Apple Pro and Tell, Genie 4, The Gurmanator Joins | Mark Gurman, Zach Rash, Ben Liu, Amiram Shachar, Jasper Zhang

(00:31) - Apple's Pro and Tell Event (16:29) - Genie 4 (25:50) - OpenAI Eyes Custom Google Chips (31:24) - 𝕏 Timeline Reactions (55:14) - Mark Gurman discusses Apple’s latest products, drawing on his expertise as a Bloomberg technology journalist and prominent Apple reporter. He praises the iPhone Duo and OLED MacBook Pro while addressing manufacturing defects, software quirks, Apple’s lagging AI strategy, and its broader challenges with innovation and acquisitions. (01:29:58) - Zach Rash discusses Coco Robotics’ autonomous sidewalk delivery robots, including its fleet of roughly 1,000 units, UK expansion through DoorDash and Deliveroo, and plans to manufacture 20,000 next-generation robots for 2027. The co-founder and CEO also explains Coco’s fleet operations, regulatory challenges, AI benchmarks, and goal of safely navigating complex urban environments at scale. (01:40:28) - Ben Liu discusses his role as co-founder and CEO of Formation Bio, an AI-native pharmaceutical company focused on making clinical trials and drug development faster and more affordable. He explains how its $600 million fund will finance promising individual drug assets while using AI, specialized talent, and a hub-and-spoke model to bring more medicines to patients. (01:52:37) - Amiram Shachar discusses Upwind, the cloud and AI cybersecurity company he co-founded and leads as CEO. He explains its platform-based security approach, the team’s cloud infrastructure expertise, AI’s impact on software development and cyber threats, and the company’s recent $400 million Series C funding round. (01:59:22) - Jasper Zhang discusses founding Hyperbolic, a unified platform that aggregates and orchestrates global GPU compute, allowing customers to flexibly access, manage, and resell unused capacity. A math prodigy who completed his UC Berkeley PhD in two years, Zhang also explains Hyperbolic’s broad hardware support, strategic investor relationships, and recently raised $30 million. TBPN is made possible by: Ramp - https://ramp.com Public - https://public.com Cisco - https://www.cisco.com Console - https://www.console.com CrowdStrike - https://www.crowdstrike.com Figma - https://www.figma.com MongoDB - https://www.mongodb.com NYSE - https://www.nyse.com Railway - https://railway.com Shopify - https://www.shopify.com Codex - http://openAI.com/codex Follow TBPN: https://TBPN.com https://x.com/tbpn https://open.spotify.com/show/2L6WMqY3GUPCGBD0dX6p00?si=674252d53acf4231 https://podcasts.apple.com/us/podcast/tbpn/id1772360235 https://www.youtube.com/@TBPNLive

• 0:01Speaker 00

  Today is Wednesday, October 28, 2026. We are live from the TVP and Ultravol, the temple of technology, the fortress of finance.

• 0:10Speaker 01

  The capital of capital.

• 0:12Speaker 00

  Let me tell you about ramp.com. Time is money. Save both. Easy use corporate cards, bill pay, accounting, and a whole lot more, all in one place. How was your morning, Jordy?

• 0:23Speaker 01

  It was great, but this morning is all about you.

• 0:26Speaker 00

  Mine was even better because I got.

• 0:28Speaker 01

  Well, I was getting a fantastic workout.

• 0:30Speaker 00

  Oh, yeah?

• 0:31Speaker 01

  You were at some computer event.

• 0:33Speaker 00

  I was. I went to the Apple event this morning. Pro and tell, very exciting. Speculation was all over the place. John Turnus shared a video of a, an cutout, a spotlight of the Apple logo shining on a wall brightly. You can see this video teaser. What do you think he's teasing? Christian Reber times in? Apple car. Obviously, they're just going to drop the apple car on you.

• 1:00Speaker 01

  Out of nowhere.

• 1:01Speaker 00

  On a Wednesday in a 10 minute keynote. No, it was a touchscreen, touchscreen MacBook Pro. And it was cool. It was, this was my first Apple event in person. Obviously, I've been watching keynotes since I was a young child. I remember watching like the Steve Jobs keynotes in college. It was amazing. And seeing like those clips, we've watched a bunch of these on the show. And we've seen the evolution, the COVID era, the highly produced big keynotes. This is a smaller event. You know, they're, they're doing a number of launches. They launched their iPad. But man, John Turnus, the star power is incredible. I'm not joking. Like, he is so good on stage. He's incredible.

• 1:42Speaker 00

  What are you trying to hit the guys?

• 1:47Speaker 01

  No, I was hitting the, the chat button.

• 1:49Speaker 00

  Okay. I was like, why can't I hear? No, he's really, really good. So, I mean, the whole script was very tight, very quick recap on what they've launched so far. Like 15 seconds iPad mini, 15 seconds on the MacBook Pro with the M6. And then straight into the touchscreen, they play a little example video, which had a hilarious theme to it, which was if you give a mouse a cookie, they took the story of if you give a mouse a cookie and apply, we, we should actually play the, the full ad because Tim Cook shared it on X. He said, we brought touch to the Mac. We brought touch the Mac from the new iPad, from the all new iPad mini to the MacBook Pro. Let's play this.

• 2:33Speaker 02

  Like a Pro OLED. Yeah, a Pro LED XDR. So it'll have more vivid colors. Brighter brightness, blacker blacks, razor sharp contrast. Then you've got a totally portable reference monitor. So you'll need anti-glare and reflection, burn and protection, and real time pixel correction. It'll look so real, you'll want to reach out and touch it.

• 3:09Speaker 00

  Boom. Mic drop.

• 3:11Speaker 02

  And you can touch it, you can swipe and drag and pinch.

• 3:15Speaker 00

  No, they clearly listened to you. You were making fun. It's always just thinner, lighter, faster. Boom. You can't say that.

• 3:21Speaker 01

  Our most touchable.

• 3:22Speaker 00

  Touchable, touchscreen, that's not thinner, that's not lighter, that's not faster. That's a different vector of new features.

• 3:28Speaker 01

  Our most touchable.

• 3:29Speaker 00

  Okay.

• 3:29Speaker 01

  MacBook Pro yet.

• 3:31Speaker 00

  It is. No, I have no idea if this will be useful or I will want the touchscreen version. I can't imagine a situation where I want that. And I would be very worried about fingerprints. I'm already like constantly battling smudges on my screen. And I don't really like using window, wind, windex cleaner. I guess I should get the official apple polishing cloth, which is like.

• 3:54Speaker 01

  Full employment for computer detailers, which we've pitched.

• 3:57Speaker 00

  Yes.

• 3:58Speaker 01

  A bunch of time.

• 3:58Speaker 00

  No, no, I, I agree.

• 4:00Speaker 01

  As a great business.

• 4:01Speaker 00

  Yeah.

• 4:01Speaker 01

  At least in San Francisco, you know, schedule different companies, go by the office, take the computer for 10 minutes.

• 4:08Speaker 00

  We actually haven't solved that here. Like, we should, we should figure.

• 4:12Speaker 01

  No, you tried to clean your computer and your hands were burned by like a very toxic, something off of Amazon.

• 4:17Speaker 00

  Nick was really skimping on the, on the, on the cleaning supplies. And I got chemical burns all over my hands. No, we should, we should figure.

• 4:24Speaker 01

  So, so yeah, my reaction to the launch is like, why the heck not? You know, you should be able to touch the screen of your computer.

• 4:32Speaker 00

  Yeah.

• 4:32Speaker 01

  Why the heck not? Seems, you know, seems pretty cool. I don't, I don't think I'll care one way or another when I, when I upgrade my Mac next, if it has a, if it has a touchscreen or not.

• 4:46Speaker 00

  Yeah. I don't even know if it's an option that you pick. The, the actually bigger thing that's an update is the, is the OLED display, which if you, if you've watched an OLED TV, like it is brighter, the blacks are darker. It's actually a nicer contrast ratio. And a lot of people have been upgrading to OLED TVs. And they're like, yeah, this is better. And so bringing that to, to the MacBook Pro makes a lot of sense. It's been.

• 5:11Speaker 01

  Yeah, it's interesting because it, you know, the Mac, the, the Pro line.

• 5:14Speaker 00

  Yeah.

• 5:16Speaker 01

  You know, historically was for professionals. Still is. But I don't think any of the, like, the most hardcore power users are like, yes, like, thank you for delivering the thing that I cared about.

• 5:29Speaker 00

  On the touchscreen or on the OLED? I, I, I do think video editors, anyone doing important graphics or color correction, especially if they're delivering to an OLED TV, like if you're, if you're editing Netflix, like you know that a lot of people, so the idea of a portable reference monitor, most people that are doing professional work would normally have a separate monitor to see how it looks on an OLED. And now it's just on the main screen, which is great. And like, you will be able to feel the difference in the screen quality. The touching thing, again, like, that's a very specific use case to be playing piano on your, on your MacBook Pro. But yeah, why not? Fun, fun teacher, fun feature. They already have the technology, super mature in the iPad and the iPhone. Why not put

• 6:14Speaker 00

  it in the MacBook Pro? Yes. But I would be, I would be surprised if apps are like scrambling to, you know, take the most advantage of the touchableness because who, who, who's really going for it?

• 6:26Speaker 01

  What did you learn about the hinge? Are we going to be experiencing this?

• 6:29Speaker 00

  Yeah, yeah, no, no. I, I think it is better. Is that an Air or a Pro?

• 6:33Speaker 01

  Pro.

• 6:34Speaker 00

  Okay. Cause mine's kind of similar. When I moved it around, it, it seemed not to do the springiness thing. You see how like, when I do this, it like springs back and forth? It seemed not to do that. So it seems like they improved the, it seems like they improved it.

• 6:46Speaker 01

  Cause if you have to hold it, reinforce it while you're touching it, it's going to be a little rough.

• 6:50Speaker 00

  Well, I think the default, the default way you'll use the MacBook Pro touchscreen will be like this. You put both hands. Then you can scroll Instagram and TikTok right?

• 6:58Speaker 01

  And you're like dri, like, like you're driving.

• 7:02Speaker 00

  Yeah, like you're driving. Like a huge iPad. I don't know. But John Turnus, very, very good on stage. Whatever he's been doing, practicing, it didn't, I think he has confidence monitor with maybe some script, but like, it felt very memorized, like, and delivered perfectly. They really take that seriously. And we've seen a lot of keynotes. I mean, the Tesla Roadster is probably the best example where there was like a lot of dead air and like Elon was sort of riffing and deciding the top speed live on stage, which is cool. It's like, it's on thing. But Turnus was very, very dialed in. Like, he knew exactly what word to say, where to put the emphasis, how to deliver it effectively. Really, really great. The only problem, little underdressed. So, you know, Steve Jobs is going out with

• 7:44Speaker 00

  the iconic black turtleneck. I would have liked to see him in black tie. I would have liked to see him with some white gloves, maybe some tails, maybe a bow tie, maybe three piece suit, something like that, top hat.

• 7:53Speaker 01

  Really make it about him.

• 7:54Speaker 00

  Yes.

• 7:55Speaker 01

  Say like, I am the star.

• 7:56Speaker 00

  Yes, exactly. In the foreground, in the foreground, a little too casual. I think I was the only one in suit at the entire event. But, but no, it, it, it was great. And we have Mark German from Bloomberg coming on at noon to discuss more of what's going on in the world of Apple, the MacBook Pro moment, and everything else that's going on with the foldable iPhone. We haven't talked to him since the duo actually launched. He predicted it on our show. He was correct. And here it is. I did get to see a couple iPhone duos. One guy had two iPhone duos rigged together for content creation somehow. I don't know exactly why, but the guy was rocking four screens, which is elite. There was another thing that was funny.

• 8:40Speaker 01

  Did you ask him why he stopped at four?

• 8:42Speaker 00

  No, I, I.

• 8:43Speaker 01

  Like, cause if you're gonna go there, why not have more of like an accordion?

• 8:46Speaker 00

  Yeah. The accordion would be good. You know, he was pretty locked in. There was one person.

• 8:51Speaker 01

  What was the thing that you said people said they were using Instagram less because it's kind of a hassle to use?

• 8:57Speaker 00

  Yeah, I did sort of like an informal polling of the iPhone duo users. And I noticed that some of them said that their screen time was maybe not lower overall, but tilting away from vertical video because they use the phone most, most of the time when they pick up their phone. Then this is very common in the brain rot era. You pick up your phone because you got a text message or you need to check something important. And then you're like, ah, I should see if I got a message on Instagram. And then you're like, well, I should see if there's a good reel at the top of my stack, you know? And pretty quickly you scroll a couple reels. This is a, this is the curse of modernity. And everyone's sort of feeling this. And, and at least anecdotally, there were a few people that were saying

• 9:41Speaker 00

  like, yeah, because I don't like the way Instagram looks, it's less addictive on the small main screen. I can answer text messages there. And then if I do want to watch Instagram, I will like open it up and it's like an iPad and like watch it. So who knows where that really nets out? I mean, people will be able to see it because Apple's been tracking the screen time for so long that you could export the data and actually look at the before and after and see like, on a week to week average, is your Instagram use going up or down? Or is your vertical video use going up or down? There's a few different ways to slice and dice it. But the other thing was that the really elite content creators. So basically, everyone had an iPhone that they were taking videos of the event. Couple of people had like rigged out Sony cameras. There was that one guy with the two iPhone

• 10:24Speaker 00

  duos, ultra elite. But the, but the real, the real power move is bring the selfie stick that goes all the way to the sky. There was someone there with a selfie stick that went so, so high. I can, I can actually show you a video. I'll send it into the, to the team because it is next level production team. Let's see if this.

• 10:46Speaker 01

  The bird's eye view.

• 10:48Speaker 00

  It, it was, it was one of the most aggressive selfie sticks. And it makes me think if I ever go back to this, I need to get one that goes, you know, 40 feet high and requires three tripod. It'll be like Tyler's.

• 10:59Speaker 01

  Yeah. Did they ever sort of pointed at Turnus and sort of like, almost like a fishing rod extended out right into his face?

• 11:05Speaker 00

  I mean, they had, they had like security screening on the way in, but this was a threat. Started at the beginning. Start at the beginning. So I this is this. So they're filming the event as the reveal happens. And then you zoom out and you just get more and more selfie stick until you get like you're truly so far above the crowd. I don't even think you need to be that high, but hopefully good. We got a nice video of Tyler, too. I put it in the in the group chat. Maybe we can share that as well. How many days until?

• 11:38Speaker 01

  More importantly, chat wants to know, there are 57 days until

• 11:43Speaker 00

  57. Keith Morris.

• 11:44Speaker 01

  Christmas. There he is.

• 11:46Speaker 00

  Fantastic. 57 days.

• 11:48Speaker 01

  Looking good in the red. Very festive.

• 11:51Speaker 00

  Um, yeah, it's right around the corner. So get ready. Let me tell you 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 from start to finish. Tyler's clapping for that ad. He was also clapping at the Apple event. We can show a video of him going crazy. They set up this little photo booth.

• 12:16Speaker 01

  I don't think I was even in the top 50% of how excited people.

• 12:19Speaker 00

  No, no, there were people that were going way harder than you. Yeah, this was this was a mild performance. The other interesting thing about the the actual keynote was that they didn't put the price up on the screen. Typically, there's like, where you do all this, and it's a it's a phone, and it's a music player, and it's a blah, blah, blah, and it has the best screen ever. And it's only $199 or $999 or $2,999. We're no longer in that era because of the memory crunch and the and the price of components going up. And so the price was not on screen. You can actually see it on apple.com. Something

• 12:58Speaker 00

  around starts at what does it start at $2,900 for the 14 inch. The 16 inch is going to be $3,400 to start, which is on the price here and but.

• 13:14Speaker 01

  And the way you spec'd it was 60?

• 13:17Speaker 00

  $6,000. $6,000.

• 13:19Speaker 01

  They also have an own nothing and be happy option. You can lease it.

• 13:24Speaker 00

  Yeah, you can lease it.

• 13:25Speaker 01

  36 months.

• 13:26Speaker 00

  I mean, it's not that crazy. Yeah, I mean, a lot of people are buying these for work. They're expensing them. And so there's there's way. And I mean, at the price for performance, like, you're in the right realm. It's not that dramatic. We can talk more about the strategy that's merging with Mark Gurman. The interesting thing was that that that initial like light shined on the wall. That was a teaser for the fact that the Apple logo now lights up on the back of the screen.

• 13:59Speaker 01

  Yeah, I completely I hadn't really been thinking about the fact that that had gone away.

• 14:05Speaker 00

  Yeah.

• 14:06Speaker 01

  Seemingly to save money.

• 14:07Speaker 00

  Yeah, exactly.

• 14:08Speaker 01

  Now they're bringing it back.

• 14:09Speaker 00

  And this.

• 14:10Speaker 01

  And it's crazy in hindsight, because it's like, yeah, they probably saved a little bit of money. But it's so such an iconic part of the Mac experience that was just gone.

• 14:19Speaker 00

  You're pro. You're pro having.

• 14:21Speaker 01

  Pro bringing it back.

• 14:22Speaker 00

  Cool. Well, they brought it back.

• 14:23Speaker 01

  No utility.

• 14:24Speaker 00

  So that might be the thing that gets you this cycle.

• 14:26Speaker 01

  I'm sold.

• 14:27Speaker 00

  You don't want the touchscreen. You don't care about the OLED. But the light up back is interesting to you. No, I mean, the the applause from the audience at that moment was the biggest it was of everything, even more than touch. Like, even more than anything else. That was what people were excited about.

• 14:45Speaker 01

  Maybe maybe they're bringing it back because, you know, a lot of people are live streaming. Sometimes, you know, MacBooks are are.

• 14:54Speaker 00

  Oh, yeah, you can't really tell what we have. I need to position it like this if I really want to be nice to Apple comms. Here we go. Boom. Nailed it. That's good, right? That looks great. Now that's that the reflection there. No, people were very excited about this. And yes, my theory is that it probably costs like $2 per unit to do that. And back when they were selling MacBooks for $1,000, that was, you know, a fifth of a percent of the total price. And maybe, you know, half a percent of the total component cost or something like this. These are very estimated numbers. But when the top line price goes up because of memory going up and because of chips going up in

• 15:34Speaker 00

  price, all of a sudden, like, the the technology that it costs to light up the Apple logo probably hasn't changed in price. It's probably maybe even gotten cheaper. Might just be still $2. And so now they're like, well, most people are going to be spec'ding these at like, $4,000. $2 relative to $4,000 is way less than $2.2 relative to $1,000.

• 15:54Speaker 01

  Internist feels like a guy that's like, yes, it's not necessary. Yes, it has no utility other than like, sort of, brand.

• 16:03Speaker 00

  Yeah.

• 16:03Speaker 01

  And just sort of like, enjoy like enjoyment for the user. It looks nice.

• 16:06Speaker 00

  Yeah, no, and I think that's the premium Apple brand that they need to like, stick with and stay with and stuff. So I thought that was that was a nice touch. Anyway.

• 16:16Speaker 01

  Google. They got a couple new models.

• 16:19Speaker 00

  Okay, first, let me tell you about Railway. Railway is the all in one intelligent cloud provider. Use your favorite agents to deploy web apps, servers, databases, and more. Well, Railway automatically takes care of scale, monitoring, and security. What did Google launch?

• 16:31Speaker 01

  Google has a couple new models. Sundar says excited to share Genie 4, today our most realistic and interactive world model.

• 16:37Speaker 00

  This was really cool.

• 16:38Speaker 01

  Steer dynamic environments in real time, shaping events, camera angles, and actions. Generate native audio, match the world from realistic background sounds to characters who can converse in real time. Simulate photorealistic scenes and real world physics. Create more coherent multi-step events through reasoning from Gemini. And they've started to roll it out as a research preview. Looks incredible. I think that the primary use case for world models for me personally so far is, you know, making me think that it's more likely that we're living in our own simulation.

• 17:12Speaker 00

  Yeah.

• 17:13Speaker 01

  I don't know. I wonder, like, who's using, you know, who's using these?

• 17:21Speaker 00

  Well, right now, it's mostly research previews. So researchers are using them. And the immediate the immediate thing is like, okay, it could be fun. It could just be something that you want to spend time in. So it's like a video game. It's a substitute for video games, or even just a full video game that you want to pay for or or subscribe to or upgrade your plan to be able to tinker in. But the the immediate case for for content creators is like, set extensions, virtual backdrops, like, you can you could effectively composite yourself into this world, and your camera could be moving you around. You've seen those videos of people larping with Higgs field. You've seen this? Where people

• 18:06Speaker 00

  would be like, look, look at me. And it's like that. And it's like, this is my Lamborghini. And they've dynamically replaced the Honda Civic with Lamborghini. And then they're like, and I'm on my private jet. And it's actually just their back staircase or whatever. This allows you to do it in real time, if it's this high quality, because it can it can interpret where the camera position is, and then dynamically generate the world. So even if you're even if I'm filming you, and I have a virtual backdrop on you, and I'm like, okay, we're in we're on the beach, and then we go outside, it can it can imagine what's outside. And it says, okay, now we're now we're on the boardwalk of the beach, and it can continually recreate this. So there will be some interesting like, real time

• 18:46Speaker 00

  applications as this rolls out. Everyone is probably familiar with the simulation hypothesis proposes that what one experiences as the real world is actually a simulated reality, such as a computer simulation, in which humans are constructs. Everyone's thinking we're in a simulation in a simulation now. You had a different thesis, different hypothesis, the the racing simulation hypothesis, which is that our world, what we know as Earth, what we know as the human experience, in fact, exists in a racing simulator run by a higher level organism that exists purely to propagate F1 and the F1 circuit. And everything

• 19:29Speaker 00

  else that we do is in service of simulating F1.

• 19:33Speaker 01

  Motorsport.

• 19:34Speaker 00

  Motorsport. Yeah, motorsport broadly. It's not just it's a it's a multifaceted racing simulator. So you include NASCAR.

• 19:40Speaker 01

  Some rally.

• 19:41Speaker 00

  GT3. Yeah, exactly. The racing simulation hypothesis is is one that's it is gaining a lot of ground among among the the futurists.

• 19:50Speaker 01

  Yeah.

• 19:51Speaker 00

  And the and the science fiction fans. The racing simulation hypothesis is getting getting popular. This is cool.

• 19:57Speaker 01

  That's not all.

• 19:58Speaker 00

  Wait, wait, really quickly, on Genie 4, is it available yet? Because you've played with some of them before.

• 20:04Speaker 01

  Yeah, I've played with Genie 3 a bunch. It was very fun.

• 20:06Speaker 00

  Yeah.

• 20:07Speaker 01

  I don't have access to Genie 4 yet. I think it's still on. You have to like, apply to their like, early access program.

• 20:12Speaker 00

  Okay. I'm excited for that to roll out and be able to play. Jordy, what is the next Google?

• 20:17Speaker 01

  They got another model. Multiple models this week. Sundar is back. He says, I did say we'd be releasing our models periodically. Introducing Gemini 4 Carbon, their next frontier model, which follows Aragon's release just two weeks ago. Carbon shows the pace at which we're pushing the frontier. And it ranks number one on the recently updated V5 of the Artificial Analysis Intelligence Index at 59. And it also leads across cybersecurity evaluations, scoring 88.2% on SEC Bench Pro for finding vulnerabilities, and 54.8% on Exploit GIM.

• 20:55Speaker 01

  Yeah, so very excited to see the reaction here. Seems like a great launch. Elon says it's impressive. And I think, again, they have to, at this point, have been learning from the sort of series of benchmark disappointments that they seemingly have had for about a year.

• 21:18Speaker 00

  Semi analysis taking shots, right?

• 21:19Speaker 01

  Yeah, just about a year straight where every single time it's amazing on benchmarks. And then the sort of lived experience that people have is less impressive.

• 21:30Speaker 00

  I like Gemini 3. I thought Gemini 3 had like a good flavor to it. Maybe it was overly benchmarked, but it felt like it wasn't too verbose. Felt like it sort of answered questions in like one page of text. And so I could like see it and see the answer and kind of parse it easy as opposed to getting like pages and pages and pages. But it was something LLMs do. It can be a little bit much. I was confused by this because it feels like the hierarchy, like with like Luna, Terra, Sol, Astra, I kind of understand that like the moon is smaller than the earth, the earth is smaller than the sun, the sun is smaller than the celestial bodies. Same thing with Anthropic where it's like a sonnet is shorter

• 22:15Speaker 00

  than an opus, a mythos is bigger. I kind of understood that. This is interesting because they're going in maybe the opposite direction, depending on how you rank them. So argon is the 18th atomic number where's carbon's number six. And so does that leave them open to do a lithium, a helium, a hydrogen model that's like bigger? Are they going to go that direction? I like that they have a name.

• 22:43Speaker 01

  You think they're going to do a helium model?

• 22:45Speaker 03

  Yes, I think the helium model could be the biggest model from DeepMind in years. It could change everything. It would be significantly bigger than their new carbon model. And the helium model will sell extremely well. You'll see it showing up on the ramp index any moment now when they launch the helium model.

• 23:11Speaker 00

  But it does sort of beg the question, yeah, like what is the taxonomy here? It's kind of cool that they left a lot of space between carbon and argon for sodium, magnesium.

• 23:24Speaker 01

  Sodium. Honestly.

• 23:26Speaker 00

  Neon, aluminum.

• 23:27Speaker 01

  I love salt.

• 23:28Speaker 00

  Silicon's in there, phosphorus, sulfur.

• 23:30Speaker 01

  I would have liked.

• 23:30Speaker 00

  I don't think you want a sulfur model.

• 23:32Speaker 01

  I would have liked some more fear mongering. They could do plutonium, fluorine.

• 23:35Speaker 00

  Well, that's really deep. So in the, this is what I'm saying, in the DeepMind taxonomy right now, plutonium would sound like a really scary model, but it would actually be much smaller than carbon because it goes from carbons the biggest and best, then argon at the 18 atomic number is smaller. And then you go way down to plutonium and it's actually like the least effective, the least smart model in theory. I don't know. It's cool.

• 24:03Speaker 01

  They should do Gemini 5 nuke.

• 24:06Speaker 00

  Nuke. Yeah, we, don't you have a button for that? I thought we had a button for nuke.

• 24:11Speaker 01

  No, I think we, I think we removed.

• 24:12Speaker 00

  I think it's too aggressive for the show. Oh, here we go. Tyler has the truth there. Gemini plutonium is coming in any minute now. No.

• 24:22Speaker 00

  The truth nuke is good. Really slow still. The longest soundboard effect we have. But this, this all feels like, okay, a year ago, people were talking about like the naming schemes at the labs don't make any sense. OpenAI was probably the worst offender with like 4.0 and then like you'd go to 4.5, which was a bigger model, but it was slower. And then there was like Omni versus Fastmo. And it was like too much. And everyone was like, I don't get it. And then, and then it was like 5.4, 5.5, 5.6. And then eventually OpenAI said, okay, we're going to use a human legible, not just purely numbers. We're going

• 25:04Speaker 00

  to go Terra, Luna, Sol, Astra. And that I feel like makes a lot of, a lot more sense to people. It's just more intuitive. And so I like that OpenAI has adopted that format. I like that Google's adopted that format. And I hope that the version numbers kind of melt away. And you're just in the world of like, okay, like big model for this task, small model for that task. And, and it's a little bit, it's a little bit easier to just intuitively get a grasp of what every model is capable of.

• 25:33Speaker 01

  Anyways, amazing benchmarks. Excited to see the reactions.

• 25:36Speaker 00

  Yeah.

• 25:37Speaker 01

  And.

• 25:37Speaker 00

  And the ramp data.

• 25:39Speaker 01

  Yeah. In other news.

• 25:41Speaker 00

  Yep.

• 25:42Speaker 01

  OpenAI.

• 25:43Speaker 00

  First, let me tell you about Cisco. Critical infrastructure for the AI era. Unlock seamless real time experiences and new value with Cisco. Take me through it.

• 25:50Speaker 01

  OpenAI plans to buy Google's chips as part of its AI build out.

• 25:56Speaker 01

  The new, like first we had, so now we have TPUs. Early last week, it was, it was OpenAI models on GCP.

• 26:06Speaker 00

  Yeah.

• 26:06Speaker 01

  Feels like start of a new era.

• 26:10Speaker 00

  Yeah. I mean, sort of a new era. I don't know. I mean, yeah, I guess OpenAI was founded because they were like, Google's bad, right?

• 26:19Speaker 01

  Yeah, that's what I'm saying.

• 26:20Speaker 00

  Yeah, that's a good point. Now it's like, they're not that bad. But this is the same thing with like Anthropic and Elon. It was like, you know, very, very diametrically opposed, you know, politically.

• 26:30Speaker 01

  Well, yeah.

• 26:31Speaker 00

  It's like, hey, they got compute, we need compute. Yeah. So, business above all. Business solves everything. Let's give it up for business.

• 26:39Speaker 01

  Let's give it up for business.

• 26:41Speaker 00

  Business solves everything.

• 26:42Speaker 01

  Dylan Patel says semi analysis clients had this already.

• 26:47Speaker 00

  Whoa.

• 26:47Speaker 01

  Screenshot.

• 26:48Speaker 00

  Little victory lap.

• 26:49Speaker 01

  AI allows largest lease ever more massive deals in the pipeline. AMD taking share? Question mark. OpenAI's quarter is one 8 gigawatt multi-phase commitment in Ohio with Nvidia backstopping 4.25 gigawatts. It's the largest single leasing commitment we have ever seen.

• 27:03Speaker 00

  Cool.

• 27:04Speaker 01

  In addition, we believe that OpenAI is discussing with NeoClouds for both GPU as a service and TPU as a service. We think there's a 1 gigawatt TPU commitment and OpenAI is looking for sites.

• 27:15Speaker 00

  Yep.

• 27:16Speaker 01

  Anthropic leased 1.7 gigawatts this quarter with notably 360 megawatts in Queensland, Australia. Let's give it up for compute down under. Signed on the 16th of September with operations starting in the first quarter of 28. The largest overseas lease by an AI lab this quarter. The 1.7 gigawatts includes 1 gigawatt backstop by Google, but also a large site where we aren't sure yet who will provide the credit support.

• 27:42Speaker 00

  Do you think AI researchers are going to be okay with doing inference in Australia? That's kind of controversial, right? Because they're known for throwing another shrimp on the Barbie.

• 27:50Speaker 01

  Oof.

• 27:51Speaker 00

  That's rough. You're going to have to do a truth. No more throwing shrimp on the Barbie. Just fosters. Just enjoy your fosters.

• 27:59Speaker 01

  Enjoy your fosters.

• 28:00Speaker 00

  And stick to inference and fosters. No more shrimp. Anyway, let me tell you about CrowdStrike. Your business is AI, their business is securing it. CrowdStrike secures AI and stops breaches. Yeah. Oh, sorry. Just to close out that, it's, it's just like every lab is using every part of the animal. Like it's the, you know, they're fully.

• 28:19Speaker 01

  They're going whole hog.

• 28:20Speaker 00

  They're going whole hog. They're using everything, everything they possibly can. Every lab is using every chip and every data center and every hyperscaler and every NeoCloud. And they're using it all for all different things. And different chips will be better at different models and different part pieces of the inference stack. And like everything will be used. Very good for the, like, if you've been in semis and, you know, NeoClouds and all this stuff, like Rising Tide will like lift all boats as everything becomes, like, there was a moment where it was like, oh, if you made the wrong architectural decision, like you would get smoked. That was the theory. And now it's like, oh, actually, these models, at least a piece of the models can be like used in niche areas for different

• 29:02Speaker 00

  things. And so even if, even if one chip is better at something, faster, but less smarter, smarter, but slower, like, like there will be an optimal way to, to get a model running on that. And with AI agents, it's pretty easy to port to different things. I think we saw that from the AMD coverage we did with Lisa Su where Anthropic vibe coded their AMD implementation like pretty quickly. So good for anyone who was betting on oligopoly and silicon, rougher if you were like, this will not be a car-like market, but still very, very early, still a lot of dependence on Nvidia, obviously. So nothing too scary there, but just interesting that everything is getting used by everyone. So that's kind

• 29:45Speaker 00

  of my update. Anyway, open source, open source.

• 29:50Speaker 01

  Yeah. So Dax over at OpenCode posted yesterday, he says, to put what I'm saying about demand in context, OpenRouter is the most successful model router. You can almost serve 100% of their DeepSeek traffic yesterday on a single GB300 rack. This is worth 6 to 7 million. Even if there's 10x this demand secretly, that's still less than 100 million of compute.

• 30:14Speaker 00

  This is crazy. This is crazy because I don't believe, I mean, I don't know, maybe people are inferencing DeepSeek from NeoClouds directly and not going through OpenRouter, but I would be very skeptical that there are large enterprises out there that are saying, oh yeah, we have a million dollars a day of inference.

• 30:33Speaker 01

  Yes, and a lot of people.

• 30:34Speaker 00

  And let's rack our own GB300.

• 30:35Speaker 01

  Well, yeah. So a lot of people were saying, well, this doesn't, yeah, this doesn't factor in people that have.

• 30:40Speaker 00

  Their own cloud.

• 30:40Speaker 01

  GPUs on prem.

• 30:41Speaker 00

  Yep.

• 30:41Speaker 01

  But the elephant in the room is the, the base tens, the modals, the fireworks, the Together AI.

• 30:50Speaker 00

  So it's going through that and it's not going on.

• 30:51Speaker 01

  Lambdas, you know, like a very, very long list of other clouds that can.

• 30:56Speaker 00

  Okay.

• 30:57Speaker 01

  That can serve these models. So I would expect that that is where, that is where the, that's where the billions of dollars of inference spend is happening.

• 31:06Speaker 00

  Sure. Do we have a, do we have a reference point for what percent of DeepSeek inference is going through OpenRouter versus going direct?

• 31:14Speaker 01

  I think it's impossible to know.

• 31:16Speaker 00

  Okay. Well, if you want to get in the game, there's a 300 rack AI supercomputer powered by 9,600 Nvidia A100 GPUs. It's on eBay. You can just buy it.

• 31:27Speaker 01

  Head over to eBay.

• 31:28Speaker 00

  If you were thinking about starting a NeoCloud, getting in the token selling business, hit up your local banker, get a credit line for 60 mil, go on eBay, click it. Fair market value for this?

• 31:42Speaker 01

  eBay up 16% over the last six months, likely because people are realizing that this, it's a compute marketplace.

• 31:49Speaker 00

  Apparently, yeah. Fair market value for this would be closer to 50, maybe 55 million, but you could still.

• 31:56Speaker 01

  I love that. The screenshot, it says in six carts.

• 32:00Speaker 00

  Six carts?

• 32:01Speaker 01

  Like six people.

• 32:02Speaker 00

  Oh, like people have added it to cart.

• 32:05Speaker 00

  Two available. Wait, they have two available of these? Wow. Um, yeah, interesting. Uh, pretty, pretty solid depreciation too. Uh, this, um, this hardware originally sold, uh, for, uh, something like, uh, what was, what, what was this originally? Originally went for, let me, let me find it. Um, it's selling for, uh, 60 million now. The original hardware was almost 300 million. So there is depreciation, but there's still some value there. Of course, there's the A100 is a seven year old chip now, something like that. Um, and so, um, you're banging the table. You're, you're hitting buy? Are you bidding?

• 32:47Speaker 01

  I was getting excited.

• 32:48Speaker 00

  Oh, you're just getting excited. Okay.

• 32:50Speaker 01

  No, very old chip, still worth a lot of money.

• 32:53Speaker 00

  Yeah, totally.

• 32:54Speaker 01

  It's the price of one dinosaur.

• 32:56Speaker 00

  Oh, yeah. Which one would you do? Jordy, which one are you going with?

• 33:00Speaker 01

  Are you going Gus, the, the T-Rex? Or are you going these GPUs?

• 33:03Speaker 00

  That's, that's the one.

• 33:04Speaker 01

  I think, um, I think I'm still to, to my oldest dismay, I'm still going, I'm going for the GPUs.

• 33:12Speaker 00

  You're going for the GPUs?

• 33:13Speaker 01

  Just because I think Malibu inference would be a great name for, uh, a Neo cloud.

• 33:19Speaker 00

  Yeah. And everyone's been making that claim. Oh, like put the data centers in like luxury towns. And you're like, yeah, I got one right here.

• 33:26Speaker 01

  Yeah.

• 33:26Speaker 00

  Um, also you could stack GPUs up in the shape of a dinosaur.

• 33:31Speaker 01

  Yes.

• 33:31Speaker 00

  So that you have a beautiful architecture in your foyer. Uh, you don't, you of course need to add a lot of cooling. Cause all of a sudden your entryway is the hot aisle. That'd be rough, but, uh, it would make a statement. People would come in. They know you mean business. They know, they know you do, you do local compute. You're not paying API pricing, you know, you're saving dollars every.

• 33:53Speaker 01

  Why on earth does Malaysia have so much compute, John? Can you take a guess?

• 33:59Speaker 00

  It's a mystery.

• 34:00Speaker 01

  Nobody knows.

• 34:00Speaker 00

  Nobody knows.

• 34:01Speaker 01

  Nobody knows. It is a total mystery.

• 34:03Speaker 00

  This was the first story we talked about on this show, the very first show.

• 34:07Speaker 01

  Wow.

• 34:07Speaker 00

  Very first show, I opened up the Wall Street Journal and, uh, and there was, uh, an article right on the cover about Malaysia being the new, uh, the new data center hub.

• 34:17Speaker 01

  Tyler, we gotta send you down there.

• 34:19Speaker 00

  They are the Switzerland of AI. That was sort of the, sort of the framing was that they can buy compute from the US and resell it in China. And so that was an incredible adv- advantage and they had a lot of power, uh, between the different, uh, locations there. And, uh, the, the, the government was sort of set up to allow a lot of compute build out. Um, pretty, pretty interesting. Uh, they, they have a lot. It's not that much compared to America. I mean, uh, Malaysia is not the green bar here. They're not the red bar. The red bar is China.

• 34:52Speaker 01

  I'm actually surprised.

• 34:53Speaker 00

  They're not even the blue bar.

• 34:54Speaker 01

  I'm, I'm surprised that the EU has more compute than Malaysia.

• 34:57Speaker 00

  Okay. Yeah.

• 34:58Speaker 01

  Aren't, aren't you?

• 34:59Speaker 00

  Uh, no, because the EU is huge and Malaysia is like a pretty small country.

• 35:04Speaker 01

  I know.

• 35:04Speaker 00

  But yes.

• 35:05Speaker 01

  Like they, they, Malaysia seems to be going so long.

• 35:07Speaker 00

  Yeah. Also, uh, Malaysia is interesting. These are sort of like the, uh, the AI industrial hub to Singapore. So Singapore is the, the global financial hub, the port city. So everything's coming in there. You get stuff delivered through Singapore. And then you set it up in Malaysia, just a hop, skip and a jump over like a, a small river or small, uh, I don't know. Small waterway to get to where you actually need to deliver the final product. So, uh, interesting. But yes, I imagine that a lot of, uh, a lot of Chinese AI lab compute, if they're doing training, they're probably using like a Malaysian Neo cloud that's racking Nvidia chips. Uh, and they can, because there aren't as many trade restrictions, I think, something like that.

• 35:49Speaker 01

  Yeah.

• 35:49Speaker 00

  Anyway, uh, let me tell you about Shopify. Shopify is a commerce platform that grows with your business unless you sell in, in seconds, online, in store, on mobile, on social and marketplaces. And now with AI agents, anger.

• 36:00Speaker 01

  Uh, two Sequoia Capital portfolio companies.

• 36:04Speaker 00

  Okay.

• 36:04Speaker 01

  Citadel Securities and OpenAI have teamed up with Sequoia.

• 36:08Speaker 00

  That's a true sentence.

• 36:11Speaker 01

  Very, it is, uh, almost unbelievable.

• 36:13Speaker 00

  Yeah. It doesn't read like that. It reads like we're Sequoia and we're brokering a deal between two large, important institutions.

• 36:20Speaker 01

  No, these are just port.

• 36:21Speaker 00

  These are just portcos. They're just like, we got a portco collab going on. We made the intro at our CEO summit.

• 36:26Speaker 01

  Yeah.

• 36:27Speaker 00

  Ken, have you met Sam? You should, you guys should do something.

• 36:30Speaker 01

  You guys really gotta talk.

• 36:31Speaker 00

  You guys gotta talk. Yeah. Yeah. I'll, I'll intro you over email.

• 36:33Speaker 01

  Sequoia announced No Limit, a bot building tournament.

• 36:37Speaker 00

  It's cool.

• 36:37Speaker 01

  Uh, hosted by Citadel and OpenAI and Sequoia. Teams of one or two get 60 minutes to build a poker bot from scratch. Agents play thousands of games across six scored rounds. Open to university students in recent.

• 36:49Speaker 00

  Are they taking a shot at Tyler with that?

• 36:52Speaker 01

  Absolutely.

• 36:53Speaker 00

  It's brutal.

• 36:54Speaker 01

  Because he's neither a recent grad or a university student.

• 36:57Speaker 00

  Neither. He's neither. That's crazy.

• 36:59Speaker 01

  Specifically set this up.

• 37:01Speaker 00

  They could have just said, they could have just said, young people, open to young people.

• 37:04Speaker 01

  They literally are like.

• 37:05Speaker 00

  Open to young go.

• 37:05Speaker 01

  No, no, they specifically don't want dropout.

• 37:07Speaker 00

  They specifically don't want to.

• 37:08Speaker 01

  They're like, anybody that like, what had a bright future in academia and quit to work on a podcast is absolutely not welcome to, to, to compete in this. We don't want you.

• 37:19Speaker 00

  I'm, I'm sort of offended. I love Sequoia, I love OpenAI, I love, I love Citadel Securities, but, uh, yeah. Rough. Give us a break.

• 37:29Speaker 01

  No, it's, it is really funny. Dropouts, not welcome.

• 37:32Speaker 00

  Not welcome. Also probably.

• 37:34Speaker 01

  No, I'm sure, I'm sure they do. I'm sure they do allow that.

• 37:36Speaker 00

  Yes. Uh, but, but yeah, this is, this is going to be very fun. Um, I, uh, yeah, I wonder, I wonder where, where you could go with that. That's interesting. 60 minutes. That's pretty quick. Uh, and I wonder what, what tools you'll be able to use, how much prep you can do. And, uh, and also like what the actual API is. Cause, uh, I, I'm, I'm increasingly interested in, uh, in the, building these types of bots, but doing it with computer use and, and having it only see what the human sees. I want basically a camera pointed at the screen with like button clicking. I, I want as much, uh, friction in the system as possible. And I, I, I'm, I'm unimpressed now.

• 38:16Speaker 00

  I've moved the goalpost since, uh, you could just, uh, interact with the API, but who knows? Uh, this is still very, very interesting. And, uh, will probably be superhuman as I, as I anticipate. Anyway, let me tell you about Consul. Consul builds AI agents that automate 70% of IT, HR and finance support, giving employees instant resolution for access requests and password resets. Uh, can I talk about TurboPuffer for a second?

• 38:40Speaker 01

  Yeah.

• 38:40Speaker 00

  Okay. So.

• 38:41Speaker 01

  Please do.

• 38:41Speaker 00

  So TurboPuffer, you've probably seen the puffer jacket that Jordy has worn on the show a number of times. It is now available for sale, I believe, or, uh, I think it's only going to cost you a hundred million dollars of, uh, contracted value with them. You need, I think you might need to be a customer or it might be for sale, but, uh, it's a beautiful puffer jacket from TurboPuffer. There's a great profile on TurboPuffer in Colossus Mag today. It's really cool. It, it frames TurboPuffer as like the only normal tech company in the world, while everyone in San Francisco is like crashing out and like being like AI doom. Like the TurboPuffer folks are just like.

• 39:17Speaker 01

  They're just puffing.

• 39:17Speaker 00

  We're building a business. Yeah. Yeah. We're building a business.

• 39:19Speaker 01

  Yeah. And, uh, we, yeah, we started working on that probably over a year ago.

• 39:23Speaker 00

  Simon's been on the show a couple of times.

• 39:25Speaker 01

  Uh, no, I'm talking about.

• 39:26Speaker 00

  Oh, oh yeah. The puffer jacket. Yeah.

• 39:27Speaker 01

  Started working on it. Making clothing is very, very hard.

• 39:30Speaker 00

  It takes a long time.

• 39:31Speaker 01

  Very, very, very hard. Uh, but, but, uh, finally.

• 39:36Speaker 00

  AI will never get a product in Target. It will take so long. They'll just be like, ah, I lost my context window. I can't, I can't get it. Um, yeah. TurboPuffer, uh, is profiled in Colossus. There's some interesting, uh, there's some interesting, um, uh, strategies that, uh, Simon's in, in, employed. Uh, he, he has this term, a coinage. He calls it ambition bullying. Uh, so early investors kept asking Simon, how are you going to build a unicorn? You only worked at Shopify before, but, uh, he thought it was ridiculous to pretend he could know whether the business would become a billion dollar company. Spoiler alert, it's well on its way, but, uh,

• 40:19Speaker 00

  he calls the pressure ambition bullying. So VCs come and they'll, they'll bully you into being more ambitious than necessarily you want to be. Distracts you from actually just putting one foot in front of the other. Uh, finally, Lockheed groom was the investor willing to accept the founder's

• 40:34Speaker 01

  The lockinator?

• 40:35Speaker 00

  Plan, the lockinator, locked it down. Uh, he said, work hard to find product market fit, but shut the company down if it's not working after a year. And Lockheed said, that's fine. I accept your pitch. Uh, and he invested 700K at a $14 million valuation. What a great investment.

• 40:53Speaker 01

  Printing.

• 40:53Speaker 00

  They also, they also, uh, bought a fiber line just to avoid, uh, seven milliseconds of latency. Like that's how obsessed they are with performance. So they discovered that traffic, um, was traveling through Seattle, uh, like talking to AWS and then getting back to their servers or something like that on Google cloud. And that this created a 15 second milli- uh, 15 millisecond round trip. So instead of accepting the routing problem, uh, one of the, uh, employees at TurboPuffer located a private internet exchange and purchased a direct fiber connection, reducing the round trip to eight milliseconds. And, uh, yeah, I mean, this stuff matters because TurboPuffer does, uh, a lot of search products. It has to be fast. It's, it's, uh, it's a vector database, but, uh, it's, uh, it's, it's effective.

• 41:37Speaker 00

  Not for like the big inference where you're cooking a pro query, deep research query. Uh, it, it's much more important that it's, uh, very quick in terms of knowledge retrieval. You can't be waiting for it. So every millisecond matters. And that's what the TurboPuffer team prioritizes. And so now they're doing over a hundred million dollars in ARR with roughly 50 employees. And so they've been, they've been crushing.

• 41:59Speaker 01

  Trey says 15 milliseconds additional is too high. That's the type of ping that'll get you killed and counter-strike.

• 42:06Speaker 00

  That's true. That's true. You need under 10 seconds for ultimate fragging. I agree. Uh, well, let me tell you about public. 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.

• 42:21Speaker 01

  Uh, Lucas.

• 42:22Speaker 00

  Oh, please.

• 42:23Speaker 01

  The grubber, grubberinator. Lucas Gruber.

• 42:26Speaker 00

  Grubinator.

• 42:27Speaker 01

  Says, uh, Instinct is currently raising at 30 billion, a hundred to 500 to two and a half to 10 to 30 in sub one year. If you're still not inside the private market, this should be a good signal to bid aggressively. Wealth creation is going to be heavily skewed towards private investors over the coming decade.

• 42:48Speaker 00

  Yeah.

• 42:52Speaker 01

  Yes, yes.

• 42:53Speaker 00

  There is more risk and there is more reward. And so that's why venture capital firms diversify and

• 42:59Speaker 01

  Yeah.

• 42:59Speaker 00

  And LPs diversify across venture funds. But it is a

• 43:03Speaker 01

  Yeah, so people, people are naturally crashing out over this. Not, not even the take that's that's, that's embedded in this post, but just, just that trajectory.

• 43:14Speaker 00

  Yeah.

• 43:17Speaker 01

  It is insanely aggressive if it's true. But again, this business has been growing so rapidly.

• 43:24Speaker 00

  Yeah.

• 43:25Speaker 01

  And the interesting thing is like, it's very hard to judge Instinct's growth relative to Muse, simply because Instinct doesn't have an app. So it'd be really fascinating to see like,

• 43:34Speaker 00

  The app store.

• 43:35Speaker 01

  Where would they sit in the App Store rankings, right?

• 43:37Speaker 00

  Yeah, yeah. What is, what is Muse doing on App Store ratings right now? Cause it was at the top of the App Store for a while.

• 43:45Speaker 01

  Right now you have TikTok Pro.

• 43:47Speaker 00

  Muse is

• 43:47Speaker 01

  Invented, of course.

• 43:49Speaker 00

  Muse is being like promoted in the App Store. If you just go to the search button, Muse is right there. It's doing well. It has 380,000 ratings, almost all five stars, 4.9 average. So it's number one in productivity, but not number one overall because TikTok Pro is

• 44:05Speaker 01

  And vinted, of course.

• 44:09Speaker 00

  Yeah, vinted. I was on vinted. I was looking at some stuff, picking up some, picking up some Rick Owens. Head to town, it's coming.

• 44:19Speaker 00

  Geordi said I couldn't be a lifestyle influencer, but I think if I pull up in the Rick Owens, the Supreme, the Bottega top hat, you're gonna be apologizing.

• 44:28Speaker 01

  I didn't say, I didn't say couldn't.

• 44:32Speaker 00

  You literally said, you literally said, you should lean into comedy. And I'm like, and then, and then my counter example is just a bunch more comedy. So I'm, a hundred percent. Anyway, you can now text your dot. Yeah, so there are a ton of these AI agents that everyone seems

• 44:47Speaker 01

  Anyways, anyways, so, so the other thing with Instinct is like so few people have real information around the business.

• 44:54Speaker 00

  Yeah.

• 44:54Speaker 01

  That it's impossible to, everyone can give a take.

• 44:57Speaker 00

  Sure.

• 44:58Speaker 01

  Right? Because the number has, the valuation has grown so quickly, so fast. But it's just really tough when you don't have the information, right? And so you're just, it's entirely, investors are making a decision based on what they're seeing in the business. Everyone else is giving takes based on how they feel about the number going up really fast.

• 45:18Speaker 00

  Yeah.

• 45:18Speaker 01

  And yeah, it's hard, it's hard to, it's hard to see exactly like what, you can only imagine kind of the way that they're positioning the business, how they're,

• 45:30Speaker 01

  what, what Noah's vision is and all that kind of thing.

• 45:33Speaker 00

  It also highlights like when new capability comes online from a frontier lab, it can be at the frontier of that new capability. But unless there is a commensurate marketing push, it not, it's not guaranteed that it gets adoption. Because you only have so much screen real estate in the ChatGPT app, in the Claude app. Anthropic has Duet now. I saw the announcement, but I haven't seen a lot of advertising around it. And so there's plenty of people that are like, oh yeah, I have Claude code in my terminal, but like, I'm a Muse guy. Or like, I use Grok Bot in Slack, but then I use Codex for coding. And, or I'm Cursor over here and Instinct in my personal life.

• 46:16Speaker 00

  And there's so much effort that needs to go in. And so Instinct is at least focused on like, let's get this value prop in front of people. And, and I think that's sort of like the value of like, maybe it's oligopolistic, maybe it's a big market, maybe there is a winner take all. Like, all the investors are still like piecing that together, but they're just seeing like incredible growth. Everyone we've talked to has been like, the charts are incredible. Like every, every month is just huge growth, even on, even while being attacked by Mark Zuckerberg, being attacked by DOT and, and Duet. Like there's a whole bunch of other companies going after this. I have to imagine Google's gonna do something soon. They've launched similar products in the past and it's so in line with them. They already have your email. They have Google Maps that has all your reservations.

• 46:59Speaker 00

  They own, they own the Michelin guide, I think, or it's not Michelin, it's the some, some star rating service that's like Zagat, is that it? You know that one?

• 47:12Speaker 01

  Uh, yeah, I don't know.

• 47:13Speaker 00

  Okay. But they bought, they bought like a, like a restaurant review, like brand. And so they have a ton of advantages there. They have Google Voice. They've done service to make calls on your behalf already, like years and years ago with AI.

• 47:27Speaker 01

  What is it? Zagat.

• 47:28Speaker 00

  Zagat, is that how you pronounce it?

• 47:29Speaker 01

  Z A G A T.

• 47:30Speaker 00

  Yeah. And so they, they have so much infrastructure, but they're always a little bit slower on the product side than like an Instinct, which is just a couple of people moving really fast. So

• 47:39Speaker 01

  Yeah, the crazy thing is to compare Instinct.

• 47:42Speaker 00

  Yeah.

• 47:42Speaker 01

  Instinct, if they're raising a 30 billion to where Cursor was when they were raising a 30, which like massive competition from the labs.

• 47:49Speaker 00

  Yep.

• 47:51Speaker 01

  And some questions around how they would, how, yeah, how they would handle that, that competition, basically competing with their suppliers.

• 47:57Speaker 00

  Yep.

• 47:58Speaker 01

  Right, in some ways. But I believe at the time Cursor had like north of a billion dollars of revenue, right?

• 48:04Speaker 00

  They were worried about the margins, right?

• 48:05Speaker 01

  Yeah, their margins weren't great, but at least people were giving them like billions of dollars.

• 48:09Speaker 00

  Totally.

• 48:09Speaker 01

  Right?

• 48:09Speaker 00

  And that's not happening.

• 48:10Speaker 01

  And so with Instinct, part of what's jarring is that you have the same level of competition and potentially even more, right? Cause it's like you have the major AI players. You have the major AI players, then you also have Zuck saying

• 48:23Speaker 00

  With Cursor, it was like Cursor was the lead IDE. And then you had Windsurf coming from behind, nipping at their heels. But you didn't really have a product that was a direct competitor from OpenAI or Anthropic or Google, really. Right? In like the

• 48:37Speaker 01

  Yeah, until Claude Code and Codex.

• 48:39Speaker 00

  But those were, those were terminal agents and, and they did

• 48:42Speaker 01

  Yeah, I'm saying they're ultimately competing over the same

• 48:45Speaker 00

  Totally, totally.

• 48:46Speaker 01

  You know?

• 48:46Speaker 00

  But like in terms of like the actual product experience, like if you wanted to at least review some of the code, still be in that hybrid Centaur model. It was really Cursor was far and away number one. And then Windsurf was kind of nipping at the heels, but no, none of the big labs had direct competitor products. And then all of a sudden it was like, but flipping it around, if you're looking at the Cursor example, you might say, okay, yeah, they don't have the revenue, but they have the, the consumer adoption. And look at what happened with Cursor, felt crazy at 10 or 20 or whatever, but got the acquisition done. And you're looking at Amazon and Apple and saying, well, I see the money.

• 49:27Speaker 00

  They, there's the money to buy this company at a valuation above where I'm investing at 30. And so it could happen if Apple or Amazon make a crazy move.

• 49:37Speaker 01

  Yeah.

• 49:37Speaker 00

  I don't know, maybe Google too. But the price is getting hot.

• 49:41Speaker 01

  Yeah. Like Elon was a unique position. How would that work from an antitrust standpoint? Like all these companies.

• 49:46Speaker 00

  It should, it should 100% be allowed. It's the most competitive industry right now. It's like not a knife fight for the personal agents.

• 49:52Speaker 01

  Yeah.

• 49:53Speaker 00

  Like, like you're going up against the monopoly guy, Mark Zuckerberg. Like, oh wait, you're like Amazon, you're, you're acquiring Instinct to, to take a shot at OpenAI, Anthropic and Meta, like go, go to town with that, I would think. But again, I can never predict antitrust. So like you, you might be right that like, they might just say, no, you're big. You can't make a big acquisition.

• 50:18Speaker 01

  Yeah.

• 50:19Speaker 00

  Like it, it just, it might be framed differently. But as a, as a consumer, as an analyst of the industry, I see them as like, this is the most competitive tech category we've had.

• 50:30Speaker 01

  Yeah, so taking out an upstart doesn't actually change the competitive

• 50:34Speaker 00

  Yes. At the same time, we didn't see Perplexity get acquired. And that was a similar one where, where Apple and, and Amazon, they didn't really have like a chatbot experience. And of course Anthropic launched Claude, that app, and then Google launched Gemini, which was like a direct competitor to ChatGPT, the classic LLM experience, the chat app. Apple and Amazon never picked up, just a consumer home screen chat app, period. So they might not jump at this category either. I don't know. But if they did, if they, like if, if Apple bought Perplexity, I think there's other reasons why they might not. But if they did a year ago or two years ago, it shouldn't have turned antitrust heads. But again, the FTC is very unpredictable on these things. Anyway.

• 51:20Speaker 01

  And the largest Apple acquisition in history was Beats, right?

• 51:24Speaker 00

  That's right. Yeah. I mean, just, it's not in the

• 51:26Speaker 01

  And, and so, yeah, it's not in their DNA. It'd be a crazy

• 51:30Speaker 00

  Turnus might just be like, I'm a

• 51:31Speaker 01

  But Turnus

• 51:31Speaker 00

  I'm a wild man.

• 51:33Speaker 01

  He could be, let's ask, we'll ask the Gerberator.

• 51:36Speaker 00

  Yeah, we will.

• 51:38Speaker 01

  But yeah, I, I think like the, the, the challenge with, with Perplexity and Instinct is like, once you get in the 10 plus billion dollar range, it becomes like you, it takes like a founder led company to make a deal like that, right? We've seen this with Zuck.

• 51:53Speaker 00

  Yep.

• 51:53Speaker 01

  We've seen this with Elon, right? With Cursor.

• 51:56Speaker 00

  Yep.

• 51:56Speaker 01

  And, and so like doing these like talent acquisitions at north of 10 billion.

• 52:02Speaker 00

  Yep.

• 52:02Speaker 01

  You have to have exceptional levels of control and conviction and be able to be very bold.

• 52:08Speaker 00

  Yep.

• 52:08Speaker 01

  And I just don't see that Jassy or Turnus doing anything like that.

• 52:12Speaker 00

  No, I completely agree. Cause the shareholders will be like, wait, even if you're paying in stock, you're diluting me and I don't like that. Whereas Elon will be like, I don't care. And Zuck will be like, I don't care.

• 52:20Speaker 01

  No, it, it was an amazing play.

• 52:23Speaker 00

  And critically, like Grok did not have an enterprise motion. They did not have developer adoption. Cursor gave them that. Grok has Grok Bot. Meta has Muse. Like they don't need Instinct. So yeah, where, where it goes. But then again, you know, maybe the market's so big that they just become a really big player in personal AI agents. And they figure out how to make 5% on every transaction through whatever. They do some ads, they do some credit card stuff that we've discussed. They do some, you know, they, they, they prove Eric Sufort wrong. Agent of Commerce does not turn out to be a mirage. And then they start printing.

• 53:01Speaker 01

  Yeah.

• 53:01Speaker 00

  But who knows? Or maybe

• 53:02Speaker 01

  Yeah, but it's very possible that the, yeah, the aqua hire horse has left the barn.

• 53:10Speaker 00

  Potentially, potentially. Speaking of horses, should we get to monkeys?

• 53:15Speaker 01

  I think it's not.

• 53:16Speaker 00

  Well, we're gonna bring in Mark German in just a second, but one more fact. So Joe Lonsdale, who was recently pictured, we said it was a private jet that he was calling into CNBC on. We got a correction. We have to issue a correction. It was actually a private jet, or it was a private bus. He was on a bus with a private jet Zoom background. I think that's what was going on, right? It looked like a private jet behind him, but-

• 53:41Speaker 01

  You're accusing him of LARPing?

• 53:44Speaker 00

  I don't know. All I'm saying is that we call it a private jet background and we were corrected by the media. And so, we need to do an issue.

• 53:51Speaker 01

  I stand by, I stand by-

• 53:53Speaker 00

  You're standing by-

• 53:54Speaker 01

  My original stance.

• 53:55Speaker 00

  I'm going, it was a private bus with a private jet background. Zoom background. That's what's going on. Anyway, he's breeding monkeys. He's breeding 30,000 monkeys. He's getting in the monkey business.

• 54:05Speaker 01

  Do you think this came from like Satriani talking about the monkey bottleneck? And he was like, "That's a good idea."

• 54:12Speaker 00

  No, I don't think it was downstream of that. I think it happened simultaneously. I think, I think at one point, it was like, it was like, there's a company out there raising. There's rumors about that. 8VC starts funding it. It gets out a little bit more. Couple of people see the deal and then, you know, the semi-analysis folks are talking about it. The Satriani folks are talking about it and then Satriani like writes the piece about it. I don't think it was like, open up your phone, read Satriani and then like go find, go incubate a company or something. This feels like it's been in the works for longer. I don't know.

• 54:41Speaker 01

  You have to appreciate how this, the title of this article is probably for billions of people, the most enraging headline there could ever be. Why this Valentine billionaire is backing up-

• 54:58Speaker 00

  30,000 monkeys, just to start. Anyway.

• 55:01Speaker 01

  What are they gonna do with the monkeys?

• 55:03Speaker 00

  They're gonna test drugs on them, Jordy. That's what they're gonna do. It's gonna be animal testing. That's gonna happen. Anyway.

• 55:08Speaker 01

  Why this Valentine billionaire wants to get the-

• 55:11Speaker 00

  Agents, meet the canvas. Your agents can now create and modify your figure files with Design System contact. And we are overjoyed to be joined by Mark German from Bloomberg, here with us live in the TVP in UltraDome. Welcome back. Thank you so much for coming on down.

• 55:25Speaker 01

  Great to see you.

• 55:26Speaker 00

  Grab that controller. I'll grab mine. Have you been playing the new Call of Duty? Are you a gamer?

• 55:30Speaker 04

  No. I game when I have a chance, when I have time.

• 55:34Speaker 00

  There's a lot of work to be done with all the Apple news.

• 55:36Speaker 04

  There's a lot of work to be done. They've got MacBook Pro this morning.

• 55:39Speaker 00

  Yes.

• 55:40Speaker 04

  MacBook Pro yesterday morning, iPad Mini Monday morning.

• 55:45Speaker 00

  What's coming tomorrow?

• 55:46Speaker 04

  What's coming tomorrow? I don't know. I don't know. Maybe more cracked iPhone duos. Who knows?

• 55:51Speaker 00

  How bad is this? I saw like five iPhone duos today. None of them were cracked. Seems like a little bit like people are trying to crack them.

• 55:59Speaker 04

  Okay.

• 56:00Speaker 01

  Yeah, like, how do you rate the durability test?

• 56:03Speaker 04

  Let's take a step back.

• 56:04Speaker 01

  It's like, I did everything, I did everything, I did everything I possibly could to destroy this item.

• 56:09Speaker 04

  Let's take a step back. No. You guys sound like you're owned by Apple right now, okay?

• 56:15Speaker 01

  I'm not. I'm like the number one-

• 56:16Speaker 04

  Not OpenAI. You sound like you're owned by Apple right now.

• 56:19Speaker 00

  They rolled out the red carpet for me this morning. It was the best experience of my life.

• 56:21Speaker 04

  Really?

• 56:21Speaker 00

  No.

• 56:22Speaker 04

  No?

• 56:22Speaker 01

  I wanna be, I wanna be clear that I have not been impressed with a single thing Apple has done in like a decade.

• 56:29Speaker 04

  Okay, well, anyways.

• 56:31Speaker 01

  I just wanna be really clear. John, John is a steel man. He can steel man everything.

• 56:35Speaker 00

  I like the Apple Vision Pro, okay?

• 56:37Speaker 04

  That's insane. Okay.

• 56:40Speaker 01

  I'm a big, I'm a big home view guy. I'm a big-

• 56:44Speaker 04

  All right, it's my turn. There are people who have done iPhone Duo teardowns. People have purposely broken their iPhone duos, pouring a bag of sand on it, taking it to the beach, dipping it in Diet Coke, whatever, right? Then there are people who've used their iPhone Duo as normal. And they have run into this issue where the display cracks. The way it works, they're using the phone and they're opening it, it feels a little tight. And then they keep going and then boom, there's a big crack. And the cause of this is a manufacturing defect related to one of the components in the hinge, where there's this one bar in the hinge that gets stuck under the display and then just punctures the screen and creates that crack.

• 57:22Speaker 00

  Okay.

• 57:22Speaker 04

  Now, Apple has come out and said, "A small number of people have had this issue." But for Apple, you know, a small number, that could be tens of thousands of units. We don't know exactly how many units it is, but there are people facing this issue. This first started cropping up on Twitter, then on Reddit and some of the Apple news forums over the weekend, starting when this came out on Friday. I saw this and I immediately started investigating. I can't do a story based off of people just saying that this happened. So what I did is I put a tweet out Sunday night. You guys may have seen it. I said, "If anyone has had this iPhone crack issue, shoot me a note." I wake up Monday morning, my inbox is filled with people sending in pictures, videos, stories,

• 58:02Speaker 04

  proof of receipts, correspondence with Apple, support, Genius Bar and Authorized Reseller paperwork. And so I did a story. 18 minutes after I put my story out, Apple finally responded, right? They didn't respond until then. I think they know what's up.

• 58:15Speaker 00

  Feet to the fire.

• 58:16Speaker 04

  I think they know, I think they know who people listen to when it comes to Apple.

• 58:21Speaker 04

  So yeah, it's a real issue. And what's bothered me is-

• 58:28Speaker 04

  I feel like I'm at Universal Studios, right? Like going through Jurassic Park. There's influencers coming out and making these comments like, "Oh yeah, my iPhone's fine because I haven't purposely broke it." I just seen, said this. It's completely ridiculous. It almost feels like this is something that Apple wants to say. Like, "You're folding it wrong," okay? But they look really crappy for saying it. And they're even coming out saying, you know, "This is a real issue. You can go to Apple Care and get it replaced." But here's the problem. You buy this phone, you wait in line 12 hours for it. You have this crack issue. You go back to the Apple-

• 59:01Speaker 01

  Hard for me to relate to this.

• 59:02Speaker 04

  Fine. Right? You go back, you go back to the Apple Store you bought it from. You're SOL. We have no inventory, right? You can't get a replacement.

• 59:11Speaker 00

  And these are the most diehard fans because they waited in line. It paid the high price.

• 59:14Speaker 04

  So they're pissed off. In fact, one of my good friends, he hasn't, he bought an iPhone Duo and his device didn't have the screen crack problem, but it has a broken camera. So from day one, the diagnostic message popped up and said, "You replaced the camera on your iPhone Duo with a, with a foreign part, with a fake iPhone camera part." And I love my friend. He has no idea how to replace a camera on his iPhone Duo, okay? So I can vouch for that. He didn't do it.

• 59:44Speaker 00

  Yeah, yeah, yeah.

• 59:45Speaker 04

  This, you open the camera, it just shows black. And so he goes back to the Apple Store to try to get it replaced. They have no inventory, okay? And then they tell him, "We'll reserve an iPhone Duo for you at the Apple Store and we'll call you when it comes in."

• 1:00:01Speaker 04

  And they've got an inventory since. There's like these online trackers. You think they've ever called him? I told him the other day, I'm like, "Listen, Joe, they're never gonna call you, okay? You're gonna have to just return this thing and buy it on your own." So the process of fixing these issues and getting this resolved for people from the beginning has just been a dumpster fire.

• 1:00:17Speaker 00

  Yeah.

• 1:00:18Speaker 04

  But they're working through the quirks and hopefully this gets resolved. At least I'm happy that I've had no issues. I love the iPhone Duo. I'll tell you that right now. I have it. Oh yeah. This thing, this thing is, is, is-

• 1:00:29Speaker 00

  What is that? Is it, did you have a case on there? What's that matte texture?

• 1:00:33Speaker 04

  This is the, yeah, this is the Apple Duo case. Where's the camera? You know, we don't wanna do free advertising. But anyway, I bought the-

• 1:00:40Speaker 00

  You look like you're owned by Apple.

• 1:00:41Speaker 04

  Yeah, sure. I had the Folio, the Folio case initially.

• 1:00:44Speaker 00

  Okay.

• 1:00:45Speaker 04

  Piece of crap.

• 1:00:46Speaker 00

  Okay.

• 1:00:46Speaker 04

  Okay, it's a-

• 1:00:47Speaker 00

  And that's the official Apple case?

• 1:00:49Speaker 04

  Official Apple.

• 1:00:49Speaker 00

  And then what are you on now?

• 1:00:50Speaker 04

  I'm on the official Apple standard case.

• 1:00:52Speaker 00

  Okay.

• 1:00:52Speaker 04

  So if you're gonna get the official Apple case, don't get the Folio.

• 1:00:54Speaker 00

  Okay.

• 1:00:55Speaker 04

  The problem is there's so many features-

• 1:00:56Speaker 01

  Why do you think you need a case? These things are so durable.

• 1:00:59Speaker 04

  Fair point. Fair point. I bet you won't throw it across the studio right now in front of everyone on live.

• 1:01:05Speaker 00

  Challenge accepted, dude.

• 1:01:07Speaker 04

  I bet you won't do it.

• 1:01:08Speaker 00

  You gotta do it. Like the softest throw, yeah, frisbee. Right onto the carpet.

• 1:01:11Speaker 04

  And I'm not paying for the cracks.

• 1:01:13Speaker 00

  Onto the carpet.

• 1:01:14Speaker 04

  No, no, don't do it. Don't do it. Don't do it.

• 1:01:16Speaker 00

  Oh, did it make it?

• 1:01:19Speaker 01

  I think it survives. It's good.

• 1:01:21Speaker 00

  That was a good throw. You didn't-

• 1:01:24Speaker 00

  Oh no.

• 1:01:25Speaker 04

  What are you guys doing?

• 1:01:27Speaker 00

  Tyler, why would you throw this phone like that?

• 1:01:29Speaker 04

  Okay, so-

• 1:01:31Speaker 01

  It's good, it's good.

• 1:01:32Speaker 00

  No, no cracks?

• 1:01:33Speaker 01

  Yeah, no crack. Very good.

• 1:01:35Speaker 04

  Yeah, good.

• 1:01:35Speaker 00

  I mean, that is a machine, that is a product that's been going strong for 20 years, you know?

• 1:01:42Speaker 01

  Who throws it?

• 1:01:43Speaker 04

  I can't believe you did that. I'm not doing that. My do. Anyways, the Duo has so many amazing features that use both displays.

• 1:01:49Speaker 00

  Sure.

• 1:01:50Speaker 04

  Such as taking a FaceTime and showing the person on the other side. But that flap on the folio, it just covers that up. It makes it pretty cumbersome to use. And so I've had a great experience with the cheaper $80 Apple standard case. So if you get a Duo, I'd highly recommend that.

• 1:02:04Speaker 00

  Cool.

• 1:02:04Speaker 04

  I also saw ZAG has a nice case, too.

• 1:02:06Speaker 00

  Yeah.

• 1:02:07Speaker 04

  It's like a clear case with a.

• 1:02:09Speaker 00

  What has been your experience on the apps properly getting the iPhone Duo updated?

• 1:02:15Speaker 01

  Well, yeah, that was my thing. That was my thing. How did Apple allow this phone to launch without all the most popular apps being like at least moderately optimized?

• 1:02:24Speaker 04

  Well, I'm long enough. I'm around long enough to know you guys are too. This happens every time there's a form factor change. They went from the 3GS to the 4. Retina display.

• 1:02:32Speaker 01

  This is the biggest.

• 1:02:33Speaker 00

  But why? I mean, can't they just call all the developers and say, here's the prompt that you put into your AI agent to update it?

• 1:02:40Speaker 04

  Apparently, that's funny. Apparently, the SDK for the iPhone Duo didn't come out until like a week or two before the phone went on sale. So this was, you know, a pretty tight.

• 1:02:47Speaker 00

  Okay, so even if you're, even if you're best friends with Apple, like you're Disney Plus developer or something and you're close. I don't know if Disney's.

• 1:02:54Speaker 04

  Disney Plus, they were updated from day one, but. How long have we had?

• 1:02:58Speaker 01

  Was it like Instagram and YouTube?

• 1:03:00Speaker 04

  Instagram, day one, YouTube, day one. It's been less than a week.

• 1:03:02Speaker 00

  Okay, the big stuff's, the big stuff's good.

• 1:03:04Speaker 04

  I can count on one hand.

• 1:03:06Speaker 00

  The number of apps that feel weird.

• 1:03:07Speaker 04

  That feel weird at this point. And so the reviews missed the mark on the iPhone Duo.

• 1:03:12Speaker 00

  They jumped the.

• 1:03:13Speaker 04

  They jumped the gun. The app thing is a non-issue.

• 1:03:15Speaker 00

  Because you just wait a week and it's all refreshed.

• 1:03:17Speaker 04

  It's pretty much refreshed.

• 1:03:18Speaker 00

  It can't be that hard.

• 1:03:19Speaker 04

  The reviews focused on a few things. The reviews focused on camera quality. The reviews focused on Touch ID and the reviews focused on the app issues. The app issue is BS. It doesn't exist.

• 1:03:29Speaker 00

  Yep.

• 1:03:30Speaker 04

  The cameras are not as bad as people seem to think. I've done a lot of FaceTimes on the interior camera and it's perfectly useful for the, for the, for that use case.

• 1:03:39Speaker 00

  Yeah.

• 1:03:39Speaker 04

  Touch ID has been a complete non-issue for me.

• 1:03:42Speaker 00

  Okay.

• 1:03:43Speaker 04

  You can enroll up to five fingers. And so I've done multiple fingers twice.

• 1:03:47Speaker 00

  Sure.

• 1:03:47Speaker 04

  I have.

• 1:03:48Speaker 00

  Oh, yeah, doing twice. Wait, you did there?

• 1:03:50Speaker 04

  Yeah, I did there. So you just put it on the side. And so for me, I'm already used to Touch ID after less than a week. It's a complete non-issue.

• 1:03:57Speaker 00

  Okay.

• 1:03:57Speaker 04

  The real problem with the iPhone Duo is there are some software quirks that unfortunately made it through version 1.0 and hopefully they fix these with a software update relatively soon. Unfortunately, the cadence of software updates is so slow for Apple compared to what you see in the AI world that they need to shake that.

• 1:04:16Speaker 00

  Does it have Apple Intelligence yet? Like, can you?

• 1:04:18Speaker 04

  Yeah, it has. Well, you call it Apple Intelligence. I call it. It's Apple not so intelligent, but.

• 1:04:22Speaker 00

  Still?

• 1:04:23Speaker 04

  No, it's done.

• 1:04:24Speaker 00

  Gemini in there. I like the Gemini model. Like, if I ask it a question, like, tell me the history of the Roman Empire, like, like Gemini is gonna give me like an essay about that that's gonna be written clearly.

• 1:04:34Speaker 04

  It works, but the damage has been done on the Siri brand.

• 1:04:37Speaker 00

  Okay, yeah.

• 1:04:37Speaker 04

  That most people are probably just gonna go to ChatGPT or the Gemini app.

• 1:04:40Speaker 01

  No, they honestly, the damage has been done to Siri. It's been done to Apple Intelligence. They need, they need a new.

• 1:04:45Speaker 04

  I don't know why they didn't use Siri AI as a, as a sort of fork in the road moment to rebrand their voice assistant to something new. I thought it was really weird. They just tacked AI onto the Siri name. Obviously Siri is in popular culture, but not for a good way.

• 1:05:01Speaker 00

  Yeah, no, it's a punchline.

• 1:05:03Speaker 04

  The thing that the reviews missed on the iPhone Duo or the software quirks. So for instance, there's no picture-in-picture FaceTime when you have the thing opened.

• 1:05:10Speaker 00

  Oh.

• 1:05:10Speaker 04

  Which means like I want to be using two apps at once and then I want to box in the corner with the FaceTime of the person I'm video con.

• 1:05:16Speaker 00

  Oh, yeah, you can do that on the normal phone.

• 1:05:18Speaker 04

  Could do that on the normal phone.

• 1:05:19Speaker 00

  Interesting.

• 1:05:20Speaker 04

  I believe you can do it on the outer screen on this guy, but for some reason it's not working on the inner screen right now. Hopefully they fix that.

• 1:05:25Speaker 00

  Exactly what you want to be doing.

• 1:05:26Speaker 04

  Well, the other thing.

• 1:05:27Speaker 00

  If I'm FaceTiming, Jordy, I want to be scrolling reels.

• 1:05:29Speaker 04

  There you go, exactly. And then the other thing I do all the time is I'm doing multiple things and then I get a text come in.

• 1:05:35Speaker 00

  Yeah.

• 1:05:35Speaker 04

  Maybe it's from you and I want to just reply really quickly because you do pull down on the notification. I think that's a great feature.

• 1:05:40Speaker 00

  Cool.

• 1:05:41Speaker 04

  Has not worked for me on the inner screen. And so they need to fix that bug too. So once they get those bugs resolved, I think they're in good shape. I love the Duo.

• 1:05:48Speaker 01

  Normal thing now, though, for Apple.

• 1:05:50Speaker 04

  I guess. I would never, I would never use a non-foldable iPhone at this point. I'm.

• 1:05:53Speaker 00

  Really?

• 1:05:53Speaker 04

  Completely sold on the bigger screen.

• 1:05:55Speaker 00

  Okay.

• 1:05:55Speaker 04

  I think it's absolutely home run product.

• 1:05:58Speaker 00

  And walk me through like, when are you using?

• 1:06:00Speaker 01

  Let's go.

• 1:06:01Speaker 00

  Yeah, that's, that's huge. When are you using the outer screen versus the inner screen?

• 1:06:07Speaker 04

  Uh, when am I using?

• 1:06:08Speaker 01

  Probably need a new phone.

• 1:06:10Speaker 04

  When am I using the outer screen versus the inner screen?

• 1:06:14Speaker 04

  For quick responses, quick interactions, phone calls, where have you? But for the most, most of the time, 75% of the time, I'm using the inner screen.

• 1:06:22Speaker 00

  Okay.

• 1:06:22Speaker 04

  I also think the outer screen is not as bad as some have let on.

• 1:06:25Speaker 00

  Sure.

• 1:06:25Speaker 04

  I think it's perfectly useful, perfectly usable. And this is coming from someone who would never use an iPhone Mini. I think it's totally fine.

• 1:06:32Speaker 00

  Interesting.

• 1:06:32Speaker 04

  I like the thing. I really do.

• 1:06:34Speaker 00

  Okay.

• 1:06:34Speaker 04

  Uh, but the software quirks need to be fixed and this hardware issue with the cracked display is a real thing.

• 1:06:38Speaker 00

  Yeah.

• 1:06:40Speaker 04

  And it's pure gaslighting that I've seen from some trying to blame consumers who spent two grand on this thing. Who take care of their stuff and blaming them.

• 1:06:49Speaker 00

  Yeah, it's crazy. So you.

• 1:06:50Speaker 01

  Some people say their customers are always right, but I, I think sometimes the customer.

• 1:06:54Speaker 00

  So you basically believe you're lucky. You're in the 90% or 99% that didn't get the defective.

• 1:06:58Speaker 04

  Probably 99.9%.

• 1:07:00Speaker 00

  Yeah, realistically. And so you, you don't think yours will ever break no matter how, as long as you use it normally.

• 1:07:05Speaker 04

  So what we've seen.

• 1:07:06Speaker 00

  Yeah.

• 1:07:07Speaker 04

  And what Apple has also said is that if you're gonna have the issue, it's gonna happen pretty damn quick.

• 1:07:10Speaker 00

  It's gonna happen quickly. Okay.

• 1:07:11Speaker 04

  And so I think.

• 1:07:12Speaker 00

  This is a manufacturing.

• 1:07:12Speaker 04

  Yeah, I think I'm, I think I'm in the.

• 1:07:14Speaker 00

  You're in the clear.

• 1:07:15Speaker 04

  I think I'm in the clear.

• 1:07:15Speaker 00

  That's great. What about MacBook screens? Last time you were here in person, you made fun of me because I had the MacBook Air and Jordy had the Pro. And you said that my screen was not good and, you know, Jordy picked the right one and I.

• 1:07:30Speaker 04

  Well, now both of you have the wrong one.

• 1:07:31Speaker 00

  I know. So talk us through what actually happened.

• 1:07:35Speaker 04

  Okay, so this morning they announced the new MacBook Pro. And when exactly as we've talked about, exactly as I've expected for the last four years. It's the same as this guy.

• 1:07:45Speaker 00

  Yeah.

• 1:07:45Speaker 04

  It's a bit lighter. So the one you got there is 4.7 pounds. The new one looks identical, but they reworked the internals. So now it's four pounds.

• 1:07:53Speaker 00

  That's actually pretty significant saving, like almost 10%.

• 1:07:55Speaker 04

  Yeah, it feels much lighter. You know, I've carried. It feels much lighter.

• 1:07:59Speaker 01

  Wait, and, and you were at the event this morning?

• 1:08:01Speaker 04

  No.

• 1:08:02Speaker 01

  You don't, you don't.

• 1:08:03Speaker 04

  No.

• 1:08:03Speaker 01

  They don't really want you around there?

• 1:08:05Speaker 04

  I don't know. That's, you know, you should send them a request for comment and ask them.

• 1:08:09Speaker 00

  You know, you know, we have a, we have a really good Hollywood makeup team. So maybe the next time I get the invite, you dress up as me. Well, we, we will bring in 3D scanner.

• 1:08:17Speaker 04

  I need to wear like shoes with like really thick, yeah, lifts.

• 1:08:20Speaker 00

  Stilts.

• 1:08:21Speaker 04

  You've got probably what, three inches on me? So, yeah, I mean, you're six eight?

• 1:08:25Speaker 00

  I'm six eight.

• 1:08:25Speaker 04

  Okay, you've got eight inches on me then.

• 1:08:27Speaker 00

  Well, that's fixable. That's fixable.

• 1:08:30Speaker 04

  You know, Lakers could use another wing defender.

• 1:08:32Speaker 00

  Do the limb lengthening surgery to really pull this off.

• 1:08:34Speaker 04

  What was that movie?

• 1:08:35Speaker 01

  No, here's what we, here's what we do. Here's what we do. We have to, we have to all go and then we'll do a reenactment where.

• 1:08:42Speaker 00

  Oh, like David Fielder. You'll do Turner's. Okay, great.

• 1:08:45Speaker 01

  With the full Hollywood makeup.

• 1:08:46Speaker 00

  So we'll do a full reenactment here for you.

• 1:08:48Speaker 01

  You'll get to be in the audience.

• 1:08:49Speaker 00

  Yes.

• 1:08:50Speaker 01

  Like you're, it'll feel like you're actually there.

• 1:08:52Speaker 00

  Yeah.

• 1:08:52Speaker 01

  And we can just, we can just do repetitions like that until.

• 1:08:55Speaker 04

  Turnus will be looking out from the stage and he'll be like, what?

• 1:08:58Speaker 00

  Yeah.

• 1:08:58Speaker 04

  What's going on? Did you see the Steve Kerr thing?

• 1:09:00Speaker 00

  No.

• 1:09:00Speaker 04

  The other day on Twitter? Did you guys see this? Steve Kerr is coaching the game and all of a sudden he has his doppelganger in the audience. What's going on? I'm sure people watching this saw that.

• 1:09:08Speaker 00

  That's funny.

• 1:09:09Speaker 04

  Yeah.

• 1:09:09Speaker 00

  Yeah. So.

• 1:09:14Speaker 00

  Touchscreen, OLED, which one's the bigger deal?

• 1:09:16Speaker 04

  Oh, yeah. Okay, so looks the same, lighter.

• 1:09:18Speaker 00

  Yep.

• 1:09:18Speaker 04

  Couple centimeters thinner.

• 1:09:20Speaker 00

  Yep.

• 1:09:20Speaker 04

  And then it has three new features: Touch, OLED, and the Dynamic Island.

• 1:09:24Speaker 00

  Oh, yeah.

• 1:09:25Speaker 04

  Uses the same M5 Pro and M5 Max chips from the March ones as, as expected. There'll be new M7 Pro and M7 Max versions at the tail end of next year. Looking forward to those as well.

• 1:09:34Speaker 00

  Upgrade if you're on an older.

• 1:09:36Speaker 04

  Yeah, it's a good upgrade. I got one. I'm still on an M1 Max from 2021. Yeah, but I got the base version. I got the M5 Pro version. I'm not doing hardcore.

• 1:09:48Speaker 00

  Local inference.

• 1:09:49Speaker 04

  Yeah, well, I am, but not enough local inference that I need a $14,600 version, which is the top end now.

• 1:09:56Speaker 00

  18,000?

• 1:09:57Speaker 04

  Yeah, 16 terabytes.

• 1:09:59Speaker 00

  6,000, but I wasn't.

• 1:10:00Speaker 04

  Oh, you ordered one?

• 1:10:01Speaker 00

  No, not yet. But I was just, I was just in the specer, like, what would I actually do?

• 1:10:05Speaker 04

  Yeah, what are you getting? Are you gonna get the M5 Max?

• 1:10:06Speaker 01

  Well, here's what's crazy. Take, take, take him, hit Apple with the bra when he saw the pricing page. So if you want the RAM, so it has 96 gigs of RAM included.

• 1:10:19Speaker 04

  On which one?

• 1:10:20Speaker 00

  On this one. He specked it higher. It starts with like.

• 1:10:22Speaker 01

  No, I'm saying, but, but wait, wait. So 256 is another $3,600.

• 1:10:28Speaker 00

  Oh, you can go way up.

• 1:10:29Speaker 01

  512 is 13,350.

• 1:10:32Speaker 04

  Wait, are you on the MacBook Pro or the Mac Studio?

• 1:10:34Speaker 00

  That's, that's gotta be a studio.

• 1:10:35Speaker 01

  Is this a studio?

• 1:10:36Speaker 00

  That's gotta be the studio.

• 1:10:36Speaker 01

  Okay, okay.

• 1:10:37Speaker 00

  There's no way you get half a terabyte of memory on the MacBook Pro.

• 1:10:41Speaker 04

  I think it pops out at 100, 120.

• 1:10:43Speaker 01

  But either way, so it's now 14 grand if you want 512.

• 1:10:46Speaker 00

  Pretty high, yeah.

• 1:10:48Speaker 04

  They put out the new Mac Studio today that was like north of 22 grand with all the.

• 1:10:53Speaker 00

  I mean, some people love that for.

• 1:10:54Speaker 04

  Yeah.

• 1:10:55Speaker 00

  Information and stuff.

• 1:10:55Speaker 04

  Nobody needs it. I'll be honest with you. There's literally nothing you need it for. It's just marketing.

• 1:10:59Speaker 00

  And it's a lot of tokens on a plan with a, with a Frontier Lab. It's a lot of.

• 1:11:03Speaker 01

  Sometimes you want to drive a Ferrari to the grocery store.

• 1:11:06Speaker 00

  Sometimes you want to drive a Ferrari to the.

• 1:11:07Speaker 04

  I mean, you did go to Air One today, so maybe you did.

• 1:11:10Speaker 00

  Yeah. So if I upgrade, it would not be because of the performance. It would be because I did get the MacBook Pro on your recommendation.

• 1:11:17Speaker 04

  In March?

• 1:11:18Speaker 00

  In March.

• 1:11:18Speaker 04

  I recommended that?

• 1:11:19Speaker 00

  I think so.

• 1:11:20Speaker 04

  There's no way I would.

• 1:11:21Speaker 00

  You just made fun of me for having the Air, which has the LCD screen and not the mini LED screen. Now I got to upgrade to the OLED screen.

• 1:11:29Speaker 04

  Can we pull up the tape? I must have said something like, wait six months and get the Touch One.

• 1:11:33Speaker 00

  You probably did.

• 1:11:34Speaker 04

  Yeah.

• 1:11:34Speaker 00

  Anyway.

• 1:11:34Speaker 04

  There's no way I didn't.

• 1:11:35Speaker 00

  Uh, I have one of those at home. I would not upgrade just for the Touch.

• 1:11:39Speaker 04

  Interesting.

• 1:11:41Speaker 04

  Well, I think it's

• 1:11:42Speaker 00

  Sell me on the touch.

• 1:11:43Speaker 04

  I have no, I'm not going to sell you on the touch because I think that it's going to be little used in the end. I think people are used to using their Macs with the keyboard and trackpad. However, we're in a new generation. Everyone's growing up now with touchscreens. They had to do it. They had no choice. It is table stakes. It is spec sheet 101. It is expected. They had to do this. But there's nothing innovative about it. It's just pinch to zoom, rotate, swipe, etc. There's nothing new about it. They spent 20 years gaslighting people about not needing touchscreens. And then they did it. And they did nothing to solve the ergonomics problem that they are the ones said that existed.

• 1:12:19Speaker 01

  Well, yeah, which is

• 1:12:20Speaker 04

  Yeah, you know, let's put it, let's, let's, let's see. Put your hand on the screen.

• 1:12:23Speaker 01

  No, this is what I've, this is what I've been testing.

• 1:12:25Speaker 00

  I, I touched it today.

• 1:12:26Speaker 04

  Well, how does your forearm feel?

• 1:12:27Speaker 00

  I thought it was better.

• 1:12:28Speaker 01

  Oh, yeah. It's a good workout, right? Yeah.

• 1:12:30Speaker 04

  And you're a thick guy.

• 1:12:31Speaker 01

  It's a feature for me.

• 1:12:32Speaker 04

  Yeah.

• 1:12:32Speaker 00

  Yeah.

• 1:12:33Speaker 01

  A lot of people that might be.

• 1:12:34Speaker 00

  You're going to be able to see people with the, if they have a huge Popeye style forearm, then you'll know that's a touchscreen.

• 1:12:41Speaker 04

  That's a touchscreen.

• 1:12:41Speaker 00

  It'll be a badge of honor.

• 1:12:42Speaker 04

  Badge of honor. Well, I'm looking forward to taking it for a spin. I got the glossy. Okay. So I like the nano matte texture on the iPhone Duo, but I went with the glossy because I think glossy is better for OLED overall.

• 1:12:54Speaker 00

  Yeah. So OLED is a bigger selling point for me because I got an OLED TV. I think it looks great. And I could see the OLED being more vibrant.

• 1:13:02Speaker 04

  Yeah.

• 1:13:02Speaker 00

  And so while I don't think I will be touching it, in fact, I think I will try not to touch my screen ever because I don't want fingerprints all over it. I want it to be clean. I think the OLED will look better. So that's the selling point.

• 1:13:13Speaker 04

  Yeah, the OLED is a selling point. And the dynamic island is pretty nifty. You know, we're all used to it on our phones and whatnot. When is it going to finally come to the iPad? You know, I would imagine.

• 1:13:21Speaker 00

  That is interesting that it's behind there.

• 1:13:23Speaker 04

  Right?

• 1:13:23Speaker 00

  Yeah. Please.

• 1:13:25Speaker 01

  Samsung has been in the news, not for the best reason.

• 1:13:29Speaker 04

  What did I miss?

• 1:13:30Speaker 00

  Smart TVs spy on you.

• 1:13:31Speaker 01

  Smart TVs, like, you know, like, basically, like, the most insane, allegedly, the most insane piece of spyware.

• 1:13:39Speaker 04

  Is this recent? Because I know LG had this issue recently.

• 1:13:41Speaker 01

  Yeah, yeah. It is, feels like, like, I know, I've, I've asked you this like a million times, but, like, I now want a privacy focused TV. And it feels like Apple's the company to do that. But they're still not going to do it. They're going to do a Home View, which no one ever asked for or needed, has no real purpose.

• 1:14:01Speaker 04

  I ordered two.

• 1:14:02Speaker 00

  You did? Have you gone all in on the ecosystem?

• 1:14:06Speaker 04

  No, I've not.

• 1:14:07Speaker 01

  But, but my, my criticism is like, I would love if Apple was like, we're making a product that is, replaces all of the technology in your home. It actually is, Apple, it's going to do everything from like light switches to shades to security cameras.

• 1:14:22Speaker 04

  They have that with HomeKit and they've got the new Home, Apple Home Collection, which is basically a starter kit. You've got the cameras, all the types of cameras you need.

• 1:14:29Speaker 00

  LG with like the tight integration.

• 1:14:31Speaker 01

  But why are they doing it with LG? That's my, that's like

• 1:14:34Speaker 00

  Who would you, who would you prefer? Schneider Electric?

• 1:14:37Speaker 04

  Well, I think people

• 1:14:38Speaker 01

  Why not? They make hardware. They make cameras.

• 1:14:40Speaker 04

  They do it. You're preferring Apple does it.

• 1:14:42Speaker 01

  Yeah, these, these systems are

• 1:14:44Speaker 04

  One of the big gates to getting people into HomeKit is the cost of the accessories. And if Apple starts coming out with this stuff and self branding it, they're going to be at a very high price point and they're not going to be able to get market penetration.

• 1:14:56Speaker 00

  But it's such a flex. You have someone over and they're like feeling like the Apple doorknob that you got.

• 1:15:00Speaker 04

  Well, there'll be an Apple camera. There'll be an Apple in-home security camera sensor. I don't know if they'll call it a camera, but that's coming. But, you know, working with a third party was a way to do it. LG is a weird, weird partner though. It's just completely random. Like, if you were to list 100 companies, I would not put LG on top.

• 1:15:15Speaker 00

  Well, all the other ones are bought. They can't do Ring. That's Amazon. They can't do Nest. That's Google. Who else would you go with?

• 1:15:21Speaker 04

  LG didn't make any of this stuff.

• 1:15:22Speaker 01

  To me, Apple is at a point where they can no longer innovate. So why not just create, like, great commodities across the entire

• 1:15:31Speaker 04

  What do you think of Apple's AI efforts?

• 1:15:34Speaker 01

  I mean, I don't like, it's been, it's been talked about enough on this show and I believe over time they can make a great experience.

• 1:15:41Speaker 00

  I thought the Google partnership is great, just in that they've worked together on the browser before Safari routing to Google Search. It makes sense to set up the

• 1:15:48Speaker 01

  I thought the pain, the most painful thing is like, right, is they're getting around to creating like a competent chatbot, like agents and like Instinct came around and Muse came around. And now that's what consumers want.

• 1:15:59Speaker 04

  Muse is really good.

• 1:16:00Speaker 00

  Yeah. And so they, and so, and when will they have that? Three years?

• 1:16:04Speaker 01

  Funny thing is they're not, they're not seeing Muse and being like, oh, this is an exciting new paradigm. They're like sigh of relief, you know, sigh, like, let me smoke a cigarette because like, this is going to take us another three years, kind of thing.

• 1:16:16Speaker 04

  The problem is, is that the

• 1:16:17Speaker 01

  But, but, but back to my point, it's like, I'd be, I'd be happier if Apple was just committed to like, we make, we make hardware, we're going to make the best version of everything in every category. The TV, like, I'm, I'm sitting here as like, sort of disappointed as somebody who grew up as an Apple fanboy, who spent my whole childhood in the Mac ecosystem, who would work, like, would, would, would, starting when I was like, probably 12, 13, would be like, working my butt off so that I could save up and buy like the best version of every Apple product. And now I no longer think of them as an innovator, but I think of them as somebody who will make the best version of the

• 1:17:00Speaker 01

  home security camera, who will make the best version of like, an app that lets me like unlock my gate when I'm having a friend over, right?

• 1:17:09Speaker 00

  So realistically, you should be able to go to Dot or Muse or Duet or whatever AI agent is actually great and say, how many deer have walked in my backyard in the last month? And it can go and pull the data from Apple's camera or, or, or adjust the screen time limits on my Apple TV.

• 1:17:26Speaker 04

  I think the Apple

• 1:17:26Speaker 01

  But, but the funny, the funny thing about the Home, the Home View is like, I, my house, I inherited the smart home system. And I, as I'm getting it set up, it's like, the app has like 10 reviews in the, in the App Store. And I'm like

• 1:17:40Speaker 04

  Is it iPhone Duo compatible?

• 1:17:42Speaker 00

  Probably not.

• 1:17:43Speaker 01

  Probably not. And I'm not even going to say it because it's like a small company that makes this like operating system. But, but in looking into what the actual cost of the entire system and all the hardware in the home, which again is a tough thing to sell because you have to sell it in when the home's being built and it's like, not easy. But like, it would cost me like six figures to remove all of it from the house, right?

• 1:18:07Speaker 04

  It's not worth it. It's not worth it.

• 1:18:08Speaker 01

  And, and, and I'm stuck with this software that's like horrible. It's not reliable. And I actually have to use it every single day.

• 1:18:16Speaker 04

  Just get a new house.

• 1:18:17Speaker 01

  Yeah, basically. Or no, I'm almost, I'm almost at the point of like, just like, I guess I got a vibe code, my own smart home operating system.

• 1:18:25Speaker 04

  That's funny. That's funny.

• 1:18:27Speaker 00

  I saw the Home View and I was like, I get that the product is not super exciting. Like, it's kind of just an iPad on the wall. I don't really want the one that moves around, but I could see

• 1:18:37Speaker 01

  Sell me the Home View.

• 1:18:38Speaker 00

  I could see those.

• 1:18:39Speaker 04

  It's not my job.

• 1:18:40Speaker 00

  But I would go.

• 1:18:41Speaker 04

  I like the idea. Well, I like it for FaceTime.

• 1:18:44Speaker 00

  Okay.

• 1:18:45Speaker 04

  I like it for FaceTime.

• 1:18:46Speaker 00

  Did you get the mobile one?

• 1:18:47Speaker 04

  Well, then that one doesn't exist yet.

• 1:18:49Speaker 00

  Oh, it doesn't exist?

• 1:18:49Speaker 04

  Yeah, that'll be next year or the year after. Yeah, that'll be the higher, that'll be the higher end one. That's like the AIified, supe-dup, pricey version. That one's sort of like the, the sort of the Holy Grail version. But Apple's had a big year.

• 1:19:03Speaker 00

  Yeah.

• 1:19:04Speaker 04

  The MacBook Neo, the iPhone Duo, Touch MacBook Pro, obviously, the, the, the new Apple Watches and AirPods and iPhone 18s were quite minor, but there was some good stuff. Next year, there'll be some pretty intriguing things too. The smart glasses are the big things for next year.

• 1:19:19Speaker 00

  How competitive are those going to be with the Meta VR glasses, which

• 1:19:23Speaker 04

  Well, no displays.

• 1:19:24Speaker 00

  They're into gear.

• 1:19:25Speaker 04

  Well, there's no new, there's no new Apple XR product next year.

• 1:19:30Speaker 01

  You could say like, it's speakers you can wear.

• 1:19:33Speaker 04

  It's AirPods in a glasses form. It's AirPods for your face.

• 1:19:36Speaker 01

  Speakers you can wear. Finally.

• 1:19:37Speaker 04

  It's AirPods for your face with AI.

• 1:19:41Speaker 00

  AI communication.

• 1:19:42Speaker 04

  AI cameras, talking to it.

• 1:19:43Speaker 00

  Oh, it has cameras too.

• 1:19:44Speaker 04

  Oh, yeah. Oh, yeah. It's a competitive product to the Meta glasses.

• 1:19:47Speaker 00

  Yeah.

• 1:19:48Speaker 04

  The big question is, do you want an Apple designed glasses or do you want Ray-Bans? Right? Who would you trust with a design for glasses? The company that's doing it for the first time? Or the company that's done it for, I don't know, how many years has Ray-Ban been around?

• 1:19:59Speaker 00

  Long time.

• 1:20:00Speaker 04

  Well, decades.

• 1:20:01Speaker 00

  Yeah, decades.

• 1:20:02Speaker 01

  I'm riding with Luxottica.

• 1:20:03Speaker 00

  Luxottica.

• 1:20:04Speaker 04

  That was super smart on many Meta's part to partner with Luxottica because nobody would buy Meta branded glasses.

• 1:20:10Speaker 00

  No way.

• 1:20:11Speaker 04

  Right?

• 1:20:11Speaker 00

  I mean, we ran that experiment with Snap.

• 1:20:14Speaker 04

  Well, sure, but I don't think the Snap issues were because of the Snap brand. I, I quite frankly think the Snap brand is a, is a trusted brand, whereas the Meta brand in some cases is not trusted.

• 1:20:25Speaker 00

  Oh, sure, sure, sure. I was just thinking of like, those original glass, camera glasses from Snap.

• 1:20:29Speaker 04

  Oh, everyone wanted those.

• 1:20:30Speaker 00

  Yeah, but they didn't actually sell

• 1:20:32Speaker 04

  You didn't wait in line for those as a kid in college?

• 1:20:34Speaker 00

  They didn't sell that well because I think that they looked good, like they were designed well, but they just weren't an iconic silhouette, like

• 1:20:41Speaker 01

  What do, what does Apple think about like Wired hair, AirPods or Wired headphones becoming like the new thing?

• 1:20:48Speaker 04

  I don't think they're going to go backwards on that. I really don't. I mean, I know it's hot right now.

• 1:20:53Speaker 01

  I know they're not going to go backwards, but they still sell the product.

• 1:20:56Speaker 04

  What are they, 19 bucks?

• 1:20:57Speaker 01

  Yeah, it's, it's ama, it's amazing deal when you're looking for me. I have been riding with Wired for years now. And I still buy them from Apple, but I like that for the price of one pair of AirPods, I can buy like, you know, five or six.

• 1:21:11Speaker 04

  Well, you know, what's funny is they probably haven't manufactured them for years, right? They probably had like a bunch sitting in a warehouse and like, all of a sudden we have to, you know, figure out, hmm, where are we building those? How do we do that? And they have to like, start the lineup again because you're right. Bose has them now. It's become part of the zeitgeist right now that people want, you know, these nostalgic products. And I think it would be a bad look for Apple to like

• 1:21:34Speaker 00

  Stop selling them.

• 1:21:34Speaker 04

  Go nostalgic in that way. But I think there's like these little like tricks, bells and whistles.

• 1:21:39Speaker 01

  I think what they should do is make just like a much more expensive version of the Wired, like just make, because what I want is like

• 1:21:47Speaker 00

  Great sound quality.

• 1:21:48Speaker 04

  Like, it'll never happen.

• 1:21:49Speaker 01

  Yeah, yeah, yeah. It will never, never happen.

• 1:21:51Speaker 04

  But what I'm saying is they can bring back these little bells and whistles that, that, you know, give us some nostalgia. Like, for example, I forgot there's a fourth new feature on this new MacBook Pro.

• 1:21:59Speaker 00

  What's that?

• 1:22:01Speaker 04

  When you were back in, in high school and college, what happened on the back of your MacBook Pro?

• 1:22:05Speaker 00

  Light up.

• 1:22:06Speaker 04

  Yeah.

• 1:22:06Speaker 00

  Yeah.

• 1:22:07Speaker 04

  So these, these new boys have the, the light up Apple logo.

• 1:22:10Speaker 00

  Why is that happening? Is that because the cost of the laptop and the memory has gone up so much that the relative price of the light up on the back is cheaper now on a relative basis?

• 1:22:21Speaker 04

  My, when I saw that,

• 1:22:22Speaker 00

  Yeah.

• 1:22:23Speaker 04

  my immediate speculation was the, using the tandem OLED panel, so the, the thinner panel. Yeah, and the, the brightness benefits. Um, you're able to use the display, uh, the back of the panel as a way to light it up. But we'll see when these things get torn apart, what happened here.

• 1:22:41Speaker 00

  Sure.

• 1:22:41Speaker 04

  I, I, it's also possible they just did it, like I was saying, as like a nostalgic element.

• 1:22:44Speaker 00

  Have people been asking for this for a long time?

• 1:22:46Speaker 04

  The glowing Apple logo?

• 1:22:48Speaker 00

  Logo?

• 1:22:48Speaker 04

  No, whatever.

• 1:22:49Speaker 00

  People were cheering at the event. They were like super excited.

• 1:22:51Speaker 04

  It's ridiculous. What everyone was asking for was to bring back MagSafe,

• 1:22:54Speaker 00

  Sure.

• 1:22:54Speaker 04

  which they did.

• 1:22:55Speaker 00

  Yep.

• 1:22:55Speaker 04

  Was to bring back ports, which they did.

• 1:22:57Speaker 00

  Yep.

• 1:22:58Speaker 04

  And was to return to, um, function over form, or at least matching function to form.

• 1:23:05Speaker 00

  Sure.

• 1:23:05Speaker 04

  If you remember the 26 Mac, 2016 MacBook Pro, they got rid of a lot of the ports. They got rid of the SD card slot. They got rid of MagSafe.

• 1:23:12Speaker 00

  That one had the touch bar now.

• 1:23:13Speaker 04

  That one had the touch bar. They got rid of the physical function row.

• 1:23:16Speaker 00

  Yeah.

• 1:23:16Speaker 04

  Um, with this one that you've got, the 2021 MacBook Pro design, they fixed the MacBook Pro. Like, this is the perfect MacBook Pro.

• 1:23:26Speaker 00

  Yeah, yeah, yeah.

• 1:23:27Speaker 04

  They didn't redesign it completely.

• 1:23:28Speaker 00

  But it's boring because it's just at the end of history. It's like the best and there's nothing you can do.

• 1:23:34Speaker 04

  Maybe.

• 1:23:35Speaker 01

  Yeah, I mean, I, I really want the 16-inch iPhone Duo.

• 1:23:39Speaker 04

  16-inch iPhone Duo.

• 1:23:40Speaker 00

  That would be huge.

• 1:23:41Speaker 04

  7.6 inches is not enough for you?

• 1:23:43Speaker 00

  Yeah, I think you could di- you could daily drive this, just, you know.

• 1:23:46Speaker 04

  Can you imagine?

• 1:23:47Speaker 00

  If you had a big enough pocket, you could take.

• 1:23:48Speaker 01

  Okay.

• 1:23:49Speaker 04

  A wagon.

• 1:23:49Speaker 01

  Does, does Turnus have the balls to buy Instinct?

• 1:23:57Speaker 00

  Hmm.

• 1:23:59Speaker 04

  How much would it cost?

• 1:24:00Speaker 00

  Over 30 billion dollars.

• 1:24:03Speaker 04

  You know what's really interesting? This kind of went under the radar. I broke the news a few weeks ago that they bounced their head of M&A.

• 1:24:08Speaker 00

  Hmm.

• 1:24:08Speaker 04

  And, uh,

• 1:24:09Speaker 01

  Interesting.

• 1:24:10Speaker 04

  put his deputy in charge.

• 1:24:11Speaker 00

  Hmm.

• 1:24:12Speaker 04

  And I think there's a couple of factors here. Your good friend Eddie Q.

• 1:24:16Speaker 00

  Mm-hmm.

• 1:24:16Speaker 04

  Uh,

• 1:24:17Speaker 00

  Former guest of the show.

• 1:24:18Speaker 04

  Former guest of the show. That's what I mean by good friend.

• 1:24:20Speaker 00

  Of course.

• 1:24:20Speaker 04

  I don't know if you guys are actually boys, but who knows? Um,

• 1:24:24Speaker 00

  We both like cars.

• 1:24:25Speaker 04

  Both like cars. He,

• 1:24:27Speaker 01

  He's with us.

• 1:24:28Speaker 04

  Uh, you know, at some point he's gonna retire. Who knows? Three, five years, whatever. He loves his job. He's not going anywhere, anywhere soon. He needs to have like potential successors in place because services is a big part of, of the company.

• 1:24:40Speaker 00

  Yeah.

• 1:24:41Speaker 04

  And so you've basically now got, uh, Oliver Schuser, who's the head of Apple Music, Apple Sports, Apple TV Business, a few things. He's there. He's sort of kind of a shoe-in. But you never want to be in sort of a lame duck situation where you only got one successor. And then the Apple Services business is so big that it may not be a clean cut replace Eddie with Oliver. And so you may need two people. And so what they did is they moved Adrian Pareka, who was the head of M&A, under Eddie Q. And he's running a lot of the business side of services. And so that's one factor why they took M&A away from, from Adrian, uh, to sort of put him in the services sort of pipeline there as a potential successor. I think more than potential as a successor to Eddie Q eventually.

• 1:25:24Speaker 04

  The other factor is that Turnus and the hardware engineering guys and the technology guys, they've had gripes for decades that this guy buys no one.

• 1:25:34Speaker 00

  Oh, interesting.

• 1:25:35Speaker 04

  Right? The, the joke with,

• 1:25:36Speaker 01

  They're like, give us something to work with.

• 1:25:38Speaker 04

  Yeah. The joke with Apple M&A is they do nothing.

• 1:25:40Speaker 00

  Yeah.

• 1:25:40Speaker 04

  It's the easiest job in the world, right? This is how the, the hardware guys, Yeah, this is how the hardware guys think of it.

• 1:25:47Speaker 00

  You just have an autoresponder. No thanks.

• 1:25:49Speaker 04

  Basically.

• 1:25:49Speaker 00

  We're not interested in buying.

• 1:25:50Speaker 01

  Well, they do buy a lot, but it's like,

• 1:25:52Speaker 00

  Small company.

• 1:25:53Speaker 01

  Tiny teams. They're, they're like, you can't announce it.

• 1:25:55Speaker 04

  We want cool stuff.

• 1:25:56Speaker 00

  Yeah.

• 1:25:57Speaker 04

  Okay.

• 1:25:57Speaker 00

  But it's not like, there's a different version of Apple, which is like they're buying Sonos and they're buying Nest and their Ring and they're like,

• 1:26:03Speaker 04

  Which they should have.

• 1:26:04Speaker 00

  and all these things and like,

• 1:26:05Speaker 04

  Which they should have.

• 1:26:05Speaker 00

  Apple-ifying them over time.

• 1:26:07Speaker 04

  Yes, they've done nothing.

• 1:26:08Speaker 00

  They've done nothing.

• 1:26:08Speaker 04

  Okay? And so I think you have the Eddie thing as, as, you know, one reason. I think the other reason is that Turnus wanted a changing the guard on M&A, someone who's gonna sort of, you know, get cracking.

• 1:26:20Speaker 00

  Mm-hmm.

• 1:26:21Speaker 04

  Um, and so I don't know if they'll buy Instinct, but I think they have a better chance of buying Instinct that they did.

• 1:26:26Speaker 01

  That's actually very interesting.

• 1:26:27Speaker 04

  And they did three months ago.

• 1:26:27Speaker 01

  It feels like they need some way to insert, uh, people into the company that can innovate on software.

• 1:26:33Speaker 04

  Uh, that's true. The software side of things right now, I think, is the weakest part of, of Apple.

• 1:26:39Speaker 00

  Well, we had an idea for an expansion.

• 1:26:41Speaker 01

  Yeah, I'm being a little bit too hard on them.

• 1:26:43Speaker 00

  Yeah.

• 1:26:43Speaker 01

  Uh, just, they clearly are, they, they make the best devices.

• 1:26:47Speaker 04

  Yeah, but they don't make the best software anymore. Actually, I take that back. Maybe they do make the best software. It's just not as good as it should be.

• 1:26:53Speaker 01

  We all know it's not as good as, as, as they could.

• 1:26:55Speaker 00

  Yeah.

• 1:26:56Speaker 01

  Yeah.

• 1:26:56Speaker 00

  Yeah.

• 1:26:57Speaker 01

  So.

• 1:26:57Speaker 00

  If they wanna, if they wanna expand Apple Services, have they looked at, uh, Apple Markets or some sort of like casino game integration on your phone?

• 1:27:05Speaker 04

  You know, we did a story on Bloomberg Business Week six months ago or so that said that Apple makes more money from casino games, that type of game, uh, than they make on the Mac.

• 1:27:21Speaker 00

  Wow.

• 1:27:22Speaker 04

  Because if you think about how much money the App Store makes and what percentage of the App Store is these like casino type of games.

• 1:27:30Speaker 00

  Yeah.

• 1:27:31Speaker 04

  It's like, it's billions.

• 1:27:33Speaker 00

  So they could vertically integrate, maybe even open.

• 1:27:35Speaker 04

  Vertical integrate. I don't think they're gonna go in that direction.

• 1:27:36Speaker 00

  They could even open a physical.

• 1:27:37Speaker 01

  Wait, what kind of, what, so, so,

• 1:27:39Speaker 00

  Go and play roulette.

• 1:27:40Speaker 01

  So, apps that, that operate markets, Apple can't take a cut when you're doing like a stock, when you're trading stocks?

• 1:27:49Speaker 04

  I don't think they can or do.

• 1:27:50Speaker 01

  They, they don't.

• 1:27:51Speaker 04

  Yeah.

• 1:27:51Speaker 01

  We know that.

• 1:27:52Speaker 04

  Yeah.

• 1:27:53Speaker 01

  Then it wouldn't, the, the whole, all the economics would not work.

• 1:27:55Speaker 04

  They should. That'd be great.

• 1:27:56Speaker 01

  So what casino, what casino game?

• 1:27:58Speaker 00

  30% of your retirement in Robinhood because you installed it on the iPhone.

• 1:28:01Speaker 04

  Can you imagine? Would you be shocked if there was a headline out, like, "Apple Bans Robinhood From The App Store Because It's Not Using Their Payment Process"?

• 1:28:07Speaker 00

  Can you imagine?

• 1:28:08Speaker 04

  Right? That's like, that feels like a headline from three years ago, like a legit headline for three years ago.

• 1:28:12Speaker 01

  No, but what, what, what are these casino games that, that are not tied, they're not like, because sports betting apps, they're not taking a cut, they're not taking a cut of?

• 1:28:22Speaker 00

  Well, don't they kick you out and ask for your credit card to load up or take crypto or something?

• 1:28:27Speaker 04

  I'll be honest with you guys.

• 1:28:27Speaker 01

  No way.

• 1:28:28Speaker 04

  I don't use any of these apps. I don't use Polymarket or Call Sheet.

• 1:28:31Speaker 01

  Well, you're not doing your job.

• 1:28:32Speaker 04

  No, I actually, I, I, I'm,

• 1:28:34Speaker 01

  This company makes billions of dollars from casino games and you're like,

• 1:28:37Speaker 04

  I'm keeping my job because I don't want to use those things actually, so there you go.

• 1:28:40Speaker 01

  Not even for research? You can't, you can't get $100 a phone?

• 1:28:44Speaker 00

  We're watching sports. Certainly that's not a violation of the first terms.

• 1:28:47Speaker 04

  I, I get angry enough watching the Lakers.

• 1:28:50Speaker 00

  Okay.

• 1:28:50Speaker 04

  Okay? And get anxious enough.

• 1:28:51Speaker 00

  If you have money on the line, you'll be okay.

• 1:28:52Speaker 04

  I don't, I don't need to add another, uh, layer to that.

• 1:28:55Speaker 00

  Okay. Well, I appreciate your, uh, your, your, your discipline there.

• 1:28:59Speaker 01

  Very cool that Apple makes billions from casino games, so.

• 1:29:03Speaker 00

  Yeah.

• 1:29:04Speaker 04

  That they criticize and say shouldn't exist.

• 1:29:06Speaker 00

  Yeah, the digital economy. They get their cut. Anyway, the house always wins with Apple. Always wins. Thank you so much for coming on the show.

• 1:29:14Speaker 01

  I wish we had more time.

• 1:29:15Speaker 00

  Of course.

• 1:29:16Speaker 04

  Oh, it's, let's hang out soon. Yes. See you, guys. Thank you.

• 1:29:18Speaker 01

  I'm gonna, you sold, I think you sold me on the Duo.

• 1:29:21Speaker 00

  Ooh, okay.

• 1:29:22Speaker 04

  Well, you need one now because I see it's a big, uh,

• 1:29:24Speaker 01

  Yeah, there's a big, big chunk out of this one.

• 1:29:26Speaker 00

  They're really tested. He's the Whistlin Diesel of technology, Geordie Hayes.

• 1:29:31Speaker 01

  Let's get a, can we get a replay on that durability test pulled up, especially Tyler's throw?

• 1:29:37Speaker 00

  I think the replay is tied to the Gong Cam specifically. Let me pull up MongoDB. What's the only thing faster than the AI market? Your business on MongoDB. Don't just own the data platform. Uh, don't just, uh, don't just build AI. Own the data platform.

• 1:29:49Speaker 01

  Production team, now the next guest is with us. Let's, let's work on getting a replay.

• 1:29:55Speaker 00

  Because we have Zach Rash from Coco Robotics, the co-founder and CEO for his first appearance here on TVP. Welcome to the show, Zach. Thank you so much for taking the time to come out with us. How you doing?

• 1:30:07Speaker 04

  Gentlemen, thanks for having me.

• 1:30:09Speaker 01

  It's great to have you.

• 1:30:10Speaker 00

  How many of these robots you got these days?

• 1:30:13Speaker 04

  We got about a thousand of them. Uh, but we have our next robot coming out soon, so you'll start seeing those around LA. Um, have you guys gotten a delivery?

• 1:30:21Speaker 00

  What we, uh, I have never gotten a delivery from a delivery robot. I see them all the time in Los Angeles. That I know of. Can, can the robots release the item autonomously yet?

• 1:30:33Speaker 01

  We don't know that, that Coco's not live testing humanoid form factors with, you know, rubber masks that, that make them.

• 1:30:41Speaker 00

  Potentially.

• 1:30:42Speaker 01

  We don't know. We don't know.

• 1:30:43Speaker 04

  That's true.

• 1:30:43Speaker 01

  They haven't come out and said that they're not doing it.

• 1:30:45Speaker 00

  But the current, the current system is if I were to get a delivery in a Coco robot, I would go up to it and pull the bag of food out of the bucket, right?

• 1:30:56Speaker 04

  Yeah, yeah, exactly. Well, we gotta get you guys a delivery then.

• 1:30:58Speaker 00

  Let's do it.

• 1:30:59Speaker 04

  But, um, yeah, for anyone that doesn't know, I mean, this is, this is what they look like. Uh, this is a miniature, pretty printed version of it, but it's a purpose-built robot, um, designed for, uh, moving goods around our cities.

• 1:31:11Speaker 00

  Yeah. And you're expanding to the UK? How did that come together?

• 1:31:16Speaker 04

  Yeah, so we, here in the US, we work with Uber Eats and, and DoorDash. Um, DoorDash had recently acquired Deliveroo, which is based over in London.

• 1:31:25Speaker 00

  Sure.

• 1:31:25Speaker 04

  And so we've been working with DoorDash for, I think, almost four years at this point. We have a really close partnership and, uh, we wanted to bring it across the pond and bring it to London. Um, and so we announced, um, Just the week, a kind of a multi-year, bi-directionally exclusive partnership to bring cocos all throughout the UK and bring it to some real scale.

• 1:31:45Speaker 00

  How do you, how do you think about, or how is, how are these sidewalk robots different from the delivery network that you might see on DoorDash with people in cars? Like, is there a specific robot assigned to an individual restaurant because they're just going back and forth? Or is it more like a fleet, a network, it's driving around and it might stop by any random restaurant and the restaurant just has to sort of see the robot and say, oh, I should put this there. How does that work?

• 1:32:12Speaker 04

  Yeah, they're, so they're shared. So they'll, we'll get offers from the different platforms. We also work directly with some retailers. So we get a bunch of offers come into our system and then the robot will, you know, go somewhere to pick up an order, deliver it to the customer and then go on to pick up the next one. So just in LA, I mean, we have probably a few thousand merchants across grocery, food, retail that use the robots.

• 1:32:34Speaker 00

  Has there been any movement on getting the restaurants to help do some of the charging and some of the infrastructure, sort of like use their back rooms for space? Or do you have to do sort of what Lime and, and Bird did where you have trucks driving around, servicing things? Like, what's been the solution there?

• 1:32:55Speaker 04

  You had some mix. I mean, ideally, you don't want to be going around swapping batteries because it means you're, you know, doubling, at least doubling your battery capex. It takes forever to drive across a place like Los Angeles to manually swap the batteries. And there's not that many deliveries happening at four in the morning. So ideally, they go back to merchants, the merchants plug them in and then they take a few hours of downtime, right? So you run them 20 hours a day and then maybe you have four hours of downtime to do cleaning and charging. We do work with some of our merchants to do this. We work with a bunch of kind of ghost kitchen companies that also are a great fit at actually charging these and helping us do some of the maintenance. And then some of our big retail partners, they have a bunch of space. They might have 24/7 staffing, in which case we can kind of work with them to drive some income mile revenue back to

• 1:33:39Speaker 04

  their store in exchange for doing some of the fleet management. So you'll never have 100% in any one of these categories, but the idea is to kind of work with our partners to reduce the overall cost of operating the fleet.

• 1:33:48Speaker 01

  What's the bottleneck to going from 1,000 active robots? Sounds like this new deal in the UK will, will increase that. But how, how do you go to a million? Like, what are, what are the kind of things that need to happen?

• 1:34:00Speaker 04

  Yeah. So, um, you know, we're going to bring cocos in about 10 cities today across the US and Europe. And next year, we should be in between 40 to 50 cities across even more continents. And, um, you know, the first step is regulatory, you know, you need to create new policy for this. This is a new class of, of thing. It's not a vehicle. It's not a person. It's, it's a robot. And so you need to define laws around how the robots operate in our cities. So you work with different, either national transportation agencies or at the state level or at the city level, depending on where we're going, introduce policy. And then ramping up manufacturing of our, of our new robots. So this is Coco 2. It's the next gen robot that you'll start seeing rolling around the cities in the coming

• 1:34:44Speaker 04

  weeks. So we're planning on making about 20,000 of those for 2027. So you'll see a pretty large step up in the fleet size across a lot of new, a lot of new territories. And then the path to a million feels pretty inevitable from there. I mean, if you look like, you know, a platform like DoorDash probably has 10 million drivers or something like that, working, working on the platform in the US. And the volumes on these platforms is still growing 20 to 30% a year. So if you just compound this out, I mean, look at the, at the end of the next decade, right? At the end of this, in the next 10 years, you're going to have, you're going to have to have, you know, 50% of the US working population, you know, driving for these platforms. So you, you need to find a way

• 1:35:26Speaker 04

  of increasing supply without getting, uh, human drivers because you're just going to run out of people.

• 1:35:31Speaker 00

  Awesome. A lot of people in tech are worried about X risk these days. Hypothetically, uh, where's the vulnerable point? Uh, if I need to disable this with like a 556 round? Where would I shoot it?

• 1:35:45Speaker 04

  I can't tell you that, but I-

• 1:35:46Speaker 00

  Oh, wow. He's on the side of the robots. This guy's traitor to humanity. Crazy.

• 1:35:53Speaker 04

  What's funny is the actual policy in Japan-

• 1:35:55Speaker 01

  You need like a dead man switch.

• 1:35:56Speaker 00

  Okay. Yeah. Send the delivery over. Send it over. We want to test it out.

• 1:36:00Speaker 04

  In Japan, in Japan, you actually have to have a big red kill switch button on the robot, um, which I feel like would uniquely only work in Japan.

• 1:36:09Speaker 00

  Yeah. Because otherwise everyone would be pushed constantly. Of course. Uh, but separately, uh, jokes aside, uh, I, I have no doubt that, uh, that you are feeling the benefits of general model progress and AI progress. Like, just take an image, identify if there's someone there, if you can go straight on this road, all that's probably happening. But what do evals, what do benchmarks actually look like? What metrics are you measuring? And what are you seeing in those? Like, how do you know that the latest iteration of your software, of your AI stack is actually improving the experience? It feels like we sort of feel the failure mode, like the hallucination is like the viral video on Instagram of the cocoa robot getting stuck on the curb. But how are you measuring that? What's progress been like?

• 1:36:56Speaker 04

  Yeah. Well, it's, it's interesting because I think in robotics, I mean, you see a lot of demos of folding clothes. And then in the self-driving car world, you have a lot of road driving, right? It's a very rule-based environment. The interesting thing about city sidewalks is they're the most complex, dynamic, not rule-based environment in the world. And every city's different, right? Navigating downtown LA is, is, is, is unique and different than Chicago, which is very different than Finland, which is very different than parts of Asia. Um, and so you have different, uh, kind of social norms and behaviors. And so the robot actually needs to learn how to be a great, you know, user of the shared space that the public, it has to be a great sidewalk user. Um, but we also need to get from point A to point B very quickly. So we can't spend a lot of time

• 1:37:41Speaker 04

  stopping. So there's this really complex, uh, kind of social navigational problem that we have to get very good at. So I think we have, uh, you know, we have millions of hours, the millions of miles of driving in these sort of environments, you know, snows, rain, uh, sun, uh, you know, weekend nightlife, uh, all these different sorts of environments. So we probably have one of the largest data sets of this sort of data. Um, and that's used to train all of our models. Uh, but then we've actually open sourced and published some really interesting benchmarks. Cause I think as these models progress, I think some of these big labs might want to be testing, hey, how does our model actually do at navigating the, the public realm? How do we actually do it navigating, um, you know, a downtown Chicago sidewalk? Um, and so we actually have some benchmarks we've created from these

• 1:38:25Speaker 04

  different environments, um, that we've published. And I think that will be helpful to the robotics industry in general.

• 1:38:31Speaker 00

  Very cool. Well, congratulations on the progress. Thank you so much for coming on the show. Jordy, sorry, do you have any-

• 1:38:35Speaker 01

  Yeah. Excited to see more of these around. It's, it's an interesting thing. Like you got it. You want to, as a business, you want to scale quickly. But I think like the, the lesson from Flock is that like individual consumers have so much power over, over like what's allowed to operate in their area. And I think that, um, making sure that, you know, friendly robots are, are never annoying is like, has to be like number one priority. And, and that's such a hard problem versus what you were talking about with like on the roads. It's like, we have established laws. When the light's green, go. Oftentimes if I'm behind John, I have to honk after the light goes green. It's not getting any greener, buddy. But, uh, anyways, very difficult problem. Um, and, uh, yeah, I want to see

• 1:39:20Speaker 01

  the cocoa robot in Japan. I promise not to touch the big red button. I can be trusted.

• 1:39:26Speaker 00

  I imagine you, uh, hiring a couple of these just to patrol the property. This feels like an intermediary. If you, if you don't want to go robot dog patrol and humanoids aren't working yet, just having a, a fleet of cocos circling the property at all times, that's security.

• 1:39:41Speaker 01

  I like it.

• 1:39:42Speaker 04

  Well, guys, how do we get some lunch orders going to the office, going to the studio?

• 1:39:45Speaker 00

  We'll, we'll coordinate. Nick, we'll text you right now. We'll figure it out.

• 1:39:48Speaker 01

  Yeah, we'll do it.

• 1:39:48Speaker 00

  We'll get our lunch delivered with cocoa. Thank you so much.

• 1:39:50Speaker 01

  Yeah, great to hear. Thanks, guys.

• 1:39:52Speaker 00

  Have a good rest of your day.

• 1:39:53Speaker 01

  See ya.

• 1:39:54Speaker 00

  We do have the phone toss replay live on TVP and Ready. Let's see it in action. Jordy-

• 1:40:01Speaker 01

  We get a little zoom in?

• 1:40:03Speaker 00

  Jordy, uh-

• 1:40:04Speaker 01

  Boom.

• 1:40:05Speaker 00

  Oh, that-

• 1:40:06Speaker 01

  But I really want to see Tyler.

• 1:40:07Speaker 00

  Held up well.

• 1:40:08Speaker 01

  Because Tyler was-

• 1:40:09Speaker 00

  Tyler comes in-

• 1:40:09Speaker 01

  Was bolder because he thought I was going to catch it.

• 1:40:12Speaker 00

  Yeah.

• 1:40:13Speaker 01

  Let's zoom in. Let's zoom in.

• 1:40:15Speaker 00

  Okay.

• 1:40:16Speaker 01

  Let's see what's happening.

• 1:40:17Speaker 00

  Can we see this?

• 1:40:17Speaker 01

  Throws it. Oh, gets out of the way.

• 1:40:20Speaker 05

  Yeah. When Jordy moves back.

• 1:40:23Speaker 00

  You fell for that one, Tyler. You fell for it. Hook, line and sinker. Uh, our next guest is the co-founder and CEO of Formation Bio with Ben Liu, uh, launching a huge drug development fund. Ben, welcome to the show. Let's get this gong ready. How are you doing? Please kick us off.

• 1:40:40Speaker 04

  Thanks for having us.

• 1:40:40Speaker 00

  Yourself and the organization.

• 1:40:43Speaker 04

  Great. Uh, so Ben Shaplin, Lou, co-founder, CEO of Formation Bio. We're an AI native company, uh, building-

• 1:40:50Speaker 06

  The next generation pharma company with this kind of core thesis that drug discovery is getting more and more efficient. And the fundamental bottleneck of bringing new medicines to patients is the cost and time of drug development.

• 1:41:01Speaker 00

  What's your stance on monkeys? What's your stance on monkeys? You got, you got Joe Lonsdale.

• 1:41:05Speaker 06

  There's gonna be a big shortage.

• 1:41:06Speaker 00

  There is.

• 1:41:07Speaker 06

  They're important, you know, national security. We gotta increase the number of monkeys we have. It's, you know, the, the similarity between our disease systems to kind of humans is probably one of the best predictors. Right? So you can imagine like solving like rat Alzheimer's is very different than solving monkey Alzheimer's.

• 1:41:27Speaker 00

  Yeah.

• 1:41:27Speaker 06

  So, but you know, the biggest, the biggest bottleneck is drug development in humans. And so, you know, you have this impetus mismatch where drug discovery is only getting more efficient. There's been a doubling of the number of discovered drug candidates in the past 10 years, even before AI. The number of approved drugs is kept by kind of constant at 50 a year. Markets are rational, so discovered drugs aren't worth much to a post-phase two. So the company, we actually don't do any discovery internally. We look to acquire drugs that are discovered.

• 1:41:52Speaker 00

  Okay.

• 1:41:52Speaker 06

  And then we built all the technology, people, processes, AI systems to run clinical trials and drug development more efficiently. But also the best humans and AI systems that know what drugs to pick and what diseases to match them to.

• 1:42:04Speaker 00

  Yeah.

• 1:42:04Speaker 06

  And then we provide all the capital and capabilities to get the drugs from where they are. So it's usually two to four years away from a clinical trial and we provide all the capital and capabilities to get it post clinical inflection. And that's where the bottleneck is, probably underappreciated.

• 1:42:17Speaker 00

  No, no, totally.

• 1:42:18Speaker 01

  What, what are, what, what were the pre, like, talk about like costs around drug discovery pre-AI? And then where you think it'll go with new advancements and models and, and, you know, continued scaling? Because it feels like a $600 million fund. I imagine you're going to distribute it across like a bunch of entrepreneurs and teams that are focused on that discovery. But then each one of those individual drugs, if they're promising and they start going through trials, could still cost hundreds of millions of dollars to actually take it all the way.

• 1:42:51Speaker 06

  That's actually the fundamental bottleneck, right? So you had this like kind of drug abundance in discovery. But once you hit the clinic, you know, clinical trial can cost a few hundred million dollars. Like a phase two clinical trial can be, you know, 60 million to 80 million, if not more. And so what's going to happen is if we don't develop more capabilities around the drug development period, we're not going to be able to translate all these discoveries into new medicines. And as you alluded to, you're kind of looking at like all these great discovery companies and, and, you know, getting a drug from, you know, a discovered idea to a 90 is now costing just a few million dollars. And that's probably going to go down over time, right?

• 1:43:31Speaker 06

  You know, with all this automation and, you know, the, the great work that, you know, folks like Anthropic and, you know, you got ISO Chi, like design is getting increasingly commoditized. But what is pretty rare is like the ability to prove that it works in the human and that capacity hasn't changed.

• 1:43:49Speaker 00

  So, so historically, what was the origin of the capital for this particular phase? Because obviously a lot of drug discovery candidate selection might happen in the, in academia. And then once you're post-phase two, phase three, you're probably working with a big biotech, maybe got acquired. But is this the sweet spot for companies that might go public, might be traded as sort of like risk on biotech public markets investors historically?

• 1:44:19Speaker 06

  And you're able to do. That's right.

• 1:44:20Speaker 00

  Okay.

• 1:44:21Speaker 06

  That's right. So either you have to go public or you raise from a traditional like trad bio VC, right?

• 1:44:26Speaker 00

  Like a flagship pioneering or something.

• 1:44:28Speaker 06

  Or, you know, an Atlas or, you know, a lot of kind of great funds, kind of finance these drug, you know,

• 1:44:35Speaker 06

  you know, Arch Capital.

• 1:44:36Speaker 00

  Yep.

• 1:44:37Speaker 06

  You know, I think what separates our fund is we're not investing in companies, but individual drug assets.

• 1:44:42Speaker 00

  Okay.

• 1:44:42Speaker 06

  And so these are drugs that we put into a new co.

• 1:44:45Speaker 00

  Yep.

• 1:44:46Speaker 06

  And we not only provide like the 60 or 80 million needed, but all the capital, but also the capabilities. So think about like the ability to run trials cheaper and faster, best in class human drug picker. So our N H A was the R N H A at Biogen. He developed the first, you know, genetic medicine for ALS, the first, or a drug for SMA, the first disease modifying drug for Alzheimer's. And we have a guy who also discovered, had a neuroscience at Roche. So folks really experienced in drug development, but also the AI systems allow you to like look at every single drug in the world, match the right drugs, the right disease, design the right study, and then be more efficient in getting the drugs ultimately to clinical value infection. Maybe kind of share a little bit more about how I got started.

• 1:45:31Speaker 06

  You know, I trained as a computational biologist. So I think like many, really excited about how all the genomic, proteomic, transcriptomic data mixed with machine learning AI can, can transform discovery. When I was a graduate student, we had discovered a few candidate drugs in Alzheimer's and Parkinson's. And just very naively, we went to some pharma execs and said, hey, we discovered these drugs, aren't you guys excited? And they said, you know, start your burst your bubble. And this is like 10 years ago. There's already more good drugs discovered than we could afford to develop. When a single clinical trial in Alzheimer's across all the phases can cost like hundreds of millions of dollars, actually half a billion per, per shot. We just can't afford to prosecute everything we, we discover. And now with AI, that's probably going to tend, if not 100x, right?

• 1:46:12Speaker 06

  And so if we can't like ultimately solve this throughput, we're not going to get more medicines to patients. And so this fund is dedicated for, you know, let's say a biotech can discover five, 10 drugs. As soon as they have a lead asset, traditionally all the capital needs to go to that lead asset. That second, third, fourth might not getting love. And so we're saying, hey, we can provide the capital, but also all the capabilities to enable that company to take more shots. And if it wins, we kind of share upside, it could be upfront, royalties, milestones and even equity. And so it's kind of a win-win for, for the industry. But it's maybe the counterintuitive thing. I don't think people appreciate that, like, you know, drug discovery has not been necessarily the bottleneck for a pretty long time.

• 1:46:55Speaker 06

  And it's only going to get more and more acute as you have drug discovery abundance. And ultimately it's like, how can you get them into clinical proof of concept? And that's what the market values too, right? Because if you have an abundance of discovered drug assets, you really don't see value until you demonstrate human clinical proof of concept. And that's where maybe the heuristic to think about this, like a discovered drug isn't worth much. Once it has clinical proof of concept, so in human trials, in the disease, post-phase two, the heuristic is one X peak sales. So you can sell drug for, you know, if it's a $5 billion a year drug for $5 billion. Once it's over phase three, past phase three, it's two to three X peak sales.

• 1:47:36Speaker 00

  Makes sense.

• 1:47:36Speaker 06

  And so, you know, and this is interesting, like, we kind of ask the question like, why hasn't there been a big pharma company really in our industry created since the 1980s? And one of the big challenges, if you're a venture, and that's kind of the Amgen's vertex of the world. If you're like a venture backed company, it takes you like three to five years to get in the clinic and then two to four years to run a phase one and phase two. Phase two has only worked 30% of the time on average. And let's say you're fortunate enough after you spend 60 million to run a phase two

• 1:48:05Speaker 06

  and it works and a pharma company offers you $5 billion to take the chips off the table. You could, you know, not sell, run a phase three. There's a 40% chance it goes to zero, a 60% chance to phase three works, and then you can sell for two to three X peak sales. Risk adjusted, the rational thing is to sell. So we're structured as a company as a hub and spoke, so can sell in new co without selling the whole company. But if you can only take three shots and you hit a winner, you should probably sell, right? Because if the expected value is 30%, you know, it's more likely your moat and IR is going to be diminished if you kind of run those two shots versus just sell it for a shot. So as a consequence, our industry usually has biotechs get acquired before they can become the next, you know, Roche or, or Novartis.

• 1:48:47Speaker 01

  Same exact thing in consumer packaged goods, honestly. Like there, there has been like a new.

• 1:48:53Speaker 00

  Yeah, yeah, yeah, yeah. There's no new Nestle, you know, right?

• 1:48:56Speaker 06

  They buy, yeah.

• 1:48:57Speaker 00

  Yeah, they buy the protein company, they buy the supplement company, they buy the cereal company, but there's no one that's, and there's been super ambitious founders that have gone after it, but all of them went up with point solutions. Yeah.

• 1:49:08Speaker 06

  And that, and I think to your point, like the second thing that's important is you take enough shots, right? Because if you can only take three shots, you should sell for the reasons we discussed. But if you can begin to take more than 10 shots, your expected value is three out of 10. Now, we have AI systems, humans have done better three out of 10, but it's better to underwrite considerably and, and kind of exceed on the upside. Once you run 10 shots, you, you expect three out of 10. So what the fund allows us to is take more shots, number one. And we can increase the throughput of also the industry. But it also allows us to raise capital and then always have to raise equity. So it lowers our cost of capital over time.

• 1:49:44Speaker 00

  And then ultimately you'll have like diversified cash flows. So you can actually start scaling like the internal mechanism of the business. Like the OpEx at an Amgen is what makes it a biotech company, a big pharma company. It's not just that they have one, yes, they have cash cow drugs, but they have a cash cow drug every 10 years that cycles out so that they can actually sustain the business, get the headquarters, get the teams, get the, you know, the whole flow flowing. And if you're just like a one trick pony, a one hit wonder, you're going to see this, you know, sort of up and down, round tripping of your cash flow, right?

• 1:50:18Speaker 01

  Do you think, public opinion, do you think public opinion around pharma will be better or worse in 10 years?

• 1:50:26Speaker 06

  I'm very optimistic. It's going to be a lot better, specifically because I think we're beginning to pursue drugs that just help you live longer and healthier. And you guys know about the GLP ones, you know, the two top GLP ones did more revenue than OpenAI, Anthropic. And also kind of did it profitably. And we think we're in early innings of things that just help you live longer and healthier. So things that help you walk longer, like muscle, right? Carilage, bone density increases. But I think you bring up a point, like biotech's really an efficient day because, you know, you're, you're really scaling with humans and there's an opportunity to build AI systems that do the work, work of entire teams. And so a pharma company is 100,000 people because in order to prosecute one idea, you need a team of people to go after that.

• 1:51:08Speaker 06

  And, and so we built like AI systems that let you exhaustively look at every single drug and target indication pair and then, you know, design your right trials, run it kind of cheaper and faster. But you also don't have to create a separate team for each asset. And so biotechs today, like you almost have to have like, you know, a CMO, chief development officer, head of quality. And, and they're under leveraged in some ways.

• 1:51:29Speaker 00

  Sure.

• 1:51:29Speaker 06

  So if you can kind of have the best human talent with these AI systems, you can prosecute 10, 15, 20, you know, 50 drugs and not have to be, you know, 100,000 person company, just a few hundred people. And that's what we're really excited about.

• 1:51:42Speaker 00

  600 million dollars.

• 1:51:44Speaker 01

  Gong, gong, gong.

• 1:51:46Speaker 00

  Thank you so much for coming on the show. Congratulations.

• 1:51:49Speaker 01

  Final last, last, last question.

• 1:51:51Speaker 00

  This is fascinating.

• 1:51:53Speaker 01

  Would you rather have 30,000 monkeys or 30,000 Bitcoin?

• 1:52:00Speaker 06

  Um, I think the monkeys will ultimately

• 1:52:04Speaker 01

  I think you buy and hold the monkeys. I think they might. I think they might. I think the monkey market might rip harder than

• 1:52:10Speaker 00

  You think monkey, there's gonna be the flippanting.

• 1:52:12Speaker 01

  Cause you have on with with crypto, you have so much like, just like cyber risk right now, whereas

• 1:52:18Speaker 00

  Oh, so if you're, so if you're, if you're holding on a board ape, you want to get out of board apes and get into real apes.

• 1:52:23Speaker 01

  Yeah.

• 1:52:23Speaker 00

  I think that might be the future. A lot of monkey business going on. Well, thank you so much for coming on the show.

• 1:52:28Speaker 01

  Very cool business, Ben.

• 1:52:29Speaker 00

  Have a great rest of your day and congratulations. Goodbye. Uh, up next, we are running a little late, but we're bringing in the co-founder and CEO of Upwind. We have Amiram. Welcome to the show. The first appearance with a massive

• 1:52:43Speaker 01

  There he is.

• 1:52:44Speaker 00

  From Bressmer. How are you doing? Welcome to the show. Please introduce yourself and the company.

• 1:52:49Speaker 04

  Hey guys, Amiram here. Thank you for having me. Uh, co-founder, CEO of Upwind. Uh, very glad to be here.

• 1:52:57Speaker 00

  And, and introduce the business. What is the elevator pitch, the log line these days?

• 1:53:02Speaker 04

  Upwind protects the cloud and AI for the world.

• 1:53:06Speaker 00

  Sure.

• 1:53:06Speaker 04

  Uh, there's a lot of cloud infrastructure out there, a lot of AI usage.

• 1:53:10Speaker 00

  Yeah.

• 1:53:11Speaker 04

  And Upwind has, uh, built a platform that secures all of it from in the entire lifecycle. Uh, we've been around since, uh, 2023 and, uh, growing fast.

• 1:53:22Speaker 00

  Super important. I can imagine you're growing very, very fast. What is the actual go-to-market motion like today? I mean, cybersecurity is on the top of everyone's mind. Ilya Sutskever is tweeting about how Neo clouds need better cybersecurity. Is, is it just, uh, buy, buy, buy? Everyone just pulling every tool in? Or are you actually going up against other cybersecurity firms for rip and replace? Are you additive? How do you plug in and just wind up achieving the mission of securing infrastructure more broadly?

• 1:53:55Speaker 04

  Yes, I think first is, uh, cybersecurity is having a big moment with AI.

• 1:53:59Speaker 00

  Yeah.

• 1:54:00Speaker 04

  Uh, right. Let's face it. Um, every like person is a hacker.

• 1:54:05Speaker 00

  Yeah.

• 1:54:05Speaker 04

  With like open source models, like you get the damage you can create.

• 1:54:08Speaker 00

  Yeah.

• 1:54:09Speaker 04

  And let alone like the people that really want to create damage. Like, you know, a lot of enterprises are a target for attackers.

• 1:54:14Speaker 00

  Yeah.

• 1:54:15Speaker 04

  And we're like in a very like complex geopolitical situation in the world. So like, there is a lot of things like that are happening on the cybersecurity front and AI just makes it like stimulating it.

• 1:54:27Speaker 00

  Yeah.

• 1:54:28Speaker 04

  So, and the way we're thinking about it is, uh, is basically the, the way to secure it is, uh, is really about the platform approach. Because, uh, if you think about like the average CISO, like when they need to buy a security tool, like how much real estate do you really have to deploy all these like thousands of vendors?

• 1:54:46Speaker 00

  Yeah.

• 1:54:46Speaker 04

  Um, so we've been taking like a platform approach that can actually take you all the way from protecting your secure development, secure deployment in the clouds, secure runtime in the cloud, and now AI lives everywhere. So we are really coming against like to your question, against the platforms, like the big platforms. And we think that's the right place to be because this is what customers really want.

• 1:55:08Speaker 00

  Is your experience at Spot, and maybe you can tell the audience what you were doing there, uh, relevant here because you've seen the nature of uh, enterprise cloud surface areas. And so you know what solutions will, uh, be adopted most aggressively?

• 1:55:24Speaker 04

  You know, when, when I go on a, on a call with customers

• 1:55:26Speaker 00

  Yeah.

• 1:55:27Speaker 04

  And they ask me, what's unique about Upwind? What's, can you tell us what's, what's the differentiation of Upwind? So I tell them the first thing is that we're not coming from cybersecurity background.

• 1:55:35Speaker 00

  Yeah.

• 1:55:36Speaker 04

  We're building a cybersecurity company, not with cybersecurity background. Like our background is from like, you know, DevOps and infrastructure. We had a company before called spot.io. We helped thousands of organizations to scale their compute uh, infrastructure efficiently on AWS, on Azure, on GCP. And that knowledge of understanding how DevOps people think, how infrastructure is wired, how cloud ops is working. It's helping us to like, like help security team unlock context of operations that they never had before. And that's actually the X factor of solving, like taking a problem from like a, like a security finding and alert into like a real, something that I really understand the risk and I can really fix it.

• 1:56:21Speaker 00

  Mm-hmm. How is, uh, how is, like, what is the state of vibe coding right now that we, we all kind of predicted that, uh, you know, there'd be tons of slop and stuff would might start breaking and it's kind of happening, but it's not even fully, uh, I, I, I don't think it's fair to put the blame for some of these cybersecurity attacks on vibe coding. A lot of people have, are breaking into systems that are just like, yeah, actually it's just somebody who was running PHP and the normal like lamp, lamp stack. Uh, they didn't vibe code anything because they built it 10 years ago and now they're vulnerable. And so, uh, where, when you're talking to, uh, CEOs, CIOs, um, what are the threat models that they're most worried about today?

• 1:57:01Speaker 04

  You know, I, I'll just tell you like a funny story. Like we had last week, we had, uh, like a really important feature that I wanted to be deployed as soon as possible.

• 1:57:09Speaker 00

  Sure.

• 1:57:10Speaker 04

  So I went to the developer and I said, I'm gonna go on a, on a Zoom with you right now to make sure like we're deploying it. Because there's like some, you know, final things. And then I go on a call and he's sharing my, the screen and I tell him like, change this, do that.

• 1:57:23Speaker 00

  Mm-hmm.

• 1:57:24Speaker 04

  And then the guy doesn't write code.

• 1:57:27Speaker 00

  Just prompting?

• 1:57:27Speaker 04

  He's like, it's just like prompting.

• 1:57:29Speaker 00

  Yeah.

• 1:57:30Speaker 04

  It's like, change this, do this.

• 1:57:32Speaker 00

  This sounds like something out of Office Space.

• 1:57:34Speaker 01

  Shouldn't have done that.

• 1:57:35Speaker 00

  So your job is to take the prompt from the CEO and give it to KONEX and Cloudco.

• 1:57:39Speaker 01

  And I could just dictate.

• 1:57:40Speaker 00

  To the AI agents.

• 1:57:42Speaker 04

  That's a true story.

• 1:57:44Speaker 00

  Yeah.

• 1:57:44Speaker 04

  And then I asked him, so like, so that's what we're doing right now. So like, we're just prompting.

• 1:57:49Speaker 00

  Yeah.

• 1:57:49Speaker 04

  And then he said like, yeah, like my coding like time has dropped down by 90%,

• 1:57:55Speaker 00

  Whoa.

• 1:57:55Speaker 04

  but the review time is up by 600%.

• 1:57:58Speaker 00

  Sure, sure. Yeah. So

• 1:58:00Speaker 04

  So

• 1:58:00Speaker 00

  Part of the midnight all.

• 1:58:00Speaker 04

  So I think you have to be really, really smart on like how to review that.

• 1:58:03Speaker 00

  Yeah, I think Karpathy had a, a take about this where he said like your

• 1:58:07Speaker 01

  Review this.

• 1:58:07Speaker 00

  Your code writing ability atrophies while your code reading ability increases. And that's probably more important than ever in the modern era. Fascinating. Uh, well, tell us about the round, a Series C. How big was it? How much did you raise?

• 1:58:22Speaker 04

  400 million.

• 1:58:25Speaker 04

  Let's go.

• 1:58:26Speaker 00

  Congrats.

• 1:58:26Speaker 01

  He's not playing. He's not playing.

• 1:58:29Speaker 04

  That's, you know, that, that's a lot of money. That's a lot of money. And I, I can tell you that this is what's needed.

• 1:58:34Speaker 00

  Yeah.

• 1:58:35Speaker 04

  To operate in that very important space.

• 1:58:37Speaker 00

  Yeah.

• 1:58:38Speaker 04

  Like it's a super important space. You have to build, you have to innovate. We have big dreams now. Like we can translate those dreams into execution. We have the right amount of cash. Um, and lots of trust from investors, a lot of trust from customers. Um, and yeah, it puts us as like one of the most capitalized companies on the planet to secure that cloud and AI.

• 1:59:01Speaker 00

  Yeah, we need it more than ever now. We saw it with the PHP hack. So thank you for everything you're doing.

• 1:59:05Speaker 01

  Super important.

• 1:59:06Speaker 00

  Good luck and thank you so much for taking the time to come on the show.

• 1:59:09Speaker 01

  I'm sure you're back soon.

• 1:59:10Speaker 04

  You guys. Thank you.

• 1:59:11Speaker 00

  Goodbye. Let me tell you about the New York Stock Exchange. Want to change the world? Raise capital at the New York Stock Exchange. If you're trying to get more than 400 million, you gotta head to the New York Stock Exchange. Our next guest is Jasper, Jasper Jong from Hyperbolic, co-founder and CEO, his first appearance on here. Jasper, how are you doing? Welcome to the show.

• 1:59:33Speaker 04

  Hey, how's it going?

• 1:59:34Speaker 00

  Sorry for keeping you waiting. Thank you so much for hopping on. Uh, please introduce yourself and the company.

• 1:59:41Speaker 04

  Hey everyone, I'm Jasper. I'm the CEO of Hyperbolic. I go out doing math competitions and won a Fugue Meadows, a Chinese Math Olympia, Alibaba Global Math Competitions. I finished my math PhD at UC Berkeley in two years, which made me the fastest person to finish a PhD in a few show Berkeley.

• 1:59:56Speaker 01

  What stopped you from doing it? And what stopped you, just out of curiosity, from doing it in one year? Like

• 2:00:02Speaker 04

  Uh, Berkeley has a residency requirement. So technically you need two years.

• 2:00:06Speaker 00

  Hands retired.

• 2:00:07Speaker 04

  That's

• 2:00:08Speaker 01

  So without that requirement, you think you could have done it in a couple of months? Maybe a month?

• 2:00:12Speaker 04

  Um, with AI now, I think so, yeah.

• 2:00:14Speaker 00

  Oh, really? Okay, okay.

• 2:00:16Speaker 01

  Did you, have you ever faced, uh, Scott Wu?

• 2:00:19Speaker 00

  Ooh, yeah.

• 2:00:21Speaker 04

  Oh, yeah. Uh, I mean, I heard about him. Yeah, haven't, haven't met him in person.

• 2:00:26Speaker 01

  No, no, I meant like, but you guys were kind of competing in different

• 2:00:29Speaker 00

  Yeah.

• 2:00:29Speaker 01

  regions, it sounds like. I was hoping that we could find a video of you guys like

• 2:00:33Speaker 00

  Going head to head. Yeah, like, like old LeBron videos. Yeah.

• 2:00:36Speaker 01

  Like sparring.

• 2:00:38Speaker 00

  Yeah. Uh, that'd be great. But tell us about the company. Uh, what do you, what are you building? How, how, how has it been going?

• 2:00:45Speaker 04

  Yeah, it's been pretty great. I mean, uh, I have a bullet where aggregate and turn heterogeneous compute into one unified platform.

• 2:00:52Speaker 00

  Okay.

• 2:00:53Speaker 04

  So it's a platform where people can just come access compute in a very flexible way.

• 2:00:57Speaker 00

  Yeah.

• 2:00:57Speaker 04

  They can manage and orchestrate a computer globally and then sell the computer back.

• 2:01:00Speaker 00

  Yeah.

• 2:01:01Speaker 04

  When they don't need it.

• 2:01:01Speaker 00

  How heterogeneous are we talking about? Because we saw a, uh, a cluster of 9,000 A100s on eBay earlier today for something like $60 million. Then, you know, you got people with a gaming PC that might have a, you know, 5090 Nvidia card in it. Uh, how low can you go? What is the practical scale to that heterogeneity?

• 2:01:23Speaker 04

  Yeah, totally. So our technology supports any kinds of, uh, GPU chips. But then, uh, technically I think most customers want at least like the A100 plus

• 2:01:33Speaker 00

  Okay.

• 2:01:33Speaker 04

  Uh, chips. We're also starting seeing people using Pro 6000 for inference as well. But we can go as high as to Blackwell and there will be.

• 2:01:41Speaker 00

  Yeah. Uh, how much, uh, spare compute is out there? Because you, you, it feels like every day there's a new story about like, oh, there's such a compute shortage, like tokens are selling, like, you know, the, the, the, the spot market is in contango and like the, there's so much AI demand and yet you're still finding folks out there that have spare capacity to bring on the platform. Is that correct?

• 2:02:06Speaker 04

  Yeah. Say we're not just tackling spare capacity, even for like a big cluster, our goal is to maximize the revenue per G P hour.

• 2:02:13Speaker 00

  Okay.

• 2:02:14Speaker 04

  Technically, you can sell off take for B three hundred, four, four, five dollars, five dollars. But it's like very low margin business. If everyone, even if you're a Neo cloud, you want like to maximize your revenue to ten dollars, twenty dollars per G P hour.

• 2:02:26Speaker 00

  Sure.

• 2:02:27Speaker 04

  You can work with us.

• 2:02:28Speaker 00

  Yeah. And so are you seeing like, what is driving that incremental off take? Is it that there's a company that is renting a cluster and they're doing a training run, then they pause for a little bit to review the results? And so you can go do some inference right there, or is it you're doing stuff on the other side of the world? When everyone's working with the big models during their workday, but then there may be the G P U's aren't on fire in the middle of the night. So you can sell inference somewhere else. Like, what are the use cases that are, that your customers are coming to you for specifically?

• 2:03:01Speaker 04

  Yeah. So I'll say what all the cases they mentioned are valid for us. And I'll say we start with first helping customers to get access to compute first. So you know, like there are more than three hundred plus Neo clouds. The market is super fat maintained.

• 2:03:14Speaker 00

  Yeah.

• 2:03:15Speaker 04

  And it's the first time that technology company founder need to start worrying about sourcing procurement, but also the cash flow, the capital, the hedge, hedging and risk management.

• 2:03:25Speaker 00

  Yeah.

• 2:03:26Speaker 04

  And so what we do is that we kind of have like the system, which is called Forge, which is a unified control plan. That allow us to take any bad metal cluster, attach it to our platform. And now it can be shared globally with different customers in a very flexible way. So they will start using the compute from our platform. And then let's say if they lock in for three year contract or like a few, a few year contract. Then like what you said, if they are doing training run after the training run, they don't need that. They can also start back to a market and the same for this inference too.

• 2:03:59Speaker 00

  So Intel's CEO, Lip Bhutan's backing up the truck for this company. What is your interaction with him and Intel? He's backing up the truck investing. But you, you know, obviously he's, you know, an incredibly important CEO in the A I world. But it's not like he's running a Neo cloud that needs off take capacity necessarily. Like there's not a super logical business relationship that I can think of. But do you see this growing into a strategic relationship? Or is it just you're, you got along great and he had the capital to invest?

• 2:04:33Speaker 04

  Yeah, I would say fundamentally like he's outside of Intel, he's the V C, right?

• 2:04:39Speaker 00

  Yeah.

• 2:04:39Speaker 04

  He founded a world and catalyst, which is the best semiconductor.

• 2:04:43Speaker 00

  Yeah.

• 2:04:43Speaker 04

  V C in the Silicon Valley.

• 2:04:45Speaker 00

  Yeah.

• 2:04:45Speaker 04

  And so, yeah, like the business running very well and I think he see the upside of building this aggregation layer, orchestration layer.

• 2:04:53Speaker 00

  Yeah.

• 2:04:53Speaker 04

  But then our goal is like, we're trying to aggregate the world's compute. We want to put the world's compute behind the people who are bold enough to change the world.

• 2:05:02Speaker 00

  Yep.

• 2:05:02Speaker 04

  So naturally, there is a potentially strategic angle here to leverage other kind of chips.

• 2:05:09Speaker 00

  Yeah. Well, you raised thirty million dollars. I'm gonna hit the gong for you. Congratulations and thank you.

• 2:05:17Speaker 01

  Because you guys called this the A one, I feel like you might be back on in like a week or so for the A two. So.

• 2:05:26Speaker 00

  It's an exciting time.

• 2:05:27Speaker 04

  Have investors putting money for the next round. So.

• 2:05:29Speaker 00

  There we go.

• 2:05:30Speaker 01

  There you go.

• 2:05:31Speaker 00

  There we go. Breaking news.

• 2:05:32Speaker 01

  Call when you're, when you're ready to come back on to announce the, the second tranche.

• 2:05:36Speaker 00

  Have a great rest of your day.

• 2:05:38Speaker 01

  The second tranche, but.

• 2:05:39Speaker 00

  And thank you so much for coming on T B P N.

• 2:05:42Speaker 04

  Thanks.

• 2:05:42Speaker 00

  Have a great rest of your day.

• 2:05:43Speaker 01

  Talk soon.

• 2:05:43Speaker 00

  Goodbye. And that is our show, folks. We will be back tomorrow at eleven A M Pacific. What did that have to do with what I said?

• 2:05:53Speaker 01

  Show's over.

• 2:05:54Speaker 00

  Oh, show's over. You can take off your headset. Because normally I think about it as like, I heard something I didn't like, but I understand.

• 2:06:01Speaker 01

  Well, some people might be nuanced.

• 2:06:02Speaker 00

  Yeah.

• 2:06:03Speaker 01

  Just off that the show.

• 2:06:04Speaker 00

  Show's over. I can't deal with no show. Yeah, that makes sense. Okay, got it. Well, uh, thank you so much for tuning into T B P N. And we will see you tomorrow at eleven A M.

• 2:06:14Speaker 01

  It has been an honor, an honor, a privilege to be here with you.

• 2:06:16Speaker 00

  Wednesday of your life.

• 2:06:19Speaker 01

  Of your entire life.

• 2:06:20Speaker 00

  Leave us five stars on Apple Podcasts and Spotify. Sign up for our newsletter at t b p n dot com.

• 2:06:24Speaker 01

  We haven't been doing a road to Halloween.

• 2:06:26Speaker 00

  No.

• 2:06:27Speaker 01

  We're not big Halloween guys, but we might, but I am, I'm, I'm kind of excited.

• 2:06:32Speaker 00

  Taking your eye off the ball. Okay. Yeah.

• 2:06:34Speaker 01

  I'm, I am, I think Friday's show might be fun.

• 2:06:38Speaker 00

  Okay.

• 2:06:38Speaker 01

  I think we might have a good time.

• 2:06:39Speaker 00

  You gotta tune in. Hit the subscribe bell, set a calendar notification. Eleven A M Pacific on Friday is gonna be one for the ages.

• 2:06:48Speaker 01

  That's right. That's right. It's gonna be a good time.

• 2:06:50Speaker 00

  It's gonna be great.

• 2:06:52Speaker 00

  Goodbye.

• 2:06:53Speaker 01

  Cheers.

• 2:06:54Speaker 07

  Journalists on the horizon, stand by.