Jensen on Dwarkesh, Cursor + xAI, Codex Update | Kevin Hart & James Morrissey, Ben Smith, Jonathan Criss, Paul Scherer, Matan Grinberg, Akhil Voorakkara, Charlie Cheever, Victor Cardenas Codriansky, Theodor Marcu

You're watching We Are Not a Car.

We are not a car.

We are not a car.

It is Thursday, April 16th, though, 2026, and we are live from the TVP and Ultradome.

The temple of technology. The fortress of finance. The capital of cars.

The capital of cars. Absolutely wild podcast between NVIDIA's Jensen Huang and Dwarkesh Patel. So many clips. So much debate. We're gonna kick it off with the question of whether or not NVIDIA is a car. First, let me tell you about our lineup because we have an insane group of guests joining us today. Kicking it off with Ben Smith from Semaphore. I think this is his third time on the show. He's live from the Semaphore World Economy Summit, has convened a massive group of policymakers, CEOs, investors. He's gonna give us the update on what conversations are happening on the ground. Then Jonathan Chris is coming on to talk about Starlink for water, former, SpaceX employee just raised $24 million to enable off-grid desalination. One of my favorite topics these days. We talked a little bit about it with Peter Diamandis. Then Paul Scherer from Eigen raised $15 million from Benchmark. James Morrissey and Kevin Hart.

New consumer social play. Yeah, Kevin Hart's coming on.

Talk about building a tequila business.

And we will be asking him about James and I, for sure. We've got to ask. I gotta know.

Then a lot of interesting coding.

Then Abaton from Factory.

Ulysses. We have a massive $46 million fundraise today for underwater and surface drones. Charlie Cheever from Expo is coming on. And Victor from Slash is coming on as well. And then we'll close it out with Theodore from Cognition. So big show. Should be going at least 2.5, 3 hours as usual. We will be off tomorrow, but we'll be back on Monday. Let me start off. Let's start off this video from Semi Analysis because it is a treat.

Wait and pause for a second.

I don't—

I, you know, I was asking you.

Yeah.

Like what app you use to edit videos.

Oh yeah.

I was to make this exact video and then I saw this and I was like, oh, I don't need to make it.

Yeah.

I don't— you're not talking to somebody who woke up a loser. And that loser attitude, that loser premise makes no sense to me. We are not. We're not a car. We are not a car.

Such a great edit. It's absolutely— it's such a good vibe reel. And it's so funny because— I love Tokyo. This is Tokyo Drift. It's funny because it's this new type of vibe reel where the semi-analysis team, obviously, they recontextualize the two clips and added the beat to get the video going. But that final edit with the cars sort of morphing together, like it goes from, there's a match cut. These are match cuts between one stick shift move and another, or one car drifting and then another car drifting. That takes a really long time. Somebody clearly made this video just because they're enthusiasts of Fast and Furious. And then the Semi Analysis team was able to quickly recontextualize that to be about Jensen's answer on NVIDIA's moat and the CUDA moat and what's happening with NVIDIA. So let's go through, this question because it's been humming for a while and it's sort of bubbled up most recently because there was a whole bunch of news that Mythos might have been trained on Trainium, then maybe TPU, then maybe Blackwell, and sort of a mix. And it just feels like more and more of the AI labs are capable of making other chipsets work. In the early days, it was all about NVIDIA, and now it feels like the incentives are really high to figure other options out and that creates a different competitive dynamic. So we can run through my thoughts on this and then we can go into the geopolitical considerations as well, because that was another fascinating segment of the interview. So Jensen, the CEO of NVIDIA, spent over 90 minutes. It was almost 2 hours when you include the ads too, in the ring with Dwarkesh Patel. It was electric. And the key question, at least for the start, was whether or not NVIDIA is a car. And I'm only half joking about that. It sort of was the key question. NVIDIA's been the market leader for years. During the gaming boom, NVIDIA was the gold standard for rendering PC graphics. There was always decent competition from AMD, but once the AI boom kicked off, the CUDA ecosystem significantly sped up development of AI systems and training of AI models. And for those who aren't familiar, CUDA is a programming model that enables GPUs to accelerate demanding workloads by parallelizing computation. So, Instead of needing to work on the very low-level instruction sets, if you are a CUDA kernel engineer, CUDA engineer, you can access all the power of the GPUs very efficiently while staying up in the more mathematical AI research, more standard Python, C++ programming paradigms. You don't need to dip down too low, but it's getting easier to dip down lower, and that's what we're seeing. So that, created the CUDA moat because developers were way more productive and the biggest bottleneck to progress was allowing AI researchers to quickly test ideas and scale their experiments up to whole fleets of GPUs and eventually entire data centers. So at the time, researchers really liked CUDA and really liked NVIDIA and they did not wanna have to spend hours and hours figuring out other hardware systems. They just wanted to, run their tests and see if the model was getting better and continue to scale. But recently the biggest cost center for AI labs flipped from researcher time, more or less, to compute capacity. And this creates a much larger economic incentive to figure out a way to drive down the cost of chips used to train AI models. I wrote about this back in, on Tuesday, October 22nd of last year, so almost 6 months ago. I said not every link in the supply chain of AI can be completely commoditized. NVIDIA has an insane amount of power having ramped full year revenue over the last 3 years from $27 billion to $60 billion to $130 billion. Absolutely insane top line revenue ramp at that scale. And that's why Jensen is so confident about how dominant this business is. It's the world's biggest company for, a reason it has been growing spectacularly at immense scale. And not only did they grow the top line, but net profit margin grew insanely. So it grew from 16% to 15%.

Yeah, you do that when you have pricing power and massive leverage, because, you know, in this case, demand massively outstripping supply, plus developer love and kind of just like—

And I believe the forecast net margin was 70-something percent. We were hearing about 80% potentially. And so, Nvidia, you know, the plan, and the plan is still to make an incredible amount of profit off of these chips because they're incredibly valuable. But all the hyperscalers and the AI labs, they are sort of incentivized now to form a bit of an anti-Nvidia alliance to commoditize the accelerator market and drive down those margins at least a little bit. And so today the AI chip market is starting to look much less monopolistic. AI coding agents can make it easier to write software that works on non-CUDA chips Sachs, and the teams behind competing chips have plenty of resources and economic incentives to bring performance in line with NVIDIA. Even if it's gonna be a big hassle, even if you're gonna just spin up a team to get AMD or TPU working, it's going to be worth it because you're talking about billions and billions of dollars spent on chips. For the past few years, NVIDIA—

Yeah, another example of like, you know, an instance where an AI lab had so much urgency that they were willing to spend whatever it took was Meta rebuilding like, yeah, exactly, like they were willing to outspend pretty much any other lab on talent because they didn't have time to find a homegrown talent or go through a normal recruiting process, especially considering a lot of those engineers were like happy doing what they were doing.

Yeah, yeah, yeah, exactly. Yeah, the incentives flip when you get to this scale, or the incentives just get so big you can build a whole team for a specific thing, solve any problem, and so For the past few years, NVIDIA has sort of looked like SpaceX's launch program. It's incredible technology with very few viable alternatives. And so that creates great margins as we've seen with SpaceX's launch capacity, and they control something like 90% of the launch market. While the products have not degraded, quite the opposite actually, Blackwell and Vera Rubin are incredibly powerful chips. They're clearly on the leading edge there, and they have an incredible amount of, of supply chain guarantees from TSMC and across memory and all the other different pieces of supply chain, like NVIDIA is ready to make more chips, but increased competition makes the category look a little bit more like the car market than the rocket launch market. And so that I think is where Jensen is pushing back and saying, no, there's a lot more that we bring to the table with our customers that don't think of us like a car. This is not, you know, the difference between a Ford and a Toyota they all sort of get you to the same place. And you can swap one out. You can be driving a BMW one day, go to the dealer, turn it in for a Mercedes, and you're going to have a pretty similar—

Yeah, or the other example that I was using with the team earlier was this idea of, if you're a delivery company like FedEx, and you have a lot of Ford vans, and then Hyundai comes to you and says, hey, you've been spending like $50,000 per Ford van, but would you consider a Hyundai van? We'll sell you— to you for $35,000. It's just as good. And it could be mildly inconvenient for the company because they're used to using Ford. They maybe have an internal team that does some maintenance. But when you start looking at that cost differential, it can start to get pretty interesting to say, hey, why don't we try out some Hyundais? Let's move over some of our routes to Hyundais and see how that goes. They try it and they're like, hey, this actually, this works pretty well. Maybe there's like, you know, higher maintenance, but it actually like maths out. And like, we're gonna actually start adding more Hyundais to our overall fleet. And so Jensen's point on the podcast is that like, we're not selling cars, right? This is not like something like you can just kind of swap out. And Dwarkesh was obviously pushing back.

And for a lot of companies, swapping out NVIDIA for TPU would be very difficult. Some workloads that Jensen Focussat, focuses on. He says we're not a tensor processing unit, we're an accelerator. There's a whole bunch of scientific computing workloads that work particularly well with NVIDIA. The problem that—

and of course, Harcash was just saying like, yeah, well, it's perfectly fine to have like a specialized chip for specialized workflows because the biggest companies in the world, the biggest buyers here have like a single type of workload that they're trying to do. And that's why they're using your competitors.

Yeah. And so because the AI buildout does not seem to be slowing down as long as power can continue to be brought As AI gets online and data centers can continue to find towns that will approve of them, demand for chips will presumably grow, but every chip designer and AI lab has to be praying that those net margins come down. How quickly will it happen? Very early to tell. The market did not react negatively to this back and forth in the short term, although the price of NVIDIA has been basically flat since last August, and I think that's why, Jensen is trying to sort of like reset on the narrative because it's possible that, you know, with the— there's a lot of movement up and down. They've been sort of flat. There's a desire to sort of like reset and recontextualize.

Yeah. Inference demand is scaling.

Yeah.

In a very significant way. And the Nvidia stock price is relatively flat.

Yeah. And there's— and since August there's been like so many different moments where fears of the AI bubble, fears of the products not finding product market fit, that revenue might stagnate. Like we've seen a ton of bullish signals for like AI demand broadly. Demand is there. And so if you're a supplier, you should also be going up. But there's been this overhang of what will happen to margins and market structure. And so that is what people are going back and forth on. Tyler, did you have anything else to thoughts on this?

I mean, personally, I think I probably am more on the DoraCache side where, yeah, I mean, at some point, these margins are so high. There's so much opportunity here. We've seen that actually, if you're a big lab, you actually can just put a little— maybe it's a lot of resources, but you actually can just train a model on a different architecture. You can serve them on a different architecture. You actually can figure these things out. As models get better, you can go lower and lower on the stack. You can write the kernels semi-autonomously. These things get faster and faster. And it's like, I'm probably much more on the Dwarkesh side.

Yeah, it'll be interesting to see how Groq fits into this, the new CPU integration between different pieces of the puzzle. And then also, I mean, Dwarkesh makes this point that the the supply agreements that Nvidia has might be a bit of a moat for the next few years while TSMC line time is so constricted.

Yeah, but even then, like, you know, Jensen was saying the kind of supply constraints, this is like a 2 to 3 year problem after that. Like you can just solve these things.

Yeah.

So it's like almost like I don't know how to think about this because, you know, to me it seems like, yeah, so much of the value of Nvidia is just like they have such an incredible relationship with TSMC.

Yeah.

But if— and it's so valuable because of how constrained it is.

Yeah.

But if that kind of constraint is maybe going to go away to some extent.

TSMC?

Yeah, that's what he says. Like, they're going to increase, you know, they can build a new fab in whatever, 3 years.

I mean, the big takeaway from the conversation is like you have one person who seems incredibly AGI-pilled, which is Dwarkesh. And then you have Jensen, who doesn't seem AGI-pilled in that sense at all.

Yeah, right.

He Dwarkesh asked about, you know, he was kind of getting at early on the idea of like, can, can, well, will you be able to just like prompt your way to Nvidia chips? You know, like you basically sell software.

Yeah.

Yeah. And he didn't put it quite as aggressively, but yeah. Yeah, that's of course like, you know, more, more on the nose. But Jensen was basically like, no, I don't, I don't think that'll happen. And then, and then when you get to the whole geopolitical conversation too, Again, it's like Dwarkesh is like, you're selling nukes. And Jensen is, in my view, like, I'm selling computers, right? And that was like the big rift. It was like these two kind of totally conflicting worldviews. And it made for some very—

Pay-per-view.

It was a pay-per-view. It should have been a pay-per-view.

Yeah, Shriram Krishnan summed it up pretty well. He said, every person hears reaction to the Jensen plus Dwarkesh podcast can be extrapolated directly from whether they believe in the frontier labs achieving short timelines for AGI, ASI. If you believe in the labs achieving RSI and then AGI, ASI for some definition of all three in the next few years, you're probably sympathetic to the frame Dwarkesh adopts. If not, you're probably more sympathetic to the arguments from Jensen. And so we can go into the export controls next and talk about that. Metacritic Capital sort of summed up a little bit of the why, just Jensen's rhetoric and how he wasn't conceding a lot of things. Dean Ball said, "Dwarkesh Jensen reveals how inconsistent and un-battle-tested AI acceleration talking points are, especially when they're filtered through the prisms of corporate comms and mass politics. Strategically coherent accelerationism is possible." He says, "I try." "But not currently prevalent." And Metacritic Capital says the problem is that Jensen doesn't concede anything. Compute spending going to the moon, $1 trillion revenue in sight, models keep getting better, no unemployment, software codes are good, other Western accelerators are bad, Chinese competitors are good, NVIDIA makes token costs decline 90% per year, but Chinese compute scientists are capable of making all the necessary algorithmic improvements. He also can't be AGI-pilled enough because at the end of the day, he is an intellectual property company in the business of sending a file to TSMC. I think it's part of Taiwanese culture to want to be loyal to all clients and don't have favorite winners. He doesn't want to betray his software co-customers. He has antitrust concerns.

Yeah, he was making the bull case for software, which was that AI agents are gonna use tools. So he is like, there's gonna be more users of software than ever.

Which is something I'm like somewhat sympathetic to, but yeah, it definitely—

It's still very easy to take the counterpoint.

The flip side of that, yeah. Well, let's play the distilled recap from Dwarkesh Patel of the back and forth with Jensen on export controls. It's about 4 minutes and we'll, we'll watch this and then discuss.

So if Chinese companies and Chinese labs and the Chinese government had access to the AI chips to train a model like Claude Mythos with these cyber offensive capabilities and run millions of instances of it with more compute, the question is, oh, is that a threat to American companies to American national security?

First of all, um, Mythos was trained on fairly mundane capacity and a fairly mundane amount of it by an extraordinary company. And so the amount of capacity and the type of compute that it was trained on is abundantly available in China. And so you just have to first realize that chips exist in China. They manufacture 60% of the world's mainstream chips, maybe more. It's a very large industry for them. They have some of the world's greatest computer scientists, as you know. Most of the AI researchers in all of these AI labs, most of them are Chinese. They have 50% of the world's AI researchers. The question is, if you're concerned about them, considering all the assets they already have, they have an abundance of energy, they have plenty of chips, they got most of the AI researchers, if you're worried about them, what is the best way? To create a safe world? Well, victimizing them, turning them into an enemy likely isn't the best answer. They are an adversary. We want the United States to win. But I think having a dialogue and having research dialogue is probably the safest thing to do. This is an area that is glaringly missing because of our current attitude about China as an adversary. It is essential that our AI researchers and their AI researchers are actually talking. It is essential that we try to both agree on what not to use the AI for. With respect to China, we want to have— of course, we want United States to have as much computing as possible. We're limited by energy, um, but, you know, we got a lot of people working on that, and we, we have to not make energy a bottleneck for our country. But what we also want is we want to make sure that all the AI developers in the world are developing on the American tech stack in making the contributions, the advancements of AI, especially when it's open source, available to the American ecosystem. And it would be extremely foolish to create two ecosystems, the open source ecosystem, and it only runs on the Chinese tech, a foreign tech stack, and a closed ecosystem, and that runs on the American tech stack. I think that that would be that would be a horrible outcome for the United States.

I mean, I think the concern, going back to the flop difference in the hacking, is yes, they have compute, but there's some estimates that because they're at 7-nanometer, they don't have UVs because of chip-making export controls, the amount of flops they're able to actually produce, they have like 1/10 the amount of flops that the US has. And so with that, could they train eventually a model like Mythos? Yes. But the question is, because we have more flops, American labs are able to get to these level of capabilities first. And because Anthropic got to it first, they say, okay, we're gonna hold onto it for a month while all these American companies, we give them access to it, they're gonna patch up all their vulnerabilities, and now we release it. Furthermore, even if they train a model like this, the ability to deploy it at scale, you know, if you had a cyber hacker, it's much more dangerous if they have a million of them versus 1,000 of them. So that inference compute, really matters a lot. In fact, the fact that they have so many AI researchers who are so good is the thing that makes it so scary because what is it that makes those engineers, researchers more productive is compute.

We should always be first and we should always have more. But in order for that outcome for you to, what you described to be true, you have to take it to the extremes. They have to have no compute. And if they have some compute, the question is how much is needed? The amount of compute they have in China is enormous. It's— I mean, you're talking about a country. It's the second largest computing market in the world. If they want to deploy, aggregate their compute, they got plenty of compute to aggregate.

Very, very tense back and forth. Dean Ball says it's a shame Jensen mostly fails here because the monoculture on export controls is bad. If you're a young AI policy researcher trying to make a name for yourself, it's almost impossible to be taken seriously unless you are pro-export controls. Monocultures are usually bad. And I am sympathetic to Dwarkesh's points there for sure, especially on the inference side, even if models exist in both worlds, like having a whole bunch of, you know, good guy compute that can go and patch bugs while the amount of attackers is much smaller. It's just a matter of how many resources you have on each side. That's a great point. The only thing that they're like talking around is just Taiwan as a particular turning point and how their various positions flow through to Taiwan policy and where the, how China's stance on Taiwan policy Taiwan is something that I've always puzzled, and I wish that both of them had articulated their sort of philosophy on actually wargaming out what export controls do to likelihood of Taiwan intervention or blockade or anything like that. I don't exactly know. I've been trying to like work through it, but I don't have a complete thesis. But we've been debating it back and forth all day. I don't know if you have a strong take on any of this, Jordy.

I appreciate Matt Zeitlin's point. What'd he say? He said, I kind of appreciate that Jensen Huang seems relatively normal about non-business stuff compared to other tech founder CEO types, but then when it comes to NVIDIA's actual operations, he's a complete sicko.

Yeah, yeah, I saw a couple takes around this that he had like some very, very, very, very strong points deeper in the supply chain. I haven't— I couldn't really assess how he did on that, but some people were— there were definitely people that were in Jensen's camp. It was divided, which I think is why this went so viral. Elliot Arledge summed it up here. Jensen Wong on Dwarkesh today. Most combative interview he's done in a while. The biggest regret, not finding—

Has there been a more combative interview? Ever with Jensen? It seems unlikely.

I think so. No. And I'm trying to think of just like in terms of combative, like people were sort of comping this to like the Elon DoorCash interview, which wasn't combative, but it was more just like digging into all the different projects and trying to elucidate the strategy and how they all fit together. There's actually some interesting updates on where XAI might be going with a partnership with Cursor that we can go into. There is some funny details in here. Apparently the Larry and Elon begged Jensen for GPUs at dinner story, that never happened. We absolutely had dinner, at no time did they beg for GPUs, which is funny. I wonder what happened. Quietly folded Groq into the CUDA ecosystem. Reason, premium ASP tokens where latency beats throughput, higher throughput, used to always win. That trade-off has bent enough that NVIDIA is expanding the Pareto frontier towards fast, expensive, low throughput. And it will be interesting to see how all of that comes together. On ASIC cost savings, NVIDIA margin is 70%, ASIC margin is 65%. What are you really saving? Someone still pays Broadcom. Open challenge to Google and AWS, publish TPU and Trainium on MLPerf and Inference Max, I would welcome Trainium to demonstrate their 40% claim. And so Jensen threw down the gauntlet and said to Amazon and Google, head over to Semi Analysis and put your chips on Inference Max so that they can be properly benchmarked. So we can see.

Yeah, that was a good— that was a very, very good call out.

Yeah. Yeah, it was smart.

I just don't know why they don't go over and—

I've been excited to see the Inference Max results. Dwarkesh posted tomorrow and it's him and Jensen standing next to each other. And Alex Volkov said, hey, Dwarkesh, was this picture taken before or after the pod? Because it does feel like it was a tense situation. Although, you know, to both their credit, like Jensen, it felt like he loved being in the arena, getting asked hard questions, like working through this. There's this back and forth where Jensen's pushing back and Dwarkesh says, oh, I can drop it. And he says, you don't need to drop it. I'm enjoying this. Like, let's hash this out. And I thought that was very diplomatic and just good overall. Well, Intel is up on the news, up 4% today, 10% over the past 5 days, almost at all-time highs. I think we're very close to the 2000 peak for Intel, which it was also around where they were trading in 2021, $67 a share. $330 billion company, clearly with all of this backdrop and just the idea of more chips and maybe the CUDA ecosystem being something that you can work around. Can an American fab run by Intel produce a chip that's viable for an AI lab? It feels like increasingly, yes. That's certainly the argument that's being put forth by Dwarkesh, and it would be very exciting. I think it's something that everyone would support, an Intel resurgence. And there's some news around TeraFab potentially getting involved. But first, let's start with the scoop from Grace K over at Business Insider. She says, scoop, Cursor plans to use XAI's infrastructure to train its Composer 2.5 coding.

Where's the golden scoop, Tyler?

According to people familiar with the matter, Cursor will use tens of thousands thousands of XAI's GPUs, they said. And we got a scoop for Grace K.

This one's going to Grace. Grace, congratulations.

You win the golden scoop. Congratulations.

Now, interesting to see something we talked about probably midway through last year. XAI's shown a tremendous ability to, on the kind of infrastructure data center side, spinning up a huge amount of compute very, very quickly ahead of any timeline that any reasonable party would have probably expected. And demand hasn't exactly followed in the way that they would have liked. And so opening that up to a company like Cursor who has all the demand and what they really need is their own model.

Yeah, it was also interesting because I don't know if If it was thrown out as a potential project for other companies, I feel like MSL mentioned it at some point, maybe OpenAI, but there was some talk of like, okay, if you're marshaling all this compute and you wind up with too much, like what do you do then? And the idea of becoming a cloud provider if you have a data center and everything—

Yeah, that's been the big question of like, you know, with, everything that SpaceX is doing and now TerraFab, they're going to be creating all of this capacity. But what is the, what is the, like, where's the demand for that capacity going to come from?

Right.

And so you can imagine a world in the future where SpaceX has a bunch of space data centers. They open up that capacity to a bunch of companies other than just Elon Inc. businesses.

So Grace says the setup effectively turns XAI into a kind of cloud provider by renting some of its GPUs to other country companies. XAI could start generating revenue from its massive infrastructure while still developing its own AI models. The arrangement could help the company offset the costs of building and operating data centers while also deepening ties with a startup that has access to valuable coding data. And so there could be some sort of trade deal going on. Ed Ludlow at Bloomberg has a report from the TeraFab. Musk team is actively requesting price quotes and delivery timelines for a wide range of chipmaking equipment, photomasks, substrates, etchers, deposition, cleaning, testing tools, according to sources. Elon Musk lieutenants have reached out to chip industry suppliers for his envisioned TeraFab project. Remember, he was pictured with Lip-Bu Tan from Intel, I think, last week. Early steps in an audacious and likely arduous attempt to break into the production of cutting-edge chips. That is a very, very tall order, but maybe there's never been a better time to break into the cutting-edge chip market, given that you don't need to— I mean, you sort of need to reinvent CUDA, but it's becoming easier potentially. Actually, that is the story. Well, there are two big releases from Anthropic and OpenAI today. Claude announced Claude Opus 4.7, our most capable Opus model yet. It handles long-running tasks with more rigor, follows instructions more precisely, and verifies its own outputs before reporting back. You can hand off your hardest work with less supervision, they say. Very good score on on SWE-Bench Pro, 64.3%. Excited to test it. They say Opus 4.7 has substantially better vision. It can see images at 3 times the resolution and produce higher quality interfaces, slides, and docs as a result. On the API, the new X High effort level between High and Max gives you finer control over reasoning and latency on hard problems. I'm excited to see how these do on Arc AGI V3.

Yeah, I think the most notable thing here is you have a model a model card that shows a model that's not publicly accessible.

Yeah, yeah. They share the Mythos benchmark.

Mythos logging Opus 4.7 just sitting there. But of course, unless you're one of the select companies, you won't be getting access to it at least yet.

Well, Adi gave Opus 4.7 the tiny man bench, which is a very funny one where you ask the model "I am a tiny man. When my son was born, they handed me to him." And Opus 4.7, "That's a nice little riddle. The answer is a bar of soap. No, wait, let me think about it. Tiny man handed to him when his son was born. This sounds like it's pointing to a cigar, the tradition of handing out cigars when a baby is born, and a cigar can be thought as a little man-shaped object." That is very, very odd. This This is like the new how many Rs in strawberry, where all the models seem to have weird results for this. We, of course, have shrimp fried rice bench, which no one has really solved. Even just humor bench and tell us a novel joke has been increasingly difficult for these companies. Let's go over to OpenAI.

I think we're going to have to wait and see. On OpenAI's 4.7, what the reactions are. I will be keeping an eye on Dan Shipper, who always does great analysis immediately after launch.

Yeah.

Yeah.

I mean, I feel like generally benchmarks are like almost completely meaningless at this point. Like there's so many like open source models where they can like really, I don't know if you know if they're like actually benchmarking, but like you just talk to a model for a while and you say, okay, this is a good model. Okay, or this is not a good model. Like, it's very hard. It's very hard to get like a quantifiable signal at this point.

Yeah, I mean, increasingly, I mean, it feels like all of these models will be sort of like tested through like enterprise rollouts and you'll just do a demo project, swap in something, test this model, test that model, just see, okay, well, our goal was to deliver value value, and we had some KPI like reduce cost, increase sales. Like, were we successful when we implemented this model, rolled it out to our team, used it for a couple weeks? Did we see a tangible result or not? That's at the end of the day all that matters for businesses. So I could imagine that that's where all of this goes. Well, OpenAI announced Codex for almost everything. It can use apps on your Mac, connect to more of your tools, create images, learn from previous actions, remember how you like to work, and take on ongoing and repeatable tasks. With computer use on macOS, Codex can now use any app by seeing, clicking, and typing with its own cursor. It runs in the background without taking over your computer, working on tasks like frontend iteration, app testing, or any workflow that doesn't expose an API. You can now generate and iterate on images with GPT Image 1.5 and Codex to create front-end designs, mockups, game assets, and more without leaving your workflow. Usage is included in your ChatGPT account. No API needed. Automations can now run in the same thread. Lots of updates here. And Thibault says Codex just got a lot more powerful. Computer use, in-app browser, image generation, editing, 90+ new plugins to connect everything, multi-terminal, SSH, lots and lots of stuff. So go give it a try. Test, go take it for a spin, openai.com, of course. You can download it for macOS. Well, speaking of websites, DJB went to the Oreo website and hit accept all cookies. Now we wait. I love this. Very funny. What else is going on in the timeline? The West creates the internet. Try nailing that Jell-O to the wall. CCP nails it. The West creates LLMs. All right, try nailing this. And the CCP picks up a hammer. There's a piece in the Wall Street Journal opinion section. AI is bound to subvert communism. This is a very contrarian take because people, at least with the internet, there was the perception of like decentralization, permissionlessness, anonymity, a lot of things that felt very democratic. AI is very centralizing by default. This is the teal take of like AI is communist and crypto is libertarian. And so this is a pretty wild thing to argue, but read the opinion piece and see what they say. The nailing Jell-O to a wall, I believe that's from the Clinton administration, the idea that the internet would spread so widely that the Chinese Communist Party would not be able to control the population. Everyone would be coordinating. It'd be sort of like an Arab Spring type moment. But of course the firewalls went up, the surveillance happened, and nothing really changed. And the Communist Party seems stronger than ever. But this does sort of undergird a lot of what Dwarkesh has been saying about the risk of China and having strong AI and stronger control over the population. It's always hard to get a read on exactly how things are rolling out in China. There's some people that seem to like it over there. Of course, Dwarkesh took a whole trip to China and made it back okay. So it's not all doom and gloom, but there are— it's a tricky, tricky thing to argue, but we'll see.

YouTube, according to The Verge, now lets you turn off Shorts.

Mm-hmm.

It is getting community noted.

What does the community note say?

This feature is a new setting for YouTube time management. Does not hide the Shorts tab in the app, nor does it hide Shorts from appearing in recommendations. When set to off, the only change is that a dismissible warning screen will appear when the Shorts tab is selected.

Oh, interesting.

Yeah, I've tried, I've tried. I land on YouTube on my phone, it immediately recommends a bunch of Shorts. You can hit the three dots.

Yeah.

And say like, show me less Shorts. And that feature does not work at all. And I don't watch Shorts on—

You like Shorts.

And I don't watch Shorts on YouTube.

The Shorts in the feed have been a lot. I wonder if they're seeing, yeah, I mean, they must just be seeing data that shows that the Shorts are more retentive or keep people on there.

Well, there are, when you're using YouTube as a video search engine, there are so many, there are so many types of searches where it can be vastly preferable to watch a 60-second video on something. So an example is like if you're doing research on cars.

Yep.

You wanna like quickly understand how the seating. I need to get a new like family car, right? So I've been looking up different cars, trying to get a sense of like how spacious the cabin is, what the seating arrangement is like. And so I'd rather just click into a Short and just get a quick 60-second overview and then bounce. I don't need to see like a 15-minute review of the entire car. But yeah, it's—

[Speaker] That YouTube Shorts creator, Forest Auto Reviews sort of pioneered the perfect 60-second car review where he talks really, really fast and says like, inside you get Apple CarPlay and Android Auto and, you know, the heated leather and massaging seats and vinyl on the dash and carbon fiber here. And he sort of runs through the whole car in 60 seconds, shows you how fast it takes off if it's a fast car, opens the door, shows you all the features. And it's sort of like a 20-minute Doug Gilmour video cut down to 60 seconds, which of course is, is doing very well. Doug's response has been to create 1-minute cutdowns of his content. And so you can enjoy that as a Short as well if you want. But if you do want to turn off Shorts on the YouTube feed, the way to do it is to uninstall the app, use the mobile version, and then get a plugin that there's an extension for Safari, I think, that will just go and and remove the actual Shorts as they pop up. But it's a hassle and it's way slower than using the app. So hopefully this feature becomes more real in the future since it does seem like some things that people are demanding. But always tricky to ask a platform to give you an option that would, in theory, increase churn. Very, very tricky. Well, we have the beginning of our our guest lineup starting now. We have Ben Smith from Semaphore. He is the editor-in-chief. He's in the waiting room, and let's bring him in to the TVB and Altium. Ben, how are you doing?

What's going on?

Doing great.

Can you guys hear me all right?

We can hear you clear. We can see you, but we don't know where you are.

You're looking very sharp, dapper, I might say.

Thank you. That's, uh, you know, we're convening a biggest gathering of business leaders in the US this year here at Semaphore World Economy.

Wow.

How many business leaders is that?

Yeah. How many people attended?

500 global CEOs and then lots of political figures from the US and Europe. Had Scott Bessant earlier. We got Lutnick later. I assume they all just kind of came over from Silicon Valley. Yeah, we do our best.

That's great. What's the structure of the conference? It's a few days. Are there specific breakout sessions, talk tracks, certain themes, or is it sort of more driven by the individual who's on stage at the moment?

You know, it's 5 days, 3 consecutive, 3 simultaneous stages, a bunch of really interesting kind of Chatham House rules, breakfasts and dinners and things like that. And I mean, the core topic is really, I mean, kind of where we live is the intersection of politics and sort of political power and finance and technology. And so, I mean, that's, that is what But that is what people here are obsessed with.

Yeah. And within that intersection, what are the top topics that people are discussing in particular? Obviously, the conflict in the Middle East is big, but then AI is big. What's actually driving the conversation today?

You know, every conversation winds up coming back to AI.

Really?

And yeah. And I think whether that's, you know, Besant— actually, I had an interesting conversation with Besant on stage where he really downplayed the conflict with Anthropic.

Okay.

I said they have, they have kind of like some minor technical issue with the Department of War. Oh, the Treasury is working with them to make sure, you know, the banks don't go down.

Sure.

But just, you know, I think there's, there's just a sense in particular that like really big enterprises and the kind of big brand name American companies are the ones most at risk.

Yeah.

In the transition. And, and I think people who work for those companies, who advise them, who compete with them are thinking a lot about that.

Yeah. What about on the labor side of big business? We've seen some reasonable strength in the job market recently. At the same time, there's a lot of layoffs going on. How are people grappling with the, with the unemployment situation in the economy?

Yeah, I mean, you know, the numbers are good.

Yeah.

And yet there is this sense, you know, among the American public that AI basically exists to put everybody out of work. And they hate it.

Yeah.

And so that's lingering over things. And I do think there's also a sense that like the class of 2026 is not a great time to get out of college.

Yeah.

And I think there is a, I would say like both like kind of authentic nervousness about, you know, what is everybody's labor force going to look like? Because of course class 2026 are also the ones who are native to these tools.

Yeah.

And are the ones you ought to be hiring.

Yeah.

And then also of like, are they going to come and chop everybody's heads off when they realize they aren't getting jobs? So I would say those two, two competing questions about the youth.

Has there been any discussion around resolutions to that or economic policy that can strengthen the safety net or, you know, deal with taxes or more weekends or, you know, anything that can sort of offset the anxiety about weakening labor markets?

You know, every member of Congress is like thinking about this all the time, taking questions about it all the time, and yet also aware that this is an institution that spent like 20 years talking about regulating social media and only now is sort of getting to it. But most of it's happening in the states.

Yeah.

And so I think there's a lot of pessimism that Washington will act. But I think there's also is a sense that I think a lot of companies are seeing unexpected jobs coming out of AI. I talked to somebody who operates like basically the sort of call center dealing with financial fraud that you would think would really get wiped out by AI.

Yeah, totally.

The good news is there is so much more financial fraud being driven by AI that they're keeping people very, very busy.

You can, you can see something like that emerging in in the legal space is like you just get way more contracts because contracts are easier to create or you get more lawsuits. So a lawyer is spending less time per lawsuit, but there's a lot more of them to work on. And so you still have full—

the good news. Yeah. How are you using AI in the newsroom at Semaphore?

Are there—

I imagine you're not just writing AI articles, but is there internal tooling that's been built, or you understand that.

Well, and I feel like you guys are well set up for this because there's like, hey, here are the facts, and then here's like takes, and like separating those two things out. I don't care what you use here. I want the takes to be generally pretty organic and coming from the author. But how do you think about it?

Yeah, you know, we use it. And it's interesting because I think like a lot of media, there's these two sort of huge trends. One is AI and automation. And the other is creators and individual connection. And people do not want AI in the way of that. I mean, when they realize that you're just avatars, you're going to be in trouble. And so we use it all over the place behind the scenes to try to like help amplify our journalists. I vibe-coded something that predicts mean tweets about stories that I'm personally very proud of, though I cannot say that I have driven wide adoption with my team.

That feels like that would just create like anxiety in the newsroom.

If you want to get pre-dunked on, use this tool.

That is what the tool— I called it pre-dunk.

Pre-dunk. No way.

Yes. They spent a whole day. It's in Google Apps Script. I love it.

That's awesome.

I would say more usefully, one of my colleagues— because actually one of the challenges with an event like this is you've got 3 simultaneous stages and the reporter who's most qualified to write the story or even to know what's news is often the same one doing the interview and sitting on stage.

Sure, sure.

And so we did build a great tool that, you know, just takes the transcript of every interview instantly and pops out what's new.

Oh, yeah.

So that a reporter who's maybe more of a generalist totally will know, oh yeah, this is news, this isn't, which is incredibly useful. And we, you know, obviously we check it, but that's a great starting point.

Yeah. Yeah, that makes a lot of sense because there's so many times when people fall back to a talking point, but you're looking for something that's actually novel or a new data point or a new prediction or something announcement, and that just might sneak in.

Like, how do you do that? You do an interview and you think there's one part that's interesting, but you're on the air and something else goes viral, right?

Totally, totally. Yeah. And we've— and we— that's a very interesting tool. We should probably think about integrating something like that because we— yeah, we found that it's very, very hard to predict, even with the best AI tools, what content, what clips will actually perform perform, what is news. And we see a lot of coverage downstream of our show where something that's said on our show will then turn into coverage on other news sites. And we've never really figured out how to grapple with all of that, but that's an interesting point.

Yeah.

And it's not just performance, right? Like, it's also you might think this is the new part, this is the interesting part. But if you're doing it live, it just sort of floats by.

Yeah.

So that's probably the thing that I'm just this week that I'm most kind of jazzed about.

What about overall market sentiment?

Capoluto, who made it.

Sorry.

What about overall market sentiment? I don't think if you told me last year we would be, you know, over a month into a war with Iran and the S&P 500 would be at a record high. Yeah, I would not have believed you.

Yeah, I mean, it's amazing. And, you know, people who— CEOs, people who show up at events like this tend to be optimistic people. It's sort of a job quality. And so there is like a pretty positive vibe here and being driven to this or parallels and is driven by the markets. One thing about Washington now, though, is there's also a lot of self-censorship and there's a real difference between what people say in public and what they say in private. And I think, you know, in our kind of Chatham House rule gatherings, you're hearing a lot more concern about a kind of— but just about the long-term direction of U.S. power, the U.S. economy, the U.S. markets, people hedging away. And no one's saying that stuff in public.

Yeah.

Well, yeah, biggest ever disconnect in my life between like group chat conversations and like timeline conversations among people who have something to lose is how I would put it.

Well, yeah, interesting times. Hopefully, what's the plan to turn the synthesis of what you're hearing in the Chatham House context into reporting or content? Will you be writing op-eds that are informed by what you learned? How do you think about actually distributing that information?

I mean, the basic Chatham House rule is that you can steal the ideas and pass them off as your own, but you're not allowed to attribute them. So that's the plan.

Okay, well, we'll look forward to it. I'm sure everyone can follow along at Semaphore. Thank you so much for coming on the show.

And we'll make it next year.

Yeah, it'd be fantastic.

Yeah, thank you.

Have a great rest of your day. Have a great rest of the conference and congratulations.

You guys too.

Thanks for having me on.

We'll talk to you soon. Up next, we have Jonathan Kress from Vital Life working on desalination with a fantastic new fundraising round. Jonathan, welcome to the show. How are you?

I'm doing fantastic. How you guys doing?

We're great. I'm so excited to talk to you.

Insanely fired up.

We were talking— Let's go. Yeah, we were talking with Peter Diamandis about desalination and it's always been, such a fascinating, just not even a sci-fi technology, but just an underutilized technology that has incredible promise, but hasn't felt like it's hit the takeoff that it needed to. And so excited to hear your plan, but why don't you kick us off with an introduction on yourself and the company and how you're thinking about things?

Yeah, so I'm John, CEO and founder. I was at SpaceX for 13 years before this. Started off in Dragon.

Amazing.

Started off in Dragon. Elon was like, stop using Dragons or building Dragons, you idiots. And I was like, oh man, I'll figure that out, boss. Started off with a small team to go figure out Dragon reuse.

Yeah.

The whole program became reuse vehicle, so just took over as product manager/RE. And then kind of got bored of spaceships, wanted to take on a new challenge. Jumped over to Starlink team to figure out rate manufacturing where we got really plugged in on how you design for really high volume manufacturing. And then my co-founder and I started really talking about how do we make a meaningful impact in the world? What's the next big challenge? And then got our heads wrapped around answering that question that every 8-year-old kid has. How could there possibly be water scarcity when there's oceans everywhere? And then just like you said at the top of the segment, it's like, man, this technology feels like it's really cool, but it's just right on the cusp of becoming awesome. And we think if we apply the lessons that we learned on Starlink with high-rate manufacturing, driving down the cost, and really delivering this technology to everyone, we can really make that meaningful impact.

So walk us through the first product. I was always thinking that the next big desalination company would be a huge desalination plant, but you're going much smaller. So walk me through the thesis there.

Yeah, I love that because that's pretty much where everyone in the industry has focused, right? We got to clean a ton of water. These systems are super freaking expensive for upfront capital costs. So the only way to make your unit economics in that use case work is by going really large and cleaning a ton of water. We just looked at it from the other end is like, man, if the upfront cost is negligible and you can keep your energy and maintenance costs low, you can probably do about the same and just scale it in terms of number of units, not a large centralized system. And then it's just kind of blown up the thought process entirely inside of that because now you're looking at infrastructure resilience and people being less reliant on centralized systems. And then you can open up to people that are more on the move, people that are island-struck. And really, man, the customer base kind of exploded from there. Like, oh man, like, if you could just deliver a low-cost product, people will want to use it.

Yeah. So, so walk me through like a textbook customer. Where are they?

Can you explain first? Can you just explain the product?

Like, I'm 5. Yeah, like how big are we talking? Is this diesel generator size?

Like, yeah, so we launched Access this week, which is our first consumer product. Uh, we're a little bit more in the middle market, uh, going into, uh, customer bases that already exist. So maritime industry has these, these types of products. It's a reverse osmosis, uh, is at the core of our technology. Uh, what we do is we're able to do that manufacturing process at a much, much, uh, lower cost and then a much higher volume. So the product that you see on your screen, Axis, the facility that we got here in Torrance, we're able to produce more of these devices in a single month than currently exist. So we're kind of really taking that whole process and scaling it.

Okay, but explain, but sorry, this is, you put a hose into salt water and then it goes into the machine and out comes potable drinking water.

Exactly.

So we always say we design for Terry.

I just, for the audience, I wanted to be clear.

So how many people off-grid actually have access to saltwater or dirty water?

Well, this should work for springs and lakes and—

Is that the idea? Yeah, yeah, yeah.

So you can clean what we say is any naturally occurring water source. Saltwater being the hardest one to clean and why desal has been kind of out of reach for most people until now. Well water is super common in the United States. I think it's like 11% of homes live on well water in the US. And then brackish water is super common. I mean, in marshlands, I mean, lots of, lots of South, pretty much all Florida.

Yeah. How do you deal? One of the common things we hear about desalination is that, yeah, you get clean water out, but you also get all of the dirt and all of the salt and, and whatever else you don't want in your clean water. What do you do with that in your case? What happens normally?

Yeah, so the stigma is in your, what you call brine, or your concentrate. So when you run a very large plant, you want your recovery as high as possible, so your concentrate is much, much higher, like 2 times the salinity in your salts. We actually run our system efficiently at a much lower recovery rate, 15 to 20%. So as long as you're putting your brine or your concentrate back into a source water greater than 5 gallons or so, it almost immediately dissipates into that volume. So we get the wins inside of there in terms of, hey, you can use this product. It's not making that environmental impact because it's smaller form factor, lower flow rate, and lower brine concentrate as well.

What about battery? Is this bat— is it battery power? Do you plug it in? What's battery life like? Could you eventually create some type of solar array so that it's entirely self-sustaining?

100%, yeah. So it runs on AC/DC. It's got an integrated battery. If you're cleaning ocean water, you're running for about an hour. If you're using fresh, fresh water, anywhere from about— probably closer to 3 hours. So it all depends on the source water that you're cleaning. So ocean water has a ton of salts inside of it, so it's really difficult to remove those at higher pressure. Freshwater is much, much easier. You're essentially just killing bacteria and viruses. So yeah, you can run it on AC, DC. We do integrate with solar as well. So, you know, if you have a decent-sized solar panel, you can run this thing as long as you have sunlight.

Amazing.

Talk about what you've learned about the traditional large-scale desalination operations. There's actually an article in the Wall Street Journal today about San Diego producing surprisingly a ton of water from desalination. San Diego now has so much water that it's selling it. Once a drought poster child, the California city now generates enough water to rescue parched states like Arizona and brew beer from recycled sewage. Are you optimistic?

What about that last part?

Recycled sewage. So they take sewage water and they clean it so effectively that they can brew beer in San Diego. They might have to disclose that on the label, but I think we can get past the stigma at some point.

Though wastewater once upon a time— Bill Gates solved this problem a long time ago and was like, no one wants to drink poop or pee water, right?

Yeah.

Luckily, the ocean is arguably the most abundant resource on Earth.

With.

Yeah, uh, and that's, that's kind of the, the, the resource that we want to tap to, to bring this technology forward. But, uh, large plants are, you know, they serve a purpose. We want to be a partner with them, not, not a disruptor in the near term, right? Similar to a Starlink, you had your, uh, what was it, $10 billion infrastructure bill to give internet to rural areas that service zero people, uh, yet Starlink could do that, uh, over at a fraction of the cost. So we look for those people that are kind of high and dry right now and need access. And that's kind of our initial customer base that we're going through.

Yeah, how much does the product cost? Where's the business? Give us an update on the round.

Yeah, yeah, so $749 is our all-around cost right now, but you can pre-order for $88, totally refundable.

Wow.

We're getting into production here over the next— actually, our first PCBs show up next week, which I'm very excited about. The round was towards the end of last year. We a great table, really awesome partners with GC and Interlagos. Yeah. And that capital has gone a really long way to help us accelerate. It's like the business has only been around for a little over a year and we're about to step into production here over the next few weeks.

Yeah, it looks like you're in a production facility. How big is the warehouse? How many people do you have on staff?

Yeah, we're 37,000 square foot facility here in Torrance. We're actually right next to the K2 guys.

Oh, cool.

I keep seeing all my friends on your guys' show, man. It's kind of a rite of passage at this point. Uh, we have 37 full-time employees right now, plus a ton of contractors that help out. Um, and we'll probably be closer to, man, 70-ish when we're at rate, with a bunch of associate and production staff just waiting to hire once we get into production this summer.

Yeah. Do you think that the business will remain almost entirely direct-to-consumer, or do you think there will be applications for business-to-business, construction crews, military, government, all sorts of different things?

Yeah, I use this a lot. Gwen gave us a talk in 2013. She was like, yo guys, we got to get NRO certification because NASA is pretty much supporting this entire business. So when we started the business, we wanted to have multiple revenue streams that in the event of a macro event, we could— we can be reliant on. So the US military has a water logistics problem. Half of all casualties coming out of Iraq and Afghanistan were related to convoys moving water and fuel. So they're excited to be able to open up where they can get water, water from. And then, you know, NGOs, humanitarian groups are almost solely reliant on bottled water, which is another logistical nightmare. So these are both groups that we've engaged with and we have a lot of exciting conversations with, but then also some partnerships that we're going to be announcing pretty soon. And then direct-to-consumer is the hardest problem. Elon used to tell us all the time when we were starting Starlink, like, go after direct-to-consumer because they're going to give you the biggest feedback. You're going to get roasted on Reddit and you're going to know quick whether or not you made a mistake. So after being personally roasted on Reddit a number of times, you learn really fast. So that's who we want to go after first. We want to solve the hardest problem first and then open up to, to those other larger customers afterwards.

Yeah, that makes sense.

Awesome. Well, very, very, very, very cool product. And I'm going to preorder one just to play around with it, bring it to the beach just to have some fun.

Come by the factory. I always say bring a bucket of your favorite ocean water. We'll clean it and drink it. It's a lot of fun.

That's amazing. How long does it take to clean a bucket of ocean water?

Well, we do 6 gallons of ocean water an hour. So about 1 hour, Well, a little under an hour because most buckets are 5 gallons.

5 gallons. Cool.

You're going to need 2. John's like horse-sized, so he could easily go through like 10 gallons an hour.

Okay. The chat wants to know what happens if you put a Diet Coke through it.

It'll separate all. We do this all the time.

You do?

We did a reverse Jesus the other day. We removed all the alcohol and water from wine.

That's amazing.

Yeah. You'll pull out all the gross stuff and just have a pure glass of water. Some water. We did Monsters the other day where we removed all the water from 5 Monsters and then drank the concentrate, which was just a super caffeinated syrup.

Wow, that's insane. Sounds like a fun time.

That's crazy.

Well, congrats on the progress and thank you for your time.

Yeah, great to meet you, Jonathan.

And your company.

And we'll talk to you soon.

Such a pleasure, guys.

Have a good one.

Great stuff.

Goodbye. Up next, we have Paul Scher from Ego. Eigene raising $15 million from Benchmark to build a mutual friend platform. Paul, how are you doing?

What's going on?

Welcome to the show.

Hey, so good to meet you guys.

Great to meet you.

Great to meet you. And excited that you're not building another agent for the enterprise. You've got something new for us.

Yeah, break it down. Introduce yourself and the company.

Yeah, I mean, so great to meet you. Thanks for having me. Of course. We're building, you know, we call it the world's mutual friend. Instead of all of these things that are out there trying to optimize our own bubbles, we're sort of, they're diverging more and more. And the world that's super hyper-personalized for all of us is also really isolating because it's different for all of us. And we want to bring these bubbles a bit closer together again.

Yeah, so walk us through the experience. Well, first, maybe like, where are you in this journey? You've raised some money, but do you have a team? Is the product live? How are you thinking about actually building the company?

Yeah, I mean, it's super early. We're really, really grateful for all the support and all the partners that we've been able to meet. I'm from Germany. I literally grew up in this tiny village, less than 1,000 people. I moved here like less than a year ago to San Francisco. And I've been, I literally landed, I had a one-way flight here, stayed in a hostel. And a few months later, I've been meeting some of the greats and they've been so incredibly helpful. And we've built a small team. It's very early. We don't, we haven't launched anything when we were working with early users to build something really special and something really meaningful.

So what can you share about—

Living the SF dream.

Yeah. Well, maybe take us back first And what were you doing before this? How did you get into startups? Walk me through that.

I've been building things, you know, forever. I grew up, you know, in this really middle-class town and I never knew anything about entrepreneurship. So I've always been building things. And then during the pandemic, I actually just went on Twitter and I started to spend a lot of time, like 12 hours a day on Twitter. I think I have like 25,000 tweets and replies from like a—

Founder.

From like a 3-month period.

Yeah.

And that's when I first, you know, I didn't know what a VC was until I was like 18. Yeah. And yeah, it's like during the pandemic, like about to turn 17, I left high school to do that. And so I've been doing startups for 5 years now, you know, spent some time in Paris where I worked with the incredible people of a company called Augment. And then, you know, I could finally get a visa and go to the US, which I always kind of felt like was my real place to be.

Yeah.

All right, and then take us through the idea maze and how you landed on this like general kind of space.

You know, I think I've always been like really interested in sort of this intersection of arts and culture and, you know, technology and like, you know, growing up in this like 'cause it's a place where I grew up and I really love German children's literature. And there's this book that my dad read me when I was 10 about this little girl called Momo who kind of connects everyone in her little town and helps sort of save the world from the time thieves. And I've always sort of been inspired by this. And so it's been something I've been thinking about for a really long time. And I think now more than ever, we really need this.

Feels very early, but you got Benchmark to come in and invest $15 million. They are the consumer social fund. They're willing to bet big and early, but typically I see them investing after there's like a spark or like some sign, real like sign of life. Do you feel like you're there yet in that you have something that early users love like at what point do you, and then I guess at what point do you think you'll launch?

So I think with Benchmark, you know, and just more broadly, we already see, I think, moments of magic and it's very, very early, but it's been just really incredible to see how people create these moments of connections, you know, early users. And I think just partnering up with Benchmark has been, an incredible experience. And I think I don't feel like I've ever pitched Benchmark. I think we just met and we sort of very quickly, I think the deal got done like 3 days, and we very quickly just realized that we had very similar ideas of what the future should look like and that we wanted to go on this journey together. And everything else kind of became secondary to just trying to make that happen together.

Do you think there's a risk of becoming a dating app, or is that the wrong way to think about it?

That's funny.

Um, is that just like historically, historically, the like, you know, meet, meet people in your area apps, even if they're intended, you know, they're intended to create friendships, you know, you may get power users that have other intentions.

I mean, totally. I think for us, we care more about strengthening already existing relationships, right? We live in a time where we have all made way more relationships than ever before. 100 years ago, you might have 100 friends. Today you have probably like about 600 relationships. And we really care about making these relationships meaningful much more than like adding just random people on top of that. And so I think it's less of a thing for us.

¡Gracias!

Have you been surprised that, you know, we're this far into the consumer AI boom and yet the only AI apps topping the App Store charts are basically language models?

I think it's natural, maybe. I don't think we're in the consumer AI boom yet because, you know, the great product people usually take it's gonna take a bit of time to sort of come out of the woodworks and start imagining what the new paradigms might look like. We've seen an incredible thing, which is that the sort of single-player experiences are more powerful than ever, ever before, multiple orders of magnitude. I think now the interesting thing is how are the enduring experiences built? How do you build a product that is so different that it couldn't have been done 5 years ago because it's just a new paradigm and sort of new primitives? And I think it takes time to find and invent that.

How do you think about Dunbar's number? I remember Dave Morin had this app Path that you were supposed to add no more than 150 friends, I believe. And I'm wondering if you think that number's changing Or you think that there's some sort of like optimal ground truth to that idea of Dunbar's number? Have you reflected on that at all?

We think a lot about it, and I think there's a lot of interesting things in Dunbar's research. To me, the two things that stand out are that the headline number is growing, right? We have way more relationships than ever before. It doesn't mean that we're, you know, our brains are getting better at computing the social dynamics, but the real big problem is that we have way less close people, right? So, and I think that really is the most interesting thing to me is how do we get that back?

Yeah. What do you think, do you think there's anything to be done on the social networking side to make filter bubbles less prevalent? There was some work done, it felt like, on YouTube where you would watch one video and immediately get served another video that was similar. And so you'd wind up sort of down a rabbit hole. And then YouTube seemed to have changed the algorithm to sort of show you the opposite side of the argument. If it was a video for, in favor of something, you'd see a video that took the other side pretty quickly. But I agree with you on the general trend that like there, there is these like, you know, niches and niches and niches, and that's been good in some ways, but it's also created some isolation. Have you just reflected on like, if you had full control over all the social platforms, what you might do differently?

It's an interesting question. I think what's missing is sort of opinion, right? We went from, you know, 5 newsletter outlets, you know, maybe 12 radio stations, a bunch of TV shows, TV stations to infinite nuance. And everything, every take that you could possibly have, you can find it on Twitter. And what we want is not the average of the world. You don't actually care about the recommendation or the take of the average of the world, but you care about the take that your people care about. And so we think there's something in the middle, an in-between space that's really interesting that brings closer with actual people versus just makes you the average of the world or leaves you alone in this hyper-isolated bubble where you end up seeing things on TikTok that maybe are generated on the fly and no one else will ever see that video. And you can't go to your office anymore and be like, guys, did you see this insane thing? And talk about it.

How do you think about just app growth? There's obviously viral loops that can happen with anything that's somewhat social, even if it's a little bit tighter. There's launch videos and paid marketing. Have you explored sort of like the current landscape of what it takes to actually make it in the App Store, which feels like more competitive than ever?

I think maybe the answer is to not make it in the App Store. What does it look like to build something? I don't know how the— I mean, we have obviously strong opinions on how the world might look like. And I'm not convinced that it's going to be another app. I have this wonderful story that one of our investors, Ben Silberman, told me when I first met him about sort of his son describing his job, like Ben's job. And it was just like, oh, he was asking you know, all these squares on your phone and my dad makes one of them and I don't want to make another square. And I think we have enough squares competing for our attention. And maybe the next thing isn't another square, but something completely different. And I think that's what's exciting about this time. You can build something that just truly has never been there before and that sort of escapes any previous incumbent.

Yeah. Yeah, it's funny because like even within Pinterest, the app that Ben's Silverman founded. It has more squares within the square.

Squares all the way down.

The layout tends to be a bunch of squares.

Well, we're extremely curious to try what you're building when you're ready. I know the chat is. I think the mystery is good. Everyone's coming out, they're launching with the 3-minute launch video explaining, and you're just coming out with pure mystery.

Where can people go to— do you have a waitlist at least?

We're on teameigen.com. You can sort of visit our office.

Visit the office.

If you wanna learn more, come in person.

Yeah, we love in person. That's why we built our office as the website.

No email capture on the website.

No.

I think—

You guys are post-email, post-app. Postsoftware.

This is amazing. I love it.

Yeah, come visit us on our office website or in person in our office. We're hiring exceptional people and we're excited to build something really special.

I like this typewriter that you can play on. This is very fun. Beautiful website. I absolutely love it.

Thanks.

The bowling balls and everything.

Yeah, great website concept. Unfortunately, it will be copied relentlessly.

Yes, but.

But you did it first and that's what matters. Anyway, great to meet you, Paul. Very excited to see what you guys ship and come back on when you launch.

We'll talk to you soon. Have a good day.

Thank you.

Did you hear that Alphabet is poised for a $100 billion windfall on the SpaceX investment they made? No. Google owns around 5% of SpaceX and could get $100 billion from SpaceX's IPO if they manage to pull it off at the $2 trillion valuation. Not bad. That is a crazy stat.

They needed a windfall.

Ashay Sanghvi from Haystack says, "What are the odds that this was a result of this?" And he's showing the debate between Eric Schmidt and Peter Thiel where Peter made the argument that Google is printing too much cash and needs to innovate, needs to invest in more things.

It's funny, Dwarkesh actually asked a very similar question to NVIDIA. He was like, oh, you guys have all this extra cash. Are you gonna train a new model? Like do something internally, right? 'Cause they've said they're gonna do open source models.

Yeah, aren't they? Wasn't the answer just yes?

I mean, he didn't say like, yeah, we're going after the labs.

Yeah.

It's like very different. They're doing open source, but it feels like a very similar question still.

Yeah, yeah. They're also doing self-driving cars at NVIDIA and obviously Google's been very successful with that. It feels like the Google investments that burned money for a very long time, although they obviously stayed very profitable, a lot of those penciled out very, very well. And it does seem like the end result of this discussion was Google should invest more, and they did, and they invested in SpaceX and Waymo and a bunch of other projects. And some of them didn't pan out, but certainly many Many of them did. What else is in the timeline?

Mark is noping out. Zach says, okay, actually in the same paper published yesterday, a research group in Korea built a gene switch you can control wirelessly using EMFs.

Wow.

EMFs.

Would you look at that?

They exposed mice to 60 Hz EMF, same frequency as your wall outlet, using a pair of large coils that generate a uniform magnetic field around the animal. For cyclic 3-day or 4-day off pulses. This showed that you could activate OSK to do epigenetic reprogramming in the progeroid and aged mice, extending lifespan and reversing aging markers across multiple tissues. Conditionally switch on mutant amyloid genes only in aged mouse brains, letting them separate aging effects from amyloid to study AD biology in a way previous models couldn't. No drugs, no impacts, just a magnetic field from outside. The body, uh, I'm close to taking a victory. I'm, I'm very paranoid about, about, uh, EMFs, but, but this could be positive.

There could be, it could be a positive outcome.

Could be positive, but we have—

responsibly, we don't have— anyway, what? Oh, we have our next guests. We have James Morrissey and Kevin Hart. Welcome to the show, folks. How are you guys doing?

Gentlemen, what's going on, guys?

How are you?

Welcome to the show.

Thank you for taking the time.

Big day.

Thanks for having us.

Big day. Um, uh, would love to get— I think everyone knows Kevin Hart, but, uh, well, I think everyone knows James.

I was gonna ask the opposite. I was gonna say, yeah, it's not about me here.

I'm gonna take a second and I'm gonna highlight my guy.

Okay.

Uh, yes, you know, some may be familiar with with the, the, the idea. Yeah, Kevin Hart. But in this space, man, it's more about my business. And, you know, business is great. We have great partners. So take a second to highlight my guy, man, James. James has done a phenomenal job in self, with self, with his company, with his own entity, and the idea of finding cool ways to create and develop partnerships with faces that are and can be more than just an ambassador, but more of a brand. He's done a great job, man. He's done a really good job of vetting out the space. And I'm lucky to have found someone with the mindset, the understanding to help me do the things that I want. So execution only happens when you have people that can do so. James is that. He's an executor. So give it to him.

I'll take that. How did you two meet? What were the first meetings like? Where did the meetings take place? Like, where did all this come from?

We met about 5 years ago in the depths of COVID through mutual friends. We're in the space of building brands in joint venture with well-known individuals in entertainment, brands that they want to own, not just endorse. Yeah. And we wanted to go into the tequila space for a long time.

We like the category.

We thought it was very compelling, but The celebrity tequila space is completely oversaturated. At the time it was very noisy. We needed something really compelling to break through and to make it right for us. And when myself and Kev started talking during COVID it became very clear very quickly that we matched each other's energy.

Absolutely.

In terms of understanding the responsibilities of being not just a face for the brand, but being a business partner. Yeah. And the responsibilities of owning a brand and building one of these businesses for the long term.

But you, I think you understood early on what my wants and not wants were. Like, you know, the biggest, I'd say the most important rule that I have, I'm never slapping my name on anything, right? If I can't embrace it, or I can't do it on a daily and really have an authentic response to it and engage with it as if I engage with everything in my life, and I don't want to do it. So all of my partnerships across the board, um, whether it's ownership plays, equity plays, or, um, ambassador-like roles, I truly am embedded and invested into the thing. So with the wine and spirit space, having my own tequila, it was necessary because I said, look, I only drink tequila. I'm drinking everyone else's product. So developing my own version of a product that I can and then drink at the same level that I drink my own when I am in a space of comfort or celebration could be dope. But let's put a story behind it. Like, let's make it— let's make it different, and let's exist in some rare air. What, what can be defining or redefining in this space that basically represents me in the best way? And he was very diligent in, in answering those questions and helping me, um, navigate on that road, right? So the idea of hard work and hard work tasting different, and the idea of a celebration, um, being, being attached to Grand Cormino, because we believe that life should not only be celebrated, but what you do on a daily should be celebrated. So hard work in whatever, whatever way, however it fits to you, how do you choose to celebrate it? My job was to give you a choice.

You mentioned that the space was crowded. How did you think about finding differentiation on on the product side, the distribution side, what really stuck out as like, okay, it is a complicated space, but there's a big opportunity here?

I would say, man, you know, you have to have access. So Juan Domingo Beckmann Jr., of course, the family, what they've done in this space, you know, you're talking about generations on generations of success and growth in the business, right? From a distribution, you know, outlet opportunity, there isn't a bigger option. So I think for us, it was getting Juan to understand the real want, getting him to understand the passion behind my want, and that this is not a celebrity play. This is not a check grab and run. This is a want to build something that can literally be attached to my family name Give me an opportunity to build generational success, wealth, visibility, whatever you want to call it. But I want that. How do I get that? You have to go to the people that have it. So I think that was our biggest one. And as you talk about separating yourself, well, when you have that machine and that machine understands your real energy and those two things connect, well, you're already so much different from anything else in the space, right? Like the space is crowded because people believe that, oh, let's get a famous person and put the famous person's name on a bottle and just put the bottle on the shelf and it's yourself. That's not true at all. Like, why did you make it? What's the story behind it? What do you care about? Do you really drink this? And if so, why? And how did you develop the liquid? What is your plan for a year, 2, 3, 4, 5? Like you People really love a story, and if you have an authentic one, I find that people respond to it. So the, the years of operation and configuration as to what we want to do were the best parts of the business, because after the liquid came out, well, it wasn't a shock to us of why we were happy and why we loved it. We did the work.

Yeah, we did the work.

Yeah. Question, question from the chat. What— walk walk us through the actual product development process. I'm assuming trips to Mexico, a lot of tastings, iteration.

Yeah.

But what did that look like?

So there's a lot more that goes into this business than meets the eye for the consumers, right? So to your point earlier, the liquor business, the alcohol business is, it's a very complex industry to navigate based on the three-tier system dating back to the years of Prohibition. So there's certain criteria and ways you can navigate And ultimately to build success. Our perspective on it within consumer products and goods, and particularly within alcohol, where we saw the biggest opportunity over the last 5 years, is large corporations typically do not disrupt. They innovate, they innovate well with liquids, they innovate well with package, but generally they're not disruptors. And that's not saying the alcohol industry alone, it's every large industry. So it takes independent entrepreneurial companies to be real disruptors. Like, you guys were speaking to the guy from Grooms the other day. Yeah, you guys are speaking to, you know, John from Happy Dad. These are independent companies that are bucking the trend. And our model has been, let's be that independent-spirited business. Let's make those bold, calculated decisions in real time and be really fast and agile in terms of how we operate in the market day to day. But let's partner with the the best large corporate in the business being Proximo and Juan Domingo Beckmann from the Beckley Corporation. And let's bring some real scale to the table. So that, that, that, that muscularity of the Proximo machine and the agility of us in the market every day leading on, on what consumers see, that's been, um, it's been a very successful partnership for us and it's helped us buck that trend.

And I also, I also think just to add to that, right, like You know, you're not dealing with rocket scientists, right? Literally, like, people that are successful and people that have won, they've done it for a reason. So you're not trying to recreate the wheel or redesign the wheel. You're trying to better service the wheel. So sometimes the wheel, it falls into a space where everybody jumps on and the expectations are of norm, and they don't understand that you got to energize the wheel. You gotta, you gotta go in with some new energy, which you'll find those people will respond to that. I think our energy— yeah, yeah.

Yeah, you need—

you guys, we'll get you guys set up with a soundboard for some of your other meetings for when you're on Zoom. That just did something to me.

I was like, I took an espresso in real time.

I don't know what you said. For sure. Okay, so yeah, to reset on strategy, you have a massive platform, but you have to get the product to a place where people can buy it when you're promoting it. What was the thought between distribution, getting it in stores, and then starting to push the promotion funnel versus just telling everyone about it and they're like, yeah, I'm excited, but where do I get it? Oh, it's only available in a few stores.

Well, A, it's patience.

Okay.

And B, it's actually realizing the real work that goes with that. Like, yes, I have a large platform and a huge social media following, but that doesn't mean that as soon as I post something and say do it, that people respond. Yeah. You have to like be on the ground. Yeah. You have to do the real work. So within distribution, you gotta go and you gotta go talk and meet and shake the hands and build the relationships, right? Um, the work that you're looking for is a response of what people feel the, the, the reason for your, your implementation in this space. Like when, when the partners meet you and they say, oh, he's not here for fiction, he's here for real, this is not fake. Yeah, we will support and we will back And we will suggest this to the new customer when they walk in. Try Kevin Hart's tequila. It's really good because Kevin came in here and Kevin sat in front of us and Kevin made us understand not only why we should taste it, but why we should back and support it. Yeah, it's no different from a new artist. If a new artist is really hungry, you're showing up at every radio station and every DJ outlet and you got your you got your CD or you got a hard drive, because I want you to hear my sound, listen to it. And you're going to get way more nos than you are going to get yeses. But the breakthrough yes, when it goes on the radio waves, makes the work so much worth it. In this space, the work of getting every restaurant, every, uh, brand, every chain, every wine and spirits liquor store, independent pendant, chain, etc. Like, yeah, I went.

Yeah.

So I expect to see the results of my work. I expect to see people responding because I know what I did to get it into a space of conversation. And I think for me, that energy is an energy that I'm not going to let go of. And my partners have responded to it tremendously. That's why we sit where we sit today.

Where is the business today? How big is it? What are sales? What can you tell us about the shape of the business?

The business today, Gran Cormino is now the fastest growing celebrity tequila brand in the world.

Yeah, there we go.

And we grew last year by 100% year on year. We've done $200 million in retail sales.

Hit that gong, John.

Congratulations.

Turn into a DJ Khaled video. This is better than anything I've ever done.

Yeah.

Uh, should, uh, uh, zooming out, should every celebrity launch a product? Should every influencer have a product? What advice are you giving to other celebrities?

Absolutely not. I would say that from a celebrity perspective, most celebrities should not create businesses that they want to own. The endorsement model is a good model for most celebrities.

Yeah.

But for entertainers who truly have the understanding, the know-how, and the commitment to put in the work and prioritize said project over everything else outside of their day job, that's when it's compelling. And that was what was compelling in this partnership.

Well, I don't even like the word, the word celebrity when it's used and attached to me, because it's, it's It's, it's, it's not what we— underwhelming. Yes, it's underwhelming to, to what I really am, right? And I get it. I get what that is. Yeah, I get the star celebrity, etc. It's not why we partner with each other. But as an entrepreneur, as a, as a real business mind, as a real like worker that's not afraid to do, build, etc., um, you, you're, you're so much more, right? Like The idea of a mogul or concept of that is just a person that wants so much and is willing to do so much. And in doing so much, it means I'm not afraid to partner or align with people who have done. So in this space, slapping somebody's name on something and just thinking that it's sell— it doesn't work. It doesn't work. You're in rare air of opportunity. The celebrities that have had amazing success in the space of business brand portfolio, they do the work, right? Like, I just don't know names. You look at Kim Kardashian, I don't think Kim gets the true credit that she deserves at all times of actually doing the work. People don't understand, Kim shows up. Yeah, she doesn't just have the idea, she shows up. Like, the people know that they're gonna see her on a daily. The office full of employees know that Kim walks through the halls. They know that her office is there. And she's in meetings, she's on call, she does the work. So I myself am a do the work individual. And I think when you are, and you are committed, there is no work of loss. You're always going to win because you're doing what everyone else refuses to do. Because they don't have the patience, they don't have the, uh, strength and mental ability to stay true to something through the ups and downs and see it all the way through. It's not easy. Yeah, it's a very, very hard space to operate in. And yes, we're in an amazing era right now, and I love it. Will it stay this way? Who knows. But no matter what, if you're committed to it, whether it's up, down, whatever, you're true to the process. You know that ultimately sun is light is going to be at the end of the tunnel. That's where we are.

That was the most compelling part of this partnership, having a partner who understands, who's willing to be on the phone every day, the good days and the bad, but ultimately understanding the business behind the brand.

Yeah. So Kevin, how are you thinking about maybe not work-life balance, but work-work balance? You have multiple roles, multiple projects. If you're going into a, a movie, are you telling everyone, okay, I need to, you know, I need space, I need focus for a couple weeks, or are you trying to have, okay, I'll do something in the morning and then something in the evening? What is your workflow like on a day-to-day basis?

You're gonna be pouring shots for the whole cast and the director.

This is an example of why things need to fit your lifestyle, right? If things fit your lifestyle and fit your day-to-day, You'll find that you're never forcing and truly fighting for time. Sure, right? Everything can be done correctly. I am a product of structure and operation. I have an amazing team around me, and within that, if I'm doing a movie, I'm doing a movie. But while doing a movie, well, how do I make my partners a part of said movie? Is there spaces for me to amplify partnerships or relationships? If I'm on Sure, can I position or present certain relationships or partnerships that I have to be visible while I'm doing the things that I'm doing? Hey man, I'm golfing. What partners can I align or, or, or place in within what I'm doing on a regular? Hey, I got vacation time coming up, but when I do go on this vacation time, man, it's relaxing, but I have certain relationships and partnerships that service the idea of relaxation. And what it looks like and what it should feel like. As a partner, you are always thinking of how to service those that are aligned with you. And when you have a mind like that and you operate like that, it becomes a systematic thing. It's never a fight. So yes, I am 365. I am, uh, you know, uh, a sunup to sundown person with But because of my system, nobody gets left out. Wife, kids, partners, business, comedy, film, you know, company. Let's just say within company employee relationship, like all of these things are embedded into an idea of my day-to-day and what I have to do. So it's never left behind. It's never skipped. Or overlooked. It's implemented and rightfully so.

Yeah. Everyone's been tracking a million different changes in media and entertainment over the last decade, throughout your career. Like, what has been the biggest crucible moment? What has been the biggest trend change that you felt, "Okay, I need to adapt my strategy. I think everyone in my community needs to adapt our strategy." How have you processed the evolution of media over your career?

I'll say getting older.

Yeah. Right?

I mean, I'm about to be 47 this summer. Overnight success. Sitting and witnessing a shift in marketing and a shift in entertainment, right? This younger generation and how they're navigating and operating within social media, within storytelling, uh, within finding ways to present themselves to the masses. I mean, you know, the concept of a live streamer and, and a person that, that is literally talking to a screen all day and looking at a chat with comments but finding a way to, to build revenue to where it's coming in droves, right? Like, that to me It's exciting. Yeah, it's exciting to see a shift. So you don't fight that, you find a way to be a part of it, you find a way, um, to support it. So where I think I've had an amazing lift is in amplifying that younger generation, right? Like when I have a chance to sit beside, uh, some of these younger guys that are doing these amazing things, well, it's dope for them to have stars like myself on their platforms or supporting them in that space. I don't need nothing or want anything in return. I just want to see people win. But when you're supporting the new, it just makes you aware. Not being aware is ridiculous. Fighting what, what is, what is in real time a new, um, space of success. You're seeing ad revenue crazy spins from the biggest brands in the business and you're seeing where they're now how spending it. Television is changing. The on-demand feel and want has changed. How we watch movies and what movies have changed. Live entertainment still exists today, but it's bigger than ever because people still want to go out, but you realize they still want to be home. You gotta go and you gotta say, how do I basically fit into these pockets, and how do I deliver in a manner to where it's real and it's never forced? Well, you do that with support first and, and elevation for them, and then things come back to you. Right now it's all about me saying, I see you young guys, you young women, I love what you're doing, I want to see you win, how can I help you? And then in return, their audience gets to say Kevin is cool. But it's a, it's a them first, me second now.

We've been tracking what you guys have been doing for the last 2 years. It's incredible. And when the business outlets came up in terms of the list of targets, TVPN was on the top. We want to be in the story. We want to know what's new. We think it's bold and exciting what you're doing. Congrats on the deal with OpenAI.

Thank you.

But I mean, you guys are just getting started. But us knowing you, us knowing that we want to be on that platform, that's what it's about. And big companies don't often think that way. But we're empowered to be able to make those decisions day to day to be here with you.

It's a great point.

Really good.

Last question, guys. James, is there a competitive dynamic between some of your different partners? The chat is talking about your work with Post Malone and ASAP Rocky. I know all these guys are very competitive. Is there any kind of dynamic internally who can build the biggest brand? No, I know you don't. I know you don't want second or third, Kevin.

It's not a competition. But there is no shortcut to success. And it's hard work.

Oh, it's a— it's looking like it's a competition.

Whenever somebody says it's not a competition, you know, it's a competition.

I will tell you this. We should all look at what each other are doing. And you figure out ways to take small pieces of the recipe that, that's working, right? Like, I think ASAP and Polo have done a great job in building businesses and building brands, myself included. And when you see what's working for each other, you find ways to take pieces. Like, ultimately, I want to see everybody win. I would love to see us all win. I would love to see his company succeed and be everything that it should be and more.

But you would just hate for somebody to win more than you. I want everyone to win, just slightly less than me.

I'm winning.

I love it, guys.

We said we picked you.

Hey, if I was a betting man, I'd be betting on you. I know your work ethic is going to be insane. I know while they're sleeping, you're going to be at every restaurant, every club, getting everybody on the program. So I'm excited to follow along and thank you guys for sending. We will enjoy it this weekend.

Yeah.

The next time I'll be in the studio.

Yeah, let's do it.

I want to get all the effects in real time, man. Yeah. I need to— your energy is unbelievable, man. You guys deserve the hype.

Scoot over a little bit there. So you Yeah, yeah, there we go! That's what I'm talking about.

Thank you guys, thank you guys, man.

Uh, great, great to see you guys.

Congrats, we'll talk to you guys soon. Have a good day.

Uh, out of control.

Fantastic response from the soundboard, really working overtime. I love the soundboard.

Uh, poor Matan from Factory. That's a tough act to follow for sure.

But he'll be joining us in about 10 minutes. We can go through—

well, speaking of entertainment, Variety has a post. Val Kilmer has been— is this a movie? Is this—

Val Kilmer is an actor, John, who passed away recently. He was in Top Gun and in the new Top Gun as well, but that was his final movie.

Very sad.

Uh, has been resurrected by AI to star in the new movie As Deep The Grave. Here's a first look at the film trailer. Kilmer's digital return has the support of his daughter Mercedes, who previously told Variety he always looked at emerging technologies with optimism as a tool to expand the possibilities of storytelling. This spirit is something that we are all honoring within this specific film, of which he was an integral part. I'm curious, does this mean he was involved with it prior to his passing, or he— that she's just saying—

I mean, there's been a There's been a couple companies in Hollywood that have done facial capture, high-res digital scans, trying to capture as much performance data. But of course, there's just the raw training data of every movie and every shot that was recorded while they were on set. And a lot of those, even the outtakes are captured and saved, even if they don't make it to the final movie, the final cut of the movie. So this has always been an option. And you imagine with the advance of models, you'll be able to train a really high-quality output even just on what's, like, out there in the final cut. You might not need special capture or special data. But this is a very interesting time. I imagine there will be a ton pushback on this broadly, but if the audience opts in and the actor opts in, I imagine this will be written into a lot of wills and a lot of, you know, contracts and agreements. I imagine that all of Hollywood will be grappling with how they process this going forward. But if it makes for good entertainment, you know—

Yeah, the reason that I think it will become extremely frequent is that just the financial incentive for the family.

For sure.

For better or worse. I'm sure a lot of people will be upset about that, but—

Do we have our next guest in the waiting room already?

Before we get Matan in, we got to talk about Zach. Zach has this app, ShareAurora. It's a running-focused app. They now have a feature from what I've been seeing where people can basically go on a run It's like a livestream of your run.

Oh, interesting.

So people can come in and like heckle you. They can cheer you on. You can come on with like video or it just has you on the map. So it's like tracking your pace, your mileage, all this stuff. It's very, very cool. We gotta have Zach on to talk about it.

This used to be such a hard feature to build. And it just, I mean, of course there's services and APIs that allow for livestreaming to be bolted on pretty quickly, But remarkable that Zach was able to roll this feature out with such a small team and so early in the journey for the app. So seems fun. Feels a little hard to walk, to run with your camera there to show your face, but I guess certain runners will adapt if it's good content. Well, we have Matan from Factory. He is the co-founder and CEO.

Matan, you dog.

How you doing?

You dog. You absolute dog.

I'm doing well.

What happened?

I am doing well.

Tell us the news. How you doing?

How big is this market? Because—

This is the most important market there is.

It seems like it.

It really seems like it.

Break down the news for us first. What happened?

Hit us with the news and then we'll get into the details.

Well, first of all, so I'm reporting live from, and let's play a quick word association game. I'm gonna tell you where I'm calling in live from. You gotta give me a first word that comes to mind. Coming in live from the Rosewood in Menlo Park.

Oh, no, Jason Horowitz.

Uh, enterprise software.

Come on, enterprise software.

The Rosewood, the temple of selling enterprise software. I'm in here closing deals, but more importantly, here to share with you guys about our latest fundraise. We raised $150 million from—

there we go—

from the great folks at Coast Tesla Ventures, Sequoia Capital, Blackstone, Insight, NEA, and some other great partners. And we're very, very excited to have their support.

That's amazing. Talk about the last, I feel like it's at least 3 months, maybe a bit more since the last time you were on the show. Talk about how the space is evolving. I mean, there's just so much noise, like every single day, you know, somebody's saying, you know, this company's over, this entire paradigm is dead. You guys clearly have just kept your head down and are making a lot of progress. But yeah, walk us through kind of how the space is evolving, how your business is evolving.

Yeah, absolutely. So I guess, first of all, I think there's obviously a lot of excitement in this space because there's just so much to be done in terms of software development. There are so many things that software developers don't enjoy doing, but they unfortunately have to be doing. And I think there was kind of an early phase of excitement where everyone was just thinking about all the possibilities and kind of investing kind of the resources both from the financing side, but then also in the enterprise side in terms of making sure they stay up to date. But I think there was a bit of a lag where first, they were like, hey, here are all the things we want to do. Let's go and tell our developers to actually do it. And there's kind of a lagging period where they didn't do it. And then in the last 6 months, really everyone started adopting, uh, agentic AI. You see that in the revenue numbers of every, you know, model provider. Um, and I think now there's a bit of a, a bit of a hangover where, you know, some people are realizing that they wanted their engineers to go and adopt things and adopt they did. However, they might not be doing so in the most efficient manner. You know, there are large enterprises that we work with that found that on a monthly basis, is they spend on the order of hundreds of thousands of dollars on developers saying things like hello to Opus 4.6 fast or GPT-5.4 like ultra high.

Yeah.

Which, you know, it's probably not the best use of those tokens. And I think, you know, a tool like Factory where we can be model agnostic and dynamically route them to the appropriate model for the appropriate task.

Sure.

That has been something that's been And that's been really getting a lot of enterprise excitement.

Yeah, this token kind of token maxing trend. Meta, of course, they love to spend money on AI. They had their internal leaderboard to see who could basically produce the most tokens. There had been some chatter that people had effectively just created loops to just like go to get to the top of the chart, even though I'm not sure it's an award worth winning. Do you think that like large corporates are already like, hey, this— like, do you think this is a period where it'll be like for the first half of this year people are like, yeah, just try a bunch of stuff and then we'll see where we land? Or do you think there's already going to be more of a pullback? I know the Uber CTO had went on the record and was talking about Like, hey, we basically spent our whole inference budget in the first quarter and we got to figure out what our plan is on a go-forward basis.

Yeah, I mean, I think what we're seeing is that every company kind of has to go through these phases. We're like, phase 0 is reluctance to adopt because developers have their workflows that they've had for the last 20 years. They might not want to change it. And phase 1 is kind of throw the kitchen sink, just use as much AI as possible. Whatever you do, just change your workflows. And then I think a lot of people, and maybe more of the frontier companies like Uber, who's always kind of very ahead of the curve as it relates to these things, they're now getting into the phase 2, which is— Meta obviously is an example as well. Okay, great. People are now adopting. We're proving that they can change their behavior and use these tools. Now we need to make sure we're actually doing it efficiently and effectively. And I think that's, it's fine as like a natural process. Yes, you'll overspend in that phase 1, but the point of that is to get to the phase 2 where then you're actually really efficient on a per token basis, moving the needle for the business, delivering software faster.

What's going on international? We saw you, we saw you in SF, I believe the day of the Super Bowl, and you talked about going international. How are companies companies and developers abroad thinking about CodeGen and the category broadly?

Yeah, absolutely. I mean, I think one thing that's really important is that people build software around the world, even though the best software is probably based in the US. There's still some fantastic software out there. And so I think the way we've built Factory in particular is amenable to places like Europe or Asia or Australia, because they have different rules about where the data needs to live, where the models need to live. And Factory is one of the only solutions that is fully on-prem, fully modular. So my co-founder, eno, likes to say you could deploy Factory on a nuclear submarine as long as you have GPUs down there. And in places that tend to have a lot of regulation like Europe, that actually works quite well. And so with some of this funding, we're opening up an office in London and—

Air horn for European regulation.

There we go. Can you talk about some of the projects that you're seeing speed up on the back of AI agents? Because there's this weird dichotomy where we see huge token budgets, huge AI spend, a lot of, and then you dig in and you see people tell stories about building a lot of internal tools, a lot of new dashboards, automating workloads, workflows being more efficient, but people, it feels like they haven't really felt the external facing, I don't know, it just feels like if you, like Meta is using a bunch of AI, if you open Instagram, it's sort of the same app. It's not like, oh wow, they have like an entirely different paradigm and it feels like an entirely new app. And maybe that's just because that particular platform is mature, but where is, In terms of internal tooling, customer-facing software, entirely new ideas, automations, where are you seeing the most adoption, the most impact?

So, I love the framing of that because I think there's kind of two separate types of usage that we see. There's one that's like the fun and cool stuff of like, let me build all these new apps from scratch, which probably doesn't move the needle for the business. And then there are the left less sexy things but actually save developers a ton of time. And that's where we're seeing kind of under the hood a lot of the ROI coming from. And I think generally the name of the game there is, you know, developers are really smart. They spend years of their lives becoming experts and they get paid a lot of money. They should not be spending their time on low leverage things like documentation or testing or, you know, to spend 2 years doing legacy code migration. So generally the orgs that we see get the most ROI are the ones who basically play whack-a-mole with what is the lowest leverage thing that our developers are doing right now. Great, let's use droids to automate it. Or, you know, some language that's really emerging is let's build a software factory that goes in and automates these low leverage tasks. And I really think we take a lot of inspiration and it is even where our name comes from is what Elon and Tesla did to the physical factories. Like if you go to Tesla factories, they're mostly like, you know, machine arms going and doing things. And there's no reason that software can't be very similar. Like obviously you still need humans involved, but those humans tend to play a role of more designing that software factory, figuring out what is the most efficient way to configure that factory so that you can move the needle on your business, you know, produce more output.

How important is wide diffusion of models? We've been following the latest model from Anthropic, Mythos. That's only available to a few companies. Is the strategy to focus on being cross-platform or get access to that model earlier and then act as a diffusion layer for that? How are you thinking about the future where more advanced models are sort of gated?

Yeah, I think for us the biggest thing is enterprises need to be model agnostic. It's just, it is a non-negotiable that they cannot standardize on just one provider for a number of reasons. One might be, you know, if that API endpoint goes down, which, you know, some models these days haven't been the most reliable. And if you're a highly regulated industry and you get your developers to become agent native and, you know, start delegating tasks asks to these agents and that endpoint is down, what are you going to do? Like, it's just, it's non-negotiable. And so being model agnostic for every serious enterprise is kind of table stakes. And for us, what we can do is make sure that we get those models, you know, day zero that they're released just so that they can stay at the frontier. Same with the open models because also different models end up being good at different tasks. They're better at different languages, perhaps. And making sure that we kind of give the optionality to the enterprises to adjust accordingly is pretty important.

How are you thinking about the forward deployed model? Is that more important now? I know that you have a very small team. It's what, 70 employees? Probably growing very fast, but how much about enterprise adoption is actually spending time on site with the customer answering questions, integrating deeply into these like large enterprises?

Yeah, the way we think about the deployed engineering is they should never be doing the same thing more than once. Because with the tool that we've built, if they've learned, hey, here's something that enterprises care about, we should be able to build that into the product very quickly. Like our core competency is not providing services, but it's building product. And we treat our deploy team kind of as like the tip of the spear where they're figuring out live, you know, shoulder to shoulder with the enterprise engineers, what are the things that if we put into our product would allow them to become agent native faster?

There's some news from Cursor today that they're teaming up with xAI to potentially train the next version of Composer. Are you thinking about training your own model at some point? Is that interesting to you?

I think it's interesting at some point. I think it doesn't make sense right now because I don't think there is a— I don't think right now enterprises need another open model that you fine-tune.

Sure.

I think there's sufficiently good ones out there. And I think most of the alpha is actually on the research side as it relates to the agent itself. So for example, Droid, which is our agent, ends up— it's model agnostic, but it also outperforms all of the agents that are coming out of the model labs. And so as an agent lab, most of the alpha on every incremental hour of our time is from the agent itself. And then we'll probably get to a point, you know, where eventually maybe there's just— there's sufficient alpha, uh, in, in, you know, RLing a model for ourselves. But right now, every incremental hour on the agent ends up being very, very high ROI.

So yeah, Jordy, anything else?

No, great update.

Congratulations.

Great progress.

Congrats to the team. Thanks for having me, guys.

Cheers.

Thanks for hopping on. Goodbye. And up next, we have Akhil from Ulysses. He's the co-founder and CEO. How you doing?

Hey, fellas.

Good to be here.

Look at that. Look at that beautiful drone.

Massive. Day today. Kick us off with the announcement. What happened today?

Well, you know, I've got, uh, you know, my, my friend, my Mako here, one of our, uh, pet robotic sharks. And, uh, we're going to be building a hell of a lot more of them. We've just, uh, announced our Series A led by Andreessen Horowitz, American Dynamism Fund, at $38 million.

Congratulations!

Very, very, very, very, very Very cool. And yeah, why don't you give, yeah, what have the last like 6 months been like? Will's been on the show before. This is your first time, but kind of walk us through what you guys have been focused on, what the opportunities are and what the future looks like.

Yeah, for sure. So, you know, as you guys might already know, you know, we're building autonomous maritime robotics to solve the most critical challenges in what is arguably the most critical domain on the planet. You know, we're talking commercial applications ranging from just monitoring and maintaining infrastructure for offshore energy, oil and gas, renewables, telecoms, you know, maintaining and securing the critical communications infrastructure, also shipping, terminals, ports, that kind of thing as well. And then as we recently started doing a lot of work in the defense space as well, taking our commercially available and deployed robots like this Mako behind me that's been on probably a dozen missions for various commercial customers till date and taking the exact same technology and helping to fill capability gaps for allied forces, you know, You look at shipbuilding where China is ahead of the US by, you know, 200 times or something like that. Unmanned underwater vehicles, the gap isn't that big. You know, it's a gap that can be easily closed. We're the ones closing it. And it's the underwater domain where most of the work happens. It's where all the infrastructure is. It's where all of the communications is. And it's a place where, you know, we have a chance to deliver and asymmetric advantage. And over the last 6 months, we've been hard at work taking our commercial tech and bringing it in front of allied forces to get it in the water with them.

So talk about the commercial applications. If I have a, I don't know, oil rig, and I want to inspect and make sure that the barnacles aren't getting out of control or something, I drop this in the water. Am I piloting it remotely? Does it have an autonomous sort of path that it can swim around on or, you know, drive around on and collect data imagery? What are the sensors stack like? How am I actually getting value from this on day one?

So that's the beauty of the product. It's modular, which means that you can put, you know, it's made up of all these sections. You can change the thruster configuration. You can put different sensors on, you can take other sensors off. So It depends on what you want, you know, are you trying to do a magnetic analysis of your structure? We have a magnetic, a magnetometer payload. Are you trying to just see what's there? We've got cameras like we have right here. Are you trying to see through really dark and murky water, really wide areas? Then we have a variety of acoustic mapping sonar payloads and things like that. And so then in terms of, you know, Can it work autonomously? Can you remote control it? Well, you can do both. That's the fun of it. You could have it just drop it in the water with a pre-programmed path and be like, hey, go and search this wide area or scan all these pylons supporting my rig. Come back, pull the data off and have a look, and then you might see something interesting. You might see something, I want to take a closer look. But then you can connect a cable, livestream the data, and drive it down yourself. And then you can look at it. And we support robotic payloads as well. So then you can start interacting with it. You might want to scrape something off, you might want to cut something that's entangled, you might want to place something. We have payloads that do all of that.

Yeah. How are you thinking about range and battery life and just all the different trade-offs that go into actually getting something that can have an impact for meaningful amount of time or across a meaningful amount of space. It doesn't look like the biggest ship. How far can it go? How deep can it go? What are the options here?

Yeah, this is one of our more compact vehicles. This one's about 60, 65 inches in length, but we can go in excess of 140 inches because you can stack batteries. You know, this one here in this configuration, it's not the most hydrodynamic configuration, which means it's not the most efficient. So it's only going to go maybe 20 nautical miles with one battery box. But if you want it to go 60 nautical miles, put 3 battery boxes on. Or if you take the thruster pods off and you just have this big tube, that's the most efficient. And then you can actually go, you know, 4 battery boxes and land because you take the pods out, that gives you space for another battery box, then you can go up to 250 nautical miles.

It's going slower but farther, if that makes sense.

Yeah, yeah, and it's also a more efficient shape.

Yep, more, yeah, more efficient. I saw that there's two cameras on the front. Is that for stereoscopic vision, or is one of them telephoto, one of them wide angle? Why two lenses on the front?

In this case, it doesn't really show up, but they are actually two separate lenses, two different sensors as well.

Oh, sure.

Just different use cases. But they're modules. You can see this port is actually blank here, but we actually make our own cameras as well, and you can plug them in with different lenses. You can have a stereo configuration. We've actually got ones on top here as well because this one, the idea is you can have it swim underneath structures and you can be looking up and doing inspection. Up above. So you're making your own cameras.

How vertically integrated are you? This feels like with defense applications, it's really important to be onshore, but I would imagine trying to take advantage of as much of the supply chain as you can. What's been the build versus buy strategy?

Well, we found, you know, coming into this, there are a lot of things we honestly wanted to buy and we looked into buying, but we found the The idiot index in maritime is insane.

Oh, yeah. What is the idiot index again? Can you break that metric down?

Yeah, that's the metric that Musk came up with where what's the ratio of the cost of the raw materials in a part—

That's right.

—versus what you're actually paying for it. Yeah. And there's been sensors and motors and other parts that we get in, we take them apart. We're like, why are we paying $150,000 for this thing? It's got like microcontroller in here worth $10, some networking and power switches. And it's not even like, okay, this is really new, really advanced technology and they need to amortize the development costs. It's like we're buying a sensor that's 20 years old. Why are we still paying this much for it? And the thing is the tech and the knowledge and the fundamentals are being publicly available for a long, long time for a lot of the stuff that we're bringing in-house. So, you know, for us it's like, if you look at the maritime industry, it's been the same for centuries. You want to do something big and important, you get a big ship, lots of expensive equipment, and you're really CapEx and OpEx heavy, right? But that's not the case anymore in space, in the air, on land, you know. Um, we want to bring that moment to to the, to the ocean as well. And what that's necessitated is a ton of vertical integration. All the metal that you see on this, all our pressure vessels, they're machined in the machines just over there. Wow. All the internal brackets and mounts, everything also machined on the same machine. The plastic fairings and propellers and everything 3D printed on the 3D printers over there. All the internal printed circuit boards with all the electronics. We don't make the boards ourselves yet, but we do have the ability to actually assemble the boards. We get the raw components in, place them on the boards, and we've got an automated line doing that. That's massively brought our costs down, made it much easier for us to have a product that's fundamentally very easy to scale, but also massively saved us tons of money in terms of development time. And iteration speed. You know, there's things we've done that we did in 3 days that before we brought machining in has probably would have taken us 3 months.

Well, that's very exciting. Thank you so much. Incredible progress and breaking it down for us.

Have a great rest of your day. Congrats to the whole team.

Yeah, you too. Thanks for having me on.

Goodbye. Our next guest is Charlie Cheever, the co-founder of Expo. Previously, he co-founded Quora where he served as CTO and helped scale the platform. Platform to millions of users. How are you doing? Hey, I'm great. How are you guys? I'm good. Welcome to the show. I think most people are familiar with your background, but would love to get the introduction on Expo and go through the news today.

Yeah, we just raised a $45 million Series B from Georgian.

And yeah, what's the shape of the company now? What's the plan?

Yeah, we're about 62 people right now and we're all over the world. We've got great people everywhere. And the plan is just that we've got a ton of stuff to do and this will just let us do it. What's basically happened over the last year or so is that Expo's sort of become the best way to make apps with AI. And so, it's just gotten really popular. And so, that's meant that there's just a ton of stuff to do. And so, we're working on it.

[Speaker:JASON] Yeah, I mean, coding models are clearly aware of React Native and Objective-C and Swift to some degree, but there's always been a disconnect between Xcode or, you know, actually delivering the product. And it feels like when you click on some Vibe-coded app, it's almost always in the web. And, you know, this— it feels like there's going to be a boom. We're already seeing some stats on the App Store. But walk me through what it takes to actually deliver value in app creation that's differentiated.

Yeah. So what we've been— our basic philosophy for the last couple of years has since the beginning has been, you know, that we want to take everything that's good about web development and bring that to mobile so that it just is easy and everything that's just easy and fast about building and distributing on the web, we bring that to mobile. And so a year and a half ago, I would have said, you know, what's great about Expo is that you can take your, you know, React web developer team and you can point them at mobile and they can build a great mobile app that feels really, really good. Sure. And today the pitch now becomes like, these AI models are so well trained on React and JavaScript, TypeScript, etc., that they're really, really good at this for this. But the other really important thing there is Expo isn't just a way to build mobile apps with JavaScript. We saw that attempted in the past and it just didn't work out that well because people would use HTML5 as a content delivery mechanism and it just wouldn't feel right.

It just wouldn't be good enough. PhoneGap was one of the popular ones at the time.

Yeah, stuff like that. So what's different here is just that Expo is really a way to make apps with TypeScript and React Native and stuff like that, but also lets you drop down really easily to Swift and Kotlin and stuff like that. So wherever you need to, you can go make that polish. In the same way that a developer making server-side software might write a codebase in a mix of Python and Rust or something like that, you can move in between them seamlessly. That's a huge deal and means that The people who are making apps with Expo are making sort of like stuff that's, you know, top tier, top of the App Store charts, that kind of stuff, because you can hit that quality bar that people expect on mobile these days.

What's your view on how, like, maybe I won't ask you that, but I would say like, what do you think Apple, how do you think Apple should be approaching— The Web Coding Boom? —this explosion of new software applications? Applications, like they are— there's been so much frustration with them. They've been under pressure because of billing and payment policies. But they certainly, specifically just looking at like if you just— if you 10x or 100x or maybe someday 1,000x the amount of software, that is a real challenge for them. And so they're not going to get everything perfect, but I'm curious like what advice you'd give them.

One place where I think they've done a nice job is on the Macintosh, where— or Mac, I guess they call it these days— where if you want to get software on the Mac, you can go to the App Store on the Mac, or you can get it from the web, or you can just— a lot of places, like if I install Slack or Notion, I typically go to their website and download an installer, and it works. And they've done a really nice job of tightening up the security model around the kernel and the operating system and things like that. And there's not really any significant security problems I'm aware of on the Mac. And I think that's I think that should be a really good model. And also, you know, nobody's on the developer side. I don't know of a lot of developers who are super frustrated with distributing software on the Mac either. So I think like, from my perspective, I feel like they've actually figured out a pretty good model and they just should apply that to iOS. And, you know, it's true. Like, you know, I was reading an article the other day that I think App Store submissions are up sort of 84% this quarter. Obviously, a lot of that is new Expo apps, a lot of those are vibe-coded. There's a lot of new app building tools out there. A lot of those are built with Expo, most of them. So yeah, a lot of people are frustrated that they're getting rejected for strange reasons or they don't know what's going on or why. But I think overall this will get sorted out. It's too possible and there's too much demand for building new software and building customized software and building software with bigger surface areas. Like, it's just— it's clearly going to happen. I don't know. I feel like sort of like when Uber launched, there was a bunch of regulatory things and policy things to find out, but like, it was just such a good idea and just needed to happen. And so like, eventually the dam kind of— I don't know if it burst, but it like— things got sorted out and people take Ubers now. And I think people will be able to— I think people will be able to make software and distribute it in the future.

Yeah. What percentage of your growth do you think is like agent-led, as in, you know, various coding agents effectively choosing Expo or helping guide the team or entrepreneur to using the product?

Yeah, I think it's still word of mouth. Like, when people make a decision big enough to sort of pick your stack that you're gonna build your mobile app on, that you're probably gonna be stuck with for a bunch of years, And for a lot of companies, it decides their business. They're not going to use one thing to decide that. So I think that, like, it's a combination of word of mouth, which is the strongest thing. Like, the story that we hear over and over again is sort of like, oh, you know, our company needed to build a new mobile app because we had this older thing that just wasn't cutting it anymore. Someone on our team came in on a Monday morning with a prototype that already had 6 screens built and felt really great, and the CEO was really impressed. And so we just kept going with that. But a lot of times that person found out about it because an AI told them about Expo, or they just asked an AI, hey, how can I build a mobile app prototype really quickly? And it directs them to it. Or they also find out about, like, YouTube is a huge referrer for us. Like, there's a lot of great content out there. Yeah.

Is this, is the story about Facebook writing a compiler from PHP to C++ apocryphal, or is that roughly correct?

I think that— I believe that's correct. This guy Haiping Zhao made something called like the HPHP compiler. And, um, they're actually kind of doing some— like, one thing that's really relevant to Expo is for React Native, they've, they built this, uh, custom JavaScript virtual machine called Hermes. And they're actually working on something called Static Hermes, which will come ahead of time, compile TypeScript and JavaScript to native code. And so you'll be able to take, you know, any React Native app, assuming this project completes, which it hasn't yet, but hopefully will soon, you'll be able to take any part of your JavaScript-driven app and convert it into native code seamlessly and have that, if you want to speed up that part at the cost of a more brittle, bigger binary, you'll be able to choose that.

So, that's kind of the trajectory we're pointing at. Yeah, that's exactly where we're going. We're moving up these levels of abstraction. You say that You think people will be able to build apps and develop software in the future. Do you have a view on sort of like real-time UI, instantiating things at runtime, and whether or not Apple embraces that? It feels like that's something that could be coming with artifacts that are generated on the fly based on a query. Doesn't make sense for every app, but does that seem like something that might happen in the future?

Basically every big app that you know, like, does something like this. Like, even if you just think about something, you know, that you think of as kind of basic, like Yelp. Like, at first it's like, oh, like, this is a really structured— every restaurant page, like, this looks the same. But then they start to add more features, and there's like, oh, well, this restaurant wants to post its hours in this way, or there's a chain here, so we need to link to other locations. And they start to, you know, make some sort of like JSON format that describes the complexity of the UI and then they keep extending and extending and sending it. There's an old saying that every sufficiently complicated C++ program contains a poorly implemented half-baked version of Lisp in it or something like that. I think every mobile app that gets sufficiently big basically ends up with some custom dynamic on-the-fly rendering system as their content becomes richer and richer. And I think what's nice about using JavaScript and React and React Native, it's a fully featured, real programming language, takes care of all the problems, makes sure things are correctly updated, et cetera, et cetera. And why not use the real thing once you're going in that direction? [Speaker] Yeah.

What does it take to get a job with you these days if you're a software engineer? There's a lot of people that are stopping they're not studying computer science anymore. There's a bit of a reckoning. There's a question about where young people should go. How do you counsel someone who's in college right now and interested in technology?

I personally, there's 3 things that I look for. I would say like taste is probably the most important thing. I mean, and like I've said this for years, but now with like AI, this becomes even more important where it's just like, we still need somebody to say, like, this is good, and like, this is what we should be doing, and what, you know, like, what you're actually trying to build. And that's really hard to replace. It's, you know, it's rare to have someone who's really, really good taste. And so someone who just has good judgment, that's incredibly important. Then the second thing is, like, we call it like high APM. APM is like a term from like real-time strategy video games, where it's basically like, um, if you ever played like, you know, StarCraft back in the day or something like that, part of it is a strategy game, but part of it is just like, if you just do more stuff faster than other people, you're gonna beat them. And so like, we look for people that just like do a bunch of stuff really quickly, but like with a precision, and you know, that you're not, you're not just doing like totally random stuff, but you're doing like effective stuff. And then the third thing we talk about is just high agency where, like, you know, you can see right now that, like, writing code has become not a big problem for a lot of companies. Yeah. But shipping good software is still a big problem for a lot of companies. Yeah. And that's because the problem of, like, actually making stuff and getting over the finish line, out the door, into your users, and, you know, properly messaged and all these other things and supported and, you know, iterated on to match what your customers actually need, etc., etc., etc. Those are all pretty hard, and a lot of times they involve doing annoying stuff for solving problems nobody ever wrote a manual on how to solve before. And just so people are just like, oh, I'm going to get this done and I'm just going to figure it out and I'm just going to run through walls to figure that out. That's always really important, I think, whether it be a coding engineering context or any other role at the company. I think that's super important.

Yeah. It's great. I like it a lot. Jordy, anything else? No, this was great. Yeah, thank you so much. And congratulations on the round and the progress. Very excited. Have a great day. Thanks, good to meet you. We'll talk to you soon. Our next guest is Victor from Slash. I take it back.

I did have one more question, but it's too late. The question was gonna be around payments. Oh yeah. Like, would they— this feels like something they would build in, helping you bring your product to the App Store. App Store now has more flexibility around payments. But I'll save that one— For the next time. —for the next appearance.

Well, let's bring in Victor from Slash. He's the co-founder and CEO. Victor, how are you doing?

What's going on? I'm doing great.

Thanks for having me on.

Long-time watcher of the show, first time being here, so—

I know. Long, long overdue. Long overdue. You guys have been probably 100x'd since we started the show. Yeah, um, so better, better late than ever.

But for those who don't know, introduce yourself and the company first, and then we'll get into questions.

Of course. Um, I'm the founder and CEO of Slash. Slash is one of the fastest, if not the fastest growing business banking platforms in America. Over 5,000 businesses spend nearly $10 billion a year on our corporate cards. And the reason all these business owners are coming to Slash is quite simple. We live in an age where you can talk to machines, where cars can drive themselves, but the vast majority of entrepreneurs in America are still banking with an institution that has an interface that was last updated in the year 2003. So, core to the Slash thesis is this idea that your bank can do so much more for you than just let you hold and send your money. It can be the place where you invoice your customers, where you reimburse your employees for out-of-pocket expenses, where you figure out how to just run a much more efficient operation. And we have ambitions to be this generation's JPMorgan, and we're going to do it by building the most powerful platform imaginable. And John, I'm not sure you remember, but we met one time. Oh, really? Very cool. Yeah, uh, awesome. Like around Napa or somewhere like that.

Very cool. Um, so yeah, what is the go-to-market strategy? Obviously it's a crowded space. Do you want to go bottoms up, top down, startups, enterprises, somewhere in between, SMBs, uh, Silicon Valley darlings, or Main Street everyday companies? Totally.

I would say a very underappreciated fact is that fintech is very underpenetrated in the US. The vast majority of businesses in the US still bank and spend with a legacy financial institution. Less than 5% of American businesses work with a fintech company like Slash or others. And so there's a lot of market up for grabs. The crux of the Slash thesis is that business owners deserve serve banking products hyper-tailored to the needs of their specific industries. So I would say we work with companies that have, you know, a very small amount of employees. We work with very large companies, but usually it's very sector-focused. So we figure out how do we deliver a ton of value for a business in one segment, and we really tune out the rest of the markets.

And in terms of actually getting those customers on the platform, is it some sort of like magical demo that is AI-enhanced Is it wow them, or is it just gumshoe, getting to know folks and reassuring them that this will be just a reliable banking platform for them?

Yeah, I mean, banking is a very relationships-driven industry. All else being equal, people prefer to bank and spend with an institution that they have a great relationship with. And so the crux of our go-to-market is depends on building amazing referral flywheels. So part of this vertical-by-vertical strategy exists because we have a phenomenal product that works on a vertical-by-vertical basis, but also because once we penetrate and acquire a few customers in one particular sector, willingness to refer is quite high because we deliver very strongly on customer support and on just going above and beyond. Yeah.

How are you thinking about integrating AI tools, AI agents, AI chief of staff, anything that can help a business owner understand their spend, understand their business, actually manage their finances better? Absolutely.

The future of software, generally speaking, is moving away from click-and-drag UI and towards just textual interfaces, natural language. Right now, today, we actually announced our AI agent. It's called Twin, and it allows you to take every single action you you can take from within the Slash dashboard, but in natural language. So that's where everything's going. That's where all software is going. And we decided to practically take that step today. That's number one. And then number two.

That's the most Gen Z name for an agent ever. Twin. I love it. Twin. That's my twin. Twin. That's my twin.

That's my twin. It's my twin.

It's an underutilized term.

It's super underutilized. Yeah.

And the really cool thing— I didn't know digital twins were a thing.

Yeah. Yeah.

Is it can buy things on your behalf because we're the issuer. Of the card you use to place all your expenses, your employees can actually go out and tell, hey, twin, you know, make this DoorDash order for me, place this order on Instacart. And I think that's a super, super powerful thing, an agent that can actually go out and make purchases on the internet on your behalf.

You just raised a bunch of money. What is VC sentiment around fintech right now? I know you guys have stablecoin functionality. I'm sure that that was interesting. And at this scale, like, you have I know you have a bunch of real, really, really positive metrics that you can lean on, but I'm curious what overall sentiment was from the investor base.

I think there were two things that investors found quite compelling about Slash. The first is that we're a very revenue-efficient company, that our business is approaching $300 million in annual revenue. We have a team of just around 70 people, so our business does over $4 million in revenue per employee. And the reason? Thank you. We've been able to do that is because so many of these processes that legacy banks and the fintechs that came before us throw a lot of headcount at processing disputes, parsing documents when somebody applies for an application, submitting a SAR filing to regulators, we've agents automated away. So Slash's leanness and this differentiation of we're building this backend AI operating system was super attractive to investors. And then the other thing that's super interesting to investors is we're living in a super interesting time in fintech history where for the first time ever, an American fintech company can serve businesses all over the world. It used to be the case that if you were a fintech company and you wanted to expand internationally, you had to get licensed on a jurisdiction-by-jurisdiction basis. But now stablecoins are this enabling force that allows you to deliver USD banking products to businesses all over the world. So a lot of this funding is going to go towards making Slash this global product and bringing the caliber of product we've given to American entrepreneurs, but to business owners all over the world as well.

Amazing. Well, congrats on all the progress.

And yeah, great to finally meet.

Great to have you on the show. Yeah, super impressive. We'll talk to you soon. Cheers, Victor. Have a good rest of your day. Powerful nominative determinism. Victor. Yes, he will be the victor. I like that. Well, our next guest My next guest is Theodore Marcoux from Cognition. He's the head of product growth.

Theodore, how you doing? Hey guys, how's it going?

It's good. Good to see you. It's great. Obviously everyone here on the show is familiar with Cognition, the makers of Devin, but take us through sort of how you're positioning the company, the updates, and any big announcements we should be aware of.

Of course. Yeah. Thank you again for having me. Of course. This is a very exciting day. So yesterday we had a huge, huge moment for the company that I think a lot of us are very excited about. We had the biggest launch since the acquisition of Windsurf last July. This is something that the team has been looking forward to a lot. We launched Windsurf 2.0, which did two big things. It brought Devin to Windsurf. So finally, the world's software engineering agent is available in Windsurf. And we have now an agent command center. So our sort of vision for the future of software engineering, which is managing a team of agents, both remote and local, that works sort of alongside you, you have an army at your back. Windsurf 2.0 makes that easier than ever.

Got it. So how, what goes into an agent command center to make it effective? We were talking about Gastown, and having different agents for different tasks. I think the buzzword is like orchestrators. There's a various amount of open source projects and sort of the idea of orchestration is percolating in the AI industry. How do you think about educating customers and enterprises about why they should be using an interface to manage agents instead of just having a bunch of different terminals open?

Yeah, so first of all, this is something that I think frankly arose as a need internally at Cognition. What we're seeing is some of the best engineers here are— and we think this is sort of where the future is going— the best engineers are using local agents. Yeah. And they are planning their tasks. They're going very deep into the codebase, architecting systems, creating, coming up with a plan, and then And they're handing those off to cloud agents. And when we launched Cognition, Scott and Walden and Stephen and a few other folks had this sort of, they saw into the future and they saw that cloud agents were going to be the future. But we also saw that engineers are sort of around and they are working very closely with their code and they want to make, there's sort of like golden age of engineering where We can go very deep into what you're excited about and then hand out after cloud agents. So with Windsurf 2.0, the bottom line is that, and what we're telling a lot of our customers with the command center, is that you get to have this overview of your agents so that your limits on your attentions are no longer there. Because as you're working with dozens of agents at a time, and you switch context, it gets really hard. So we've built, we actually were very intentional and we built this Kanban view. Sort of anyone that's familiar with project management can, you guys seen a Kanban view before? Where you can see all your agents working on different projects at a time and then you can sort of like switch in between them and quickly check on them and then spawn or create new sessions when you need to and then those Those sessions are actually grouped into spaces, which is sort of this way that we built to make it easier for agents to share context and also share sort of states together. So from our perspective, what goes into a great command center is just making it very easy for a single engineer to work with a team of agents and have sort of like this, like, army at their back.

For big enterprises, they have so much different development work to do. What is the most low-hanging fruit? If you're talking to a customer who wants to get spun up with Windserve and Cognition and Devon, are you trying to understand their backlog, the hairiest tasks, the most miserable? Are you trying to get them excited about greenfield projects and new dashboards and automated workflows? Are you trying to get in the hands of their best engineers, their youngest engineers? What is the greatest foothold for you right now?

Yeah, that's a great question. I think that is something that has been evolving ever since the beginning of Cognition. I think early on with Devin, what we found was that we would go in and talk about all these sort of very specific use cases. Cases that we'd find inside companies, whether it's migrations, whether it's building internal tooling, whether it's these sort of like big backlog projects that they've been wanting to work on. I think more recently, as models have been getting better and as our agent harness has been getting better and better, what we're finding is that there's a lot of— frankly, everyone can use a software engineering agent and the universe of possibilities has expanded greatly. And I was talking about how you can start working locally and there's this sort of like gold. The way I think about it is there's this golden age of engineering where like the best engineers in the world can do more and they can offload their tasks and not just the best engineers, frankly everyone can offload the tasks that they don't want to spend as much time on to agents in the cloud that can sort of like handle them sort of very quickly or over many minutes or long hours. And then they can sort of stay in control and work on the things that they care most about and the hardest problems that are most exciting to them. So Scott always uses this sort of like this idea of you're going from being a bricklayer to an architect. And I think a lot of what we're seeing in a lot of our customers is that entire teams and individuals on those teams teams are moving from being bricklayers or sort of like writers to architects or directors where they're sort of like managing an orchestra of agents. Sure. Yeah, that makes sense.

Jordy was asking this question earlier about just model agnosticism, how valuable— like, do you see demand from customers and companies for wanting to switch between model providers? Use open source for things? Like, how much is using the right tool for the job, understanding the Pareto frontier, and not blowing up your budget in one month if you're token maxing?

Yeah, that's something that comes up all the time. And as you know, Cognition has been model agnostic from day zero. We've always worked with the best models and all of the models that are available, evaluated them, internally, figured out how to make the best agent harness for Devin, how to make the best agent harness for Windsurf as well. And what we're seeing from our customers is that there's a lot of demand from trying out different models for different tasks. We're constantly working with them, actually, to make sure that we can advise them on what are the things that— models have this sort of jagged line of intelligence where some of them are really good at specific things, others are less good at those things. For example, Opus 4.7 came out today, and congratulations to Anthropic on a great model launch. That model is very good at deep investigations. So we're sort of trying to look at how can we use that in our, for example, dev and review workflows where dev can go in and look at a PR and try to figure out all the bugs and all the issues that might be associated with it. Yeah, that makes sense.

Well, congratulations on the progress. Thank you so much.

Great to meet you and give our best to the team. Yeah, we'll talk to you soon.

Thank you guys. Have a good rest of your day. There's a ton of breaking news. The big one is Reed Hastings.

Reed Hastings is stepping down. Stepping off the board of Netflix and the stock is down tremendously. This is good. He's not stepping off the board.

He's stepping off the board in June. He announced that he's stepping off the board.

But yes, I mean, it's insane. Why is this good? It's good because— it's good for Reed specifically. Oh, yeah. Because it shows that people have confidence in his leadership and his vision. Oh, sure, sure. And I'm sure that— I would expect Netflix to make a quick recovery.

Yeah, it's up 12%, 13% in the last month, down 8.5% overnight, after hours. But we'll see where the stock settles tomorrow after the market processes this.

The alternative is a nightmare for Reed, because if he announced this and the stock popped 20%, he was handicapping the company. Yeah.

And I mean, the flip side is that Ted Sarandos, it seems like he put on a masterclass over the last 6 months with the Paramount negotiations, not getting over his skis. The shareholders wound up really liking how that all penciled out. And so it seems like the company's in good hands and all of the different strengths that Netflix has continue to show across advertising and subscriptions. And while they've kept their The big headline with Netflix is that they've kept their content budget essentially flat or slightly growing while they've grown subscriptions and revenue and top line very precipitously and very consistently even at a time where they haven't needed to invest exponentially more money in content. Obviously, they spend a fortune on it, but it's not growing as fast as their revenue is growing, so their profits are growing, which which is good news. Hastings' departure marks the end of an era for Netflix, which under his leadership transformed from a DVD-by-mail business to a juggernaut in subscription video streaming and disrupted Hollywood. He said, "My real contribution at Netflix wasn't a single decision," Hastings said in a statement that was in a company letter to shareholders. "It was a focus on member joy, building a culture that others inherit and improve, and building a company that could be both beloved by members and wildly successful for generations to come. Well, we wish him the best on his next chapter, whatever he winds up doing. What an absolute run. Well, Jordy, is there anything else we should talk about? TSMC, of course, has earnings. We can cover those later. The chipmaker TSMC is more bullish than ever on AI despite the Iran war. So some good news there. And lots more stories to talk about. But we will be back with you on Monday at 11 a.m.

It's been an honor and a privilege. And we'll see you here today. Leave us 5 stars on Apple Podcasts. Best of your day.

Sign up for our newsletter at tbpn.com. Goodbye. Rolling flashbang. Rolling flashbang. Goodbye.