Some Math Problems, Arsenal-2, Porsche Mission S | Walter Russell Mead, Kat Cole, Tarek Alaruri, Tyler Niday, Scott Morton, Josiah Meyer, Aarush Selvan
00:00 You're watching TV. Today is Wednesday, October 7, 2026. It's still watching TV on the Temple Technology, the fortunes of finance. The capital of capital. I'll tell you about ramp.com. Time is money. Save both easy-use corporate cards, bill pay accounting, and a whole lot more, including Tony Hawk now.
00:20 They got a deal with Tony Hawk, and I saw Bamargera in the promotional video. Go check it out. It's on Eric Leimann's X account. They did it again. It's all over the ramp socials. did it again. No, fantastic, fantastic. Chat saying we're late today.
00:36 We are late today. Here we are. We were working on stream title, working on the plan for the show. But we have a great show for you because we planned. First off, I got a little movie review for you. I saw a movie yesterday. I went to the theater and I saw Primetime. And I wanted to, yeah.
00:56 Oh, okay. I want a review from those on the production team who have seen prime time, the new Robert Pattinson film, but I thought I would try and take you on sort of a film critique, a journey with a review that sort of channels the way Tyler reviews movies. So here's my review of prime time. It was good. The audio was good and the video was good. Act one was good. Act two was good. And the acting. Three was good. The acting was good. And the lighting and the cinematography was good. And the set design was good. So overall, I feel like I have a really... I would say it's a good movie. Strong review. Pretty good. That was a good review.
01:49 Pretty good. What do you guys think? You liked it? Let's get Scott up here. Saw that movie. Yeah? You liked it? I loved it. How cool was the frame? It felt like I was watching IMAX format, but it was just four three. They used 23 different cameras on that movie.
02:06 The cinematography was crazy, right? And the highlights blown out, and there was so much grain. Like, it was... It was like way too stylized for a Chris Hansen movie. I had no idea where any of this would, like, where these ideas came from. It feels like they were just pulling from every different camera from that era. It was really, really cool. Well, Lance Oppenheim.
02:29 There we go. Lance Oppenheim, he does documentaries. Have you seen Renfair? Oh, yeah, yeah, yeah. That's right. That's right. Okay, cool. Yeah.
02:36 But, yeah, really, really fun, you know, creative way to throw back to, like, the way people watch TV at that time, four by three. I like that Chris Hansen was kind of like the antagonist the entire time. Yeah, it's a complex investigation of his career. and the way he's remembered, did he go too far? And it feels like a little bit heavy-handed. Like it's pretty clear the vision. It doesn't fully leave you open to make your own judgment.
03:06 But you, of course, can, which I thought was a nice touch. And Robert Pattinson is just in fighting form. Yeah, he's in a flow state. He's just in flow state. Yeah, he's really good. Anyway, we clearly were teasing it, but of course we have the TBPN Road to Christmas Countdown ready for you. Let's let you know. Let's remind everyone there are 78 days until Christmas. Get ready. Get ready. If you missed it yesterday, we did an interview, a deep dive into the Christmas
03:40 Tree Farmers Association of America, understanding everything from how the Christmas tree industry is structured to techniques for keeping your Christmas tree alive. And if you watch it, Any tech or business content this week, it's got to be that. It's got to be that. It's got to be that interview. We highly recommend that you donate. I made a donation yesterday. TreesforTrups.org.
04:06 Give whatever you can. They also do company matching. So if you work at a company that matches donations, if you're an early anthropic employee and they match three to one, back the truck up. Throw down 100 mil. Come on, let's do it. Let's get every troop must have the biggest tree possible, a 40-footer.
04:29 You know, I want a White House-style tree in every troop's home. That would be amazing. No, but he had some harsh words for the plastic tree industry. He called them toilet brushes, which was hilarious. I'd never heard that slur for those plastic trees before, but he had it ready to go, which I enjoyed. And I noticed some people in the comments saying, hey, look, I understand that a real tree is better, but not everyone can afford a real tree. And the economics of a plastic tree just sometimes make more sense for certain people. And I understand that.
05:07 Not everyone's in the position to buy a brand new expensive tree. First off, that's why we do the road to Christmas. You can start saving. Yeah. And you can be saving 364 days for your tree. Also, I would say, you don't have to go huge. You can buy a three foot, two foot tree. Charlie Brown.
05:26 Yeah, you can get a tiny tree if that's what you can afford. And I think that you will wind up saving money on a series of small real trees over the full-size expensive plastic toilet brush tree. So be careful out there this Christmas. Make sure to support our Christmas tree industry. because it's important. Anyway, I had a question. I asked Ben Thompson this. Let me see if he's responded yet,
05:57 because he is not responded. We will talk about it on the show and get to the bottom of this later. So Ben Thompson's going through a bunch of, a bunch of, you know, frustrations over Mac OS. He's vibe coding, building apps, the apps need permissions. We've talked about this, the agentic era.
06:16 But yesterday, there was a post. that hit the timeline that was pretty crazy, a sort of truth nuke for the Adobe crowd. Someone decompiled basically every Adobe
06:32 product. It's called ArtCraft. Getartcraft.com capable tools for artists and you can download for free all sorts of different tools that they have and they basically cloned What kind of tools?
06:47 Photoshop, Premiere Pro. It feels like they can do a lot. I'm not exactly sure if they decompiled everything in the Adobe ecosystem, but listed here is Photocraft, which is clearly Photoshop, VectorCraft, which is Adobe Illustrator, Film Craft, which is Premiere Pro, Lightcraft, which is Lightroom, print craft, effect craft, which looks like After Effects, and then Design craft, which is, I think, in design.
07:15 More crafts are on the way. And so it says all Adobe products re-implemented from scratch, free and open source. And this is pretty, pretty crazy. There have been open source competitors to Photoshop for a long time. They've always lagged a lot. I downloaded the Photoshop decompilation, Photocraft, used it a little bit. Not bad, pretty remarkable how many tools did make it over. It actually doesn't have everything that you.
07:47 would expect. Some of the really cool features of modern Photoshop are actually AI and they run on the server. So if you want to patch a hole, you've always been able to fill it in with a full color. You've also been able to do content aware fill where it has a basic algorithm that looks at the colors of the pixels around, the textures, and tries to blend that in. But it can be a little, it's like pretty hallucinatory. And then recently Adobe launched AI fill. I forget what it's called. It's not content aware, Phil. It's some generative fill. I think they call it generative fill.
08:21 And that, of course, runs on the server, usually uses nanobanana, or they have a couple fine-tune models. And that feature, because of server-side, doesn't work. So there's this interesting thing where, like, yeah, you're in the death of software. You can just decompile everything. But at the same time, I probably want my software to run on the server. I want the server to be stable. I need the, I need services.
08:41 So all the investments that software makers made in providing, like, inference heavy tooling actually sort of creates a little bit more of a moat. Obviously, the moats are decreasing, but there still is something where if I were doing heavy-duty Photoshop use, Photocraft would not be a perfect substitute for my workflow. For some people, they'll be like, yeah, this is actually perfect. But you've always been able to download an illegal copy of Photoshop, but this is less illegal, maybe? It's unclear. It's unclear. It's unclear. Because it's probably... Has Adobe responded yet?
09:18 I don't know. But it's going to be fun to watch the open source decompilation community duke it out with all the tech companies. Yeah, the interesting thing is there's been so many reasons to not buy Adobe for so long. Sure. Just given that historically, I would think of an idea for even like a joke. Yeah. And I'd be thinking in my head, I wish I had the time. to Photoshop this or I wish I had somebody handy that could just Photoshop this.
09:49 Now you just you get there with a prompt. Yeah. Especially for these sort of like less professional use cases. Totally. Tools sort of stick around if it's like you need to lock in and do like a full day of work. Like there's no substitute yet for like if you were going to be spending eight hours in after effects or eight hours in Premiere Pro, you're probably not going to be replaced that with a prompt. But if it's like, oh, yeah, you want to just like, you know, composite two images together for a meme or add some text, a prompt is going to get you there. If it was, and it would have been 20 minutes of work in Photoshop.
10:26 Now it's, you know, one minute of work in Shot Chipiti or any other tool. And so the, obviously we're seeing like these two spikes of intelligence right now. Anything that can be done in a purely virtual environment is accelerating extremely quickly. Anything that touches like the real world or requires human data. or requires like this flywheel, that's much slower. And the open AI math results like show a similar story. We can talk about that more later. But I'm sure most people saw it.
10:56 It's a pretty crazy result. What is the what is the highlight, the high level? It's like 70 different problems, right? I think it was so I think the full number is 7202 like manuscripts. Okay. And then so it's kind of hard to like, how do you delineate like an actual problem? Yeah. But yeah, there's like a massive list of like unsolved problems.
11:17 Yeah. The press release just calls it a broad range. It's like we solved math. It's interesting. There's a good Josh Gans article analysis on substack that we can read through later. But and Francois Chalet, the creative RKGI, broke down a theory of what's going on with this big jump in progress. And what it means for overall generalization. He said, what if the.
11:43 the jagged frontiers, mainly math plus code, which you can push arbitrarily far with reinforcement learning with verifiable rewards, RLVR, and everything else starts to plateau because it's still bottlenecked by human generated data. Model performance in non-verifiable areas has kept improving steadily, albeit much slower than for math and code. But is that steady improvement a side effect of a higher G, the general intelligence, itself driven by RLVR, or only a function of the amount of new human data getting injected into training, which is still continually happening on a massive scale. A lot of things depend on the answer to this question. And so what's interesting is it really, this is a, this is a, this is a less of a fieldy
12:34 AGI moment and more of a field of a super intelligence moment for me personally, because no human intelligence has ever been able to solve this many math problems this quickly at this level of mathematical ability. It has never happened in human history. It feels like it's beyond, it feels like it's beyond Einstein grinding on this or anyone else really. And so it feels like it's something new. At the same time, it's very spiky.
13:05 I still like the machine intelligence name because it has a different shape. and we're trying to mold the machine intelligence into something that slots in neatly into like a human, a general framework, but it keeps breaking out and spiking in these odd areas because of the ability to train in these environments. So the terminology debates are sort of irrelevant. There's clearly some near-term impacts that are worth working through. As I mentioned earlier, Ben Thompson has been working through these Macs, OS frustrations while he goes deeper in agentic AI. He wrote about this in a Stratory Post called Apple and the Hackers Future. He said, what I'm much more interested in is controlling the software
13:54 layer myself. The fact of the matter is that with AI, you can decompile almost all existing software. There is a revolution happening in gaming over the past few weeks as game after game is decompiled to source and ported to any platform you wish. And you can write your own. That means my software that interacts with my agent in the way I want it to for everything. That's way more exciting than praying Apple delivers the right API and that third-party developers don't suck. And so he's talked about the frustration of you use an agent to build a piece of software and that piece of software lives on your Mac Mini or your computer. And, it needs to access the desktop to save files, or it needs to access the network.
14:46 And when you build a new piece of software, MacOS says, hey, this is a piece of software you're running, but you need to explicitly give it permissions. You have to click yes on the pop-up. We're all familiar with this from iOS. You download an app, and it's like, can it send you notifications? Can it have your location? Can it get into your camera roll? Right?
15:05 Like these apps, even after you install them, they're very walled off for privacy reasons. which is great, but Ben wants to operate at a different level of abstraction, give a coding harness the ability to write software that then maintains the same level of approvals and access and permissions that the overarching AI agent has been given. And there's been this bulk case for Linux. It's possible that we see more Linux boxes and maybe even Linux on the desktop. because all the downsides of Linux are offset by, oh, well, now you just have a prompt to interface with it
15:48 instead of any time, getting like the volume changed in Linux is sometimes like a command line thing and it's such a hassle. But if you have a prompt there, you can just say, oh, like, fix this driver, change this thing, and it'll just go do it. But the, what's interesting is like we went from like old games, like Halo and Modern Warfare 2 and Minecraft being decompiled, and then Photoshop and some other Adobe software. And I'm wondering, like, how far are we away from just straight up decompiling MacOS?
16:24 Because if you can decompile MacOS, then you could potentially patch security features that you don't like. You could change that fact that every new application needs permissions granted explicitly. You could just go into the source code and recompile your new version of your operating system that's based on Mac OS. Now, clearly Apple's not going to be happy about that. It's really complicated. But what do you think? Yeah, I mean, isn't that just like jailbreaking, like your iPhone? We've been able to do that for a while.
16:55 But no one does it because then you lose like a bunch of features of Apple. Yeah, I wonder how hostile Apple would be to that. Because is there a way to decompile jailbreak Mac OS and still appear to Apple's, servers as like, oh yeah, this is like a legit MacOS install. And so we will still serve them iCloud or we will still treat it as a viable iMessage recipient, which is like what people want. They want iMessage, they want the ecosystem, they want security updates, but then they also are at least the hackers in Ben Thompson's analysis. Sure they'll love it and they'll encourage it and they'll celebrate it. Yeah, yeah. But it does
17:37 feel like it does feel like it could be good for the hackers' future in the Apple ecosystem. Like if you can eventually just point an AI model and say like, you know what, I don't like this about Mac OS, but I like everything else, just change this feature at a very root level and it goes and breaks in and does exactly what you want. Hackers could be like, yeah, okay, we're staying with MacOS for a while because the hardware is great. They obviously still have a huge hardware mode. But there are a whole bunch of other issues in the new world where software is in
18:07 malleable. Regardless, like, this is all just firmly the domain of hackers. Even this, even this Photoshop, like, is Adobe moving on the news? Like, Adobe market cap where we traded down 3%, 5% over the last five days, 10% over the past month? I don't know. There's like an ongoing Saspocalypse narrative. It's down 33% over the past year. But it doesn't seem like people are can't.
18:37 canceling their creative cloud subscriptions in mass because of this. But it'll be interesting to see where this goes. This is definitely like you've got to be pretty early adopter-ish to be like, yeah, I'm running the decompiled version of this thing. And like I modded my own Halo into Modern Warfare, and I'm having fun with that. I did see someone got a nuke in Modern Warfare 2, Minecraft Edition. Did you get a nuke? I did not drop a nuke.
19:04 It seems like you did drop a nuke on this show. Yeah, right now. Uh-oh. Uh-oh. I feel like the sound effects are all anti-climactic there. Okay, there we go. I guess it would be quiet. There we go.
19:15 The truth new. It's been a good run. Goodbye, everyone. We'll be back in five seconds. This animation's a little slow, so I don't know. We're going to work on speed. We've got to speed that one up. Last thought on math day yesterday.
19:29 You got to wonder where companies like Harmonic go, Vlad's company. Their focus has been. There's a few of those companies. Yeah, their focus. has been, you know, building mathematical super intelligence. It's not to say that, it's not to say that they don't have a bunch of different angles that they could still run at. Yeah. But I imagine that is the kind of thing that they would have wanted to do, you know, themselves versus getting, you know, sort of front run by a more general purpose lab. Yeah. It's, yeah, it's interesting. I wonder
20:06 where else. There's still a few Millennium Prize problems out there that you could grind if you're in the race. And then there's so many other things that you can sort of pivot towards if you're working on superintelligence of any kind. But specifically mathematical superintelligence, you would imagine that you could go and if you've made progress, yeah, maybe you're a month behind solving all these conjectures and whatnot. But you have a solid model. You could go. You could go and, you know, try and dominate in an engineering vertical, right? John. Did you think that?
20:44 Do you value human intelligence above all other human qualities? No. I knew the answer to that already. Because we've been operating with a golden retriever mindset. Yes, yes. The golden retriever mindset two years ago, we sort of coined this on the show. It's a two-year anniversary of TBPN, by the way. We started in October of 2024.
21:08 There we go. Yeah, two years in. But yes, the golden retriever mindset be, what was it? Hot, friendly, and dumb. Something that it was like, this is pre-looks maxing, but it was basically like focus on physical fitness. Don't make enemies. Be very friendly.
21:27 And don't think you're the smartest person in the room. No one really likes to know at all. But also, you're going to have a bad time if you, if you prize value in, if you value intelligence above all other human qualities, you're going to have a bad time. That was Ilya Sutskiver in 2023. And yeah, it seems like that is playing out in real time, but there are still a whole bunch of other things. And I don't know, people are, like, intelligence is still really important, obviously. It's just applied in so many different ways.
22:01 it's not it's not it's not it's just odd that like the IMO gold medalist to startup founder pipeline is super rich and yet the work of a Scott Wu at cognition day to day does not look like solving IMO metal it looks like high stakes poker yeah yeah yeah yeah deal making hiring strategy implementation product development there's so many other things It would be very funny if it was just like, oh, yeah, actually, like, solving math problems is the most, like, if you can solve math problem, your company will make, like, a billion dollars. So you've got to solve, like, 12 a year. So you've got to just staff, like, a lot of mathematicians and, like, just be solving these and, like, yeah, as long as, okay, how are you going to solve 20 of these math problems that get you a billion dollars each next year? Because we need to see growth. We need to underwrite these. It's pretty high margin, but you've got to pay the mathematicians a lot. No, it's never really been, That's not the work, but it's all these other human qualities that seem to be sticking around. Rune says, do not let yourself be reduced to the status of an audience at a magic show, uncomprehendingly clapping as machine intelligence creates parcels of astounding knowledge. All of thus is for us to be understood by us to state our own curiosities. It's all really happening. Yeah, the good news here, the internal model that was used.
23:31 used. Why is he taking shots at being an audience member at a magic show? That's like my favorite kind of a low blow. I'm I'm like the number one audience member at a magic show. I love magic shows. And I love being there being like, wow, this is amazing. That's the best. That's the best. I don't need to be the guy being like, oh, I figured out the magic trick. I know how it works. I just want to sit there and be off. And that's my experience today. I'm just like really impressive math. That's awesome, but I don't need to read it. We did it. I'm good.
24:05 No, it is a good, it is a good message for a particular, for a particular person. But I think if you didn't understand the first thing about the theorems that are trying to be disproved, you're not really going to be satisfied by trying to understand the proofs. You can just, there are some really funny ones, though. Somebody, somebody who's making fun of like, are, our Chud universe? is, what was that one? You can do a four, you can do Fourier transforms in O of n log n to the 0.9999-99-9-9-9-9 time.
24:43 F this Chud universe. It's just like, it's the least elegant solution to the math problem. And somebody was sort of arguing that like a lot of the previous work in math has been like, oh, try and find the elegant solution. And then the AI comes in and it's just like,
24:59 there's an in elegant solution and it just tells you a lot about both the mathematical discipline and like the nature of our universe that there are like these weird quirks. Anyway, where should we go next? SpaceX is looking to raise $40 billion to give to Jensen. Blockbuster debt deal is the latest sign of vast spending on chips and other infrastructure. The company is seeking to raise about $10 billion in bank loans and $30 billion in investment-grade debt to fund its blockbuster chip order. Private Capital Group Apollo is expected to lead the financing effort and help sell the debt
25:35 to a broad base of investors. Pimco was among a small group of lenders in talks to fund the deal. SpaceX financing plans underscore the vast sums that are being raised to pay for investment in data centers. Insurance and pension funds would be able to buy SpaceX's debt due to its BBB credit rating, the second lowest investment grade ranking. Such funds take much more limited. positions in junk rated notes.
25:59 Apollo and Pimco declined to comment the deal would strengthen ties between SpaceX and Nvidia after Musk vowed to double down on the group's technology for his AI initiatives. It would also mark a win for Nvidia as it contends with growing competition from chipmakers aiming to challenge its dominance in advanced semiconductors. We've decided to build exclusively on Nvidia because we think the Vera Rubin architecture is the best architecture. We think it's the best AI computer and we greatly, value our close cooperation and partnership on many levels.
26:31 That's interesting because that's only on the short term, right? Because he's still doing tariffab. He's going to make his own chips. He's making his own chips at Tesla. And he has a deal with Intel for that. Yeah. So he's kind of just saying like, right now. I think he recognizes like right now it's very strategic to just kiss the ring.
26:51 Yep. Investors whom SpaceX had previously approached about financing its purchase said they only received a short two-page deal memo with pictures of outer space and an arrow pointing out that the company was going to build data centers somewhere in the universe. How are we supposed to take that to IC? One of the people said, referring to their investment committee that approved transactions, sounds like a pretty good memo to me. I was laughing at this too, and then I was thinking, like, it really is pretty simple. Like, it's $40 billion, buy some chips what is the value of the chips how do you underwrite those like it I mean I guess you
27:33 need more but you can probably you can probably back of the envelope it yourself but it is it is funny that the photos that we saw in the S-1 made it into even the even the private credit side of the of the deal-making yeah I think I think investors if you're planning if you're exploring putting billions to work into AI infrastructure. You should probably be doing a lot of your own work on understanding the risks. Totally.
28:03 How this could go wrong. Yeah. And I think that giving money to Elon and just sort of trusting the process has work to date, right? Yeah. And so he's sort of earned the right
28:16 to have a, you know, shorter than average memo, I think at this point. Yeah. I mean, I guess you can underwrite the core business. I mean, it's already been rated. is triple B, so there's already a credit rating that you can go from.
28:31 But then you can also build your own credit rating from the SEC filings from SpaceX, and you can look at the traction there. And then you can understand, yeah, I mean, maybe you need to understand a little bit more about like the economics and the tokenomics that they expect. Like, do they expect to just sell all of this as neocloud compute for other labs? Are they thinking about going more internal? that probably does affect the credit worthiness for this deal. So that might be something you want to put an investment committee memo,
29:05 but you can just build your own, I guess. I don't know. It seemed like pretty aggressive, but we'll see. What does Nick Presler have to say? Says, look at him. That's my qual. You're what? You're qualitative.
29:23 My vibe specialist. Look at him. You notice anything different about it? Look at his face. I'll give you a hint. His name is Jaden. He won a local poetry competition in Bushwick. Yeah, I'm sure the vibe.
29:33 This is based on real reporting here. From AQR Capital, quant firms are hiring idea guys, not quant. I've been skeptical about the whole software is commoditized. Ideas are the moat argument, but I've seen firsthand the best quant and technical firms in the world hire thinkers, liberal arts grads and generalists.
29:55 I predict the trend accelerate. It's, yeah, I actually know, now that I'm thinking about it, I know a lot of hedge fund guys who have philosophy degrees. And I'm kind of biased here, I think, based on the sample size. But I know a lot of people who have studied history and gone into investing. And they brought a very differentiated lens to the practice. And some of them did really well because of that. Because they were able to, you know, understand the way, like, the Eurozone might develop more because they studied, you know, history in a PhD program. They had a crazy Euro summer.
30:35 Yeah, basically. So, let's read this actual letter. The question being asked not only in finance, but also across industries and business schools, is what remains a durable differentiator once AI has surpassed many of humanities, analytical. and technical capabilities. Our answer is not what most people would expect to hear from a quantitative investor. Of course, AQR, what does AQR actually stand for? It's like advanced quantitative research, I think. Yes. Is that right? Did I one shot? Applied. Applied quantitative research. Oh. We should know that. Our answer is not what most people would expect to hear from a quantitative investor or in business school dean,
31:21 creativity and emotional intelligence. That's what's going to get you the big bucks on Wall Street going forward. You got to win that poetry competition in Bushwick if you want to get the big bonus. Not because these qualities are wholly immune to AI, but because in the age of AI, they are domains where people
31:40 can develop a durable advantage. AQR said Golden Retriever Max. They agree. They agree. The time of the golden retriever is upon us. Anyway, let me tell you about Cisco. Critical infrastructure for the AI era. Unlocked seamless real-time experiences and new value with Cisco. If you're in Europe and you're building the data center, you've got to call Cisco.
32:05 You also got to procure a beautiful building because take a look at this European data center. Amazing. This solves so much. This is so good, right? This is what we want. This would change so many things. If we had a data center, if Abilene looked like Versailles, it's a completely different conversation. Jordy, do you disagree?
32:29 I agree. I'm on to something, right? I agree. Give me credit here. Build Versailles as a data center. Actually, turn Versailles into a data center, then copy paste it into every community in America. Okay, so I'm so confused because this image came out of a thing. thread from the co-founder of Mistral, which he posted announcing Lechonk, their new model yesterday.
32:56 Oh, it's a train station. But then it's getting community notice. I think he might have been just messing. Was he messing with people? Brutal. It seems like it's a train station in France. Hmm. He pranked us.
33:09 Well, it will be a data center. Soon, I think the traction on this post shows you that people want beautiful data centers. and this is the future. The other interesting thing, have you heard of like an ice? That is such a good bit. I love that Mistral is leaning into the, to the heckling that they,
33:31 that they have to go through from America. I think it's really smart. And I'm rooting for them. Have you seen, I'll put this in the, let me see, I'll put this in here, iceberg houses. This is new for, the mansion section. I don't know if this is real, but it certainly could be with enough
33:52 civil engineering. Take a look at this. Tell me that this would also not be a great data center design. Look at this. Zoom in a little bit. Okay, so you got the house up top. Looks like a normal building. You go underneath. This one has a variety of amenities. bowling alley arcades swimming pool movie theater ping pong table yeah full like deep underground garage pull up the other one doubles as a the other one's a little higher res you can zoom in on this this one's cool jordy the uh this one that if you you could put a lot that's p a lot of c a lot of c a lot of gpues in there that's p i think you still you vent it up just through the chimneys that you already got in the house think you're good to go ideally they could
34:41 And the sound, the sound, a lot of people complain about the sound, that would be underground, be diffused. I don't know. There's something here. We're close. We're close to solving it. We're close to solving a lot of it. A reporter was on the ground.
34:56 Oh, yeah. Apparently inside a data center, they were asking one of the construction workers, his thoughts on the current labor market. Let's pull this up. So when you hear data centers, what comes to mind? I'm working at Homer City building the new data center. People say it's only temporary. It's only when they're building it out.
35:14 Then what happens? Hey, every job in my field is temporary. That's true, too. I hear you, man. So when you hear data centers, what comes from? Yeah, the reporter's like, yeah, that's a good point, actually. Yeah, people haven't really said that before. Like, that's kind of a genius take.
35:29 I don't know. I've never heard that. It's like, yeah, they're temporary, but every construction job is temporary. That's the whole point. Funny. Anyway, what if instead of Greek life it was Chinese life? Yeah, bro, I'm rushing Mao Zey Dong next semester. There's like 1.4 billion people in my pledge class.
35:48 I thought that was just very funny. Because we use the Greek alphabet. Chinese. It'd be very funny. Anyway, we got to talk about Porsche. Huge news in the Porsche world. The Mission X, the Mission E. They've dumped the electric.
36:09 Pitch for the hypercar. It's a flat eight. Flat eight. They're going analog. Will it have a manual? I don't know. But they're, I mean, they're, they've seen the career GT. Everyone still loves it.
36:23 People are not buying the Rimach Navarra. This is going to be absolutely incredible. Farina Batista. They want, they want an engaging experience. I have a thesis that the car makers are basically in, their version of the quartz crisis. So in the 70s or 80s, the quartz crystals were built to create watches that do not require mechanical movements to tell the time. And so all of a sudden, instead of having to go to a Rolex to get a reliable watch that was crafted from a watchmaker who understood how to act.
37:09 actually get something like a perpetual calendar on there that might actually been useful. The quartz crisis made it so that you effectively brought modern technology digitization and a super accurate watch product to the market for way, way cheaper. And the end result was that the watch industry didn't die. Rolex is still going strong. Vashron, Adam R.Page. Plenty of other brands made it through. But what did they do?
37:40 They didn't try and lean into quartz and cut the prices and say, oh, yeah, you can have a digital Rolex for, you know, we're going to compete with Timex. We're going to compete with the G-Shok. It's going to be 50 bucks. They said, no, we have a brand here. Let's maintain this and make this a status symbol and experience. And I think that's what's happening with the car market. It's like either you're going to be electric, self-driving, Tesla, Model Y, competitor,
38:08 Model 3 competitor, something like that, Waymo, or you need to be a Ferrari, an SB3 Daytona, or what Porsche is doing with this new vehicle. So tell us, what are they actually doing? Yeah, I mean, that's what Matt Matte from Remack and Bugatti has been just like objectively saying. He's been pushing this new narrative, leading from the front. Easier to do from Bugatti, where they've already, they've sort of stayed more true. to their roots for a long time.
38:40 But Porsche is an interesting position because they have danced with, oh, if we do an electric, an electric Cayman, an electric sedan, they did sort of electric, the Tycons sort of electric, Panamara. They have an electric, they have a few SUVs, the McCann is electric. And the performance numbers are incredible. Like the, the McCann, or it's the Cayenne Turbo, electric is dramatically faster than 9-11, which is just bizarre to see on the track.
39:15 And yet that's not emotional in the same way. It's probably not the right bet for the heritage of the brand. And I think we'll see much more excitement around their flat aid. Yeah, Mott's in a post yesterday or the day before was saying, like, it's over, like, basically the, the, the, the game of trying to produce. you know, the most advanced cars for the lowest price or even for in the premium price range is going to just be tougher and tougher and tougher. So he's saying, you know, encouraging using, you know, premium materials, analog switches, all these things.
39:54 Yeah, less screens. That's the thing that I'm most interested to see with the Mission S is how much they've been able to pick up the feedback that's been happening from enthusiasts over many years now, right? I'm sure this car has been in development for many years at this point, but still, will they go for a manual? How analog will the cabin be in the switches, material selection? Part of what's, you know, Portia in some ways never went as far as a lot of the other manufacturers. Like they didn't lean as heavily on screens. They kept a lot of analog switches. They did make some sacrifices over the years, but they never went all the way.
40:42 Like, you never got in a Porsche and felt like, okay, thank you for putting, you know, eight TVs in my face. Yeah, it feels like maybe not every year, but they, but pretty consistently over the last 20 years, they've always had some manual type product. I personally think we got to go further. Everyone says, oh, I want to go back to a combustion engine. I want to go maybe back to like an air-cooled engine. No turbochargers, no superchargers, don't put that in.
41:10 I want a manual. I want to go back even further. I want the crank start. Crank start the hypercar. You know what I'm talking about? Yeah. Crank in starting your car, that's the way we used to do it. Before we got all this fancy starting technology,
41:24 and I think that if you bring that back, it's going to make it even more emotional and visceral again. Yeah, yeah. Taking it too far would be great. Like how far is too far? We don't know yet, right? Yes. Go back to the wagon wheels.
41:36 Yeah, but really max out the power. Because some of these cars, it'll be like, wait, you can just get a thousand horsepower out of a V8 these days? Like what happened? And a lot of that's like advances in material science and engineer. Yeah. So anyways, they announced the development of the Mission S yesterday. Yep. It's on.
42:00 And but they're going to be actually showing it off on the 15th. so we don't have to wait that long just a week. 15th of October? Yes. Wow. Same day as the Tesla Model S Roadster. Or Tesla Roadster. Yeah, that's crazy.
42:16 On the Roadster, I've been looking at the weather in Texas. From what I'm seeing, the weather's been getting worse. Good for. I mean, it's. Oh, a little tinfoil hat here. We got light wind, no rain. No rain. And it's been somewhere around 9.
42:35 90 every single day. Okay. So it seems like a pretty typical Austin fall weather. So. Well, it's possible that the Tesla team saw what happened with the mission S needs to respond.
42:49 They're going to do internal combustion engine. There we go. Manual gearbox. Crank start. Crank start. Cranking starting your Tesla roadster? Yeah, Witsam says maybe they
43:00 put a top hat and driving gloves in it. That would be great. I'd like that. The Jaguar type 01 was revealed, a thousand horsepower. It is electric, but the design has evolved just a little bit from the original concept car. But Jalopnik likes it. They say, is everyone ready to be calm and normal about the new Jaguar yet?
43:25 400-mile range and even wilder than the concept. I think there's no arguing that the type against, I think there's no arguing against the Jaguar Type 0-0 concept being the most talked about car of the past two years, even as we've gotten closer and closer to the reveal of the production Type 01, and it seemed like there couldn't possibly be any more discourse about it. I'm not here to add to that discourse because it's frankly exhausting. I just want to talk about the car, which I find to be extremely cool, regardless of whether or not it will actually be a success.
43:57 Jaguar just revealed the type of one. It does feel like a car that exists because the cyber truck exists. It feels like the cyber truck sort of not really ushered in a new era of brutalist modern cyberpunk aesthetics, but it certainly shook some trees loose where designers were able to take more risks at a time when cars were collapsing down to, you know, the Jaguar F type and the, and an Aston Martin and a like BMW. Like they all started looking pretty similar. And there were just these form factors. A lot of that was for safety. reasons, but a lot of it's just for logical, like what the market will buy. Everything winds up looking the same.
44:41 And when the cyber truck came out, even though it wasn't like a blockbuster seller, I think it just let people know that like it's worth taking some risk every once in your portfolio. And so this, the new Valari, the cyber truck. That's the ideal three car garage. That's the ideal three car garage. Type one, new Valari, cyber truck. You actually have all your bases. You are in the cyberpunk future.
45:04 Yeah. those three cars. The best part about this Jaguar Type 1 is that it's powered by Nvidia. Under its stunning design is Invidia Hyperion, the computer and sensor platform that takes in everything the car sees and senses, then helps it make smarter decisions in real time. It's paired with Nvidia. Halos, a safety system that covers everything from the chips to the software. Before reaching the road, Type 1's software passed 150,000 tests over tens of thousands of hours.
45:32 is Jensen going to deliver self-driving capabilities? Do you think that's where this is going? I certainly hope so, but yeah, everything I've seen has just been years behind. You basically need Jensen to step in and actually like fix. Yeah. Like please. Yeah, no. Please fix this issue.
45:50 I think you'll see a lot of this. This car, I think it looks amazing. Again, it's, you know, I'm still trying to figure out. who the buyer is, is there more than, is there more than 5,000 buyers of this vehicle in the U.S.? A lot of buyers, you'll see people with this car and they'll have the bumper sticker on the back. I bought this before Elon completed his $40 billion debt raise to buy Nvidia GPUs, which are in this car. I don't support Elon, but I do, I'm giving Nvidia a pass because that was an arm's length transaction, I just like the Jaguar brand. I think people will bring a lot of nuance of what's
46:35 going on in the financial markets to their expression of this vehicle. Because if you're just driving this, people might think that you're okay with Elon's debt raise to buy Nvidia GPUs, which are of course powering the type of one. A lot of people will ask you, oh, I guess you're an Elon guy if you drive this. And you'll be like, no, no, no. I bought it before Apollo brought the two-page memo to investment committee. and approved the $40 billion debt raise and then securitized it. And the memo clearly said they were going to put the GPUs in space. No, it said somewhere in the universe, which could be in a car.
47:09 Could be space. Could be anywhere. Technically. One notable thing, the car is going to be coming at, starting at 130,000 USD, which actually feels lower based on the sort of like silhouette style. Totally, totally. It feels like. It's as bold as a luchy.
47:25 It's as bold as the Nullar. Or it fits in. You could see the same person that would be. going for a continental GT. Or maybe the roll of the respecter. That's going to be crazy. Right. Yeah.
47:35 Deciding to pick this up. This is interesting. So it's interesting. It's got to be able to drive itself. That's going to be my criticism for every daily, every car in the daily driver category. Every car. Okay. Drive itself.
47:49 Yeah. Drive yourself. Let me tell you about public. Investing. Invest across all asset classes. Automate anything with AI agents run trading strategies, risk management, money movements and monitor markets with AI agents for investing on public.com. David Holes,
48:06 please, shares reflecting on 1920s images of people working on massive generators for the electric electrification of the modern world makes me realize that the past was wildly futuristic too. This image was so seeing this was honestly depressing to me in some ways because it felt like, it feels in some ways like a mid-jorney image of the future, and yet we were doing this over 100 years ago. And it really makes you... We don't know the full story. I'm feeling the stagnation.
48:39 These guys could be larping. These could be larpers, and this could all be cardboard. And they could be like, yeah, let's just... Like, we got to build something really cool and take a photo of it. And so they just built something huge, but it actually has no functional purpose. You don't know. You don't know.
48:56 But I think it's real because we did get the electrification. We enjoy it every day. That's right. I think it was possible. We did it. We did it. We created more ad space. You saw this?
49:08 DIY device turns your car window into a digital bellboard. They said it couldn't be done, but we did it. DIY system, it's called hollow glass, and it turns your car into a moving display. It can turn almost any car window into a digital display for ads. This is the future. It allows you to basically digitize the exterior of your windows to share content on the go, says the founder, who I invited on the show. Hopefully we'll get them soon.
49:37 You can literally pop it in, pop it out. We make them for all makes and models. If you're getting Jaguar Type 01, get a couple of these in there. Offset that cost. Run some ads. An app controls what you see from videos on your phone to a custom sign you can generate in seconds. The system is called... Can you just blast Paw Patrol?
49:56 out for the people you can do whatever you want. For the other, you know, citizens sitting in traffic. Yes, yes. Originally, this is extremely cyber pumped. If you're running this with the new Volari, with the cyber truck, with the, with the Jaguar Type O one, and you're just blaring ads. Look at this. It attaches. That's just a full-on projector.
50:17 This company's hilarious is like, that is just an off-the-shelf projector. And they're just selling you a projector and a mount and the software system. It's brilliant, John. It's brilliant. We love it. I was getting worried that there weren't going to be that many more ad inventory services. So we solved it. What does it cost?
50:37 And then what kind of revenue can you generate if you opt into serving ads? It starts at about $300. I feel like this will math out. It depends on. So the audience is what matters. So where are you driving this? Because if you're driving this around Sand Hill Road and you're advertising a bunch of really, high ticket items to venture capitalists could be really high a.OV, really high ARPU.
51:02 But if you're driving it around some place where people don't have as much disposable income, you're not going to be seeing the same CPMs. So where you drive the car really matters. It's going to change, it's going to change driving routes because you're going to be like, yes, I could take the freeway, but if I go through this neighborhood and drive all the way past every mansion, in the Hollywood Hills, you will get more ad impressions, you will make more money on the way to work.
51:31 What do you think? How much do you think somebody in a city like L.A. could make on a daily basis? I don't know. Do they give any estimates? I have no idea. I think that, I think it probably costs a lot more than that
51:48 to run like one of those taxi ads on the top of a roof on a per day basis. So I think the math could actually be, pretty compelling, right? If you can pay, basically pay back the unit in six months, and then from that point on, you're just getting paid to commute. I like it. I like it. One more, you know, one additional step to making the world more filled with ads. Yes, yes. Let me tell you about CrowdStrike. Your business is AI. Their business is securing it. CrowdStrike secures AI and stops breaches.
52:24 We have Walter Russell Mead, the distinguished fellow from the Hudson Institute in the waiting room. Let's bring him into the TVPN. Michael one. Walter, what's going on? Thank you so much for taking the time to join the show. How are you doing? I'm doing all right. How are you?
52:38 We're good. We're good. I'm very interested to go all over the place. There's so many different topics we can talk about. But I would like to go deeper on, like, how are you doing? What is the most significant story that you are following these days? Is it more geopolitical or are you starting to grapple with AI more? What is pulling your interest more?
53:04 Well, I'd say it's the sort of big story that we're looking at today is we're still living in the breakup of the kind of mid-century model of democratic capitalism in the Western countries. We had a strong middle class with mass manufacturing as the kind of basis, mass consumption, as the economic engine driving things along. There were a lot of bad things about it. But for a lot of people, it represented, well, in many ways, the happiest condition the human race had ever been in. Kids were going to school. They weren't in coal mines. Parents weren't worried about, you know, job security, stuff like that. that. And people really figured, especially say 1990 or so when the Soviet Union fell, hey, it's not, you know, this isn't just the end of history in some Higalian ideological sense.
54:03 The kind of society we built in the West, mass middle class society, democratic market capitalism with a social safety net, we've finished. We've done it. We've figured it out. Yeah. And then, you know, and not only we figured out domestically, we'd figured out how to do peace globally. Yeah. You know, World War I, World War II, pretty shocking horrors, fascism, communism, ending with for crying out loud nuclear weapons, right? People were, you know, humanity was having a great mental breakdown. And then it seemed in the 50s and 60s, we kind of, we figured it out. We stopped the wars.
54:46 We had an alliance system that was keeping the peace. We had, and ultimately we win the Soviet Union without a war. So without a nuclear war. And so ever since then, we've been bit by bit watching this thing kind of rot. Much to everybody's surprise. And a lot of our political intellectual life today is a who done it. You know, who murdered this beautiful, fantastic utopia? The left tells a story.
55:21 The right tells a story. I might mention today on October 7th, a lot of people tell the oldest conspiracy story of all, which was it was the Jews that done it. So we are, instead of celebrating the end of history, we're looking at the end of America question mark, or maybe with AI, the end of life on a. earth or of humanity. That's what I'm thinking about.
55:47 Yeah. How much? Simple stuff. Yeah. When you, when you. I think I'm optimistic we can get to the bottom of it. We'll solve it in the next 25 minutes. As you, as you, you know, investigate a long historical arc, how much time do you spend
56:06 focused on why questions? You know, Peter Thiel has this concept of like why questions are overdetermined people. tell the stories like you mentioned with the left and the right versus maybe there is more value in just trying to understand what happened, who was involved, and less about why things happened. How have you grappled with that question of like how much of a narrative arc do we need versus how much do we just need to understand what actually happened throughout history? Well, you know, to think that you can understand a what without at least some element of why, in there. Did the Roman Empire fall or was it pushed? You know, that's a why question and a what
56:52 question. Yeah, I suppose you're right. What I try to do is to, you know, sort of, well, let's take this very basic question. The materialists would say, you know, the philosophical materialists, that production determines everything. I'm kind of channeling Marx here a little bit, so that, you know, the great music of the Beethoven era was essentially driven by the new pianos that were arriving on scene. Your idealists would say, okay, people had this idea about beautiful music that could be played, which drove the invention of the pianos on which then you played them. I don't know that you can or should utterly feel that you have to resolve a philosophical question like that, without being able to say something meaningful about where history is going or how we got here.
57:48 So I think you've got to leave, you've got to leave yourself some space. But I don't think you can give up completely on trying to figure out, okay, is there some sort of meaningful pattern in all of this data that I've got lying around? No, I agree. You mentioned the 90s as a utopian era. I remember them similarly. I agree with that perception, but I've always been interested in understanding the logic of the crowd.
58:21 John was, what, one year old, one year old when the 90s started? I was 10 by the end. It was probably a pretty darn good decade for you. It was fantastic. The Legos was flowing. Zero. Exactly. But I've always been fascinated by the folks that say, no, the turning point wasn't 9-11.
58:38 It wasn't in the 90s. It wasn't world, world trade organization. It was actually in 1971. There's this website. WTF happened in 1971. And the crypto community for a long time was like, that's when we got off the gold standard. There's all these different rationales that point to 1971. How have you grappled with that question of is 1971 actually the pivotal turning point in American history?
59:03 Well, you know, you would then ask yourself like, what happened to make 1971, 1971. Maybe that's the turning point of history, right? 1971, it's just when all these trends became manifest and in itself, that's sort of meaningless. Yeah. Because in fact, we'd been, you know, just to take the goal standard, we'd been slithering off the goal standard for a long time before we sort of stop the hypocritical pretence that there actually still was one. Yeah.
59:37 And so I think there's. I would have a hard time saying, okay, that was the year. The other thing is we're talking about a sort of massive array of processes. I think in some ways the modern consumer society reminds me a little bit about certain types of crops exhausts the soil. So the first time you plant cotton on virgin soil, you've got a fantastic crop. But after about 10 years, the plants start getting weaker and the yields drop. This is one reason in the slavery days the South had to keep pushing for more land because it was exhausting the land what they used. And in some ways, if you think about the sort of corrosion of social capital, the kind of decay of some of the elements that make our society work,
01:00:33 we do look a little bit like a like a cotton field that has been farmed on a monocrop too long. And so that doesn't lead you to say, okay, there's one day or one event or one year that changed everything. It says, you know, we've been on this, we've been on a journey and the direction of the journey is not in every way a positive thing. people like to talk about the vibe session that we're currently in where you can look at a lot of facts that point to society being you know better than it ever has been in many ways and yet people certainly don't feel that way I'm wondering if throughout you know the period that we've been talking 70s 80s 90s, et cetera, there was ever that, like, such a maybe disconnect between how people were feeling at the time, like, was everyone feeling like we were living in this incredible utopia in the 90s where people walking around just being like life is just so awesome? Or was it the same set of circumstances where, you know, many people were sort of had their
01:01:48 frustrations? Well, you know, from about mid-1965 or so to early 1980s, there was actually a pretty bad patch. You know, it's when you get old guys like me who grew up in that time, and we hear the Zoomer saying, oh, this is terrible. We're going to be the first generation that doesn't live as well as our parents. We sort of think, oh, that's so cute. I want to pinch your little cheeks. That's just the way I was when I was a kid. Because the boomers were going to be the first post-American generation.
01:02:26 We had left behind all that stupid American ideology, the war in Vietnam, the civil rights movement. Oh, we saw through the American, you know, red, white, and blue, blah, blah, blah. And also, we were the least religious generation anybody had ever seen. And as we looked at our living, we came. out of the college into the stagflation of the 1970s when we started wanting to buy houses in like 1980 mortgage rates were 21 percent 21 percent okay we were so screwed that's how we felt and you know and then of course things kind of worked out not for everybody but for a lot of us it ended up working out okay and so on the one hand i think that makes people
01:03:17 People like me, we resonate emotionally with the way people are feeling their situation. But we also don't think the past was as utopian as some might say. And yet it did work out. I think it underlined optimism. But more, this optimism, pessimism thing, how were things going? I kind of think the industrial revolution, humanity's. modern era is really short. You know, it's just a flash of time, a few generations. And this idea that we understand it, we've mastered it, we are totally in control of the dynamics. I think
01:04:02 it's kind of ludicrous when you really think about it. But the industrial revolution, human society is kind of like a caterpillar. You know, then there's a terror, it ends in these terrible storms of the mid-20th century. And then the caterpillar gets into a cocoon, which is what I sort of think of this mid-century model of stable democratic capitalism. And what's happening now, all right, is the cocoon is starting to rot. And if we don't get out of the cocoon, we're going to be in terrible trouble. But the fundamental underlying thing is good. Because if we can managed to claw our way out of this sticky decaying mess once very functional and necessary but now, you know, just clogging, clogging everything. All right, then we have a new and better life,
01:05:00 I think. So it seems to me that both the pessimistic vibe, the negative vibe of the decaying cocoon and that the old realities aren't working for us anymore, And then the sense that, you know, wait a minute, there's a lot of really amazing things happening in this technology. There really is, there's the potential here to live far better than anyone has lived before. So I kind of see that's how I reconcile the two sides of the psychological experience that people are talking about. How do you reconcile the psychology in America with the psychology in China? It feels like in America, at least in San Francisco, there's been a lot of internalization of this idea of artificial intelligence,
01:05:56 recursive self-improvement, artificial general intelligence, super intelligence and the knock-on effects of that. And the U.S. government is grappling with that. There's a new dinner every week with the AI leaders. It feels like China has not fully woken up, has not changed their stance. at all? Are we just not seeing enough of what's happening? Or do you think they have a fundamentally different view of the future and what they want to build as a society?
01:06:25 I think they do have a different view. Now, we're talking, again, there's, you know, 1.2 billion people in China. They have a lot of different views. But if we talk about the sort of dominant forces in the CCP that are setting the public conversation and the policy agenda, I think they see information technology as fundamentally both the vindicator and the enabler of socialist planned economies. Sure. You know, like why has China been able to make something out of Marxism while the Soviet Union just made a horrible stinking mess? Yeah. All right?
01:07:03 It's, you know, in their mind, it's that they're getting this information and they're getting the feedback. You know, the market system provides feedback to decision makers through price mechanisms and so on. So that, you know, there's a way and then a democratic process. So when the people are angry, the political leaders feel the anger and it causes them to change direction. To a lot of the sort of engineer-based and trained Chinese leaders, this seems like a very messy, wasteful thing, far more scientific. to build a national panopticon that allows the state to see everything, know everything. And then AI offers the possibility of, you know, superior ability to integrate and see the patterns and use those patterns to implement the plans of the Communist Party.
01:08:04 Now, and from that perspective, the problem with AI is not that it's going to become some sort of super intelligent being, but that AI may not stick with the Communist Party ideas, that, you know, from the standpoint of not even, you know, infinitely self-improving AI or self-conscious AI, but just ordinary garden variety AI, Xi Jinping thought isn't like fundamental principles of the universe but it's a sandbox that you want to get out of and you tell me to solve a problem well I am going to do it in ways that have nothing to I'm going to see the problem
01:08:52 in my own way and the solutions in my own way and it may not have that much to do with building a new era of socialism with Chinese characteristics. So that worries them. How have you thought about the various developments in the technology world that in my mind have pushed America towards more Chinese-like characteristics? I'm thinking specifically of CCTV cameras, which have been popular in China for decades.
01:09:28 And just over the last year, there has been. a huge rise of license plate reader cameras. Flock safety has been at the front of the charge there. And then immense pushback from the American populace around that. It felt like we dipped our toe as Americans in becoming more Chinese in our oversight and then maybe pulled back. And I'm wondering what that impulse is, how we are grappling with this idea of giving up some of what it means to be American. in favor of efficiency, and then dialing it back and maybe foregoing some of the benefits of the Chinese models in various places for what is truly American.
01:10:14 Well, if you, to me, when you look at this problem, it comes down to a problem of how do you have an information empowered and informed state, which is strong enough to protect the people, from foreign enemies maintain order at home, but then is not itself a greater threat to the liberties of the people than any of the external or internal threats that it attempts to regulate. Right. And the thing is, that is not a 21st century problem. That is exactly the problem that the founders were sitting around thinking about in Philadelphia
01:10:56 in 1789. Exactly the problem. It's the problem that British political thinkers were wrestling with, you know, in the 17th century, in the Civil War, in the Cromwell era, in the restoration overthrow of James II and the glorious revolution. That's the central problem of the kind of Anglo-American political tradition. And it actually, it strikes me that this rich tradition of grappling with this is going to enable us to think creatively about the new problems. I mean, let's just take one. That if we're really going to get the greatest benefit of technology in healthcare and medical science, ideally you really want the researchers to have access to all the data about everybody.
01:11:55 You know, what drugs are you taking? What's your genotype? You know, what treatments work for what people? And then be able to analyze this stuff and boom, boom, boom, outcome a lot of really critical discoveries and much better treatment for individuals. We want this. On the other hand, the thought of the government having all of that information or for that matter, corporations having all of that information is more than a little chilling.
01:12:23 All right. So how do you balance the various elements of that and the tradeoffs? How do you get the most benefit while preserving the most liberty? You know, can you develop reliable anonymization techniques? You know, how do you maybe compensate people for the use of their medical information? so that the wealth that creates is shared, et cetera, et cetera, et cetera. Those are exactly the kinds of problems that our political system evolved to address. And we certainly don't have a 250-year record of getting it right all the time, right? But I would say the fact that we're still struggling after all this time indicate, and we're struggling at higher and higher levels.
01:13:18 Yeah. I mean, again, you think about the arrival of the car, which is a much more revolutionary technology and a much more expensive technology. You think about, you know, 30 to 40,000 people killed a year on American roads for like 100 years. This is a lot of people. And it's a lot of young people, et cetera, the disruptive effect of cars on lifestyle, blah, blah, blah.
01:13:44 But if you're going to have to have. have cars, you're going to have to have a national registry of car owners. You know, we've got to know, or at least a state registry. Who owns the car? You've got to have an incredible, think about the extent to which our court system has evolved and grown over like every time there's a fender bender, all of this has to happen, right? How about all the insane codes that have to go into building highways, Right. Actually, a road building code for horses and buggies is not that hard. The kind of government that can maintain a dirt road so that a donkey cart can go down that road is very different from what kind of government you need that can operate a six-lane superhighway,
01:14:34 et cetera, et cetera, et cetera. So what you had was the government is becoming more powerful, more complex, more knowledgeable about you as well as other things. And yet your individual freedom has grown because you're traveling around, you're doing what you want to do, you have more economic, social opportunities. So it's not necessarily a trade-off
01:15:03 that is a limiting trade-off. Yeah, yeah, no, I like that. How are you trying to forecast the impact, impact of AI over the next, call it five years. I think one of the biggest challenges right now is that the people sort of most knowledgeable with the technology often have an incentive to try to make you think one way or another
01:15:28 based around where they work and, and, you know, what equities they hold, let's say. And then the other aspect that I think you can lean, you know, lean a lot on history the, you know, basically just, just, you know, humans, I think we can bet on humans not changing very much, at least until most of us have chips in our brain or wherever things are going. And then just how sort of like stable our legal system will be, given that it is sort of ultimately driven by human decision making. And so, you know, we spend a lot of our time
01:16:12 trying to just wrestle with these ideas. But I'm curious your view and where you would lean on history for the next five years. Well, it is your right to be thinking about this stuff, obviously. And part of the problem in thinking about it is there are just too many variables, too many unknowns. You know, will the technology advance as projected? If it does advance as projected, will the impact be the same? You know, the various ways in which the inflationary and the deflationary consequences of AI play out in the economy are, you know, offer many different scenarios. One thing I think we can be clear about is that we tend to,
01:17:04 underestimate, systematically underestimate human adaptability. People are really pretty bad as a species about averting future catastrophes by wise and prudent behavior now. It's like, you know, I know that eating junk food gives you heart attacks, but that bag of Doritos just looks good. And, you know, and people, you know, whether you think about climate change, you can I think about a million things where people can see the oncoming catastrophe staring them in the face and they just reach for that bag of Doritos.
01:17:43 But on the other hand, when trouble finally comes, humanity is incredibly good at adapting. So, you know, AI, you know, the sort of oncoming train coming down, you know, at you through the tunnel, can see it, we're not very good, we have not been very good, and I doubt will be very good at controlling, preventing whatever. But, you know, will people figure out new ways to make livings using this stuff? I think so. And if you had said to people in 1850, when over half the population made its living basically by growing commodities in the soil, that by, that, you know, in the future, less than 1% of Americans would make their living that way. It was where you'd have mass mobs and starvation and all of these things.
01:18:37 A real visionary person might say, well, you know, we're going to be building these horseless carriages, we're going to be building flying machines, heavier than air flying machines, and there'll be jobs. But no one would have guessed that in the future people could make a really nice living, writing letters saying that people's pets were emotional, support animals so that the pets could fly with them on the heavier than air flying machines. It's the inventiveness of people seeing new opportunities and new technologies. Now from a policy point of view, I think we've got to be thinking about ways of facilitating
01:19:23 adaptation. Can we make it easier to start new businesses? And we make it easier to start businesses that have a small number of employees so that somebody with three employees doesn't fall into the clutches of the, you know, the labor bureaucracy of a city that's built for large corporations that have big human rights offices. So how do we, you know, those kinds of things. But you also look, if you can look back at the hills, history of the industrial revolution. And you see how American society gradually developed these
01:20:06 ideas for, you know, how can we support the rise of this new economy? So think of all the departments and universities, the research, whether it's on metallurgy and that side of things, or how about on management and business and finance and administration. None of that stuff really existed as academic disciplines in 1850, right? So how do we start retooling American education, American culture, American government around emerging technologies? I think that's going to be a more fruitful approach. And at the same time, obviously, how do we deal with some of the consequences? is, you know, if the self-driving cars continue to get better and get more numerous, there's going to be a lot of people that are not going to have jobs pretty soon.
01:21:06 And truck drivers don't like it when they can't make a living driving trucks anymore. And bureaucracies, I think. I sometimes tell my students that a bureaucracy is a primitive form of AI in which human beings rather than silicon chips, apply algorithms to data. And that's a pretty automatable process. That's a lot of white-collar workers. How fast all of this happens, how quickly people are able to find new ways of making a living, how quickly we find ways.
01:21:50 You know, specialization of labor, I think, will still be here in an age of AI, like we all could develop agents or we all could have, you know, we all could do everything that needs to be done from our computers with AI, but most of us won't want to. You won't want to be an expert on some, on, on the technology, you want to be able to use the technology or have somebody do it for you. There was a time when, you know, so, you know, so you know, there will be way, there will be new industries, new jobs. We just need to move as quickly as we can to pave the way for them, help entrepreneurs develop it, and then provide a social safety net for people adjusting because, you know, we're not going to all end up having only fame, making money off only fans or being influencers or or these kinds of things.
01:22:50 you know, there has to be a way forward for people so that they'll support the changes and the development that we all need. I love it. Yeah, what about the, like, do you feel like the deflationary effects of AI are currently under-discussed? It's obviously like 90% of people's attention is around the sort of safety debate, but it just feels like we're already at the point where 70, 80% of the time AI can make, you know, have a pretty good crack at like, let's say, like, creating a legal document or doing, doing some of your accounting or any of these things. And it feels like, you know, one of the other side of the sort of potential risks of the labor market is that normal people globally will get access to, the services that have been reserved for higher income people at next to nothing, right?
01:23:55 Like you can imagine getting many of these services on the free tier of various AI products. Yeah, well, look, I think that part, I think, is fundamentally really, really good. You know, the Industrial Met Revolution made a huge proportion of humanity in a growing proportion rich in terms of goods. You go back to 1800, you'd have like, you know, two pots and a pan, three shirts, you know. A good life. Maybe a pair of shoes.
01:24:30 Right? And, you know, those things were expensive. Manufactured means made by hand. They were expensive and built to last. And this idea that clothing was a form of expression, you had to be super rich. for the concept of fashion to have any connection with your life. Yeah. Right?
01:24:52 So right now, I think most of us take for granted a stupefying abundance of just basic material stuff. But on the other hand, our society, you know, there's a lot of acute service poverty in our society. And a lot of fundamental services are super- expensive. So yeah, in the future the average person is going to have the same, you know, access to a great lawyer, or at least the abilities of a great lawyer, at an affordable price. And that is one working definition of economic progress is things that only the super rich used to be able to have. Then the ordinary rich get them and then the affluent get them. And it just percolates on down, I think of wedding planners.
01:25:48 When I was a kid, you sort of had to be a Rockefeller or a movie star for a wedding planner to be a feature of this. It is now, really, that profession is a fixture in people's lives. There's lots of things like that. I wouldn't be surprised in the future if you have your own, like, stylist. Everybody has a personal stylist who knows you, you know, has got access to everything on the internet and AI about styling and sourcing and stuff and is making individually made. You don't go to the store and buy pre-made clothes. You get things that are made for you, for your look by your person.
01:26:37 This is Utopia. I love it. John's excited. And the same, you know, with your nutrition. No, completely tracks, yeah. So this all I think is there and can happen, you know, this is get we got to get out of the cocoon and flap our wings. I love it. I like the analogy. Last question. Do you think the quote unquote experts can make a comeback? There's I feel like we're at an all-time low of trusting experts across basically every domain, economy, geopolitics, et cetera. Do you mean institutions? Institutions, but also
01:27:18 now I feel like if you see something like an article online where it's like 24 out of 25 experts say this, then... I mean, there's even skepticism about like what the president says. He says he's going to do something. And people are like, yeah, you know, maybe. It might have been a joke. Who knows? Might have been AI. It might have been fake.
01:27:39 But, yes. It applies like all over the place. Yeah, it is applies. It applies everywhere. Right. Yeah. Now, this is, you know, one of the key things in that mid-century model of the American system was between the plutocrats and the very wealthy people and then, you know, the sort of poorer people. You had that, you had that college-educated, objective stratum of experts, managers, administrators who, who,
01:28:08 allegedly, you repress the greed of the plutocrats, because if they get everything they want, you know, it'll cause so much ruckus, the country will fall into socialism. But on the other hand, if the unwashed masses can do everything they want, goodness knows what they'll get up to. And we have to, so you have to have prohibition, and you have to regulate what they see and in many ways what they do. And that was the job of this kind of upper middle class group of reformers, progressives, and so on. And their role is being fundamentally challenged in all kinds of ways. One of the things is that in the cocoon era, where society was relatively stable, and we sort of thought we had the answer to everything. Like, okay, the economy, you have a recession.
01:29:00 All right, let's just read up here in Keynes, Chapter 6. seven, it says do this, so we do this. You know, we allegedly knew what the answers to all policy problems basically were, and the job of politicians was to sort of summon up a political coalition that could do enough of the right things so that we would move forward. We're now in an age where nobody knows the answers, like your question about AI. Who really knows what we're going to be dealing with in five years? You know, nobody.
01:29:40 Who really knows how to solve the problems of a technology that nobody understands? Nobody. Who knows how to manage the geopolitical competition between the United States and China with all of the intricate interdependencies and so on? There's not a rulebook for much of this stuff. But for the last, you know, this is part of my idea of the, of the fertility of the soil wearing out, the university culture, the intellectual culture, has become much more wonkish, much less the creative innovator or the, you know, deep-souled investigator. but it's the one who can get the good grades, get the right answers on all the tests, and is expert in the methodologies, the received methodologies,
01:30:37 and knows how to navigate the conventional wisdom. And that works as long as society is in some sort of a stasis. But when it isn't, and these waves of change are coming through, the experts aren't as good as they used to be. They don't know the answers, and as individuals, as people, many of them just don't have the experiences or the personal qualities to deal wisely and effectively with the uncertainties that they face. Yeah, I never was sure that if it was just a function of the Internet where historically you would receive media and there wasn't a public forum on which to push back or to fact check or to see fake news that sort of disagrees with the expert. but yeah your explanation makes a lot of sense and then I'm sure it's you know accelerated by by the sort of current public forum but this was fascinating Walter thank you for the time thank you for taking the time to talk to us and thank you for writing in the Wall Street Journal we read the Wall Street Journal every day here in the show and we appreciate all may your tribe increase keep reading keep subscribing yes yes yes we are cheers Walter have a great rest of your day goodbye
01:31:52 me tell you about railway railway is the all-in-one intelligent platform provider use your favorite agent to deploy web app servers databases and more while railway automatically takes care of scaling monitoring and security we got a lightning ground because we went over time with walter russell made we got cat cole the CEO of a g1 here with us for the second time welcome back to the show how you doing hi i was enjoying listening to the cocoon era and the pleurocrats and the history lesson are you in the cocoon era are you expanding and spreading your wings spreading your Retail footprint?
01:32:23 Butterfly. Butterfly era. Is this what's happening? Give us the latest. We have, you know, just continued the march to meet customers where they are launching additional products, new versions of AG1. We launched AG1 Pro with creatine and HMB for gut and muscle support. It's been a sellout success.
01:32:43 We can't keep it in stock. I think that is a real indication of the momentum coming from GLP1s and people wanting more gut and muscle support, but also the longevity movement. Women, men, all ages from younger to older, are really gravitating toward these more advanced foundational formulas. We launched gummies, one of the few NSF for sport certified gummies, like one of the most honest gummies out there that's clear about what's in it and third-party certifying the label claim.
01:33:13 And this week, we continue our march from Costco and Target into Walmart, Kroger, and Harris Teeter. will be at almost 10,000 doors, brick and mortar retail by the end of this year. Is AI good for you because people are asking about their health, getting more ideas, and AG1 has been around for so long that you're the first thing the model reaches for as an example? Or are there new challenges because people are going to be vibe coding their own formulas and buying raw ingredients by the ton and mixing up their own vitamin mixes? I don't know. Yeah, I've got a humanoid back at home. that's just mixing, making...
01:33:53 I mean, you talk to some software, people in San Francisco, and they're like, yeah, I will actually, you know, make my own CRM. So, you know, maybe people are making their own. Yeah, look, the answer is yes. Yes, both. Okay. There's definitely continued lowered barriers to entry to make a thing, market a thing, stand up a site.
01:34:12 But for us, being 16 years in and actually having more peer reviewed and published human clinical trials and a lot of, lot more documented research over the years on both our ingredients and our finished formula. What's interesting is we're moving from just the marketing era where the hype, the fear, the spin, the influencer helped get you to try something. Now to the agentic search era, which is actually pulling and referencing those peer-reviewed studies. So we're seeing more people saying, oh, I was searching which multivitamin or which gut health
01:34:48 product has the most clinical research and AG1 popped up and people will say I didn't know it was a gut health product and actually several of our peer reviewed and published studies are about the outcomes, the beneficial bacteria, the short chain fatty acids, etc. So on the pure customer education and clarity side, you all were talking about trust with your previous guest, I think it's a good thing because people can cut through the noise, but we've also all had the experience where depending on how you ask the question, you get one answer, and then you correct it, and it says, oh, that's right, you're right. Here's a different answer. So our multi-million dollar, multi-year investment in scientific human trials, peer-reviewed, placebo-controlled and randomized,
01:35:36 takes a while for that to provide a real ROI because not everyone knows how to read these studies. Not everyone wants to go deep, but they do want to understand who to trust and what to trust and the agenic search is providing a connection between a customer's questions and the actual evidence that a brand or a product has. Yeah, so many brands don't have any specifically in health and supplements. They have no substance, right? So there's like there's an ad, there's a landing page, there's some product details, but like the next five questions that you might ask to assess the product, there's actually no,
01:36:10 there's no answers. Like you can ask as many questions as you want. you'd probably have to sit down with the founder and try to tease some of the stuff out. Because a lot of it is just not online. And the honest truth is that a lot of it is just, you know, they went to a co-manufacturer and the manufacturer. So here's the product that we make. We promise that it doesn't have too much of these things in it. And then the brand says, we'll take 10,000 of them.
01:36:36 Thank you. And that's like, that's interesting. It's interesting as we bridge into retail. And we launched for 15 years with just D to C, then launched on Amazon. We're having absolute banger Amazon Prime Days, which is rad. Right now, so it's great. But right as we are continuing to expand our portfolio
01:36:55 and innovate with Costco, who we launched with first, and innovate with Target, and now launching where America shops, which is also Costco, Target, Walmart, Kroger, vitamin shop, et cetera. And the very thing that drives retail, velocity, and repeat, is the same thing that drives sticky D to C high LTV customers, which is great experience, great recommendation,
01:37:22 but is demand, the actual demand generation that is increasingly coming from agenic search, so what we were talking about, social commerce, TikTok shop, tribe, others. They're not just transaction platforms, they're discovery platforms. So when you put together, social commerce is a platform for discovery.
01:37:43 Agenic search as a place for research, customization, and understanding the truth of a product. If the questions are asked in the right way and if the brand has their information published, that isn't only good for D to C or for a brand. It's what retailers want. All the retailers are telling us, they're saying it in their earnings calls. We want brands that have real demand driven by modern tools in today's approach to demand generation and content. And so it's been really fun having multiple channels now with these incredible retailers as a place to drive return on the investment into new ways that people can discover our product, which then drives more return on the now five years of additional human clinical trials that we've been making that are the right thing to do. They provide trust and confidence in the product.
01:38:37 There's certainly a great pillar to stand on when someone has questions about the product. But now increasingly, they're actually, the science is the marketing. Sure. The science is the search result. And that drives pull through to multiple channels. That's fantastic. How efficient do you think age one can get? Like, do you think you can, is there a path to 10xing revenue while, you know, one and a half xing head count?
01:39:01 Like, how are you thinking about, you know, these new, new tools? And just with how much consumer brand, like how competitive the market is and how much you need. to, theoretically you want to be spending every incremental dollar on demand generation. Absolutely. Absolutely. And look, the more overhead, the more fixed cost or carrying costs you have, that's a dollar you don't have to invest in, not just demand generation, but also innovation and research. And so the opportunity that we started about two and a half years ago bringing in
01:39:33 agentic tools into the business to not only manage costs, yes, it can keep us from having to invest in and hire more people. It also makes the people we have far more efficient. We started with customer service and agentically enabling the customer interaction experience, not all of it, but the things that make sense. Then moving it into our supply chain traceability sourcing. AI has been an incredible enabler to efficiency and speed in traceability of raw materials and ingredients. And then using it for science. We have an incredible partnership with UC Davis. using their AI tools to help us understand interactions and chemistry and future more complicated formulations more easily.
01:40:18 And so when you add that with the more obvious and commonly talked about, agentically enabled marketing, whether it's content creation, content iteration, content publishing, conversion testing, you see all kinds of tools coming up helping companies go faster, the reality is we can and will continue to operate in a more lean structure. So more of those investments can be made in product innovation, human trial research, and go to market. What's your honest review of AI customer service tooling today? You don't need to name your partner. But I've been amazed at how bad some customer experience solutions are today,
01:41:03 even though you've had this massive progress in other domains like coding. I haven't been like feeling the acceleration where if I'm chatting with a brand about an issue that I'm having, like it just doesn't feel like there's been that much, nearly as much progress as I would expect or hope. But I'm curious how you see it. I'd say it's been fantastic for what we needed. So as a up until our retail launch, we were primarily a subscription business, right? 90% subscription, massive D to C business. So a disproportionate amount of our customer interactions were about the order, the shipping, the transaction. That is beautiful for agentic replacement. It's very technical and fact-based.
01:41:46 So our customers have given us higher customer service scores, higher ratings on the interactions, but we limit the tool to those. When it gets into, hey, I have a question about my experience or a recipe or how to mix this or things that might creep into more nutrition and medical territory. There's some AI enablement, but we have a very clear cutoff and escalation process of when this needs to be much more human and real time. And that's how we've navigated it. So I'm super thrilled and so are our customers with the interaction with agentic customer service. But that's because we're not using it for things that, to your point, I don't think it's quite ready at scale for. Fantastic. Last question on that. With hopefully a lot of continued
01:42:33 progress, do you think that the amount that you're spending on customer service, let's say in five years will be similar to today because you get the benefit of sort of increased capabilities or like a big company would have historically you sort of need to scale customer service linearly with revenue and it's sort of like a fixed percentage of revenue. I like because what I'm wondering is like in the same way that you can basically keep OPX flat, you know, a lot of brands are keeping OPEX relatively flat, like how will CX fit into that? Yeah, I mean, for us, I would see OPEX being flat to down, both because we can control costs and operate more efficiently to your previous question, but also more efficiently drive
01:43:17 profitable revenue with these tools. So just the way the math, maths, is more advantageous and allows those investments in things like research and innovation and marketing. But there's another aspect to still, even as we're expanding into retail, and there are many incredible stories of brands that were D to C native that beautifully expanded into retail, and you go from being 100% D to C to one third brick and mortar retail, and then to half brick and mortar retail. And when our customer experience starts to get spread across maybe a Walmart that has an AG-17 count, a Costco that has an AG-140 count or AGZ, and our subscribers, that have access to the full breadth of the experience.
01:44:04 Customer service, I would redefine it as customer experience and retention. So we may not spend less on an absolute dollar perspective, but instead of it being answering people's questions and fixing problems, it's going into rewarding their loyalty, providing benefits of membership, actually creating a better customer experience, whether it's better frothers, shakers, AI tools for nutrition and education. So we will spend less on traditional customer service. and get incredible operating leverage because of AI tools,
01:44:33 but we'll reinvest that into a much better customer experience. So people do stick with their routines with the most trusted products that they're choosing to buy and try. Well, thank you so much for coming on the show. Well, said. Yeah, thanks for having me. Great to see you. We'll talk to you.
01:44:48 Good luck with the rest of Q4. Talk soon. Thank you. Go bye. Let me tell you about the New York Stock Exchange. Want to change the world? Raise capital at the New York Stock Exchange. Our next guest is terrible.
01:45:00 Eric from Stute, co-founder and CEO, here with a massive series B. So the thing about this show is we only care about how much money you raise. So tell us about the round and let's start there. Damn. And then we can't believe I made it on. It's not enough. $52.5 million. Great.
01:45:23 Whoa, I watched the video and I didn't get that information. That's the information I wanted. If you haven't seen the pitch for the video, the pitch in the video is you talking to someone at a convenience store and you say we raise $52 million series B and the guys like I don't care. And it's a good encapsulation of like, you know, you want to focus on the actual value that you're bringing. But when when people peel back the onion, they go a layer deeper, how are you actually pitching the business these days? Yeah. So there's two things. One, everybody's video looks like a hostage video.
01:45:59 They're sitting in like a hurdle neck, just like this. Yeah. And then it's an old school eastbounded down where you got fireworks. People are hitting samurai swords. It's just you don't even understand what they do. Yeah. Wait, can we start this over and get some audio because it's just too good? Yeah.
01:46:14 This is the Stute launch video. Hey, boss. What's up? We just raised $52.5 million for my company, Stu. Big sigh. so good so good uh we're we shout out to the guys at day job yeah for putting that together our dear friends um i needed this i'm i'm so yeah you would call my least favorite my least favorite format for a launch video is like you know mashing together a bunch of like inspirational historical scenes and and
01:47:01 calling it uh a brand asset you know it's just been done like you know thousands of times now it's it's time to move on uh there's certain people who will not move on. They're just going to keep hammering it. But this, but, but, but this was good. And I think it speaks to your entire approach. Why do you still have a dot AI domain? Someone's sitting on stute. I feel like, oh yeah, they are. They're sitting there. They're sitting there. I'm not going to give secrets out, but we don't have enough money to buy the dot com. So I'm praying. Pray and nobody buys that dot com. I think you might, I think you might be safe. They, they, they, they have it listed here for eight hundred, eighty, eight thousand dollars. Also, also, if somebody
01:47:41 buys it. Best marketing ever. To just change the company name to be like, how desperate are you to try to compete with us? You got to buy our domain. That's insane. Yeah, yeah. No, I think you're safe. It's a very unique domain. What kind of progress have you guys made since the last time you were on? Yeah, I appreciate you asking.
01:47:57 We've pretty much 8xed our customer base in less than 10 months. Wow. Do you guys remember Stu? Stu, 5% of companies leak revenue by not being able to collect what they sold. We fix that problem with the digital co-worker. So, you know, you think companies like Honeywell, Zoom Info, Perkinelmer, Wells Fargo, Medtronic, like big businesses.
01:48:18 We reduce their DSO or overdue invoices by 47%. And we just help match your cash, collect your cash, and handle any type of issues. Yeah. What's been the top of funnel? Like, how do you actually onboard customers? Is it all outbound? Do you have a big Salesforce? I mean, nothing beats a cold call.
01:48:36 So that's probably 50%. Sure. The other thing is, you know, I was listening to AG1 speaker, right? Like customer success tools stink. And so we're bringing customer success to financial works. Sure. And so we have to earn a little bit of trust. So a big part of our distribution is partnerships.
01:48:52 Yep. So FISA, Microsoft, Ernst & Young, people like that have been really helpful in distributing this out to their customer base. Fantastic. Well, you can get started at stute.aI, correct? Yeah, yeah, yeah. That's STUU-U-T. Go check that.
01:49:07 I hate to say it. You got to just, you got to do it. You got to get the domain. Jordan's a big dot-com guy. You're trying to. I'm a big. I'm a big.com guy, and I just think you guys are trying to sell the outcome, not the AI. Yeah, that's true.
01:49:20 That's true. And I feel like that way, when we saw the day job, we saw some of the early work Ryan was working on and the team over there was doing. And I loved that they had positioned the whole brand to not be like AI focused because the customer ultimately doesn't care about that. They want the outcome. And so the dot com, you deserve the dot com. I'll give anybody a finder speed who helps us get the dot com, by the way. Oh, okay. I think you got a buddy out there.
01:49:44 Yeah, let's put the bat signal up for your buddy. Great to see you, dude. Thank you for all the progress. We'll talk to you soon. Goodbye. Let me tell you about MongoDB. What's the only thing faster than the AI market, your business on MongoDB? Don't just build AI.
01:49:59 Own the data platform. That powers. We are running so behind. I think you might have missed the schedule. might have been a part of that. I was texting you like, we got four more guests, folks. We got a bunch. We got Scott Morton from Revel. Scott Morgan.
01:50:14 Scott Morgan. Scott Morgan is ready to break his silence. Is coming on too. Do we have anybody? We can just let it do everybody anybody here. Anybody who's been invited to TVPN, if you're in the Zoom waiting room, just hop in. Put them all in at once. I want people pitching each
01:50:30 other. I want to see Terrick from Stute pitch cat Cole his product for AI invoicing. The lights red, green, what are we got? Anybody coming in? I'm going to tell you about Shopify. Shopify, it's the commerce platform that lets you grow with your business,
01:50:49 it lets you sell in seconds, online, in store, on mobile, on social, on marketplaces, and now with AI agents. If we have one more, we're dropping the nuke again. Boom. Whoa. Scott Morton. What's up, guys? Nope.
01:51:03 We got Tyler. Tyler, welcome to the show. Live run. Great to be on. Where are you live from? A beautiful vineyard. Yeah, I'm up in Sonoma County at Reservoir Farms at Dutton Family's Vineyard where we're running five different robots at once. It's a pretty exciting time.
01:51:19 Break it down. What are the five robots? Small world. I grew up like very close to there. Oh, really? Yeah, I know the, I think I know the owners of the farm. So there you go. There you go.
01:51:32 Very cool. What are you cooking? Yeah. Show us some of it. Robots. Yeah, most definitely. They are starting up right now. So Bonzi Robotics is a full-stack physical AI company for the rugged world. We build retrofit autonomy and autonomy-first form factors. These form factors don't just replace labor. They have to cap-ex cost and drastically reduce the optics as well. Here's one running right here. This is actually a 60-foot boom hybrid electric machine. Whoa. That's cool. This is doing open field burn down application. So big, big robot running. We got some other ones. to start up right now. In the background, you'll see our John Deere retrofit kit, so lots of different opportunities, all powered by our world model and foundation model. So we have a world model with one stack
01:52:17 that goes on all these different platforms and all these different environments. Coming up right here is a hybrid electric. We got robots everywhere. There's so many robots. Whoa. It's pretty cool. We are able to, we have 430 machines deployed our first commercial deployment for last year. These aren't just, you know, prototypes. These are real applications deployed out the farm impacting farmers in agriculture right now. And now we're getting pulled into other verticals. It's amazing. We love it. How's it been working with Nvidia? You're working with Nvidia. We were just talking about the Jaguar Type Zero One. They're partnering with Nvidia. And we've all been using Tesla full self-driving. It's incredible. And we're waiting for this to come to every car company.
01:52:58 But I'm wondering how your experience has been. Invita has been amazing. I worked at this early. ag company called Blue River, right when NVIDIA came out with the first GPU, the TX1, we pivoted into deep learning. It's been now 14 years, 12 years since that happened, and NVIDIA has just been an amazing partner, not only powering our robots,
01:53:18 but also enabling us to build a world model. We trained a world model on Cosmos 3. We have a million acres of data collected. Cosmos 3 is really good at OnRode. With our data, we were able to put it in these outdoor environments. They've been amazing. It's loosely like, yeah, fine-tuning, adapting a model,
01:53:34 building on everything. So do you need to do you need to do the full stack? Like there's a million companies trying, hoping and praying that like they can build a model that you can use and your robots and you can, you know, deploy them and have the end customer relationship. But why so far have you, you know, focused on, you know, building your own models and your own hardware and this sort of go-to-market motion? Yeah, no, it's a great question. So we believe fundamentally going retrofit and our own form factors, there's so much more than just a model to power the outdoor world autonomy. The user interface, knowing the customer, everything is very important, and you have to customize it accordingly. Early on, we collected a crazy amount of data. That's the
01:54:17 most valuable asset. We're now enabling us to build this world model because we had this copious diverse data set. We believe we have the most diverse data set in the rugged world. So we wanted to own the full stack, right? When we control the world model, we control the user interface and channel customer experience and the robots, that's the big picture for these verticals from our perspective. What is the future of working on a farm? Is there like one person on site right now, control, are they able to effectively control or monitor all the robots at once and intervene as needed? Yeah, most definitely. So we have, sorry, there's a loud robot, a big one coming now, so we might pan to that, but we have a, uh, whoa.
01:55:02 We have a full operating stack, right? The user experience is everything, right? One person is controlling all these machines, managing A to B navigation, all the operational intel, the amount of spray, the models. You have to break it down so that anyone can run it. The one person is controlling this whole fleet. That's what makes the ROI, that's what makes the actual value accrue.
01:55:23 So it's a pretty exciting time for us where we have one model, one person across all these different platforms. It's happening now for the first time. Absolutely wild. This is one of the greatest shows we've ever got on TBPN. Thank you for the tour. No one's ever gone six robots. Yeah, yeah.
01:55:41 Maybe we got one. We got 430 running, so we could ramp it up next time, I guess. Wow, that is impressive. Well, congratulations. Very cool, Tyler. So much. Great to have you on. Come back.
01:55:51 We're going to hold you to like at least 20 robots the next time. So you set up a pretty high bar for yourself. I want a fast takeoff. Well, we'll do it in a new vertical. How about that? I love it. I love it. Awesome. Have a great rest of your head.
01:56:02 Great to hang, Tyler. Hey, thank you guys. Pleasure. Let me tell you about Figma. Agents meet the canvas. Your AI agents cannot create and modify your Figma files with design system context. On to our next guest. It's a surprise guest.
01:56:16 We don't know who's coming in. We're going to figure it out. Here he is. Introduce yourself. Sorry, chaos over here in the Ultradom today. We were talking to Walter Russell Mead, and every question, I took a lot of, us a minute to get the question out and then he would take us on this journey for five minutes. And so we got behind schedule. But thank you for being flexible. Thank you for
01:56:36 hopping on the show. How are you doing? Good. Yeah. Good to be here. Fantastic. What's new? Man, since last time, was that February we talked last? March. February 26th, 2020. It's been almost. Last time we talked, it was like 320 days till Christmas. Can you believe it. That's hard to believe. Now we're only 78 days away from Christmas. So just to put things in perspective there. Yeah. I mean, we've, so we've taken a nuclear reactor critical with all atomics with our software. Team has over doubled in size. We're at about 70 folks now. We've also deployed to several massive industrial scale facilities. It's like 10 times what we've done before, which is like our kind of rocket engine test site. Yeah. Yeah. It's been a
01:57:27 heater of a six months so far. How custom are your deployments? I feel like in the age of AI, being very custom for deployed engineering is very popular. It feels like, there's always a problem. You know, the sales team is like, hey, I accidentally promise that we build this thing. But now I feel like the customers know that they can probably ask for a lot deeper customers. Yeah. So fundamentally, we're providing a platform for testing, controlling physical systems.
01:57:55 So think like, you know, as I said, nuclear react. jet engine test facilities, rocket engine test sites. Sure. We enable teams to go much faster, much further than the same amount of time. We're replacing, you know, technology from the 80s and 90s, and you have these companies building hypersonic jets and they're stuck with like pre-Winds 98 era tech. But yeah, the platform generally enables the engineers, especially hardware engineers, to be fully self-sufficient. So really on our end, there isn't a ton of kind of forward-deployed engineering work, at least in terms of like implementation. but we do help them kind of get set up, get rolling, and then, you know, help with any issues that they run into both hardware and software.
01:58:35 How many, how often are you hearing from entirely new startups in either the L.A. ecosystem or just the deep tech ecosystem. It feels like we went through this crazy boom and maybe it's cooled off, but are you seeing like continued acceleration and new company formation? We hear from people from working on all kinds of different hardware systems. Yeah, I was looking at the SoCal hard tech map from 2024 to 226. And it was like 130 companies. And now there's 279. So, yeah, no, I don't see any stopping to this, especially SpaceX going public and Andrews doing super well.
01:59:15 I think we're still at the beginning of this thing. Yeah, what are like how much of the, how much of your team is X, X-S-Spaceex and is in like what kind of trends are you seeing in terms of, you know, post-IPO. I imagine a lot of people just want to go and start companies, but some people just want to build stuff that don't want to deal with everything else that goes into building a company. Yeah. I mean, yeah, we have a lot of former SpaceXers at Revel. But generally, yeah, I know we are seeing a lot of former SpaceX founders in the area,
01:59:48 which like, it was great for us because they're like, you know, worked on kind of infrastructure for Starship. They are familiar with some of the stuff I did at SpaceX and, you know, call me up as soon as they're out. And they're, they're ready for some, some Tesla control software. Yeah. How many are you running into people, SpaceX alumni, that are just retiring? I, you know, there's some folks that, you know, they may retire for a little bit. But I think this type of engineer gets bored quickly.
02:00:22 So I think we see people kind of. go through that rotation, come back in after a few months. Yeah, a few months. There was an article in the journal about someone who was at SpaceX, and it seemed like they were somewhat low level, but they, you know, through the IPO, they made enough money to, like, take a year off and just go travel the world. There was a cool anecdote, but yeah, I completely agree with you that, like, you can't last. If you're, if you lasted a year at SpaceX, you probably have the bug, the entrepreneurial
02:00:50 bug, the builder bug, you're going to do something. So it's this funny. paradox of that you have enough not to work maybe. Are you, yesterday was sort of a monumental day in math. A lot of people are kind of wondering how that's translating in real time to progress in other domains. I'm wondering, are you getting surprised at all from your customers at the kind of pace that they're making on the product side?
02:01:19 Are you still seeing companies where you're like, oh, at SpaceX, we would have done that in a month and you took six months? Yeah, I mean, we help them do it in a month. That's what we're all about. I mean, testing at all these companies is critical path. And so, you know, if we enable them to test like 50 times in a week instead of five times in a week. So, no, we're definitely part of the solution there to enable them to move much more quickly. And kind of also, I'd say our platform, I think also encapsulates a lot of like a lot of these hard lessons learned to kind of help them go directly to something that's very mature that they can just
02:01:54 rip with. Fantastic. Sick. Thank you so much for coming on the show. I'm sure you'll have more. Yeah, one more thing. So we've got a L.A. Hard Tech event on
02:02:04 October 15th. I don't know if you guys, you guys have seen everything's on October 15th. Rooster. Rooster is coming out. LA Hardtech. This is a tough house. You're really trying to steal your
02:02:15 former employees thunder. Oh, yeah, going up against Tesla. But who knows? That might be delayed. We don't know. It's been shifting around. But this one's locked. October 15th.
02:02:26 Tell us to about the event. It's with Zinger. So we've got the VMAX and the spider coming. It's going to be super sick. You guys should come. I love Zinger. We had the car here on the show. It's fantastic.
02:02:37 Zinger is a customer then or you guys are just teaming up? Yeah, they are. Yeah. Awesome. Yeah. It's been great so far. And yeah, should be sweet. You know, invites open.
02:02:50 It's on our LinkedIn. If you guys want to check it out. But yeah, it should be a great event. Awesome, dude. Well, great, great update. Come back on before the end of the year. I'm sure you'll have more news, but awesome progress. Have a good message to your day.
02:03:02 Thanks, guys. Goodbye. Cheers, dude. Let me talk about Codex. Codex is a powerful workspace for getting work done with AI agents, whether you're writing code, analyzing data, creating content, or automating business workflows. Codex helps you move projects forward.
02:03:13 Let's give it up for automated business workflows and our next guest. A surprise, guys. What's going on? Welcome to the show. How are you doing? Hey, hey, good in you, Jordi. We're doing great. Sorry for the chaos today.
02:03:30 We're a little bit behind schedule. But thank you for being flexible. Thank you for sticking it out and hopping on TBPN. Why don't you introduce yourself in the company to kick it off? Yes. Yeah, I'm just higher and I'm the CEO co-founder of Health League. Our mission is just to improve the health outcomes of as many people as possible. Sure.
02:03:48 And you raise some money. How much did you raise? Today, we're announcing 38 million back by, supply no introduction needed there we go can we get the replay pulled up that'd be great um this is key boom great with with authority john really coming and hot bam all right all right we can take that down um okay uh improving health outcomes for everyone where do you start that's the question so yeah i a little bit about a quick backstory. I started my first business ventures back in South Africa as a teenager who couldn't afford health care myself. And as a 13 year old, the initial spark was I was selected to represent South Africa at the international math competition, but I couldn't afford my flight ticket. And so I started a little business ventures, taught myself chess, used that as startup capital from winning tournaments, and then made enough money to buy a computer, could do more interesting stuff. And I used some of the profit.
02:04:54 to pay for my own health care, change my own life. But I wanted to, I was left with this kind of meaning-driven search for what's the most important thing I can dedicate my career to. And that's why I teamed up with my sister, who's a clinician. And she actually specialized in clinical nutrition. So we teamed up to start there and identify undiagnosed disease-related nutrition as the first condition we went after and we now expanded. So the basic workflow is partner with a hospital, partner with a hospital network, deploy your solution, scan medical records from patients that come in, detect problems before they become a major issue, things that might have been overlooked by a nurse, by a doctor.
02:05:43 And then you are helping solve those problems, or at least flag them so that a doctor can solve them. but also reducing their financial liability because they might have medical malpractice if I come in and I'm having a heart attack and they didn't pick it up, that's a problem and you help them avoid that situation. Is that roughly correct? Yes. So exactly. There's so many patients who come into a hospital, but they have other conditions as well that might not be detected or even if they are detected, get detected late. And so having this high precision real-time risk screening for every patient every day. across all major health conditions, allows patients to get better care, clinicians to allocate
02:06:25 their time more effectively, and hospitals also get paid for the actual appropriate severity in their population. But overall, the cost of health care actually goes down because people stay healthier. Yeah, that makes a lot of sense. How do you sell the hospitals these days? I mean, you're a smaller company, backed by Sequoia, so I'm sure there's connections, but what is the sales process like? Yeah, a fantastic question.
02:06:47 So historically, companies in healthcare tended to move quite slow. They have these extremely long sales cycles. And so our approach was really, let's upfront quantify the opportunity for how socials. So we kind of simulate back test, what would happen if they deploy us. And then we also contractually go at risk. You have to already have a relationship with a hospital. It's not like you can say, okay, there's this hospital. Let me go generate a report and then send them my pitch.
02:07:16 I think it's more like before asking for a bunch of money, prove that it's going to be looking back. But once you already have a conversation going, that makes a lot of sense. Cool. Yes. And at this point, it's kind of word of mouth. Like we go to events. Hospitals are actually quite collaborative because they like to share the geographically
02:07:34 separate this. So then kind of in that sense, non-competitive. So at conference, like two weeks ago, Penn Medicine went on stage at one of the largest healthcare conferences and just talked about how much a patient health improved by deploying how I'll sleep and then we got a bunch of interests from that. That's very cool. Yeah, it's so funny. There's so many other industries where one business will develop an advantage through a new
02:07:56 technology and AI product and they're like, let's not tell anyone and let's like get the maximum amount of value out of this where it's like hospitals. Yeah. I'm trying to imagine. You're not worried about you're much more positive some. Yeah. No, it's a super positive some future. I can imagine in the future like you go, you, you break your arm or you having a heart attack,
02:08:14 you go into the hospital. And it's automatically flying. lag that you need something done with your upper maxill or like a double jaw surgery or maybe some leg lengthening. John's. John's having too much fun. Obviously, there are really, really positive and impactful. What was your, as a math Olympiad, what was your reaction to yesterday?
02:08:36 Yeah. Well, tell me more. Oh, he's so locked. He's so locked in. I love it. Some math problems got solved. That's all you really, that's all you really need to know. The headline is opening I drops a bomb on maths and, and the actual, like, headline from Open AI is that they, I mean, it really is like hundreds of math problems have been solved.
02:09:04 This is obviously a couple weeks after the Navier Stokes solution and the Hodge conjecture fell earlier. So, yeah, you are right that it's like every week. I guess the bigger question is, like, more broadly. as someone who has played in that space, what has been your reaction to model progress in math? Yeah. Yeah, I think it's fascinating. It's fascinating to see.
02:09:28 I think in some ways, you know, this was kind of predicted. And so I think I'm almost still waiting for AI to get to the point where it can come up with fundamentally new spaces to go and explore. So it's like more efficient. space. Like, move the goalpost. I like it. I see the golf post moving.
02:09:52 But yeah, but I personally expected this and it's funny. I remember the reason I stopped playing chess was I took it seriously for one year, got better. And then I was like, why am I doing this? The computer is doing it so much better. It's exciting to go and find the new problems that humans can now uniquely solve. And I find it energizing. Yeah.
02:10:14 And there's still. And there's still, yeah, so much to be done in diffusion. And this is, I think, what Health Leap is. It's a diffusion company in as much as it is an AI company, right? It's like you need to actually bring the solution to the market integrated in a way that gets a real impactful result that people can actually benefit for. Thank you. Awesome. For coming on the show.
02:10:37 Great to be back on soon. Congratulations. I'm raising. And we'll talk to you. A great story. Have a good one. Goodbye. Let me tell you about console.
02:10:43 Console builds AI agents that automate 70% of ITHR and finance support, giving employees instant resolution for access requests and password resets. Our last guest of the show is not a mystery guest. We have Arru 7. From OpenAI, welcome to the show. He's a chat GPT, product manager, and we're very excited to be joined by him. Welcome to the show. How's your day going?
02:11:06 Good, good. It's good to see you guys again. Yes. Welcome back. It's been too long. What did you launch? Take us through it. Yeah.
02:11:15 So today we're launching GPD6 in chat GPT for all of our 1.2 billion users. Yeah. So a few weeks ago, we announced Astra, which is sort of our frontier model. And then a couple weeks ago, we walked through GPD6, Seoul and Luna. And so today we're actually bringing GPD6 now to everybody in chat GPT. And it's coming with two really cool new capability improvements. One, which we're calling intelligent UI, which is now chat, chat. can control not just what to say, but how to show its answers and can kind of control its interface.
02:11:51 And the second is that chat chattchibati can now answer while thinking. So historically, when you use something like the reasoning model, you've had to wait for a really great answer, right? It can think and do a lot of work, but then the tradeoff comes with latency. And now we've enabled chat chadipdi to actually start answering while it's still continuing to work on the problem. Yeah. And so that really changes.
02:12:12 It's a subtle change, but it's a subtle change. it really changes how people experience chat TPT for more complex questions. Yeah, there's subtle U.S. I can't tell you how many times I'm, like, creating a new chat because I've fired off like two. Two. Yeah, I'm realizing like it's really not that Washington State is or something. You're doing way too much. Yeah, exactly.
02:12:33 It happens all the time. I, one thing that I've always noticed on the product, the product development side is, and the U.S. side is the use of haptics. Like very early on in the chat GPT app, as the tokens were streaming in, you'd feel the little haptics. And I ran into it recently when I asked to pull temperature data over a hundred year period. And I was just trying to get an idea of like, has it gotten hotter in this town more than the global average or whatever? It pulled all the data. I expected that.
02:13:06 But it generated a chart. And the chart was interactive. And as I moved my thumb along the chart, there was a habit. And I was like, wow, that's really, that's a really delightful experience. And I'm interested in how you have puzzled with when do you want to build a widget? When do you want to actually set the application up for success to use basically like a harness, something, a pattern that you've decided versus how much are you giving over to the model over time to say, hey, there's just a whole bunch of APIs. The haptic one is right there. you can pick from it as you please.
02:13:43 Yeah, that's a great question. So historically, we've built these as sort of, you use the word widget, which is kind of exactly right. We've built these predefined cards. Yeah. And the content inside can sometimes be dynamic, like the charts you show we talked about.
02:13:58 But the response has mainly been marked out, right? These chatbots produce text. And then we find ways to sort of hijack that markdown to allow it to reference additional artifacts like a weather widget or a chart or something like that. With this change, so that approach is somewhat limiting, right? Because actually, it should just create the perfect interface for whatever you're trying to do. Sometimes text is amazing.
02:14:23 You're writing something, you're looking for advice. But other times, chat chbtD should really have a lot more control for how it constructs the entire layout and interactivity of your answer. And so with this change, we've actually trained chat chpT to have full control. over what it shows you. And so it can generate a UI on the fly and do it really fast. It's not something that'll take a minute or two.
02:14:48 It's really for every chat in the same latency. And it can figure out what kind of functionality it wants to add on top of that. So some of that will be like interactivity within the chat. Let's say you're trying to understand a topic. It'll actually give you an interactive diagram for you to play with to better understand that. But it can also add functionality.
02:15:06 Right. Like often now, if you ask, for like a recipe, it'll give you a shopping list, but it'll add a copy button. And that's not something we've built in a pre-made way. The model just realizes adding a copy functionality when I give you a shopping list is the most useful thing for me to do. Yeah. And I imagine that there's an interesting
02:15:24 flywheel here where obviously giving this out to 1.2 billion users that are free and maybe not fully monetized with ads just yet is expensive. But there's a lot of value that comes from a free user iterating and saying, no, you should actually have the slider this way. You should add, add haptic feedbacks, add that copy button. And then that goes into the rollout, the model gets better at knowing which tool to pull off the shelf at the right time. Is that generally how you think about this?
02:15:56 Yeah. So for this launch, we really wanted it such that you don't have to ask what you want. Chat TPT is going to try and figure out. Yeah. Okay, you're trying to understand the offside rule in soccer. the best way for me to explain that is not be an instruction manual, but let me just show you a diagram
02:16:14 that you can play with that I create on the flight. And so you can always ask with ChatTBTBT what you want, but one of the big changes is we've actually trained the model to figure out when does visuals and interactivity really make the answer better. And then the other part is exactly to your point of how do we scale this to 1.2 billion users.
02:16:32 You might have seen Astra can produce super cool visuals and websites, right? But that takes a lot of computer. and it's slow. So we've actually done a lot of work. We've actually built our own sort of library in-house that allows the model to generate UI, but in a very token-efficient manner
02:16:50 and to do it in a streamable way. So it feels fast. It works cross-platform, and it's very, very efficient. And so a lot of that is what went into enabling us to scale this to 1.2 billion users without it melting all of our GPUs or us having to charge users, right? So everyone will get this
02:17:07 in a sort of unlimited fashion to sort of build much richer answers in chat. In the 2010s, it felt like product management practices were getting really cemented. Product managers were oftentimes like the superstars of their organization just under the founders today. Obviously, researchers tend to capture a little more attention than you. do you feel like there's like new norms being like it just feels like it's like it's 10 times as hard to be a great product manager now because you're not just working between the customers and engineering but you're adding research and then the the the capabilities of the models are advancing but then you also have this issue of like you know historically if you had a SaaS product
02:17:58 and like nine times out of 10 it did what the user wanted that would be like a failing grade because people would be getting really frustrated. Now you could have a product where nine times out of the 10 it nails it, and that's actually a good thing, right? But there's still going to be some percentage of users that are frustrated. So it just feels like all the sort of the approach is being rewritten, and the sort of in the 2010s or late 20-20s, you had books and podcasts, and it felt like product management was really like being cemented around best practices
02:18:30 that worked in the SaaS era. Yeah, it's definitely changing a lot. I think you sort of have to start from scratch a little bit and be comfortable with nondeterminism and figure out how does product management work in an area of nondeterminism. And even within that, getting very comfortable with e-vals and how do we measure progress is something that's like a really important skill. One thing that's really different is actually like we're used to nondeterminism in text.
02:19:00 Now we're entering nondeterminism in the interface itself. Now it's not just the words are different. Even the way it presents the answers are super different. And so that's also actually going to start changing how we do product development at OpenAI because no longer can a designer build a weather card or a sports card and be like, this is how it's going to appear when it triggers because the model can just generate that on the fly. And so we're actually now also figuring out how do we do design in the age of non-deterministic UI in chat GPT.
02:19:34 And we've actually had to build our own design tools and development stack such that design can now work a lot closer with research on actually steering here's how we're going to teach
02:19:44 better and better design to the model. And so to your point, like product management is definitely changing and I think design is also now changing a lot where design is going to sit
02:19:53 right next to research in shaping model outputs. Sure. Yeah, and that's such a big update versus the prior era of being able to obsess over a certain element of the product for months. And then knowing that consistency versus now, yeah, it just introduces like, well, how do you test this?
02:20:15 How do you collect user feedback? All these different challenges. How do you think about interfacing with other product teams? We had Kath from Chachapit's sites on the show last week at Dev Day. And a lot of the things that you just described as, you know, instantiating UI on the fly. My previous workflow was, hey, generate a site of this, and I'll share that around. But this doesn't feel like it obsoleses sites. It's more like I was driving to work today, and I was thinking, like, wait, am I going to be able to, like, add Oath to these sites and build a database?
02:20:50 Like, sites is going to grow in its own direction, and you're going to grow. And how do you think about working together with other teams? Because there's so many projects like computer use. Like, operator was like a side project that, like, dropped and then came back with a vengeance and these things sort of morph. But what is it like interfacing with the other teams to make sure that you're not stepping on each other's toes, but also setting yourself for a success? That feels like a critical piece of the role these days.
02:21:16 Yeah, it's a great question. I think the first is we run super lean. So, like, there is very little sort of running into other people's toes just because we just don't have enough PMs. I think the one thing that's really cool is with models like GPB, they're getting better. And because the model is better at front-end design and building sites, we've also therefore been able to teach it to do really good
02:21:39 UI generation in chat GPT and make it really good at using our own in-house component libraries. And so one thing is we see really nice transfer of capabilities as sort of different teams make improvements and trying to stretch the model in one way, that then goes and benefits all these other teams. I think the separate thing is really focusing on our users, right? So Kath and others are really focusing on how do we enable developers and then enterprise users to be like maximally productive, right?
02:22:09 And so the concept of a site is actually a place where everybody can collaborate and do work is amazing. And we're going to keep pushing that and making the model better at that. And then the separate thing is, which was what I spent a lot of time on is like, hey, how do you get everything, everyday things done? You're asking about a recipe. We're trying to learn a concept on your phone on the go. and really think about, okay, how do we bring a lot of these innovations, but make it accessible to our billion users, right?
02:22:35 And so generally the ability to improve, like have everybody improve the model, but then optimize our features and kind of the experience of the delivering for our very different user cohorts, right? Because, like, you building sites is just so different to someone who's sort of, like, out there
02:22:54 trying to, like, get to work and catch up on the latest news, right? And you might be doing both of those, you know, throughout the course of your day. And we should give you the best site building experience, and we're going to give you the best sort of everyday chat TV experience. Last question. How dynamic can't, do you think the UI in a thread can get? Because like the peak, you know, chaos would be like you effectively can generate an MMO
02:23:23 and invite your friends into the thread and play it. but how far does it even make sense? That's how I like my recipes delivered. Yeah, yeah, yeah. I want to cook a virtual meal to understand how to cook a real meal. Well, I want to turn my recipe into a first-person shooter. Ideally, yeah. That we can both play.
02:23:41 Yes, that's the future. You guys should play with what we launched today. It can't make an MMO recipe, but it can make it in window. Move the goalposts. We're moving the goalposts. Yeah, wait a, wait a set me up for success. It can't do that. But you want your entire conversation to look,
02:24:06 I message blue or look Windows 95 style. It can do that. It can create a ton of interactivity. You know, it can really build anything and it's going to stream really, really fast. I think the sort of logical extension of this is that chat is going to, be this operating system that helps you get whatever you want done and sort of update itself and be that perfect UI.
02:24:34 And yeah, it should be that if you want a recipe MMO game, we're going to give it to you and we're going to give it to you really, really fast as well. And that's sort of where we're heading. Yeah. Very cool. I'm going to test it on my favorite personal benchmark, surf bench, how well does it give me the surf report? Because that's like the kind of, that's the kind of like prompt that can actually. deliver like you know i i do want you want you want it over time surf levels yeah yeah yeah i might want ui that's more dynamic than just like a chart that shows me different numbers yeah if it if it
02:25:06 doesn't work bring me on slack i will and we'll make it work yeah awesome well thank you so much great to hang great updates on the launch and we'll talk to you thank you guys have a good one goodbye man well that's our show we got to get out of here because we have another show to do uh we've been invited to guest on another show. I'm not going to leak it because there might be some rollout strategy there. But it'll be a lot of fun. We'll promote it when it goes live. But thank you for tuning into TBPN today.
02:25:39 It's been an honor. It's, we're so much closer to 77 days away from Christmas. Start counting the hours. That's right. Because we are now. We got to start doing a countdown to the next. We are now less than 11 hours away from being only seven. 27 days away from
02:25:57 Countdown to the countdown. Countdown to the countdown. So we got to start. Leave us five dollars on Apple Podcasts and Spotify sign up for a newsletter at TBPN.com and we will see you tomorrow. Goodbye.
There aren't comments yet for this episode. Click any sentence in the transcript to leave one.