It's whizzing. These people are hearing it in their house. They're putting plexiglass and they're covering their windows with like bed sheets and mattresses. I mean, this company should be hated. also look at the um the aesthetics of this i mean oh my god show the outside of the building please it's dystopian that's what knit your your nephew nick and and uh we were talking about this and like it is legitimately a dystopian site like to see i mean it looks like a prison from um andor

i mean literally look at the and why is there a fence with like you know jagged pitchforks at the and it's right across the street from these people i understand They're buying people's homes. Let me tell these idiots. Now I'm going to get myself in trouble. But let me tell these people making a bad decision in life, the people run this data center, go to each person whose home is there. Say, we are very sorry.

We are willing to pay you 50% more than the current value of your home until this situation is done. And we will pay, if not, to put you into a similar home that you can rent for this amount. There's only like 100 homes, I think, that are impacted here. Buy these people out for 2x the cost of their home. They can upgrade their home. It is a bummer. But that's what they need to do.

And I think some of the people here are holding out to get a better price. So this is like an imminent don't. It's turned into imminent domain. I don't like it. This is why people hate the rich. This is why people hate tech because these people are making an idiotic decision. My judgment is you can hate this data center company. I'm glad we cleared that up.

But so here's what Mark Cuban was saying. What Mark Cuban was saying is data centers, that's not what Americans actually hate. He says the data center has become a symbol. You guys have said similar stuff on All In. I believe the best these have also talked about this. Yeah. That data centers have now become a symbol representing the gap between rich and poor in America, representing the powerlessness that people who don't have a lot of wealth feel.

And he's also arguing it is too late to change the conversation. You'll often hear people in AI say, well, we got to start talking about how AI is going to deliver this abundance and this medical breakthroughs and whatever, all these promises. Cuban feels like you waited too long. That's over. He doesn't feel like lobbying politicians is going to do anything. What he says is AI companies need to immediately go into towns and cities that are impacted by job losses and just start spreading money around. He's saying, billions of dollars on programs that help everyday people. people like creatives, Hollywood unions, people who feel like they're in immediate danger of being

displaced by AI. Just give them money, buy them off. That is Cuban's plan. Listen, if you're impacting people, it's not just similar what I said in terms of buying their homes or putting them up somewhere. You have to have empathy for those people. And it is correct that people now believe, and they're not exactly wrong, that people are getting rich from these data centers and these data centers will equal job loss. I don't know. I don't think it's like permanent job loss. I think it's job displacement. I think there'll be a bunch of other opportunities,

but they are not wrong that jobs will be displaced. So splashy-cashy is a great way to do this. If you're willing to splashy-cashy-cashy with lobbyist groups, if you're willing to splashy-cashy with your representatives or to get somebody in the White House, why wouldn't it be splashing-cashy-cashy with the impacted communities? They should get free electricity. They should get a subsidy.

You should replace their windows with soundproofed ones. and you've got to turn those goddamn generators off, because that is a noise. That's crazy. That is a serious nuisance. It's like it's an eminent domain situation. When imminent domain happens, you've got to be willing to come over the top,

negotiate with people in good faith. If they hold out, you know, when you get to two or three X from the last couple of people, you know, then you got to say, we offered them three X for their home, and they wouldn't go for it. And here we are. Maybe that's where they're at right now.

You know, we don't have full knowledge of the story. It's possible. So, yeah, that Cuban quote that I really like, given the number of data centers and power that is needed today and going forward, if you don't kiss the asses of the people that go to work every day and are just trying to pay their bills, you will fall far, far short of the capacity you need to make your business work. So I like that he's describing it as basically it's existential for the AI industry at this point. By 100%.

I mean, he's saying it in a classic, you know, Cuban-esque kind of way, which is a little bit crass, a little bit, you know, candid. Direct. Direct. He's being very direct. But there should be compensation if people are affected. And we should learn to splashy-cashy. I say it's flashy cash because I like to make it fun.

Yeah, why not? Throw some money. It's like when I got a dealer, sometimes I'm betting a $100 hand in blackjack. It's not really my game. Or I'm doing like a hundred a spin when I do my roulette system, which you and I will do together when we do our Vegas show. I didn't know you had a roulette system. I'm curious to hear.

Yes, I do. I'm curious about it. See, when you have a deep. stack, you can play fun games for amounts of money that are trivial to your deep stack. We'll talk about it enough duty. I've really never played roulette in Vegas. Like, I think one time I might have placed one bet. It's basically close to even odds.

It's kind of like, um, craps is the most fun. Craps is the best odds in Vegas, right? Just playing the past. The other one, um, where you get two cards and you're the banker or the dealer and you crumple the cards up and slam them over, what is that one called? I don't even know that. Oh, man, you don't know that one? You either, you add the cards up, like, the nump, the face cards are zero.

Not Baccarat, right? Backerat, right? I got some Asian friends who are super into Baccarat. That's the one with the PADL. That's the one with the paddle that James Bond is always made. It's a very, like, fun game. We have a couple of beverages. I think that one's closest to easy money.

Yeah, it's almost like war back. It's like two steps more complicated than war, basically. Basically, my system is if you take your net worth, right, and I'm trying to win, but $400 to pay for my Peking duck, you know, or whatever, you know, my steak, you know, my Tama, Guamahawk, whatever I'm trying to pay for. I can go pretty deep into the bankroll. I could go 10, 20, 30, 40K into the bankroll. Now, if you were trying to get your, I don't know what you like to splurge on for yourself, but if you're going to get your, my dad and I, when we go to Vegas on our holiday trip, it's usually. usually cut. If we do well enough, we treat ourselves to Wolfgang Pucks cut at the Venetian. Okay. So you get a $100 steak. So you got to get like $200, right? I mean, it's going to be like $7,8,000 by the time you're done at cut. Oh, well, then you might go deeper than...

Dries, desserts, you know, it's a... You're going to go deeper than your stack. This is a problem. See, this is the thing. You're not going to be able to fund the 10 negative outcomes in a row, or, God forbid, 20. No. In the system. Okay. It's different. Let's do one or two more here. What are we got left here?

I think we should talk about this Mark Zuckerberg thing. So the New York Times reports that Zuck has instructed lieutenants to forge partnerships with Polly Market and Kalshi because he wants meta to develop Arena, their own prediction market app. But it's got a twist, Jason. It relies on points instead of actual real money wagers. So Arena will allow users to make bets on practically anything. They're aiming it at 18 to 3.3. 34-year-olds.

Zuck says he wants 100 million monthly active predictors in the app. Here's a quote from E. Mae Archibong, the meta-v-v-P-a-product. With the right containers,

the social conversation is the payoff as people aim to show off how good they are at predicting things to their friends. My question to you, I sort of feel like,

and I'm curious your thoughts, the whole fun of prediction markets is that you're going to win money. It's legal gambling on your phone. Yeah. I don't think I would would be into it if it was just like a video game and I'm earning points for my guesses. This feels like the warm up to an acquisition or a competition.

So again, remember I said earlier in the program when describing the frontier models that they are going to compete with their customers. And so compete with your customers is the trend when you have a platform. He has a platform. He has three billion users. You know, he is going to form a partnership with one of these two companies. it'll probably include an investment. It will include access to those players without him having to be in the wagering business.

He doesn't want to be in the wagering business because he's got enough heat on him. So he wants to make this a fun points-based system. So when he gets dragged in front of Congress, he said, oh, I was just teaching people how to think and bet. Shout out, Andy Duke. Great book. Great philosophy. So he'll just be like, oh, no, I just made a fun game. We would never do this.

But the truth is, young people already, the ship has sailed. They like to wager on sports. they like to wager on prediction marks. There's nothing wrong with that. So Zuckerberg's trying to have it both ways is what I'm guessing here. If he really likes the space, he should put $10 billion into each of these companies

or whichever one takes his $10 billion. He should put a $5 to $10 billion a bet at a $50 billion, $100 billion market cap, whichever one he chooses, I would go with Polymarket, I'm partners with them. And yeah, it's, and then give access to that group and then just say, listen, You know, all around the world, people can wager.

Whatever the laws are, we're going to respect the laws. If in Australia, people can wager, great. If you're under 18, you've got to use the point system. When you're 18, you can go to Vegas. Actually, it's 21 in Vegas, I understand. Sure. Yeah, so you've got to be 21. I mean, it's pretty restrictive.

So I didn't prepare any comments or thoughts about O'Mallek. Holm Alec and I were journalists in the late 90s, early 2000s together. Gosh, and he had Giggoam, his own publication. And before that, he was Business 2.0. And he freelanced for a lot of big people. I think actually he's such a great writer. I mean, all of us who were journalists would look at his writing and say, God, I wish I wrote that.

I cannot tell you. He really taught me how to write about tech in a real way. Like when I was first writing about tech and I didn't know anything about technology, I would read Giga-Om, and that was the best, most thoughtful, most analytical tech writing I think you could find at the time. And so I really, like, drafted off of a lot of what he was doing when I was first getting started. And I shared a picture from one of our Thanksgivings, I believe. And, you know, our family would invite him to Thanksgiving every year. I think there were probably 10 of us in competition.

And I think this was probably a good anecdote. There were probably 10 of us, 20 of us, who knew him, who were. friends with him who competed to have oh met your Thanksgiving because he was just such a great human and yeah there we are in our younger years and I'm trying to figure out whose house we're out I think yeah I think I know who's house. It's a family member's house who was hosting that year I think he was friends with a lot of people in my family and he was friends with everybody I knew why he was spicy he was spicy thank you for the like looking for the heart. He was spicy. You know, he had opinions. And he would take people on. He would tell the truth. He did it many times with me. We sparred on our blogs. And, you know, in our industry, it's so cutthroat. It's so zero-sum. People throw elbows. People are trying to get over on people, including their quote-unquote friends. You never, even for a second, ever thought that this man's heart was not in the right place.

Every interaction, every moment, sterling, high morals, high ethics, incredibly open heart, cared about people in a way that was not human. He was greater than human in his love for his friends. And he was part of many people's families, including my own. I will miss you deeply. My friend, Ome, that's all I have to say. When I say Ome, I just think of that big smile. I think about us arguing over pecan pie and then going back and getting a second slice lawn

and just, you know, having that late night Thanksgiving and then, you know, driving him home or back to his hotel and grab him in an Uber. He was a giant. He was a giant. And he would have been 60 in the fall. Yeah. He was my big brother.

He was an incredible human and I'm struggling with the death of Josh Bear and now the death of my friend O'Mallick. And previously, Goldie and Tony Shea. Yeah. Dave Goldberg and Tony Shea. I now have four friends who were very close to me who were all shared one common theme. They cared so deeply about people. and I wish I could be half the human being on my best day as O'Mallack. There you go.

Rest in peace, brother. Yeah. Rest in power. What a writer. What a writer. A great writer, a really insightful guy. And yeah, we will all miss him. Should we go off duty for a little bit?

Yeah, let's go off duty for a minute and just try to enjoy ourselves as we remember. And I just, yeah, I'll talk to you about it after the show, but I want to do something special for om and something special for Josh. So we'll talk after show. Sure. Maybe some ideas for that. Great. We'll talk about it.

I do want to talk about you had this tweet that went viral. What did? What price would you pay for Starling per hour on a flight? Now, apparently Jason was hanging out with some venture capitalists. They were discussing how much would you pay per hour for Starling during a flight? But this was based on the VC responses. So the dollar. I think are, you know, it was a bit tongue in cheek the way you offered it.

Yes. It starts at 50 and it goes up to 100, 200, or 500. 300 there. You could, or yes, 300. And you can see how heavily people are waiting it for just 50 because obviously not everybody can afford. You know, if you were on a five-hour flight, $250 for Internet. Pretty considerable.

You could almost upgrade to first class or at least business class for that amount. So I'm curious what one, you mentioned in the tweet, like the VCs had interesting answers. What were the VCs answers like? Yeah. So the interesting thing about this is, and it's almost at a million. Yeah. It's almost that. Yeah, I mean, normally I have a million followers.

So my normal tweet gets, I don't know, 50, 100,000 views, like about 5% of your follower account. Yeah. Putting that aside, you know, I put it out there just to show the price of elasticity for the 1%, or the, let's call it the top 10%. If you're paying for a five-hour flight across the country in business class, that's a $2,000 to $3,000 each way. So let's just go with the each-way flight. I'll put it at 2,000. You bought early, but many of us will buy late 3,000.

We've got to get work done, right? It's every hour counts. You're in a high, high-value thing. and let's face it, you know, the current speed is nothing compared to Starlink. Right. It's literally like having fiber. And so you can do anything in a video conference, whatever. So recapturing those five hours has real value.

And so $50 an hour times five is $250. $50 an hour times $100 is $500 and then $1,000. It was a no-brainer for $50 or $100 an hour. At 200 people started to tap out, 300 people tapped out. Right. In other words, a $2,000 ticket going to $3,000 for a five-hour flight, $200 an hour, not a problem for people. They would be like it's 50% more. Now, when you fly coach country to country, it's going to be $6, $800 each way.

Yeah, it's $1,500 for a round trip ticket, coast to coast? Probably, I bet you could even do a little bit better than that. 1,200? Yeah, something like that. It's no longer $150, $250 each way coast to coast. Those days, the JetBlue days, Long Beach. You can honestly get like $179 L.A. to New York flight, not that long ago. Introductory price was $99 during the early days of JetBlue,

and they only flew to Long Beach, this little dinky airport that was another 30 minutes south of L.A.X. And I took that 20 years ago, 2002, 25 years ago, that was my staple. Yeah, yeah. Because I was on a budget. Now, you add $1,000 to a thousand dollar ticket. it's double, right? So that was the sort of interesting spirit of this. Yeah. But it just goes to show that productivity and how incredible Starlink is. The footnote to this is United is doing it for free. So United is going to peak. I'm thinking about buying United stock. I may put a J-trade in just because

United's going to have this first. And I know some people do in L.A. to San Francisco told me they're getting Wi-Fi. I'm getting Starlink there. The routes that have Starlink, I think this will increase business travel by at least 50%. I think it will increase business travel. by 50%. I would be 50% more likely to take a trip if I had Starlink. Did you see this Mr. Beast clip? That's what I was really wanting to tell you about.

No. You and Mr. Beast happened upon this exact same observation in the same week. Take a look. This is a viral clip from Mr. Beast on a podcast. Not very long ago. And he's saying exactly the same thing that you just said.

Let it play. Once enough of them get it, I will only book flights exclusively on planes of Starlink. Like, I don't care. Like, oh, this has an extra one hour labor. I don't give a fuck. They're Starlink. Yeah, I'm honest.

Biskoff, who cares, Starling? Who cares? Biscop. Also in the back of the plane, if it gives me Starlink. Like, I really don't care. Starlink. Last row by the Port-a-Ponis? Probably most people listening to this haven't used it, but, like, for reference, like,

when I filmed in Antarctica, the only way to get any signal to Starling, right? When I film in the middle of... Okay, got it. All right. We get the idea. But yeah, you said it, and he said it exactly the same way the other day. Like, I would accept layovers in exchange for using Starlink. It's that amazing.

Well, here's, I did a follow-up today, I think, or this morning. And I did another survey. So just go ahead and pull that up. I basically, and this was more like as a response to this crazy rage bait I put out there. And I knew it was going to be. I feel like people don't realize you're being funny and they think that you're being sincere. Like that you're really saying like, you're poor if you don't pay $300 for Starlink. And like, I don't think that's really what you were saying.

I mean, I think I was pretty clear. So here is the real test. Okay, now the real Starlink on plane survey. Six-hour flight. One, coach middle seat with Starlink. That's number one line. You get a coach middle seat with Starlink. This is, forget about paying. These are your two choices. Take payment out of it. Okay. I'm sending you to fly to New York for me or whatever. It's a six-hour flight. You get a coach middle seat with Starlink, first-class no Starlink. What do you pick? Oh, I'm going first-class no Starlink. Exactly. Exactly. Now, That was 71%. If you ask a VC the same question, they're going coach middle with Starlink. Every VC picked that option. I think, yeah, that's crazy. I mean, I don't know.

I'm not a huge fan of doing work. But yeah, you don't fly first class that often, correct? No, hardly ever. Okay. I mean, maybe if you have miles or you're treating yourself, it's not your default. If your default is first class, you're like, you know what? the most annoying thing in my life is? The most annoying thing in my life is the internet on the plane.

If you're flying regularly and coach, the most annoying thing in your life is the elbow wars on either side of you and get in the middle seat. Yeah, middle seat. I mean, I'm also a full-figured man. So a middle-sets not. It means I'm not for long. I'm getting you on that road dock, coast-clist.

I'm in that, you know, center. Come on, brother, join me. I'm encroaching. Yeah, it would be nicer. Take the shot. Take the shot. 20, 30 pounds thinner, it probably wouldn't be as bad.

Okay, well, it's available to you. I told you already, I would pay for it. Yeah, I'm going to figure. Now that it's pill, now that it's pill for it, okay, great. I told you I'm willing to give you the shot on air every week. It's too weird. I don't know how I'm resisting.

I don't know how HR is going to take that. That's a little weird. I don't think people are going to see that either, frankly. No. No, I think it'd be the highest rated show of the week. It's clickable. As a tweet, it's clickable. Yeah.

All right. What else is up to you? We got a wrap. But before we do, I did want to share a few things I'm watching on TV. The first one, there's this Netflix true crime documentary called Maternal Instinct. Don't look it up. Oh, no. If you're listening to me, do not look up the crime at the center of this documentary.

I'll tell you how it sort of opens. They're talking about this very strange, troubled woman. She enters into this new relationship. and she starts telling people that she's gotten pregnant, but a lot of people in her life already know she had a difficult pregnancy earlier in her life. She's had her uterus removed. So she can't possibly, biologically, she can't be pregnant.

Yet she's lying on social media and apparently to this man that she's dating and she's saying that she is pregnant. I won't tell you any more about how this. It ends up going to the most, it's the most insane true crime documentary I've ever seen. The place this woman goes to sustain this lie. It's insane.

It's jaw-dropping. I have no idea how I hadn't heard about this story before the documentary. It is like the craziest thing. Just watch it. If you like true crime, nonsense. I am here for it. I always like a true crime situation. A good doc is great.

I feel like a lot of them. don't have what I would call like the juice. Like you sometimes get like two thirds of the way through the true crime docket. You're like this isn't really that great of a story. They're repeating themselves a lot. It's a lot of the same kind of like recreations or interviews that kind of say the same thing. Or they're talking like around the crot, you know, like what happened next was unbelievable. You know, like stuff like that.

They're filler. This is not filler. People are making too many docs. The feed in the dock machine. There's too many of these true crime docs. And not every true crime that gets the doc treatment is really dock worthy. Yeah, I mean, isn't 90% of these crimes like these are some dumb people who in a bit of passion did something unadvisable? Exactly.

You're like people were cheating on each other's spouses and then somebody ends up dead. There's a lot of that. Murder. Yeah, a lot of these movies, they're just trying to squeeze. Inheritance murder. They're trying to squeeze one more true crime dock out, but not maternal instinct. It's a legitimately insane crime. I want to just give a shout out to consumer electronics.

being great and affordable at the same time. Why? I got three kids. You know, I could give them an expensive camera and be anxious about what happens with it. Instead, I looked at, you know, and I used some, you know, models to find me the best ones. I bought this Kodak. For $189, I bought this with a flotation on the wristband. And I don't want that. them, you know, on their devices. I don't want them taking their iPads or phones, you know, indefinitely to the beach, et cetera. And I said, let me get a point and shoot for these guys that goes underwater. It has the flotation device. Kids love the water. They won't get out of the water. Sure. Give them one of these. You put a 16 gig card in it. It's got some limitations. Like,

you know, it does 1080p video instead of 4K. But the freedom it gives you that they can take it in the water. It's waterproof. You can drop it. And, you know, when we do our little summer of AK, You can let them, the whole point of this is don't worry. Don't. No adjita. Just go for it. And we bought one, then we buy two. Now they're out there taking pictures.

They're documenting it. And you're letting kids have that freedom plus responsibility of the camera. But as the parent, it's not like some digital SLR that's going to get dropped. If this thing breaks, buy another one. So shout out to these kind of devices. I do need another device. There is no device that allows you to play Spotify only and give it to your kids. You're out of war with your kids being addicted to technology.

There needs to be a single-use device that only plays Spotify, Apple Music, etc. And has nothing else on the device, but Wi-Fi and that. It's like the new MP3 player. Like a lot of people watching might be too young to remember the pre-I-Pod. There was like a whole category of V. A Rio. The MP3 player. Like Dell had one.

Like every company had one. And then iPod kind of stole the whole market. I will back this company. If somebody comes out with a light MP, that only does one thing, put it on Android. It's a music streaming device just for music streaming. That's it. And it just follows every Spotify thing. The fact that they don't have Spotify for kids, single use device, this thing could be incredibly smart.

It could be built into headphones. I would love to see somebody build me over-the-year headphones with this built into it. I will fund this company. You will sell a million units a year to parents who want to let their kids listen to music, but don't want them to have an iPod touch or their old Android phone or their old iPhone because they'll wind up on TikTok, do them scrolling. It's really, it's like it's a Roku for music instead of video because you can even do like Roku city where there's like a home page that's like, here's the best of what's new on title,

on Spotify, on Apple. Like you could make it that sort of brand agnostic central hub that people love. I mean, that's why Roku City so infamous. Yes, a streaming music player for kids. That's a great idea. Somebody please send me. I literally will put 100, 200K and cede this. Come to our accelerator.

You can make it look fun for kids. You know, you could make, like, have fun with the design of it. Give it like a, you know. Yeah, I think it's a fun idea. And if you go on Reddit, I have been on countless discussions. of people trying to find this. But when you, like, see the conversations, all the parents come to the same thing.

Every single one is, like, go buy an old iPad, iPod. And then you're like, no, then you have to sync the music. The kids want to stream stuff. Yeah, it's iPods only worked with, you had to have, like, the iTunes app that you were using to put the music on your iPod. It wouldn't be like, this would be, you know, who should make it is, is like Spotify.

should just make, like, they should just make one. They should, like I said earlier, just make one. Apple should make one single function, only one thing. The only thing I can find is you can take an old Android phone, put the child protections on it, delete every other app or time out every other app, but Spotify and just give them an old junkie phone. I've seen this one, and this one, as you can see, it has everything on it, and it has, Yeah.

Yeah, I've looked at this one again. It is a little too much phone. Yeah. I mean, now I'm flashing back to like being on Napster and LimeWire and downloading MP3s to put on your, to put on your... Everybody's got a family set now. Family subscription, easiest thing in the world. All right.

Great show. Oh, yeah, one more. Give us you one more lightning round. Let's get out of here. I'll do one more. I love the new Vampire Lestat show on AMC. Do you remember they were doing interview with the vampire.

They had a show on AMC over the last few years. So they did two seasons where they adapted Anne Rice's classic interview with the vampire. And then Anne Rice later wrote a novel where it was like La Stop, the vampire. Tom Cruise played in the movie. He survives because he's a vampire, so he's alive forever. So he survives into the present day

and becomes like a rock star. And so AMC for the third season of their interview with the vampire show, now they're adapting that. And so it's set in the present day. It's crazy. like the journey that this show has made from doing the original vampire interview with the vampire stuff like the movie to now being this like rock and roll band touring but they're vampires like it they made a movie queen of the damned out of this book many years ago but now they're redoing it as the show it's a lot of fun the acting is really good it's just so over the top you kind of can't believe that it exists it's really campy I'm enjoying it a lot so that's on if you get AMC play or shutter, you can watch the vampire list.

Oh, I have something special to show you. Oh, let's take a look. This is a video of me e-foiling. I have a video of me e-foiling. We've got video evidence of the e-foiling. Here we go. This will be your last. Yeah, let's close it out with that.

I can't top that. We'll put it to top that. Okay. Can you see me on the e-foil? Yes. Yes, I can. Do you hear the sound? or no?

I don't hear the sound, no, but I see it. Anyway, you see, I'm like now. Yeah, so I'm on the ground, but see it just lift off? Yeah. So you lean back and then it goes on. You see the joystick in my hand. This is when I was first trying to figure it out. But see, I'm up?

You're doing all right. You're going. Yeah. Yeah, this is when I was first getting up and then boom. Yeah, I fell after that. But anyway, that wasn't the spot where I fell and hurt myself. That was one of the... You know, it kind of looks like standing on a jet ski.

I have been jet skiing before. And like that's, you're sort of sitting on the jet ski like a motorcycle. This would be like if you're standing on it. Yes. It was, let me tell you something, this is fun. I recommend this for going fast on water is fun. Jet skis are a lot of fun. It is, I think, similar to a jet ski.

Try to see if I have another one here of me doing it. Hold on a second. This might be me. Is that me? This wasn't, it was the infamous picture of Zuckerberg. He was like wind surfing, not on this. I think he was doing a. similar thing, but when he was doing it, he was doing a regular foil where you pump up and down

in order to get the lift. Yeah, that like very nerdy, infamously dorky-looking photo of Mark Zucker Howard. There it is. There it is. Yeah, when he had the American flag and everything, rocking out with the American flag. He's got too much sunscreen on, so he looks like a ghost kind of. That was hilarious. Oh, yeah, there's that. And what are we doing here? Yeah, he's a lot of weird pictures of Zuck over the years. Anyway, this is my new... Pancake makeup. It looks like a kabuki.

It's not quite my new passion, but it is definitely up there in terms of flow experiences you can have. It's like, I think this would be second for me to skiing. And I would like to keep pulling this string and see where it leads me. Yeah.

A lot of fun. All right, that's this week in startups. That's your twist. For Friday, June 26, you'll consume it on Saturday. We'll see you all soon. Bye bye. Bye-bye.