Microsoft’s Energy Tab, OpenAI Goes Super Bowl, Ellison Makes His Move | Marc Benioff, Brian Chesky, Baiju Bhatt, Gabriel Carafa, Alfred Wahlforss

27:32But the talks had stalled. privately bankers on each side were blaming the other side for needlessly slowing down the negotiations. You're moving too slow. JPMorgan had secured a discount on the balances, but to help cover potential losses, they'd secured a discount on the balances. We talked about this where normally these balances, they traded a premium because people expect them to be paid back with interest. But in this case, JPMorgan was going to get a discount, but they wanted more protection in in case the loans grew worse, in case they were higher than expected defaults. Goldman executives, meanwhile, didn't feel the need to bend much now that the Apple program was finally looking profitable.

28:16Some executives on both sides had started questioning whether to walk away. Diamond and Solomon got on a call on December 8th, according to people familiar with the matter. They discussed why the Apple deal was taking so long to close and agreed to break the logjam to see the deal through soon, people said. Just before the new year, the banks finalized a deal that was announced last week, confirming the Wall Street Journal's earlier report. A little patting on the back for the Wall Street Journal. We're getting the scoop. The deal brings two of the country's most influential companies together.

28:50J.P. Morgan is adding the flashy program to its credit card lending operation and strengthening its connections to the trillion-dollar tech giant at a time consumers are increasingly using phones and watches for payments and managing their finances. Apple gets a new partner with a sprawling consumer base that is eager to build the card. Goldman gets closure on its failed venture into consumer lending that has brought the firm billions of dollars in losses, a chapter it is hoping to forget. Moving the Apple credit card was never going to be simple. Card programs this big aren't put up for sale often and few potential buyers exist.

29:28Apple is famously finicky about details and control, including with its credit card. Before launching the program in 2019, Goldman and Apple agreed to unusual terms that other banks balked at taking over. Interested executives were especially worried about higher than normal delinquencies and subprime exposure and wanted huge discounts to consider a deal. Even still, almost nothing about the process of finding a bank to replace Goldman has been normal. To begin with, Goldman began looking to exit from the contract with Apple only a few months after it had extended its partnership with the tech giant to the end of the decade.

30:05Apple and Goldman had discussions with larger and smaller credit card issuers, American Express, Capital One, Synchrony, Barclays and Satander, as well as tiny fintechs. Apple even debated bringing in private credit firms to help a smaller card issuing entity take over. But it didn't go that way and they landed with JP Morgan. So very interesting. Anyway, let me tell you about LabelBox delivering you the highest quality data for Frontier AI. Get in the box. Get in the box. The LabelBox. Moving on. OpenAI is going to be running another Super Bowl ad. It's in the Wall Street Journal. OpenAI to run ad during Super Bowl.

30:48Why don't we play their last one? We'll see the Twitter's ad, react to that, and see what we think will change, what will stay the same, what will they do. The artificial intelligence company behind the popular ChatGPT tool is set to broadcast a 60-second commercial during NBC's broadcast. All right, with sound. Let's watch it with sound if we can. The ad is the latest component of OpenAI's expansive marketing offensive, which began last year with its first Super Bowl commercial. We don't have sound. Okay. Okay. We, the production team is telling us, unfortunately, we do not have the technology to play sound on a video.

31:28Let's work on that. But it's, if I remember the song correctly, it's like, doon, doon, doon, doon, like something like that. Is it a licensed song? Because you were thinking that they should license Creed's Can You Take Me Higher for this next one, right? Yes. That was what you thought? Yes. Can you take me higher? And it's like, to a higher plan. You're on the free plan. Please upgrade to the$200 a month plan. Oh, we got it. Let's go. We do have the technology. AGI is here. Okay, so there's someone throwing a spear. We create fire. Then the wheel. Then the wheel with spokes. Then the horse. The horse came out of the wheel.

32:11Man created the horse. We created the horse and corn. We created corn. Maybe they should use the corn song. Freak on a Leash. That's the song they should license. and a nod to Korn. I mean, the motion design is incredible. Like, it is a very well-done piece. So they go inside the skeleton, build the 747, take a look at DNA, invent movies, TV news, that's us. They go to the moon. Somewhat equally important, create the internet, and then you start talking to the computer. And now you got AI. next. And then you get the blue chat GPT dot. Yeah. So, so I, I love, I love this. Yeah. It was just not necessarily.

33:00I didn't need a Superbowl ad for me. Yeah. I need a Superbowl ad for the Clydesdale crowd, right? The people who are maybe on the fence about AI. Um, but yeah, as a technologist, it has the aesthetics of like, wow, this is the coolest tech company ever. And And this has the vibe of Apple. It has a lot of throwback stuff. It feels like high technology. It feels techie, elite. Very, very cool. It feels designed. It feels refined. And it's possible at the time they were like, look, we're growing so quickly. We don't really need to run a Super Bowl ad. Yeah. Let's just flex on everybody a little bit by running one.

33:36I mean, there's a certain... But this year I expect it to be much more pragmatic. About what you can use to do something. Exactly how they want. Yeah. the average American to think about ChatGPT. Yes, yes. Where it integrates into their life. Yeah. And so do you think it'll be one specific use case? Like there's been OpenAI videos about ChatGPT helping you cook dinner. And I actually got an odd email from a friend all about how he loves cooking with friends and how food is this important thing. And then he was calling out the ChatGPT video on food being like, I don't like this because it's taking away the one thing I like as a human.

34:17I actually had a funny thing. Practically, I think it's fine. Using an LLM for a pancake recipe last week. And it completely, the pancakes really did not work. Because it's basically taking the average of all pancake recipes. You made a mid-pancake. And it just made the most runny mid-pancakes really bad. Interesting. I don't know. It would be interesting. I wonder if they try to do anything on the, like anything on the, it's probably too early for the commerce side. You could see some of that, but maybe they want people to know commerce is coming. I could also see them trying to make it interactive somehow.

34:54You could imagine like when, you know, maybe this. No, no, no, no. But I was going to be like, ask, like talk, like, like chat with chat, chat, like tell chat GPT you want it. I don't know. Well, it could, like they could sort of launch like chat GPT, sports or something where you know you're like the demo of what it's showing you is that if you have chat gpt as your second screen during the super bowl you're going to pull up more interesting stories about the players you're going to pull up statistics you're going to be able to understand the the all like add more context to your experience of super bowl watching even like Oftentimes, if I'm watching the Super Bowl, I'll see an ad, and I'll go to Google and say, well, what are the greatest Super Bowl ads of all time?

35:46And I'll wind up on some listicle. And ChatGPT, of course, can generate that type of stuff all the time. You could filter and say, show me the best Super Bowl ads from tech companies back in the dot-com boom. I want to know about those. And you can go anywhere. And that is the magic of ChatGPT. There's a whole bunch of other ways that they could try and contextualize it. They could try and be emotional, go with ChatGPT Health. I wonder if they'll do anything with the new Disney integration. Like if they wanted to spark another Studio Ghibli moment, that IP is coming online. Originally they said something in January, so maybe the timing lines up.

36:21That would be really wild. Yeah, you get Darth Vader in there or something. I mean, there's a lot of characters to choose. I was saying try to use the Super Bowl ad to prompt people to go to ChatGPT and turn your Super Bowl party into a Disney theme. That would be very good. Yeah, surprise celebrity appearance or celebrity endorsement would be good. They've sort of shied away for a minute. Scarlett Johansson. Maybe they can bring her back into the fold. Reversal and getting her back on the team would be fantastic. But I expect that this will be the AI Super Bowl. You think there'll be even more ads?

37:03I'm sure there will. So for context, tech companies across Anthropic, Google, Microsoft, OpenAI, and Perplexity collectively spent$333 million on linear TV ads promoting their AI offerings in the United States just last year. And that was up 43 % from the prior year, according to estimates from an ad tracker, iSpot. They also shelled out$426 million on digital ads in 2025, more than triple their 2024 outlays. So interesting. digital is growing faster than linear. I would not expect that. I would expect the digital to have been really, really popular and then them to have just jumped into linear.

37:45But I guess they've been... Wait, are they talking about streaming ads? Maybe. It's from Censor Tower. It just says digital ads, which I assume is Facebook, Google ads, all that sorts of stuff. Anthropic kicked off its first major push into advertising in September and has been blanketing NFL, NBA, and college sports games with ads for its clawed chatbot. It shelled out an estimated$16.5 million on linear TV ads in 2025. Despite boasting a base of over 800 million weekly users, OpenAI is, and that is such a stale number. They cannot get people to download this app. What do you mean? Like they really can.

38:25I mean, they're not in the top 25 apps. Wait, which one? Clawed. Oh, you're talking Clawed. So, sorry, sorry. All right, that 800 million weekly active users is for OpenAI, for checking. No, I know, but I know. Yes. I know. So they've been spending money all over NFL, NBA. Well, you know why the cloud is not at the top of the app store charts? It's because if you generate a deep research report on the iOS app, it will not read it to you out loud. It doesn't have that functionality. And that's clearly a gap that everyone has experienced and recognizes. And as soon as they run into that, they uninstall it, and they don't recommend it to their friends.

39:02So if you're listening, Anthropic, please add that functionality. I want to be able to listen to deep research reports. Yeah, I would just think, how do they figure out how to get it? How do they figure out how to become a real player and a consumer? Because clearly they want to, otherwise they wouldn't be spending that money. I mean, they need hooks. You need some viral flywheel. So the chat GPT number, 800 million weekly active users, that feels woefully out of date. It feels like they were well over a billion and they're just waiting to actually drop that press release at a key moment. And now is not the right time to wow everyone with the real usage numbers.

39:40But I definitely think that they're beyond that. This number is very stale. Like we got this 800 million weekly number months and months ago, and it must have grown. Google obviously has a whole bunch of hooks into the Gemini ecosystem through Gmail and Google. There's so many different entry points. Claude as an independent app doesn't really have that. Also, I think just on a product sense, ChatGPT, I do think that the holidays were successful. And it was not discussed on X, the tech elite and the tech, anyone who's in tech was focused on Cloud Code. They were doing vibe coding projects. They had time off from work and time to lock in and play around with the hottest new coding agent.

40:29So people knew that ChatGPT could do images, but right before the holiday started, ChatGPT launched the image tab. And if you go to the image tab in the app, the most interesting thing there is that it's not just a prompt box. They have preloaded, have you seen these? It will show you retro anime. It will show you ideas. Like turn yourself into a bobblehead, turn yourself into a sugar cookie or a fisheye lens or ink work or pop art. And so it's taking the work out of someone. Like most people in tech, an image model comes out, Studio Ghibli thing happens. Everyone's like, what's the prompt? They were like, it's too much to ask people to have an idea.

41:15It is. We need to give them the idea and we need to make the image for them. It actually is too much. And then eventually when you have your little sweet pea, whatever they're calling their new device, it will just be generating images on the fly based on your day. Yes. And then eventually you'll just be able to stay laying in bed all day long. Yeah, but I mean, just from a product perspective, the ChatGPT images tab is, I think, an important just growth hack, which is something that they just have to do now because they're at that scale. And I saw that people were generating images and then sharing them, and they weren't screenshotting them or downloading them with some watermark.

42:01They were clicking the share button in the ChatGPT app and sharing a link to the ChatGPT app that would show you the image in a preview if you shared it in iMessage or in a group chat. And that would onboard more people. And those sorts of viral loops, they feel like, oh, the tech is so magical, it should just grow organically. But it already grew organically. Like the organic growth was a billion people. Now, how do you get the next billion? Well, you probably have to do some growth hacking and a product like images in ChatGPT with Disney characters and links that beg you to install the app for yourself.

42:39and Claude isn't quite there yet where there's an obvious thing that a huge swath of people can go to and then hook in and we can collaborate and I send you a link and then you install it. Yeah, the tough thing is like if you're spending all this money trying to drive consumer downloads, but then you don't have image generation at all, at what point, why don't they just, even if that's not a priority for them, it still feels like a product that consumers really want in their daily driver. Yeah, yeah. And so at some point, they may have to capitulate there. I mean, I don't know if they care about the app race at all, the consumer app race.

43:16It feels like Anthropik's been super focused on the enterprise, super focused on coding, and that's been - I know, I know, but they say, like, all I'm saying is you can say that you're not focused on consumer, but they run billboard ads all over LA. They're spending a lot of money on television. and so you kind of need to like make up your mind. Yeah, you don't really advertise in NFL and NBA. I mean, I guess enterprise buyers watch those shows. So 16 million, maybe it's worth it. But I agree with you generally. Their ad campaigns have been great. Yeah. That's it. And maybe they just want to rub it in the face of the other labs.

43:54They're also going public, so they want, developing like real name recognition broadly, I think is important. I mean, in theory, if they host CloudCode in a virtual machine that you can access from a phone and have it write software for you, it should be able to go and set up an API with an image generator and generate you images. and so you should be able to do that from the app. Like Cloud Code, if I ask Cloud Code to generate an image for me, it will figure out how to find an image generator, sign up for the API almost, or at least I will give it the API key and then it will be able to generate images for me, correct?

44:38Yeah, it would ask you to, like, you would have to go generate the API key and stuff. Yeah. But that's actually, I'll try that right now. Yeah, what happens if you just ask Cloud Code to generate you a picture of four cats on a boat or something? How does it actually work through solving that? At some point, it should prompt you to at least maybe make a decision about which image generator you want to use and at least ask for credit card information to pay for it. Maybe it goes to a free image generator. I don't know. But the Wall Street Journal continues, and we will continue after telling you about the New York Stock Exchange.

45:16Want to change the world? Raise money, raise capital at the New York Stock Exchange. Selling AI to the masses isn't easy, says the Wall Street Journal. Due to sustained public fear about the technology's potential downsides, businesses have already cut jobs using AI for tasks previously performed by humans, and executives have articulated plans. Why would the masses be scared? Why would they be scared of this technology? Half of U.S. adults are more concerned than excited about artificial intelligence, according to a spring survey by Pew Research. That's a while ago, but... Only 10 % say excitement outweighs their worry.

45:57Skill issue. Just be in the 10%, right, Tyler? Yeah. Yeah. Well, OpenAI's first big game ad position, ChatGPT is the next significant advance in human innovation, drawing parallels between AI. Wait, this would actually be a great bit. We should make an AI Tyler that we can go to and say, and just we'll compare the two of you guys. And this can just be a good barometer for AI progress. How do you feel about that? It could probably do a pretty good job. Because it can already do like a talking head if you just put an image of me. And then the voice you can clone. I know. But I don't think it would have your takes.

46:38I don't think it would have your takes. Just say like, or your aura. Or a cash. It would say, I'm sorry, that's too controversial. I can't hold that opinion. While OpenAI's first big game ad positioned ChatGPT as the next significant advance in human innovation, drawing parallels between AI and transformative inventions, such as the light bulb, more recent ads frame the technology as a relatable tool. They feature people using the chatbot for everyday problems, such as finding a recipe or searching for exercise tips. Which way will OpenAI go in the new Super Bowl ad? I imagine that they will go more proactive, more productive, more tangible.

47:25There will be a human shot with a camera using the app probably on the phone. But we'll see. in its ads, Anthropic has been positioning Claude as a partner for problem solvers rather than a threat to human intelligence. Yeah. Why would you want to run an ad saying that? This strategy anchored by the tech line. This is a Super Bowl ad concept. We are not a threat to human intelligence.

47:52We are not a threat. Don't worry about us. Yeah. How about a Super Bowl ad? This is - We hear you. We hear that you're concerned. Yeah. Just a massive Super Bowl ad. I'm warranted. Our PDOOM is actually really low relative to the true risk. We think that the PDOOMs should be low. Everyone's like, what are they talking about? 5 % chance of annihilation by end of year. Well, if you're running the Super Bowl, you got to get on Restream. One live stream, 30-plus destinations. I want to be able to watch the Super Bowl on LinkedIn. If you want to multi-stream, go to Restream.com. And I know the folks who have the rights to the Super Bowl.

48:33Sasha really should push to get LinkedIn the streaming rights. For sure. Be a game changer. For the Super Bowl. Okay. So Gemini introduced a more personal Gemini today designed for you. Personal intelligence, they're calling it. Personal intelligence draws insights from across your Google apps to provide truly customized responses for Gemini. Learn all about it below. They say Gemini already remembers your past chats to provide relevant responses, but today we're taking the next step forward with the introduction of personal intelligence. You can choose to let Gemini connect information from your Gmail, Google Photos, Google Search, and YouTube history to receive more personalized responses.

49:16Here are some ways you can start using it. Planning. Gemini will be able to suggest hidden gems that feel right up your alley for upcoming trips or work travel because it'll know where you've stayed before, where you've been. It'll know that I went to Austria and recommend a different city maybe. Shopping. Gemini will get to know your taste and preferences on a deeper level, help you find items you love faster. Motivation. Gemini will have a deeper understanding of the goals you're working towards, look at your to-do list. Privacy is central to personal intelligence, how you connect other apps to Gemini.

49:46The new beta feature is off by default. You can choose to turn it on, decide exactly which apps to connect, and can turn it off at any time. Personal intelligence begins rolling out today as a beta to Google AI Pro and Ultra subscribers. That's me. So I got to turn this on. I'm very interested in seeing, I imagine that I will not be able to link multiple Google accounts because I have a personal Gmail and then I have a work Gmail and then I'd sort of want to integrate my old work Gmails that I think I have access to still because that says a lot about me even though I'm not really using that stuff anymore and I'm not actioning those emails.

50:28I could see that being important in terms of like building out a. Yeah, if you were sick of switching between constantly switching between calendars and Chrome and having your personal calendar, calendar, get ready to be sick of switching between LLMs. Yeah. Some of them have been surprisingly good at bootstrapping knowledge. I feel like a few of the LLMs have surprised me by I'll ask a question and it'll just be like, and here's the context for TVPN. Here's the TVPN angle. And I'm like, oh, I forgot that I mentioned that to you. Or like you figured that out somehow. Or like, yeah, you did something.

51:04I don't know. The personalization is getting better. Maybe this is the year when it really hits. I know ChatGPT introduced it as a concept a while ago, maybe almost a full year ago. And it hasn't felt magical yet. Pulse was cool, but I never became a daily active user of Pulse. The interesting question is, is it a data source problem? Is it an architectural problem? Or is it an actual model level problem? Are the models just not smart enough? Do they not have enough data? Or are they not designed in the right way where they can actually go over all of your information? I was using the new Salesforce Slack bot that we're going to talk to Mark Benioff about at 2 p.m.

51:51And it was remarkably useful in Slack because I'm not on Slack a lot. And so I had a ton of notifications that had built up over months. and I was able to go in and write a prompt or just talk to Slackbot and say, hey, like, catch me up. Like, what's been happening in Slack, in this Slack workspace? And it did a good job summarizing all the different key things, where I was mentioned, what was relevant to me. I did prod it on what Anthropic model it's using exactly, and it's not using the latest and greatest, 4.5. It was using, I think, 3.5. It didn't seem aware of the latest models. And also, it didn't have the ability to search the web.

52:32It didn't have the ability to act as like a full-fledged LLM. You couldn't just ask things. It was very confined to Slack. But within that domain, it was pretty useful. And it did help me like sort of get to like inbox zero in just like a couple minutes. Because I was able to just scan a digest of everything that's happening across all the channels. Instead of clicking through them, scanning, reading the threads, going down, it was just able to be like, here's the deep research report on it. Yeah, I am waiting. I'm waiting for the moment where people are like, I'm not really interested in trying a new LLM that's hot because having memory in the LLM and personalization is so important.

53:12Like waiting for that lock-in moment. Don't think it's really – you have some forced lock-in where a company is saying like, we use this LLM. Or I mean vice versa where it's just like, yeah, in the course of my day I use five different LLMs. Like on my phone, I use Apple intelligence and that's routing to Gemini. And then I log into Slack and that's using the Anthropic APIs. But it doesn't really matter because I don't care that they don't talk to each other. Like it's all fine and they work. They're good enough for the particular thing that I'm trying to do at a particular time. That's another possible outcome here.

53:46Andrew Curran is summarizing some of the news here, saying that it includes Google Workspace. Google Photos is sort of interesting. I wonder if that plays into, you know, nano banana. You can have, you know, more suggested images. Hey, do you want to touch this photo up? Do you want to create some sort of holiday card? Like what Chachi Petit ultimately wound up doing with images? Go through Google Photos and turn me into a dinosaur in every photo I've ever taken. Set the GPUs on fire for sure. Before we move on, phantom cash. Fund your wallet without exchanges or middlemen and spend with the phantom card.

54:24New York Times has some coverage today talking about IPOs of this year, 2026, maybe the year of the mega IPO. Not really new news for those of you listening. SpaceX, OpenAI, Anthropic are going public. They talk about Eddie Malloy from Morgan Stanley says we're going to get into a period of potentially unprecedented IPO deal sizes. We are confident they're executable given the scale of these companies and the investor interest. In two decades, I haven't seen private companies that are this meaningful and are this impactful, said Jeremy Abelson, an investor at Irving. Investors, not only are they bigger and more relevant, but they're incredible companies with numbers that we've never seen before, ever.

55:17I think everybody's excited to just get into these S1s. I expect them to be extremely impressive for a number of reasons. There's been a number of reports about how bad the gross margins are, how much money they're losing. We're about to find out the real details. It's going to be exciting times. Exciting times. So yeah, we'll see. All the geopolitical chaos, even this ruling around tariffs, which we didn't get today, all this stuff is going to play into whether or not the window stays open. Yeah. Console. Console builds AI agents that automate 70 % of IT, HR, and finance support, giving employees instant resolution for access requests and password resets.

56:10Saks Fifth Ave. Neiman Marcus and Bergdorf Goodman filed for bankruptcy. Are you a Saks guy? Do you shop at Saks Fifth Avenue? Have you shop at Saks Fifth Avenue? I've never been a department store guy ever. You go straight to the Chrome Hearts store. You're the guy in line outside of Supreme. Go to Rack, yeah. And they don't sell Supreme at Saks, right? Maybe that's why they're going bankrupt. The hype is moving on. Zero Chrome. Zero Chrome. Anyways, they filed for bankruptcy protection late Tuesday, crumbling under billions of dollars in debt, a fraying relationship with vendors and lagging sales.

56:47The filing is the latest sign that America's luxury department stores, once landmarks that served as immersive fantasy worlds for the wealthy and aspirational, are becoming an endangered species. Jeffrey Van Raymond Donkt, the former Neiman Marcus chief executive, will return as Saks' global CEO. He succeeds Richard Baker, the chief executive and executive chairman, who in 2024 orchestrated a$2.7 billion acquisition by Saks of the Neiman Marcus Group, which owns Bergdorf Goodman. Saks said it had secured roughly$1.75 billion to help finance the company through bankruptcy, much of it coming from its bondholders.

57:27Saks said it intended to emerge from bankruptcy later this year, and it expected to honor all customer programs, make payments to vendors, and continue to pay employees. Saks has brought in additional executives to help sewer the company, including Darcy Pennick, former president of Bergdorf Goodman, as president and chief commercial officer. This is a defining moment for Saks Global, and the path ahead presents a meaningful opportunity to strengthen the foundation of our business and position it for the future. This is good messaging. Trump was messaging around the Supreme Court ruling. He said, if they rule against us, we're screwed.

58:03I was like, not that. He just was like, if they rule against the tariffs, we're screwed. We're cooked. We're chopped and we're cooked. Ridiculous things. Yeah, I mean, I feel like these stores are, if anything, just nostalgic, right? It really feels like an end of an era. essence obviously went bankrupt earlier this year. in this case, in that case, too, founders are going to actually maintain control. So we'll see how they navigate through. But again, I just wonder, you know, as these stores shut down, like who can even fill these spaces? Like there's just not, there's not companies that want to come in and occupy a bunch of this, a bunch of this real estate.

58:59You got to get Rick Caruso in there. He's like the master of this. Like there is something that's a little bit more durable about the mall that you go to for, there's restaurants, there's a movie theater, there's a coffee shop, book store, escape room, probably laser tag VR, maybe, maybe a sphere. Um, but there's a few key stores that are anchor tenants, the Apple store, and there's electronics. And then there are a few, there's a Louis Vuitton store, a coach store, a Bottega store. And those brands want to own the entire real estate experience, I think, more now. I don't, I'm not sure, but it seems like, you know, as much as we joke about like the Supreme store having lines out front, like there isn't, there is a deliberate strategy to that.

59:44Like you control the full experience, not just a corner or a display table in a larger store. So these like aggregator retail stores seem to be having a lot of trouble. Yeah. It's not good. Well, MongoDB, choose a database built for flexibility and scale with best-in-class embedding models and re-rankers. MongoDB has what you need to build what's next. Larry Ellison is being profiled in Vulture for New York Magazine. It's a very long profile that we should read through. First, we should debate this story that we almost got to from Reddit yesterday. Are there any real candy innovations right now?

1:00:30R slash candy, which I didn't know existed. Hey, everyone. I'm curious to hear your perspective as candy lovers. Do you feel like there's any real innovation happening in candy right now? Are there any sweets, brands, or flavors that are new? Someone's been reading zero to one. The stagnation theory has come to the candy industry. We're seeing progress in the world of bits, less so in the world of sugars. Of candy. Some people did make the good point that, yes, there has been a phenomenal breakout success in the candy industry. Do you know what I'm talking about, Jordy? No. There's one really big winner recently, and it is Nerds Gummy Clusters.

1:01:14These Nerds Gummy Clusters, I don't know where this got buried in this. This went very viral, I guess. But Nerds was a brand of candy that had been around for a long time and was doing, I think, fine, sort of like, you know, stagnant revenue. And then they came up with this way to create like – have you ever had Nerds? Are you familiar with this? Of course. Yeah. It's like this like grainy like sand texture or something. I don't know. It's like a bunch of little pellets. And it's sort of like – Are you mansplaining Nerds? Yeah. It's sort of like hard to – I don't know. So it's not that accessible.

1:01:50It's like you either dump it all in your mouth, and then you just have a ton of... It's not a great user experience. It's a little outdated. Really? I think so. I thought it was fantastic growing it. You did? Yeah. Interesting. Nerd ropes, too? Yeah, yeah. So basically what they did with the Nerd rope, they cut it up into these little clusters, and this product has been selling fantastically well and has done very, very well. And there were also some... But yeah, there's a bunch of people chiming in the comments. Peelers, gummy mangoes, there's a whole bunch of stuff. There were even some companies that did sugar-free candies that I think have done well.

1:02:29There's been a number of sort of new startups. There really isn't like a David's Protein Bar type of breakout success in candy that I'm familiar with. I love candy. I just assume that it all wants to kill me. Bring me the poison. Bring me the poison. What was your favorite candy growing up? Hmm. I don't know. That's a tough one. Maybe Starburst, Skittles, the classics. Oh, wow. Is there anything else? Hot tamales. Hot tamales. Okay, interesting. Yeah, yeah, yeah. Tyler, what candy do you get to? More of a chocolate guy. Chocolate guy. I don't really like any. Oh, refined. Oh, interesting. Yeah.

1:03:05I don't know. I pick up candy from time to time at the airport, something. Oh, yeah, you do. Yeah, yeah. If we're getting on a plane, I like to have some candy hanging around. But, you know, the Nerds Gummy Clusters, I've had those on plane flights a few times. But good to see that the candy community is coming out in full force in support of candy innovation. Well, I went and found this original post on r slash candy. What do they say? The post seems written by ChatGPT, M-dashes, all of it. And, yeah, basically everybody's just nostalgia maxing. What are they most nostalgic for? What do they like?

1:03:44They're just saying they like the old school stuff, Willy Wonka candy. Does a Reese's peanut butter cup count as a candy, or is that a chocolate? I think of chocolate as a subcategory within candy. Halloween candy. I've got to buy Halloween candy for Halloween. Snickers came in there. Okay, yeah. Snickers, Snickers counts. That's good. It's also an ingredient that you could use. You wouldn't say a chocolate cake is candy. There's something about the size that matters. Producer Ben wants a protein Twix. I would be surprised if that revitalizes the candy industry. Anyway, let's move on to this story about David Ellison, the sun king of Hollywood.

1:04:28Before we read through this, we'll tell you about Sentry. Sentry shows developers what's broken and helps them fix it fast. That's why 150 ,000 organizations use it to keep their apps working. So with help from his billionaire father Larry, David Ellison, is trying to become the biggest studio mogul in history. One weekend in 2006, Dean Devlin, a movie producer, was driving across Los Angeles to deal with an emergency. His new movie, Flyboys, had opened that weekend to reviews so bad that one of the stars had abruptly checked into the hospital. What? On his way to visit the actor, another potential crisis appeared on his phone.

1:05:10I was terrified when the phone rang and it was Larry Ellison, Devlin said. Devlin made a name for himself with a string of blockbusters in the 1990s. Stargate, Independence Day, Godzilla. Wow, what a run. But he'd struggled to come up with the money to make Flyboys, a script about a band of World War I fighter pilots. Then in a stroke of luck, he learned that Ellison, the founder of Oracle and one of the world's richest men, had a college-age son named David who was not only trying to break into Hollywood, but was also an amateur pilot. David even had a nascent film production company with an aerial name, Skydance.

1:05:48Larry agreed to help fund the movie, which cost$60 million, and Skydance was listed as a producer. David, meanwhile, got a plum role alongside James Franco as a pilot named Eddie who can't shoot straight. Come on, Eddie, David says in the climactic fight scene. Can we try to find Eddie in Flyboys? Yeah, let's find a David Ellison hype reel or maybe just the trailer. And do you see the photo that Vulture chose for this is sort of a collage, a rendition of David Ellison. But you know, they put him in the Royal Oak right there in the photo. You'd love to see it. So Devlin, yeah, let's watch the Flyboys trailer.

1:06:34I'd love to see this. I actually haven't seen this movie.

1:06:41Franco.

1:06:44Maybe this is the movie to watch this weekend. Except he wanted to learn to fly. No, that's a cool shot.

1:06:59We've got to turn this into a cool closet. I think that's David, right? Oh, really? Guy sure knows how to make friends. All his friends out there. This is your quarters. French is sure put on a nice walk. That's him. Yeah, wow. Let's go. Heard his good luck to rub your head. I wouldn't do that if I were you. Sorry, fellas. This room's reserved for killers. Oh. Did you have a fuck there, killer? Going to war in two days. You worried about me? The Germans are moving to a world heart. I definitely saw this movie as a kid. I love planes. There he is. We'll be back by lunch.

1:07:44This is a pretty sweet movie for 2001. Worth the$60 million. They underpaid. Are they missing a zero? Seems like a$600 million movie. I'm going to turn this into a cult classic for fans of enterprise databases. If you're a fan of hyperscale and data center build-out and data center construction like the rest of us, you've got to be a Flyboys guy. You've got to get the original Flyboys merch. You've got to get the DVD copies. You've got to start collecting this stuff. Devlin. Devlin thought he knew why Larry was calling. Flyboys had bombed at the box office, and Larry stood to lose millions. As it turned out, Larry wasn't all that stressed about the money.

1:08:26He had spent tens of millions more just trying to win a sailing race. Sick. He was, however, concerned about his son, the actor in the hospital. Wait. Whoa. The reaction to Flyboys had so unsettled David that he was experiencing an episode of atrial fibrillation. Doctors had to shock his heart back into rhythm. Wow. That's crazy. This is bullish. Yeah. This is bullish. He was so upset about losing that he was hospitalized. That's crazy. You think he's going to lose Warner Brothers? Maybe not. I don't I think this is a guy you want. Yeah, this is good. Okay. He was thinking that whole weekend, I just lost my dad a bunch of money.

1:09:11Wow. He was upset about letting his father down. Larry told Devlin to make sure David knew he was still proud of him. Flyboys was the humble beginning of the Ellison's family's adventures in Hollywood. After giving up on acting, David spent more than a decade building Skydance into a player in the blockbuster business, financing such hits as Top Gun, Maverick, Crazy. and Mission Impossible series, along with a number of largely forgettable movies. Have you seen Top Gun? Yes. Sick. Have you seen Top Gun Maverick? No. It's good. It's not as iconic as the first one, but it is a great ride. What about Mission Impossible?

1:09:43There's like seven movies in the series. Have you seen any of them? No. The first one's great. No. He was, in other words, a producer much like many others in town. Then last year, David's father put up billions to help him buy Paramount, one of Hollywood's legendary movie studios, and accompany more than ten times the size of Skydance. This past fall, the Ellisons set their sights on an even bigger prize, Warner Brothers. At 43, David is now the first millennial, let's give it up for the millennials, to control a major Hollywood studio. While running Skydance, he built a reputation as someone with an earnest love for the kinds of action-packed blockbuster movies he adored as a kid and the money to actually make them.

1:10:20But his pursuit of Paramount and now Warner Brothers had revealed an empire-building streak in line with his father, who is currently the fifth wealthiest person in the world. It's possible to sum up the divergent fortunes of the tech and entertainment industries over the past 20 years in the simple fact that when Flyboys came out in 2006, Larry was worth$18 billion and couldn't afford Paramount, which was valued at$22 billion. By last year, the$6 billion Larry spent to help David buy the company barely made a dent, which briefly peaked in his fortune, which briefly peaked in September at close to$400 billion.

1:10:53Oracle, Larry's company, is the world's most boring tech giant. Disagree. It's fascinating. I agree. It's one of the most exciting to me. Tell that to Chase Lockmiller Crusoe, okay? We'll see who's boring then. We'll see how boring Oracle is when we talk about Abilene, Texas. Go to Abilene, Texas and tell me that Oracle's boring. Come on. Who's writing this? Who's writing this? Larry had long harbored more globe-spanning ambitions. Even before starting Oracle, he dreamed up a giant conglomerate he would someday build called Universal Titanic Octopus. What? What? That's so sick. Today, he keeps some of his wealth, which includes 98 % of a Hawaiian island, in an entity called Octopus Holdings LP.

1:11:37In the past year, Oracle has become central to the AI boom, providing back-end computing power for OpenAI and others looking to build LLMs. In September, after Oracle reported a potential windfall of AI-infused revenue, its stock shot up so much that Larry's net worth increased by$89 billion in a single day. That still was such a wild moment. 15 Paramounts, basically. You get 15 Paramounts in one day, and you're like, okay, yeah, I'll take 10 % of the gain from this day and buy a movie studio. Later this month, the company is also set to acquire a stake in the American version of TikTok. All of that combined with David's ownership of one and perhaps two major studios, as well as one and perhaps two major news networks, Paramount owns CBS and Warner owns CNN, has spurred concerns that the suddenly omnipresent Ellisons are building an empire that could dwarf even the Murdochs.

1:12:28What the Ellisons want with their conglomerate individually and together will affect the future of what appears on all kinds of screens, and the relationship between father and son has provoked no shortage of psychoanalysis. Larry divorced David's mother when he was young, and the two became close only as David grew older, and especially once he got into business. David has always struck many people as his father's opposite, shy and competent, not a bombastic, world-conquering visionary. David's ambition was initially as humble as making movies he likes, but the current climate has thrown him together with his father and shown him a path to something greater if he can pull it off.

1:13:06Here, I'll continue, but first I'll tell you about Plaid. Plaid powers the apps you use to spend, save, borrow, and invest, securely, connecting bank accounts to move money, fight fraud, and improve lending now with AI. So last month, the Warner board spurned the Ellison's and Paramount's offer and agreed to sell its movie and TV studio to the HBO Max and the HBO Max service to Netflix, creating a potential streaming colossus. The true streaming octopus is forming under Netflix's reign. The Ellison's have continued to argue that their offer for all of Warner Brothers, including its cable networks is superior.

1:13:46They are currently attempting a hostile takeover and trying to persuade Warner Brothers shareholders to accept Paramount's offer instead. The battle is expected to extend well into 2026. Glad to be in the media business covering this because it's certainly entertaining. Of course, thanks to Dylan Byers over at Puck for putting this on my radar by posting it on X. The battle is expected to extend well into 2026. Some observers believe that acquiring Warner Brothers is in fact critical to Paramount's survival. Either way, now that he has made his move to become one of the industry's leading moguls, David Ellison is faced with proving to a skeptical Hollywood that he's up to the task.

1:14:22David's mentors are Steve Jobs and David Geffen and obviously his dad. Those guys are ruthless killers, single-minded people with a different DNA, a former Skydance employee who worked closely with David said on the record. A former Skydance employee said, David is a smart dude who is reasonably shrewd, but he's not that. He's not the ruthless killer. He's not the Steve Jobs, the David Geffen. He's something else entirely. During the Warner Brothers auction, the company gave code names to each of its suitors. These were the code names. Netflix was noble. Comcast was charm. Paramount was given the name Prince.

1:15:02Interesting. One of the reasons David Ellison loves blockbuster movies is the way in which a spectacle can contain a more intimate story, how a battle for an intergalactic peace or control of an empire can really be a story about a family. He has often described the formative experience of seeing Terminator 2, Judgment Day as a child. The fate of humanity was at stake, but what David remembers is Arnold Schwarzenegger rampaging across the screen to protect a young boy who never knew his father. It's emotional. It's a great movie. Terminator 2, Have you seen it? No. Brutal. It's like one of the greatest movies.

1:15:40Should I watch it? It's so good. Is that what I should add? You should, you should. And truthfully, you can watch Terminator 2 without watching Terminator 1. They have very different vibes, very different pacings. You should watch both, but Terminator 2 is just a fantastic movie. It's so, so good. And there's iconic memes and gifs that I feel like you should know. I'll be back. Okay, so Azar and some of the chat actually put together a film list for me. Yes, thank you. They sent it to my DMs. Thank you. Interstellar, Heat, Inception, Shawshank, Steve Jobs, American Hustle, Silver Lining Playbook, Almost Famous, The Talented Mr.

1:16:18Ripley, The Sting. The issue is these are so many movies you could easily be messing with me and just sprinkle in like 10 that are just like really bad. The worst movie ever. And I'm like, well, the boys said I had to watch it. You've got to watch Sheeshly. I think that's notoriously one of the worst movies. There's a lot here. It's three pages of movies. So I promise I will do my best to watch all these. It may take me multiple decades at my current pace. Yeah. But I'll work down the list. Well, so at the end of Terminator 2, David Ellison sitting in the theater. Arnold Schwarzenegger, not to spoil the story, but Arnold Schwarzenegger sacrifices himself in a vat of molten steel in front of the boy, and the eight-year-old David Ellison starts to cry.

1:17:03Wow. Larry wasn't around much when David was a kid, one of the only similarities in their very different childhoods. Larry was born in New York City in 1944 to an unwed teenage mother who gave him up for adoption at nine months old. His adopted father, Louis Ellison, was a Russian immigrant who took his last name at Ellis Island. And a foundational part of the Larry Ellison story is that Lewis regularly told Larry he would never amount to anything. Wow, what a crazy thing. Why did you adopt this nine-month-old kid just to dunk on him? This is so, so bad. But clearly it lit a fire in Larry and he went on to build one of the greatest tech companies in history.

1:17:45Maybe Larry should write a book called Hardcore Parenting. You know, people talk about soft parenting, right? You want to, you know, you don't. But Hardcore Parenting, it's like wake up every morning, tell your children you will never amount to anything crazy as a result larry likes to say that he had all the disadvantages necessary for success and colleagues at oracle got the sense that larry's seemingly endless ambition came partly from a desire to prove himself to his adopted father i'm not sure i would recommend that everyone raise their children like this larry later said but it certainly worked wow that's crazy david's mother barbara booth was Oracle's 10th employee.

1:18:25He hired his mom when he needed a 10th employee and Larry's third wife, oh, David's mother. Sorry. Larry hired David's mother, Barbara Booth. They divorced shortly after David's third birthday and right after his sister Megan was born. When Larry and his first wife tried couples counseling to save their marriage, Larry told the therapist that the very idea of love eluded him. I don't understand love. I don't understand people bonding to one another. What is it? Larry said that he spent obligatory weekends with the kids and seemed to only realize later that a family was something he wanted. One night...

1:19:01Guy doesn't understand love, but is on his fifth marriage. Something. He's like, I just need to get my reps in. Larry called Booth to say he was lonely. You're never alone because you have the kids. You always have a body, Larry said. I'm here all by myself. David credits his love of movies to his mother. The family had a collection of 3 ,000 VHS tapes, and David's ideal day as a child was to watch the Star Wars trilogy or all the Rocky movies back-to-back. I did the exact same thing when I was a kid. It was fantastic. The Star Wars trilogy on VHS was one of my prized possessions. Quickly, let me tell you about Lambda.

1:19:39Lambda is the super intelligence cloud building AI supercomputers for training and inference that scale from one GPU to hundreds of thousands. It wasn't until David became a teenager that he and Larry started taking flying lessons together that they grew close, eventually staging mock dogfights over the Pacific. These guys are dogfighting each other. This is insane. How to bond with your son. Yes. As a matter of fact, I do have that dog in me. I literally dogfight with my dad in planes that we fly over the Pacific. David went on to compete as an aerobatic pilot, whipping barrel rolls, flying at 300 miles an hour, just 15 feet above the ground and pulling off his signature move, the Dave-o-later.

1:20:26He has a signature move. This is incredible. After graduating from high school in 2001, David went to Pepperdine University as a prospective business major. He hated it and he transferred to film school at USC. He commuted from the beach in Malibu where he lived in one of the many homes Larry had purchased there on a$65 million buying spree. His neighbors included Jennifer Aniston, who was renting one of Larry's other houses. For a senior thesis, David wrote, directed, and starred in a thriller called When All Else Fails, in which he plays a billionaire's son who has to rescue his diabetic girlfriend from kidnappers before her insulin runs out.

1:21:00That's an interesting frame for a plot. His girlfriend at the time, who was a diabetic, played the part. According to people who worked on the production, David was respectful and eager to learn, if a bit oblivious to the ways in which he wasn't making an ordinary student film. The budget was close to$100 ,000, and he shot part of it in one of his father's giant homes in the Bay Area. When they ran out of film, David offered to fly his plane to Los Angeles to pick some up. Instead of graduating, David left USC to film Flyboys. He worked with an acting coach to prepare, but his performance didn't do any favors to a mediocre script.

1:21:34David spent the next several years going to auditions and landed a few more roles from Devlin, the Flyboys' producer. He played a bumbling assistant to a corrupt mayor on a TNT show and a shirtless drug addict who punches an FBI agent before leaping from the second floor of a motel in a TV movie. His most significant screen time came in a raunchy 2009 comedy called Hole in One in which he plays. John, have you seen this? No, I haven't. Should I have? Do you expect me to see every movie? I haven't seen every movie.

1:22:07he plays the owner of a golf apparel company and there's a picture of him here in a sort of golf polo whose partner loses a bet that leads two nefarious plastic surgeons to give him breast implants sending David's character on a quest to drum up enough money to help him get them removed what a wacky plot only in 2009 are you making movies like that that is wild he delivers one of his first lines in the movie after a woman takes off her shirt. That was pimp. He says, wow, that is a phrase straight out of 2009. No one uses those words anymore. That slang has fully died. His acting career more or less came to an end a few years later after he wrote a script called Northern Lights about a pilot friend of his who had died.

1:22:53Taylor Lautner signed on to play the lead, but dropped out after finding David had written a role for himself. Rough. What did Taylor have against David? I guess he was like, I don't want to act with him or something. I mean, it's like an F1. There's always like, who is this person? Do they deserve the seat? Or are they there because they just have the money? And so you sign up for this. You're in this role. The movie bombs. And it's seen as a black mark on your IMDb resume, I suppose. Anyway, before we move on, let me tell you about Vanta. Automate compliance and security. Vanta is the leading AI trust management platform.

1:23:32While David inherited a love for movies from his mother, it became clear that if he wanted to stay in the business, he needed to think more like his father. David wanted to be an actor. And when that didn't work, he was like, if I'm not going to be on screen, I'm going to be the biggest guy in the room. A person who knew him at the time told me, the writer of this vulture piece. Larry's friend, David Geffen, introduced David to Skip Brittenham, a legendary Hollywood lawyer who had represented Pixar, which was co-founded by Steve Jobs, who was Larry's best friend. Brittenham helped David make a business plan for building a production company and learn to navigate the web of agencies, producers, and studios in Hollywood.

1:24:12David wanted Skydance to make the type of movies he loved, movies with explosions. Let's go. Finally, the last Hollywood producer who understands how to getting in the theater you know how truth has their etfs their patriot etfs i think they gotta get paramount paramount skydance in there yeah because this is the movie studio that likes committed to explosions explosions very american yes something there so andy kessler was an investor who worked with david to raise money for skydance early on and they had uh a rule they said he swore to me. No rom-coms, no touchy-feely emotional Oscar bait, just explosions.

1:24:54But funding the Hollywood dreams of the 25-year-old failed actor son of one of the world's richest men was a tough sell to investors. Will people show up to these movies? And Larry ultimately made the point moot in 2010. He stepped in to provide almost all of the initial$150 million David was trying to raise, according to multiple people with knowledge of the company's finances. Their timing was good. the late aughts Hollywood was roiled by the financial crisis and Paramount, the studio behind the Godfather and Titanic, was the struggling movie-making arm of the Redstone family's Viacom cable TV empire.

1:25:29Deutsche Bank had bailed on an agreement to fund Paramount's movies and the studio needed cash. When Larry's funding for Skydance came through, David signed a deal to co-finance Paramount's movies, which meant he would help fund production in exchange for a share of the profits. David, had initially considered partnering with Relativity Media, which later collapsed in a fiery bankruptcy. While co-financiers often get stuck funding risky projects, Paramount's desperate state opened the door for David to negotiate his involvement in some of the studio's biggest franchises, like Star Trek and Mission Impossible.

1:26:02The first feature film, David backed a Coen Brothers remake of the Western True Grit. I've seen that. It's a great movie. Have you seen it? No. No. Went on to make$250 million, but they couldn't get a dime out of Geordie Hayes, apparently. And it earned 10 Oscar nominations, but couldn't get Geordie to throw it on streaming on one random Sunday night. His sister, Megan, followed him into the industry, co-financing True Grit and launching her own company, Annapurna Pictures. Oh, I didn't know that. Interesting. Annapurna has been on a run. I think they also have an interactive division, although that might not be related.

1:26:37I'm not sure. uh the skydance thesis according to a person that worked closely with david was that hollywood wasn't prepared for the overwhelming economic force of a silicon valley billionaire throwing around his wealth and ambition plenty of dumb money came into town but producers and talent mostly viewed it as something to take advantage of money in hollywood is not appropriate appreciated uh the person said this was an opportunity to reverse that to make hollywood treat money with respect we got to respect the dollar in hollywood nicholas cage was it was early here yeah Yeah. He famously said, I do not chuckle when I get a big check.

1:27:11I have respect for the dollar. That's right. Cage, Nicolas Cage. Shortly after signing the deal with Paramount in 2009, a person who knew David then said that he had declared, quote, we are going to buy Paramount one day. He called his shot in 2009. Well, here he is. He did it. David proved to be a savvy enough operator negotiating better terms with Paramount when his deal came up for renewal and working hard to pick up the quirks of a job that involved everything from managing the egos of talent to negotiating debt facilities with J.P. Morgan. He would sometimes tell colleagues about an instructive conversation he'd had years earlier about running a company in Hollywood.

1:27:51At the premiere of Steven Spielberg's Minority Report, starring Tom Cruise, he walked up to congratulate David Geffen, who had produced the movie through his company, DreamWorks. As David told it, Geffen shook his hand. Steve is getting paid. Tom is getting paid. How much do you think DreamWorks is getting paid? Geffen said. He held his hand up in the shape of a zero. DreamWorks was making a relative pittance on the movie, an early lesson in the confusing economics of the business, where a poorly structured deal could mean even a hit might generate little for the company that produced it. It was always in the back of David's mind.

1:28:28On Minority Report, everybody got paid except for DreamWorks, a former Skydance executive said. Have you seen Minority Report? That is one of my favorite movies of all time. It is a fascinating, fascinating movie. It is amazing. It's not on the list. It's great sci-fi. It's not on the list that the chat made. Add it to the list. Add it to the list. Minority Report is fantastic. Really great vision of what the future of VR and augmented reality could look like. has a whole bunch of other interesting sci-fi elements. Tom Cruise obviously puts on a masterful performance, but the interactive, how they interact with computer technology is remarkable in that.

1:29:06Anyway, let me tell you about Fin.ai, the number one AI agent for customer service. If you want AI to handle your customer support, go to Fin.ai. Okay, I think we're running out of time. Chesky is going to join in a second. There is some funny dialogue happening. I'll fill you in. The Calci market, will Elon win his case against OpenAI? Yeah. He was at a 36 % chance this morning. What's he doing now? He asked, what are the odds looking like? They responded. Now Elon is posting, I've lost a few battles over the years, but I've never lost a war. Elon said that? He just posted this. Oh, my God. Absolutely insane.

1:29:49Scores to earth. Well, we have Brian Chesky, the co-founder and CEO of Airbnb in the Restream Waiting Room. We'll bring him in the TVP at UltraJump. Brian, great to see you again. How are you doing? Hey, guys. Good to see you. Thank you for having me back on. Of course. You're welcome anytime. It's great to be here with you. First, kick us off with the news. You've been acquiring talent, building the roster. You have a new CTO. How did it come about? Tell me about this person. Yeah. Yeah, so like, you know, obviously a couple years ago, I started asking around like, who are some of the best leaders in AI?

1:30:21And I started meeting people. And one of the names that came up over and over was this guy named Ahmed Aldal. And he was head of generative AI at Meta. We met and I just was completely blown away by him. Not only was he at Meta and he essentially led the team that created the LLAMA models, been downloaded a billion times. There's been over 60 ,000 derivative models. But he was also at Apple. He actually joined Apple at a college, worked on the original iPhone in 2005, led a lot of the multi-touch technology. He led the autonomous technology group that was the self-driving car project at Apple.

1:30:57So he was somebody at the frontier. He really understood both frontier technology, but he also understood that kind of Apple design craft, which I think Airbnb also was really passionate about. And we just really hit it off, became friends. And as my current CTO, we were talking about handing the baton to somebody. we thought, I'm going to be really great. Yeah. How are you thinking about design in the era of generative AI? These apps are amazing and the fun and the models, like you see the benchmarks, you use the models and it's magical. It's like, you know, we created fire. And then one rough edge, one button that doesn't quite work the way you expected it.

1:31:34I was lamenting the fact that I was using Claude and I couldn't get a deep research report to read it out loud to me. That feature is in one app and there's different features there. How do you think about design? Yeah, it feels very notable that even in all the leading LLMs, there's still so many rough edges in the product. Yeah, little low-hanging fruit. And part of that is like a new product paradigm. We're kind of figuring out what the edges are. Yeah, maybe there's all sorts of metaphors. I remember when I was a kid, I think I got like a compact computer and I was running MS-DOS. And we're kind of more in the MS-DOS phase of AI.

1:32:12I think, not to belittle the interfaces, but the chatbot is not the end state interface. The current chatbot is not the end state interface for AI. Think about your phone and your app. Do you want to chat to get the weather? Do you want to chat to do most of the functions? The answer is no. I think the future AI interface will have a reference to chat, but it's probably going to be much more visual. Most people aren't completely text-based. and I do think it's kind of like we've been iterating the engine of the car, but we haven't really redesigned the car. And so the chatbot is really a pre-AI interface that we've plugged onto this incredibly powerful engine.

1:32:54And so I think what we're trying to do is think through, where can we take this interface? What is the multi-touch reference for an age of AI? And we're not going to do this overnight. We're not going to be the only company doing this. But I really think the models are even more powerful than people realize because we have hardly harnessed them. And if you look at where they're going to be in a year or two, we really want to skate to where the puck is going and do something really, really cool. So I totally agree with you. And I think the other thing I'd say is I'm on the board of Y Combinator.

1:33:25And I think the last I checked, 87 % of the companies are enterprise companies per batch. People are. I know. We go and we meet with every batch to have them on the show. And we're always trying to, hey, we want every consumer company to come on. Because oftentimes they give you. And the interesting thing is all of them, all these founders started doing consumer businesses. You asked them, they were doing Minecraft servers and flipping sneakers and stuff. So they actually have some consumer DNA, but they end up realizing it. You're bringing up a really good point. I joined Ycom here in 2009, and that was two years after the iPhone was introduced.

1:34:09Almost every startup was a consumer company. Enterprise is awesome, and I think it makes complete sense why AI is starting with enterprise because the models are not cheap to run. You can basically vertically integrate any business and completely automate the workflow. It's going to revolutionize how we do work. But the point is that the biggest prize is consumer because that is what's going to reach daily life for billions of people. I think a lot of people are a little bit, a lot of entrepreneurs are afraid to go into consumer. That's the biggest thing we've heard. They're afraid. Partly, they're worried that OpenAI or Google are going to essentially eat their lunch and vertically integrate.

1:34:43The other concern is consumer is just hard. It's a little bit more of a hitsterism business. Distribution's a little more mature because of the app store. There's not really new distribution channels in the same way. At the same time, we are in the midst of a revolution. and more than just a couple chatbots have to be the way that you interface with the world. And I also, this might be self-serving to say, but I also believe this independently. I don't think we're going to use just one agent. I don't think we don't have just one person in our life. I don't think we're going to have just one agent in our life.

1:35:13I don't think we'll use thousands, but I think there is going to be some level of specialization that's going to help. And there's going to be too much specialization where I wouldn't download that agent or use that agent. But I think even if you want a meta orchestrator, whether it's Claude or Gemini or Chechebiti, specialization is going to help even within consumers. So we want to do some really cool stuff within traveling and living. Yeah. What advice do you give to the 13 % of Y Combinator founders that are going after consumer, at least for now or at least at this moment? Is it, hey, keep your burn really low, run a ton of experiments, stick it out?

1:35:55Because it feels like part of the reason why 87 % wind up in enterprise is because businesses are just throwing money at them. They're saying, yeah, you're doing consumer, but if you come solve this one really small problem for me and build this API for me, I'll give you a contract. I'll be your first customer. It's very easy to get the money flowing on the business side potentially. But how can you actually make it as a consumer founder when there is a big prize, but it might be a harder road. Yeah, so many thoughts here. I mean, one of the things that Paul Graham taught us and Gary Tan teaches, which I think is very effective, is use other Y-combinator companies as your early adopters, hence why a lot of companies go into enterprise.

1:36:33If you have an artist company, don't go to like Boeing to try to use your service, go to a young startup and work your way up to the large companies. And so they've developed this incredible ecosystem where thousands of companies are adopting each other's tools and they bootstrap each other. And that's actually been part of the explanation for why Y Combinator has so successfully created so many enterprise companies. But when I was in Y Combinator, we got everyone to use Airbnb. And so in other words, like we have a huge alumni well work. I would tell people that are going into consumer, number one, don't be afraid to do AI consumer.

1:37:03It might fail, but like investors aren't going to hold it against you that a startup failed. They're going to like a founder who's failed is more fundable the next time than someone who's never started unless they really blew it because you have experience. And I I think that's what's so great about this. Number two, you got a lot of early adopters within Y Commodore who are normal people, too. They have daily life. I would agree to keep your burn rate low. I would pick a niche. I wouldn't pick a niche that's too narrow. Maybe AI to do something incredibly narrow is going to get really, really difficult for somebody to want to change their workflow and behavior.

1:37:38Just think about all the parts of daily life that are annoying. Maybe pay attention to your sister, your significant other, your children, whoever's in your life, and just look at their life and ask, how could their daily life be a little bit easier? I would just go for it. By the way, the App Store is mature. That's a little bit of an issue. At the same time, the top apps in the App Store are AI apps. If something is truly great and truly breakthrough, people will seek it out. And I do think my prediction is that AI is going to revolutionize consumer applications in the coming years. And they might not be apps.

1:38:14They might become agents. Agents might be the new apps. We have to figure out what that means, what's new platform paradigm. It's probably agents, not apps, but we'll see. But ultimately, I think that's the big takeaway here. And so what we're going to do is we want to really innovate on the consumer experience. We want to interface on, like, what does consumer interfaces look like in the age of AI? And I think they should evolve past the current design paradigm. And we'll see where it goes. Yeah, it would be, I'm not expecting to open the Airbnb app and see a blank text box purely. There's so much more that you can do.

1:38:48Have you thought more about, Airbnb app is unique, I think, because it's not this reactive, people are opening it 10 times a day. There is an intention. And there's probably a lot of value in understanding, okay, this person, they like to travel around the holidays. They're opening the app in October. Let's think that through. Agents and LLMs can probably work through a customer's history, their preferences, just do better recommendations, what you're servicing them. How much have you thought about being proactive about reaching the customer with something that's more tailored? What's been working?

1:39:28Where do you see it going? Yeah, I mean, so many things. One of the other problems with consumer, to keep coming back to it, is there's a business model challenge. These models are not cheap. There's probably one of three ways you can make money in consumer. You can put ads in your chatbot. You can pay a subscription, or you can do essentially transactions. Airbnb luckily has the business model. We have transactions, and our transactions are pretty high-dollar transactions. We actually have the business model going for us. We also have a lot of data. We do about 50 billion searches a year on Airbnb.

1:40:03That's nothing compared to Google, but 50 billion. That's that number of times somebody's typing in something into an Airbnb search box. We actually do have quite a bit of data about people, their preferences, and we can bring it in. Part of what we want to do is if the interface is a little more chat-like, again, it won't be per se a chat bot exactly like OpenAI, but if it's a little more of a turn-by-turn chat-like connection, you're going to get a lot more information from people. And I think the one benefit of the current chatbots are they have memory. And so you're putting a lot of context in the window.

1:40:39And I think memory, by the way, is something I think we're going to want to technically innovate on because the chatbots still are stateless. And so you're essentially the prompt is putting in this kind of, obviously, a lot of your memory. So I do think there's going to be an area around how can anybody Airbnb, understand people's preferences, learn more about you, understand what you care about, and just be really thoughtful. Like, what's your bucket list? Who are you traveling with? One of the other things Airbnb can be really good at is the majority of trips have multiple people. Most chatbots are single player mode.

1:41:10They have collaborative tools, but most people are not chatting with multiple people. Most people on Airbnb are searching for an Airbnb with multiple other people. So I think another thing we really want to do is really be great at understanding two people. They have totally different preferences. How do we help them figure out the right choices for them? So a lot of collaboration tools are going to be really important. But I think at our heart, what we want to do is really build out these robust profiles, deeply understand people, and give them what they want in the real world. How are you thinking about the challenges of generative AI imagery in Airbnb listings?

1:41:44I can imagine there's sort of a risk of, what is it called, catfishing, where someone takes a photo. It is their house, but they said, They patched up the walls. Let's change the lighting and make it look a little bit warmer, a little bit nicer. And there's probably a little bit of that that's fine. If you took it on a phone, you want it to look nice. But you want it to represent the actual experience. How are you thinking about the challenges with that? Yeah, I mean, it's really interesting, right? I don't know. I saw on Twitter yesterday the stranger things where you can turn your face into someone else's face.

1:42:14And so what this really means is that know your customer authentication. A lot of it is like you do a selfie and an IED or you do a video. A lot of this stuff now AI can get around. And so, I mean, we should remember AI stands for artificial. And so we're now starting to live in a world where if you see it in a screen, it might be artificial. And so I think authentication is going to be critical. And one of the best authentications is in the real world. I mean, we're going to be pretty soon living in a world where you don't know if something's real unless you see it in the real world, unless we have authentication.

1:42:45One example of the way Airbnb can solve this, for example, is we do have a network of professional photographers over 5 ,000 around the world. In fact, we built an on-demand professional photography network before Uber even launched, where you hit a button and the photographer shows up and takes photos at your home. An advantage of that is those photos are by Airbnb, and so if we took them, a customer would know those are authenticated photos from us. That's just an example of something that is possible. But I do think solving this, is this real, is this authenticated, is really important. So we're going to have to have technology solutions for it.

1:43:19What's your latest thinking on integrating Airbnb with various LLMs? How are these conversations going? Into commerce? Yeah, I'm interested. I think last time I was on, I said they weren't ready. Yeah. And I, my view, you know, we chose not to launch with ChatGPT. They launched with Expedium Booking. And the reason why wasn't like, like I was a huge fan of that project and gave some advice on it and stuff. And I think it's awesome. I would love to show up on ChatGPT. I would love to show up on Gemini. I'd love to show up on Claude. I don't, here's the thing. I don't want to be the first. I want to be the best.

1:43:59Like we want to have the best integration, not the first integration. There are some limitations. 90 % of people booking Airbnb send a message to a host ahead of time. You have to have a verified ID. There's just some things we need in integration for Airbnb to have a really good integration. We'd love that. By the way, the traffic coming from chat to GPT converts higher on Airbnb to a booking than the traffic coming from Google search. Yeah. So actually, there's a win-win. And various e-commerce brands have been reporting this too. Like the traffic is super high intent because people are informed.

1:44:32Yeah. So I get, I get this question's asked to me often as an existential thing, as in, are you going to partner with them or not? And are they going to kill you? And there's going to be no app in the future. And my answer is there probably are no apps at some point in the future. They're probably agents. That's, there's probably some platform like paradigm. We're probably designing agents. I don't know for sure, but I think that we're going to exist on those platforms, but I also think specialization is going to matter. I don't think there's just going to be a few chatbots and they do everything and you are separated from every application.

1:45:02That's going to be true of some apps. And that's going to be true if you need like maybe a commodity. But there are going to still be reasons to work with different agents or different applications. I think some specialization will make a difference. What are the big, what is your super? One sec. One thing I appreciate about your framing of like we don't want to be first, we want to be the best is because I think other CEOs in travel that are not tapped into Silicon Valley that aren't on the board of Y Combinator feel this pressure to be like, I have to prove that I'm at least trying to be AI native, whereas I feel like you don't have any, there's no ego around, can we build great products with AI?

1:45:40It's simply like, yeah, we're going to, but I don't need to prove it by rushing into something. I appreciate you saying that. I remember when the metaverse launched, I had somebody who didn't on my team that was not - Which metaverse? You mean the idea of the metaverse? Yeah, yeah. Sorry. Yes, when it was announced, Mark Zuckerberg, they were like, what's our metaverse strategy? And I remember other companies were like, we have to have a metaverse strategy. And it's like, that is definitely what someone usually would say if they're not really steeped into it. So I think it's important that you do something really, really well, but I don't think you want to be first to signal, we care about AI.

1:46:17That's kind of vanity. You don't need to do that. And so when it's ready, we'll do something great. And we want to add value to customers. And that's kind of all that really matters. If you were running Apple, what would your AI strategy be? That's a big question. Oh, wow.

1:46:41I would say that, like, the devices should be completely built from the ground up, AI native, AI first, probably not even AI first, AGI first. We should ask ourselves, what is it going to look like in a few years? It's probably going to be fully agentic. These devices are going to be able to work while you sleep, and you should work backwards from that. These devices and the paradigms haven't changed since the advent of AI. These are pre-AI devices running AI. So my question is, what would it look like if you designed it from the ground up knowing what we know today it would probably look pretty different yeah so less gen moji more actual hardware more hardware we are we are in the midst of a revolution obviously i mean i don't want to be like saying what everyone already knows but here's what i do want to say like the everyday life of the average person has barely changed in the last three years we're on chat you bt a lot we're in gemini we're on claude people listening are much more technical than the average person.

1:47:41They're doing unique things, workflows. The average person is still mostly living the life they lived three years ago using the same devices. The world hasn't finally made the shift yet. It hasn't finally made the shift. It won't make the shift until daily life has changed. And daily life doesn't change until the devices have changed, the operating systems have changed, apps become agents, and we're all living in a completely natively consumer AI world. And we're not using chatbots, at least not today's chatbots. We are still living in the DOS era of AI, and we need to move into the multi-touch era.

1:48:12The chat is begging us to ask you for an arm routine. Is there a dedicated arm day? Are there any particular machines that you recommend? Free weights, what is the secret to those arms? Oh, thank you very much. Well, first of all, genetically, I have better arms than shoulders or chest, so that's just one thing. But I actually don't have an arm day. I used to do a five-day split where I do back, legs, chest, shoulders, arms. I now do arms with shoulders. So I'll usually do two exercises of biceps, two exercises of triceps, typically a compound free weight movement and a machine movement. So maybe like a dumbbell curl followed by like a hammer strength cable with like a rope or a V-bar.

1:48:58And then triceps, always doing a push down with a cable, and then maybe like a French press or skull crusher, depending on what we all call it. Okay, that's a good one. I like to call them skull crushers. They sound a lot more hardcore. Oh, yeah, totally. Yeah, yeah. You know there's heavy metal playing. We don't know how to name exercises like that anymore. We don't. The world's become soft. It really has the best name. I do think generally if you're like under 40 for sure, you should be doing almost entirely compound movements and free weights and dumbbells and barbells. You know, I'm 44, So I think transitioning to some more machines is good just because the tax it puts on your joints.

1:49:35But I still try to put in free weights. Are you using any AI in the gym? Are you tracking movements or asking questions? He's going to the Church of Iron, John. Or is it all raw? I'm a nutrition. Nutrition? I have a trainer. He was a former Mr. Universe. He competed in Mr. Olympia. So I let him do that. But my diet and my supplements are very much like I put everything into AI. And I do things, I don't know if it sounds eccentric, but I'll get blood work and they'll give you a PDF and you can put the PDF in a chat bot and it'll tell you to analyze it. And it can even tell you what supplements you need based on your blood work.

1:50:18It's probably something everyone should do. It's not that hard. And so I don't go out enough. So I don't need more vitamin D. The blood work said that, so I take a vitamin D supplement because I'm inside working a lot. So it's stuff like that. Yeah, yeah. Have you been surprised seeing the explosion of peptides? I wouldn't have imagined a world where people are clamoring to do actual injections. And given how popular it was, I mean, in the bodybuilding world, there was always a divide between how hardcore you were, which was like, are you natural? Are you natty or not? Yeah, natty or not. And now nobody's natty anymore, seemingly.

1:50:58No one's natty. I'm not surprised. I'm surprised, actually, a little bit how much the tech communities embrace fitness. I mean, like, when I came to Silicon Valley, I was, like, slightly self-conscious about, like, wearing long-sleeved shirts because I thought if I wore a short-sleeved shirt, people would think I'm a me-head and therefore an idiot. And people just didn't really lift weights in, like, 2007, 2008. And I really think that everyone's come around probably to just the science that nutrition and weightlifting is good. It's good for longevity. It's good for your mind. You'll live longer.

1:51:29You'll feel better. And honestly, weightlifting is far superior to cardio. You should do both. The weightlifting is far superior. It increases bone density. It's just you can only do one thing. You should do weightlifting. And so I'm really excited to see people actually embracing this. I mean, one of the things I learned about bodybuilding that I think is very much consistent with Silicon Valley is I was like 120 pounds in high school. I played hockey. I remember I broke my leg, and I had to do physical therapy. And through my physical therapy, I said, I want to learn about my body. And there wasn't a lot of information on the internet, so I'd go to the Barnes & Noble and read as many books as possible.

1:52:08And I learned something about bodybuilding. And that was, if I can change my body, I can change my life. and it was about control. And it's about this idea that you don't get in shape in one workout. It's 1 % a day. And frankly, those are the lessons I learned with Silicon Valley. It's not about one idea. It's about grinding every single day, year after year with consistency. And overnight successes take thousands of days, actually, typically. And so I think there's a lot of lessons from bodybuilding that I've brought to Silicon Valley. Yeah, no, 100%. It's a really, yeah, it's a really important frame of mind.

1:52:40And we appreciate you sharing all of that with us. We know you have a hard stop, so we've got to get him out of here. Oh, hard stop. Yeah. But thank you so much for taking the time. Do we have time for one more question? Enjoy. I hope to be back soon, guys. Really, thank you for having me here. Yeah. Awesome. Awesome. Great to see you. We appreciate having you on the show. We'll talk to you soon, Brian. Have a good rest of your day. Bye. Goodbye. Wanted to ask him about Super Bowl ads. Oh, yeah. And specifically, like, how they're thinking about it. I know they've run an ad in the past. Yeah. They haven't been an annual spender on that front.

1:53:10But we've got to do this. annual showcases sort of like the WWDC style. You wouldn't expect it from Airbnb because they're not necessarily introducing a new iPhone, but Brian's been very outspoken about putting the whole team on an annual cadence, a rhythm of releasing all of the information, packaging everything up, getting everyone to rush towards one deadline. It's an interesting way to run the business. We got cut off on the Sun King of Hollywood about David Ellison and Vulture. I encourage you to go read it. Yeah, you should go read it. It's a very long piece. It goes on and on and on. So many great stories.

1:53:45You should go check it out in Vulture. I'm glad that we know that the father and son are dogfighting together. I mean, some great writing from Reeves over at Vulture. Matthew McConaughey. But first, before we go into that, let me tell you about Graphite.dev. Code review for the age of AI. Graphite helps teams on GitHub ship higher quality software faster. Matthew McConaughey is taking a novel legal approach to combat unauthorized AI fakes, trademarking himself. The actor plans to use trademarks of himself saying, all right, all right, all right, and staring at a camera to combat AI fakes. And the print headline is, actor is not all right with AI fakes.

1:54:25I love it. Over the past several months, the Interstellar and Magic Mike star has had eight trademark applications approved by the U.S. Patent and Trademark Office, featuring him starring, smiling, and talking. His attorney said the trademarks are meant to stop AI apps or users from simulating McConaughey's voice or likeness without permission, an increasingly common concern for performers. Maybe we'll have to get a trademark on. You're watching TVPN, so we will get paid when someone generates a deepfake of us. The trademarks include a seven-second clip of the Oscar winner standing on a porch, a three-second clip of him sitting in front of a Christmas tree, and an audio of him saying, all right, all right, all right, his famous line from the 1993 movie Dazed and Confused, which I know you haven't seen, but I don't think I've seen Dazed and Confused either.

1:55:20Tyler, have you seen Dazed and Confused? No. Yeah. Just a little bit before all of our times. Maybe we've got to lock in and watch that. But is there anything else in this Matthew McConaughey piece? What else is he doing here? He says, my team and I want to know that when my voice or likeness is ever used, it's because I approved and signed off on it. We want to create a clear perimeter around ownership with consent and attribution, the norm in an AI world. So lawyers, lawyers finding a way. Well, if you do pay Matthew McConaughey, like Salesforce has, for an advertisement and you want to run it on connected TV, on streaming TV, You got to go to vibe.co where D2C brands, B2B startups, and AI companies advertise on streaming TV.

1:56:04Pick channels, target audiences, and measure sales just like on Meta. Yeah, this is interesting. It doesn't seem like they're going to trademark this and then go to ChatGPT and say like, hey, you want to license my likeness. It feels like he realizes the value of doing a handful of big deals with a sales force and probably wants to keep doing that. It'd be interesting to see if every all-talent needs to go and - File trademarks. File trademarks. At a certain level. I wonder what his total all-in cost for all of those trademarks is. It's not super cheap to file a trademark. We've done it. It's not nothing.

1:56:45It's not a billion dollars. uh what did uh what did rune say he replied to the deciphering of the uh the rune elves vague post Rune said, the elves have left for Valinor. And Boaz Barak says, our internal models are so good they can decipher Rune's posts. Says Boaz, who is a computer scientist at Harvard and OpenAI. And Rune says, after a vague post gains a certain level of mystique, I lose the ability to correct misconceptions. I'll let them have it. And Brandon Gurrell says, I was just made aware of John's theory that you were literally just watching Lord of the Rings and live tweeting. No one looked to see if he had tweeted, you know, Gandalf has reached the Shire.

1:57:31The fireworks have started in the Shire. They're having breakfast in the Shire. No one checked on that. So maybe that's the next one. Speaking of OpenAI, I have some breaking news. They partnered with Cerebris. Oh, yeah. So there's 750 megawatts worth of capacity over... Oh, no. That's a big one. Oh, no. That is massive news. It was well-deserving of the flashbang. Yeah, and it seems like Cerebrus is on an absolute tear post-Groc deal. There are a ton of rumors swarming about financings and IPOs, and it seems like they're in a great, great place. Do we know anything more about the partnership between OpenAI and Cerebrus?

1:58:20It's in the journal. Okay. Multi-billion-dollar agreement. I think$10 billion. $10 billion. Yeah, through 2028. Wow. That's very exciting. I'm very interested to see how they integrate it. Are they running 4.0 on it? Yeah, I assume, I mean, it's definitely, I mean, not definitely, but I would be very surprised if it's going to anything besides just like inference, right? Yes, but what are they inferencing? Because they also do a Sora video generation. That's a different model that might need different silicon at a certain point. OpenAI is working with Broadcom on custom chips. They're co-developing stuff with NVIDIA.

1:58:57They're working with AMD. There might be an Intel deal at some point. Who knows? Like they obviously, they use every part of the semiconductor supply chain and how Cerebrus fits in. It doesn't feel like it's just like, oh, OpenAI now runs everything on Cerebrus. It's probably a particular piece of the model architecture or a particular model. Maybe when you go to a deep research report or do some reasoning, it needs to be faster. I'm not exactly sure where they would be taking advantage of it, but it is exciting for them. I don't know. It's certainly bullish. Cerebrus raising one on 22 billion is the rumor.

1:59:36So they're doing the meme. Remember the meme 2021? It was like the classic round was like one on 20. But they're, of course, doing it at 22 billion. B instead of M. And also, I mean, good news for the OpenAI folks. They have the 1.4 trillion in deals. So they do another 10 billion deal. It's just 1.41. And at that point, you just round down. It still rounds down to 1.4. So you're still good. It's a rounding error. It's a literal rounding error. You're good at this point. Plant says, every guest on CNBC is like, yeah, we are concerned with the end of the global order and death of U.S. institutions.

2:00:12But we are looking forward to double-digit EPS growth this year. So we want to remain focused on that. Brutal. Yeah. Real. It's a tough time to be navigating politics and business, but we're having fun in the trenches. We already covered this one earlier. What are the odds looking like on will Elon win his case against OpenAI? Yes. Cal, she says 36 % chance. Let me tell you about Figma. Figma Make isn't your average vibe coding tool. It lives in Figma, so outputs look good, feel real, and stay connected to how teams actually build, create, code-backed prototypes and apps. Tyler was joking around earlier.

2:00:54He was saying, I'm going to move to Oakland. Oh, yeah? And become a resident for the chance of being a juror in the trial. I feel like there's a plot of a movie with that where, you know, you try, you know, if you really are a crazy corrupt lawyer, you try and get a plant on the jury. They're making an open AI movie. Maybe they should make a second movie about this, the story of Tyler sneaking his way on the journey. And the whole time, you're just like, look, guys, we got to settle this in a way that maximizes the total amount of compute in the United States and ultimately delivers AGI as fast as possible.

2:01:33And everyone's like, what are you talking about? Someone's guilty here. Skilled AI has occurred. Well, okay, so this market is low volume, but it's now up to 57 % chance. That Elon wins. Yes. Interesting. Yes. What's at stake? If Elon Musk wins, what does that do? Is that just a settlement? Is that just some monetary penalties? Does Elon have the chance of getting equity in OpenAI? Because at a certain point, if you convert it to a for-profit and you can just have a settlement where he gets$10 billion in stock, it doesn't really change the structure. He doesn't get a board seat. That's kind of a non-issue, even if he wins.

2:02:13But then there could be the nuclear option where he wins. and he gets control and gets a board seat or gets a majority stake. I mean, if you run back the clock and you look at how much he donated, if you weight ownership by nonprofit donations, which is a crazy thing to do, that's, I don't think, on the table, you would wind up with a very, very different control structure. But it is an exciting time, and I'm sure the court reporting will be fascinating. I wonder if there will be cameras in the courtroom and if we'll get more testimonies. Or just somebody sketching. Maybe. Yeah, we might just get the court.

2:02:51Have we figured out why they still do sketching? It is fascinating. Is it just for fun? I think it has aura. It has motion. I love how the person sketching can just have so much ability to make the person look cool or really, really bad. I mean, a photographer does too. Not as much leeway as just drawing it. They can catch you picking your nose, and then you look like a fool. Or they can catch you from a really great angle with perfect lighting, and you look like a hero. They can frame things up. I'm just saying, if Tyler was in a courtroom, he'd just be mogging. And it'd be hard to catch him whacking.

2:03:30Well, didn't they do that with SPF? That is all time. That's what I'm talking about. Yeah. They just gave him the GigaChad filter. Maybe that's because. Let's pull up this courtroom sketch. Yes. We have our next guest. Oh, we do. That's probably more important. Let me tell you about Gusto first. It's the unified platform for payroll, benefits, and HR built to evolve with modern, small, and medium-sized business. Cameras are just not allowed. That makes sense. But I'm saying, why don't we? Is that not a rule that we could at one point say, okay, we can have one camera in the course? I don't know.

2:04:01Maybe there's a lobbying group. Maybe there's a union. Maybe it's just the way people like it. Big sketch. Big pencil. Big colored pencil. Well, we have our next guest. Welcome to the stream, Andrew. How are you? How are you? Kettleman, how are you doing? Oh, you gave us a preview of the wrist check. What's on the wrist today? Oh, yeah. I figured this was going to happen. The liquid. That FP Jorn on today. Whoa. There we go. That is beautiful. Fantastic. And what's behind you? The T30. T30. Very nice. And what's behind you? The liquid nitrogen. You're doing serious business. Yeah, I got a few tanks of liquid nitrogen behind me, which we're actually like making a ton of ruckus a couple of minutes ago.

2:04:44and I'm glad they're not anymore. Yeah, they were like hissing and making all kinds of noise. So you're fully funded. You have a lab, you have a team, you're building, what are you building? What's the first product? And I'm sure I'd love to go into the far future, talk about data centers in space. And then I also want to go back and talk about your career at Robinhood since this is the first time on the show. Yeah, yeah. I feel like it's been a long time coming. So yeah, what are we doing here? So start out by telling you guys about AetherFlex. Yeah. The mission of the company is to build a power grid in space.

2:05:16And the core idea behind this, when I started a little less than two years ago now, was that collecting electrical power or collecting power in orbit was this like really special and important energy source that had not really been thought of as such. And the reason is because if you place things in the right orbits, they're going to be illuminated by the sun. The solar panels are nearly continuously. This is sun sink, right? We learned this term yesterday. Scott Nolan, sun sink. Or the more specific one is called Dawn Dusk Orbit, which is also sometimes called the Terminator Orbit. Oh, Terminator Orbit.

2:05:51I like that. We were talking about Terminator 2 earlier on the show. The Terminator Orbit. The Terminator Orbit sounds like the one I want to be in. So what is special about the Terminator Orbit? And then how does that lead into the whole strategy for the business? The thing that's cool about it is that it's always facing the sun or nearly always facing the sun. So if this is the Earth, right, and the camera is the sun, then it sort of follows this orbit, if that makes sense. So it's kind of passing over sunrise and passing over sunset always. Oh, okay, okay. Got it. Yeah, and the benefit of this is if you take a solar panel, which is the cheapest form of energy as we know it today on Earth, the problem is it's not terribly useful because it's very intermittent.

2:06:34It produces peak power or an hour to the middle of the day, if you take those same or very similar panels and you put them in orbit, they're going to be producing power 24-7 at nearly peak power. So you just get so much more energy production out of effectively the same panels. And also, low-Earth orbit is actually kind of an interesting place to build electronics applications. So I started the company out with the goal of making energy and space a commercial business. And a little bit more on that, the idea was that the reason space isn't a bigger part of people's lives day to day is because it's like for the large part of history, but mostly a government thing, right?

2:07:19Where it's in one shape or form funded by the government or there's government pricing that kind of drives what the activity there is. But if you can find new ways for people to actually make money doing it, then you'll attract more young people to want to pursue those careers. and you'll just make space a more commercially relevant part of day-to-day life. So if the applications of space today are like earth imaging, telecommunications, defense, and internet, what's that next thing? And my thesis was that energy would be the next thing. A little analogy to the past company that I started. The core idea when we started Robinhood was that if you thought of mobile as a platform, it was like back in 2012, 2013 era, then the applications that were built on top of it were like photo sharing, maps, email, you know, like they were mostly digital services like that.

2:08:21And the thesis that we had was that the next industry to sort of natively exist on this platform was going to be financial services. And that's Robinhood. And the idea here is that low-Earth orbit is a platform, and that energy is the next vertical of the economy that's going to natively exist there. Yeah. How much of the energy do you think will stay in space, just be routed around? Obviously, there's new narratives about what you can do in space with data centers and whatnot. But versus beam the energy collected on the solar panel down to Earth via a laser, something like that. I've seen there's Reflect Orbital that's just doing mirrors.

2:09:03There's a whole bunch of people that have said, hey, there's cheap and abundant solar up there 24-7 in these orbits. Let's get it down to Earth. How are you thinking about where the power should be used? Yeah. Well, I think our view on it is we want to have electrical power in orbit. And you can use that for two applications. You can kind of think of the two applications that we're building as almost like appliances that you plug into this newfangled energy grid. and the first one is actually getting, they're doing subcomponent assembly like right next to me, right over there. So behind this container is our clean room, which kind of goes on for a long ways back there where the first satellite that's going to space is hanging out right now.

2:09:48So that first thing is basically the appliance that plugs into the energy grid and then beams power down to places where there's no power grid. So the use for this is going to be places like like DOD applications, excuse me, excuse me, Department of War. Yeah, Department of War applications. DOW. Yeah, where there's no energy grid, where you can basically provide resilient, useful power, where there's a contested environment and stuff like that. Yeah. The other application is, is rather than beaming the power down to the ground and doing things like powering artificial intelligence data centers, the more natural thing to do is cut out all those energy losses and take the chips and put them directly in orbit.

2:10:31Because they don't actually, in the grand scheme of things, they're not the things that are the heaviest part of this. So that's the galactic brain. And that's actually where a lot of our efforts are right now. Yeah, that's exciting. I remember when we first talked about this years ago, thinking the natural reaction was going to be, oh, what does this finance guy know about space? The math is so important. And if you get one number wrong in the spreadsheet, Maybe the economics don't work and it's 10 years out instead of two years out. But of course, you have a math background. So I know you did the math.

2:11:05I know you crunched the numbers. But I'm interested to know how did you actually crunch the numbers to understand that now is the right time with the amount of funding you're going to bring in to the orbital launch costs, to the value of what you're doing up there, to the materials, how much a solar panel costs. There's all these different variables. They all have to synthesize into something that eventually makes profit within somewhat of a venture timeline, I imagine. What was the actual process? Were you talking to a team? Were you in an Excel sheet? Were you just asking an LLM to sort it out for you?

2:11:39How did you do the back of the envelope to know that now is the right time, that the economics can actually make sense on a reasonable timeline? Well, I did what any rationally minded person would do, is I made a big old spreadsheet. There we go. I'm here for spreadsheets. I love it. In all seriousness, though, I kind of make this joke with people sometimes. When I started doing this, for all intents and purposes, I was a finance bro. You could make the argument that was arguably perhaps the archetype of the finance bro in some ways. But in all seriousness, I was approaching this as a novice, as a student of the field.

2:12:19And the cool thing about this is that artificial intelligence has made the learning curve dramatically, dramatically, you know, easier to cross. Because the hardest thing about physics I remember, because I studied physics and math in college, was you sometimes get like a textbook and you're trying to like learn a new subject. You end up sometimes just getting stuck in the first chapter of it because you don't know the formalism. Like you're, you have to, in order to learn these things, you have to learn them in the language that they're written, obviously. but the difference is now you can just ask questions as you understand them and artificial intelligence will kind of meet you there.

2:13:00So it's been a huge accelerator in learning sort of the ins and outs of doing it but beyond that, look, we're a team of like 30 people. We've got optics experts. We've got folks from JPL. We've got folks from SpaceX. We've got folks from the, what's it called? Lawrence Livermore National Lab. so we've done pretty deep foundational work in understanding kind of how all the components work together, what the economics of them are asking questions like what is the raw cost of this, right? What is the cost to buy a component versus what's the cost to manufacture it from raw ingredients? And if you see the delta in any one of these things being too big, you kind of know you have to ask the question why?

2:13:50But to kind of answer the question that you started out with, the goal from the beginning was, how do we downcost this down? How do we make this as close to the sort of theoretical minimum cost? And if we do that, does this make sense? Does this make sense at bigger scale? And I think that there's a path to doing that. But the cool thing about it is the path starts with actually showing that you can do this. So, and by the way, you guys should come out when we do this. So we're going to have a demonstration this summer, ideally, because we have two satellites going up that are actually going to beam power down from space.

2:14:29So we're going to have something between a scientific conference and like a mini Burning Man in the desert where we'll show what this looks like in real life. Yeah, that's remarkable speed. I feel like the modern space company, Varda, did a little bit of this where it's like, if you're going to be taken seriously as a space company, you got to get up into space quickly, get up there, get the experience, do something, iterate, start the process instead of, you know, 10 years of, oh, yeah, we're going to go one day, but it's got to be the perfect system. uh you know varda went up and back and i think there were a whole bunch of chaos that happened in space but they learned a lot and it obviously improved the second third and then already the varda launches have become kind of old news oh yeah i guess they're up and back again that happened and it just becomes part of the business so uh i i love that you're moving so quickly there uh talk to me those guys were inspirational for me in that regard like i talked to i talked to will brew you a ton in the early days.

2:15:29I once called him my space Sherpa. And I think for that exact reason, I think they went from concept to real thing in space. It doesn't have to be perfect. It doesn't have to be the exact version of the thing that you're going to be doing forever and ever. But like you're much better off just demonstrating it in year one, year two, year three, rather than waiting 10 years because the odds that the thing doesn't work are still pretty high. you might as well get it out of the way early. Yeah, you got to blow some stuff up if you're working in the aerospace industry broadly. Yeah, you got to be comfortable having stuff not work and iterating from there.

2:16:09What kind of weird groups are coming out of the woodwork to try to get you guys to help them do data center stuff in space? Because there's a lot of different types of groups that seem like they have an incentive to just do something like space data center just for narrative purposes, right? I mean, this whole thing kind of like heated up because you had a handful of SpaceX investors really start to get excited about it. But I imagine you've got some kind of strange pitches because like you guys are working on this problem broadly and people are like, hey, do you have power? I want to put some chips up.

2:16:45Yeah, we're cooking some stuff. Stay tuned on that. We got some good stuff we're cooking up. I will say though, I think the credit for this goes to Bezos. who was looking just like very professorial in that talk that he gave last fall, where he was like, data centers in space might happen in the 2030s, and then everybody in the world was like, now we're actually going to do this the next five years. Yeah. I will say, though, that there is something kind of interesting to this point, which is that I think a lot of the people in the space industry see this data centers in space concept as one of the first novel ways to make money in space in kind of a long time.

2:17:24I would say in about a decade. And I think that that's an important point and like one to kind of pause on for a second because the reason space isn't a bigger part of our lives is because we haven't really figured out how to like justify people working on it in mass numbers, in my opinion. Yeah, and Starlink has been a big part of Starlink, I feel like, is the best example of like, hey, if you just build the capability of going up and down a lot, then you can build a lot of other capabilities, telecom being one of them. One thing I'm curious about is how has it been like building and running this company versus Robinhood?

2:18:06You've sort of like done the meme of like, make your money in finance and then work on the thing that like you really want to work on. Of course, I'm sure you really wanted to work on Robinhood as well. But is it like, you know, at the same time, like you have so much to prove, right? Like you need to, you want to do something at Robinhood scale or bigger, right? Which is a high bar. But then at the same time, there's maybe more, I would hope like some element of like, even just like calmness because you're like, it's not so existential as I feel like early companies are where it's like, I need to make something in this world.

2:18:50Yeah, a couple of thoughts on this. You kind of hit the nail on the head. This is something I think about pretty often. Because on the one hand, starting companies and doing things at early stages when they're not really figured out is kind of my happy place. So I do like that a lot. At the same time, it's kind of a brutal reminder of when you're in the early ages of starting something, it's actually pretty damn hard. because you have all the problems of like a small company, in my case, all over again, right? Where you've got like, you've got, you know, a function within your organization where there's like one person working, right?

2:19:26And that person for whatever reason leaves, right? And then you're like, well, I've got to rebuild that thing from scratch. And so there's like a lot of those kinds of growing pains where you're like, oh, there's nobody here that's doing that. I got to go out and hire a recruiter and then that person has to go out and recruit the person that's going to do the thing. So in many ways, it's really hard. I guess that's what I'm trying to say. It's a reminder that starting companies is actually not that easy. Yeah, you get better, but it doesn't actually get easier because if it's easier, it just means you need to be doing more and more and more and pushing yourself even further.

2:20:06I also did do something that I didn't have a ton of professional experience in, which was kind of the point, right? I wanted to do something that was net new because it was intellectually very interesting for me. But at the same time, and not to throw shade at anybody out there that's building an enterprise SaaS company, right? Like that's, this is like new company on absolute hard mode. Yeah, yeah, yeah. Yeah, I mean, like there just is the reality that in enterprise software, you can get a customer to pay you pretty quickly if you build something that works. and in hard tech and anything that requires government approval or real science and R &D, like the money is just going to be further down the road.

2:20:51And so that's going to require more risk. That's just a fact. When people talk about progress in space or really any human domain, there's this general sentiment that, okay, this thing, this thing that we want to do is really hard, but robots will be able to do it in a few years. And you're looking a little concerned, which is why I wanted to ask, what about the process so far? I'm sure you've been pitched by various robotics companies that want to help speed up processes internally or help with the work that you're doing. Is there anything that's really adding a lot of value today? and there's not a lot of robotics going on here no this is like what i figured like just fire up the welder it's not to say it's not to say that that's not going to happen by the way there is a welder i think either in the corner or it got moved back to my garage have you picked up any new skills can you use a blowtorch now are you are you are you learning this stuff or i mean imagine you have a team but every once in a while you know you want to take the you want to take the crane for a spin.

2:21:58Sometimes you got to do it yourself. Exactly. Yeah. I mean, I'm really good at opening boxes. Okay. That's good. I joke around with the team about this sometimes. I'm like, yeah, I left a software company to start a hardware company, but somehow I ended up with a job that's emails and spreadsheets. Yeah. Of course. Of course. No, but it is fun because I do get to, I mean, my background is in physics. So I get to think about the actual physics of it pretty often, but we have a team of people that's actually doing assembly and doing a lot of the hardware design. But I get to live vicariously through them.

2:22:35And I can't show you this right now, but literally there is assembly happening on and off. You'll see people walking behind me because they're carrying boxes and stuff in there for the first satellite that goes up. Yeah. When we watch tours of Abilene, Texas and see data centers, There are a lot of people buzzing around. It feels like, you know, even if you just get an off-the-shelf, like, NVL72 NVIDIA rack, there's people that are maybe doing wiring or dealing with cooling issues or a variety of things, a chip that's, you know, died and you need to replace it. That all feels really hard to do in space.

2:23:13I heard one pitch that was sort of making the case that wafer scale compute could be more popular in space because you want the entire model on one chip that's just in one satellite. You don't want to put a NVL-72 up into space or whole rack. You're just going to have more problems in space. And so it's going to live on one sort of complete system. We asked the CEO of Cerebrus if this was in his forecast. And he said, no, I'm not really interested in the next three to five years. And then he went and did a$10 billion deal with OpenAI. So he kind of denied that that was on the roadmap. but what do you think the shape of like the compute workload or the compute hardware that you will eventually put into space or, or any company would put into space?

2:23:57What, what will be the norms? What will be the, you know, the, the, the full, like we, we have, uh, we very much coalesced around like one rack and AMD and Nvidia, they all work within rack scale architectures. What will be the equivalent on a satellite bus that is compatible with a Starship or a SpaceX rocket, what will the shape of the compute stack be in space? Yeah. The way that I think about it is in terms of the workloads and the energy draw or said workloads. And there's versions of this in the future that are more ambitious, right, where you say, we'll get to much, much bigger scale integrated power levels per satellite.

2:24:39But I think from my perspective, the right starting point is to say, what can you do right now in terms of satellite size, power envelope per satellite, and can you solve meaningful problems with that, right? Because if you're trying to do something brand new like this, you want to have a conscious approach to how you take on risk, technical risk more specifically. And if your idea is predicated on making something that's like the size of 10 football fields on day one, you might want to see if there's a way to do it much smaller at satellite sizes that are commonplace. So the thought process is that we'll start out in the next 12 to 18 months or so with demonstration missions culminating in a first-of-its-kind satellite from us that has, let's say, the 10 to 20 kilowatt electrical power drop that powers a set of 8 to 10 interconnected GPUs that can host a model and do inference on it.

2:25:46So inference, I think, is the primary initial use case for this. There's the idea of, like, could you do training with this or could you do stuff that requires massive interconnectivity? And it's possible, I think. I think the orbital mechanics of it might be challenging if you're trying to connect 100 ,000 GPUs together. but you don't want to start there because it's kind of interesting because we kind of ran into the same problem when we were working on power beaming from space where the old versions of that concept called for one gigantic satellite that would beam power down with microwaves and a byproduct of the frequency of light and the aperture sizes that you would need.

2:26:33That thing called for like satellites that are like one to five kilometers in diameter, right? Like that's too much. Like if your demo mission is a kilometer wide satellite in space, let me give you a little hint. You're probably not gonna get funded for that. Like there's probably not funding out there to like spend a billion dollars on something that may or may not work for the first time. So that's kind of how I think about it here too. Are you feeling any pushback yet or pushback coming on the horizon and around like space junk, blotting out the sun. It's annoying when I'm stargazing. Is that something we need to worry about?

2:27:12Is that something you should be worried about? Even if the worries are unfounded, you need to get out ahead of from messaging perspective, like how a lot of the AI data centers probably should have been talking about recycling water a year ago. So that never became an issue. It did and it got debunked, but it still was something that took over the news cycle for a few days at least. No, I think that's a good point. So on space junk in general, I think a couple of things. You have to look at space and realize that it's going to get commercialized. And that might happen this year. It might happen five years from now.

2:27:48It might happen 10 years from now. But I think that there is an inevitability that as humans exhaust or really push the boundaries on like the resources on planet Earth, that we're going to look to space to do more stuff there. So I think my starting point is the civilization that we want to build has a massive footprint in orbit. And so the question is, is how do you design these satellites and how do you design these mega constellations that are, you know, I hope in the next 10 years putting gigawatts or more power up so that actually works and it doesn't create massive problems with things like the Kessler syndrome.

2:28:26And we think about that from the beginning. We think about the sort of component choices that we have to make sure that as we're deorbiting them that they're made with components that can demise on reentry. And there's a lot of work that's actually going into this already to make sure that, you know, there's this, what's it called, where you can basically disambiguate the space that your satellites are in. That being said, there is space junk out there. There's actually a lot of effort to sort of identify where it is, to track it and to do orbital maneuvers to make sure that you can protect your spacecraft against it but again like space is really big yeah sorry uh any plans to go to space yourself blue origin it's it's got to be a little bit weird like working on something it's like working on like a restaurant but you never actually go there you know like yeah jord isaacman's been he's right i feel i feel very confident with him at the helm at nasa given that he's been to the thing that he's in charge of we got to get you up there what's the plan he's a good dude he seems very fit too he seems like he's like um seems like he's he's he was able to survive that and had to build for it um i gotta tell you guys maybe he is fit because because he went to space yeah oh you think the space did the fitness for him yeah in that case count me in it's a miracle that's funny um one last question oh yeah in all seriousness on this one um can you imagine how upset your stomach would be in space like can you imagine the pressure changes and like oh you're gonna that's my reaction you're gonna you're gonna let okay last question the the chevy corvette zr1x just posted a zero to 60 time of 1.68 seconds it goes uh quarter mile time of 8.675 seconds can we call it a supercar is america back in the supercar game what's your take on the zero x i think so okay but i'll say this um i don't think people give a shit about that okay i think what people care about is is the car fun to drive and what they should do is they should make that thing with a manual oh i think that's you guys know i love cars yeah yeah of course that's a good take yeah sort of a uh a chevy corvettest would be the move, something like that.

2:30:53Yeah. Yeah. That Corvette's really cool. It's based on a, or they, it was like inspired by the four or five, eight Ferrari. Yeah. I mean, there's a lot to like about it. They set the, the, the speed record for American cars. I really like how they have the, the Chevy executives driving it on the track, setting the records, the employees themselves. They don't just hire some, I think that's really cool. I mean, that's equivalent of like sending an Aetherflux employee to space. You know, it's like, it's a little bit of a gimmick, but it just feels like, oh, they really take it seriously. Yeah, I don't think it's a gimmick at all.

2:31:23It's way cooler to be like, okay, one of their employees. Yeah, what were you saying? If they want to do a gimmick, you know what would be a really good one? Put a manual in that. There we go, back to the manual. Yes, 100%. I completely agree. Do you see any predictions around, yeah, just cars, supercars with manual transmissions? It feels like there's certainly a lot of enthusiast interest, but it feels like potentially one of those things where the TAM is actually just like you take like people that will buy a supercar and then like people that will buy a manual supercar because you get into this you start running this calculation of like okay uh i'm getting this car and then and then i'm on a hill in san francisco maybe i'm sam altman i'm i pull up on a hill and maybe i drive stick but i'm in a four million dollar car and it's starting to slide back

2:32:17Sam Holman stalls F1 on Gary Street or something the thing that I am bullish on is more analog controls in cars anybody that's owned a Ferrari loves how the buttons get all sticky I may or may not have been dealing with some sticky buttons this morning the sticky button thing is truly so funny they're like Like, yeah, like we made these and all after a few years, everything is going to feel like a kid was eating candy in the car. And if you want to replace it, it's going to be. Yeah, you go to turn the volume knob and like the volume knob, like comes off on your hand and you're like, this is disgusting.

2:33:01I got to tell you guys, like the fact that there are not too many manuals that are being made today. And I love cars. It just takes people like me out of it. Right. I'm just like, if you make a manual, I'm in. But otherwise, there's plenty of cool cars out there. Well, one exciting thing that I've been thinking about is I feel like in 15 years from now, it is not unreasonable to imagine that it is illegal to drive a car yourself on the roads. Like at some point, autonomous vehicles will become so safe. and driving a car yourself is so dangerous as it you know one of the the number one causes you know motor vehicle accidents being the number one causes of death that car manufacturers will start making insane enthusiast cars that are effectively track only because you have some percentage of the population that's just like they're not buying a 911 to like use as a daily they're like i love cars So I'm going to get this like, you know, something like they're making like a W1 track only edition.

2:34:10Right. And I could see like track only cars having like a real renaissance. That kind of stuff is happening already. But I will tell you this. While such a thing might happen in California, Florida man will never stand for that. I agree. Like Texas will never stand for that. There are parts of this country that simply will not. From their cold dead hands. For sure. Yeah. For sure. But like Florida man's not down with that. I don't know. I just feel like I see it. But at the same time, eventually, you know, these. The safety data is going to be overwhelming for perfect self-driving cars in a decade.

2:34:50Like they're just never going to get an accident. It will literally be like moms against self-driving. But they're talking about like driving yourself. Yeah, yeah, yeah. But it's so dystopian though. Yeah. Right? It's so freaking dystopian. I'm not saying I'm excited about it. Because cars also, I mean, I'm going to wax poetic here for a second on cars. Cars are really important. They represent freedom, in my opinion, right? They represent the idea of being able to have the freedom to explore the world, the geography of the landscape, of everything around us, right? It's the feeling of the wind in your hair, the way the steering wheel feels in your hands when you go over a patch of road, right?

2:35:31like the way all of these things feel, like it's not just about transportation. It's not just about the feeling of power. For me, I think it is like the embodiment of freedom. And the idea of like going through a sort of like technological course and a course of human history where we make that freedom illegal, man, I find that to be super dystopian. Yeah, let's hope it doesn't happen. You can still ride a horse around if you want on many streets. Yeah, we might see a resurgence in horseback riding. Stay with the horse. People are like, I want to live within horseback distance of my office. Give that horse a manual.

2:36:13But I don't think this is going to happen. I don't think this is going to happen in the next 10 years. Okay. Okay. We'll have you back on and we'll see what happens. We'll check in in a decade. We'll check in next month when there's some amazing news. You'll like California. There's some, uh, there's some like mileage tax that they're working on where you'll be allocated like a certain number of miles and pass that point. Pass that point. It's like, you're going to pay your odometer. Leave it to California to come up with all these brilliant ideas. Who knows? Well, thank you so much for taking the time to come.

2:36:47Hang out with us. Come back on. As you have these, uh, launches announcements. Yeah. Always welcome. Yeah. We'll talk to you soon. We'll have you guys out too. Have a good rest of your day. Goodbye. Eleven Labs. Build intelligent, real-time conversational agents. Reimagine human technology interaction with Eleven Labs. We have a small piece of news, and then we'll bring in our next guest. CBS is preparing a new segment called Whiskey Fridays with Tony Dacopu. I can't pronounce the last name. Some staff were only first made aware of it because they encountered CBS testing set designs of a faux stocked bar brought to you by jack daniels uh so they saw cheeky pint and they were like we got to have a cheeky pint at cbs okay cheeky pint great concept they got the faux bar whiskey fridays hanging out it's casual but they're not drinking so what if you did cheeky pint oh with whiskey wait is this a this cannot be a real photo this is it has to be fake update here's what it may look like.

2:37:52Okay, because that makes no sense to just put a news desk in that type of set. Anyway, we'll have to tune in. We'll have to give it our review. But until Whiskey Friday comes out, we'll review the second post. We've got to do it, because we were just talking about cars. This post here from CDTV, The Goose says, remember when Quavo flexed driving 60 miles per hour? I go a mile a minute. Skirt. Yeah, that's actually quite slow on most freeways. The speed limit's 65, and most people are going 85. 60 might be time to speed it up, Quavo. Let's step it up. Let's step on the gas. Let's get you into a ZR1X, hopefully.

2:38:37Anyway, let's bring in our next guest, Gabe from Noise. He's the CEO, and we've got some fantastic news. Thank you for waiting. How are you doing? Welcome to the show. Good to see you, dude. Good to see you. Absolute pleasure, gentlemen. It's great to be here. First time on the show, kick us off with an introduction on yourself and Noise. And get the gong ready. I will warm up the gong. Yeah. So my name is Gabe. I'm one of the co-founders of Noise. We started this project maybe at the start of 2025 when we started building it. But the idea is to build this trading platform for relevance, right?

2:39:09So can we give users the ability to long or short trends, brands, ideas, and have these markets kind of be rooted to social data in a way. So when you combine the trading activity and you combine the kinds of things you're seeing on Twitter or TikTok or Reddit, you kind of end up with this objective measure of cultural relevance for different types of brands or trends. And for many reasons, we think that's extremely important for where the world is at today. But maybe we can jump into those that you guys think. Yeah, maybe an example. So when we started covering Labubu, we were like, this must mean it's about to be over.

2:39:47And I think it was. I think a lot of the stock at least has sold off a lot. In that situation, there's a public company associated with the product. So that is one element. You can track Google Trends, but yeah, walk me through more of how you pull out signal from all the noise around a brand. There's so many different data points that you can use to track. Yeah, so like Google Trends is a good example, but Google Trends in an era where people are spending more time in apps, more time in all these things, like social relevancy, how much something's being talked about, there's less signal there.

2:40:23Yeah, yeah. What's the process like? Yeah, so Google Trends is, you know, if you were to look at a market on noise, let's say it's quad code or it was Libubu, you would see something structurally similar to Google Trends. The problem with that is, you know, like 20 % of the information that is going into Google Trends is what actually matters in today's world, right? Which is what he was saying. You're seeing things on TikTok, on Instagram, on Twitter. So the Lububu one's interesting. We basically, we partnered with Kaido first. So they were doing this already for the crypto industry specifically.

2:40:56so they're building social graphs on twitter and they were able to tell you how much relevance does polymarket have versus call sheet in that industry how much relevance does um you know ai compared to defy have and then you can see kind of how that relevance changes over time what we care most about is kind of the market component of it which is you know it's great like you can get a snapshot of where attention is today you know can you really trust it that's that's something to discuss as well. But we want to use markets to have kind of this forward-looking nature and help basically all these different types of industries make decisions based off of trends and resource allocation, right?

2:41:34So where are trends going to go in the future? Where is relevance going to shift? We're talking about movie theaters. We're talking about, you know, is Claude, is Anthropic still going to be as important as it is today in the AI industry by the end of 2026? These are all like extremely interesting questions, right? So, yeah. If you're pulling, if you're using Signal from social and these markets get to some amount of scale, there will become an incentive for market participants to try to game it, right? I can think of somebody spinning up a bot farm and talking about Labubu millions and millions of times in an organic way.

2:42:15how have you guys thought about like parsing through all the noise and trying to find like real signal yeah for sure um it's a good question and it's often like one of the number one questions we get but you know we pull from a variety of sources and i'd also add that um the relevance on noise is very much market-based so it satisfies the criteria of any other market which is you know if the relevance of let's say clawed or movie theaters or peptides is going to go up, it's because people are inherently putting capital behind those positions, right? They're longing it. And those are good questions.

2:42:53But I think the best question that we could try to answer right now is, is social media data alone today a good measure of relevance, a good measure of understanding the shared interests of the world? Our bet is that markets are probably the best coordination mechanism for these kinds of questions. and we just need markets to be a little better about reading where genuine interests are and where those trends are uh than social media data alone so is it a solved problem no we'll continue to basically build in that direction but uh it's definitely interesting what uh what are some rising trends that you guys are thinking about building markets around for this year specifically yeah uh i really indie rock music i think people aren't really noticing this but it's it's been on the rise for the last 18 months my prediction is this is coinciding with uh this return to empty in-person meta indie rock music is probably the best kind of music for that um so that's one we'll see claude code i think one of the interesting things is with noise is you kind of get to see where echo chambers are breaking right because Claude code was kind of this thing in Silicon Valley that took over the entire timeline and everybody assumed you know nobody's using Chat2PT at this point but these open AI products are basically in the dust but the things that we were seeing is the engagement and the kinds of interest that people have in Chat2PT is still higher than Claude no matter what so when you actually compare these things side by side at a very global level it's interesting to see where these trends are actually at.

2:44:33How do you think about relevancy to particular sub-communities or niches? Because we were reading this piece in Vulture about David Allison, and we were joking that they were saying that Oracle is this sleepy, no-name company that makes a lot of money, but it's not known or it's not popular. And in us, it was like in the headlines all last year. We were following it very closely. We know everything about the company. And so, but at the same time, I understand where the Vulture writer is talking about in the sense that like Oracle is not a company that... They're like Oracle the sailboat company?

2:45:08Yeah, like if you stop someone on the street, they might not be talking about Oracle, where they might know about Libubu. So how do you think about someone's stock rising relative just to their niche? Yeah, one of the most interesting things that we found when we launched our beta and summer of last year was that people were kind of, you know, we kept seeing maybe 40 % of the users that we had invited onto the platform using it every single day, every single day, consistent time patterns. And we were like, you know, why the hell are you using this so much? Right. So we're trying to talk to these people.

2:45:44And part of it was they were, they wanted these niche trends to be part of their daily news cycle, right? They wanted to see how the relevance was changing based off things that they really cared about. And mind you, we're talking about some projects in the crypto industry that people don't really know. Your audience probably doesn't even know exist. And it was about kind of engaging with other people who are also interested in those things or have differing opinions about those things. And basically creating communities around these niche trends. So, yes, it's interesting. More interests are becoming niche over time.

2:46:20But still, the echo chambers and seeing where you're actually at is something crazy. We love it. We love Echo Chambers. We celebrate Echo Chambers. Give me the lead. Who led the round? How much did you raise? Paradigm, 7.1 million. Great hit. With authority. With authority. Congratulations. Thank you so much for coming on the show. Great to finally have you on the show, Gabe. Have a great rest of your day. Excited to track noise on noise in the near future. we will talk to you soon have a great rest of your day Gemini 3 Pro Google's most intelligent model yet state of the art reasoning, next level vibe coding and deep multimodal understanding we have our next guest already in the Restream waiting room we have Alfred back on the show for the second time from Listen Labs he's the co-founder and CEO welcome to the show again thank you so much, we're excited to be here welcome back I love the energy we're fired up we're fired up and we got a special real quick reintroduce yourself to people that missed the first appearance and then we'll get into the special deliveries that you got for us amazing yes so I'm Alfred I'm the CEO of Listen Labs what we do at Listen is an AI that speaks to thousands of people and tells you what they want and why and this helps you develop better products or fix your marketing campaigns and what you have in the room with you is some incredible products that have been developed using Listen.

2:47:55Yes, yes. So take us through these. What is this? How are you involved? What happened with this crazy, remarkable product? Did you want John to wear these? Who are you talking to? You didn't talk to me about this, but somebody's going to be happy. Are you going to put them on? Yeah. Yes. Who's happy with that? You know, products are all about finding the right audience. And so these are shorts from Chubbies, which has been one of our early customers. And they used Listen to test their prints, test their fitting. And actually one of them, they developed an entire new product line for kids, kids' shorts, where they used Listen to interview hundreds of kids.

2:48:36And they figured out the liner in the shorts that they had previously were very uncomfortable. Maybe something for you to wear. Yeah. Very, very fascinating. Maybe 10 years ago. What are some of the challenges with consumer brands when it comes to really understanding your customers? Because I think with a SaaS company, an early-stage SaaS company, you can just talk to all your customers. Yeah, but you can actually get them. Many, because there's a few. And there's probably a power law where there's one customer who's providing 80 % of your revenue in the first couple of years. Whereas with consumer brands, you could go talk to 20 customers, and then you could get a terrible signal because you just happened to pick incorrectly.

2:49:19Yeah, totally. Yeah, I mean, in the olden days, you were stuck in this paradigm between you could use a survey, but then you would get a ton of people not really paying attention and they would answer your questions, but you wouldn't get rich, deep insights. You could then have one-on-one interviews, but as you said, if you have thousands of customers, it's really hard to do them at scale. And so with Listen, you get the best of both worlds where you have an AI talk to all of your customers and then you can figure out the nuances like which segments work, which don't. And people are much more honest talking to an AI than talking to a person.

2:49:57It's something that we found out. So their feedback is more driven to how they actually will behave in the future. I have a company that I'm going to email and tell them to use Listen Labs. I don't know them at all. It's this company, they make sound machines for kids. And I've had multiple of them. I don't know why I've had multiple of them because both of them, when you're trying to adjust, if you just touch the thing in the wrong way, it triggers a motion thing, and it just starts playing music. It literally, I went, I was messaging. Quietly trying to adjust. I felt bad. I was messaging the support a week ago saying, like, please let the CEO know that your product makes bedtime worse, a disaster.

2:50:44And I will buy another one. But you need to fix it. I'll give you the third chance. And that just seems like an incident where they shipped a product and then it sells well because it's in demand, but they're not actually paying attention to how people experience it. Do you feel like you're able to pull out that emotional response from like you, you interview a hundred people, they're all like, eh, it's okay. And then there's that one person that clearly found a fatal flaw with the product is animated about it. They're aggressive. They're emotional. Maybe they're screaming at the AI. Maybe they're being rude.

2:51:18You censor the expletives, but you know that there's emotion there. Good, good on you. Uh, and, uh, and you're able to give that more weight. Like, How do you think about balancing out all the different emotions that come through in responses? Because sometimes they can get collapsed in AI models. Totally. Yeah, I mean, the CEO of Sweetgreen, I think, was on TBPN talking about this, how surveys miss the outliers. Yeah, that's the protein ball that they developed using lesson. But the other methods of collecting feedback miss the crucial pieces of feedback that really matter. and it's often about one or two or a couple of people that are part of your segment that will drive most of your revenue and they will be much more animated and care much more.

2:52:07And so the AI is able to detect the way they show up emotionally as well as the nuances of how they answer the question. We have something called response quality. So the AI can rank how well this person answers the question and how much they know about their product. So it's able to kind of surface those outliers to you and then give you actionable recommendations. What is the actual workflow like in general in order to actually start talking with customers through Listen? I imagine it's also different with like a kid's product line with Chubby's. Like how do you actually facilitate getting a parent to allow their kid to talk to a robot?

2:52:51I'm very curious. Yeah, of course you need the parental consent. But it's essentially four steps. So you start by asking your question, which could be, you know, SaaS tools also use it like Replit. So they could ask, what should we change in our product if you compare us to Lovable? And then Lyssen will go and recruit that audience for you. So we'll find Replit users as well as Lovable users. And we have 30 million people in our database. so we can find pretty nuanced recruits. And then Lyssen will go and have video calls with them where they could potentially also share their screen. So you can have thousands of those video calls, and then the AI will kind of analyze those and give you recommendations of what to do.

2:53:36Makes sense. Take me back. We didn't really talk about your earlier career, the first time you were on the show. BeFake image app, what was the story there? How did you build it? How did you scale it? What was the value prop? And how did you wind up selling it too? Yeah, for sure. So my co-founder and I, we built this AI consumer app, which was essentially kind of gibbify. You could gibbify yourself. You could do this kind of image prompts. Image transfer almost. Yeah. And the real use case was creating Tinder profile pictures, of course. And so it like blew up on Reddit. We were on the front page of Reddit and we had 20 ,000 downloads in one day.

2:54:21And that was the inspiration behind Listen because we had these many users and we were experimenting with LLMs at the time and thought it would be interesting to let an AI talk to our users. And it turned out to be really useful. So we were like, why are we making this shitty app? Let's just sell this thing. And then we actually ended up in this limbo state where it did work for us, But it still took us 18 months of making the product perfect before it became useful. And that's something I've learned, that being early looks the same as being wrong. And a lot of people we talked to told us they've already built the product that we were building, and it didn't work.

2:54:58And so we had so much rejection. But now, since we launched nine months ago, and when I was last on TBPN, we've run our revenue 15x. We interviewed a million people. Let's go. Raised 100 million in total now. So let's go. There we go. Congratulations. I'm still trying to get used to the energy of, you know, I'm Swedish. I'm trying to amp it up. No, I mean, this is night and day from the last time. You're fired up. I love it. What's next? Just scaling on the customer side, hiring. The whole Fortune 500. You got to get them all. That's right. That's right. You've got to select them all. Sweetgreen, Google, Nestle, Skims, Microsoft.

2:55:42There's a couple of names that are - We have a list. We're going after 500 companies, one at a time. But we have, like hiring is the number one priority. And we're trying to do these tricks to stand out. We did the billboard. I don't know if you saw that, where we had - One thing we've learned is that there's so many companies growing really fast, raising a ton of money. And so you just got to communicate your culture and get people excited to join. And so we made this billboard that had just black text on white that led to a puzzle, an engineering puzzle. And that made it much more attractive for folks to join the team because they realized we're a team of competitive programmers and puzzle hunters overall.

2:56:25Puzzle hunters. Yeah, Mike Isaac in the New York Times wrote a whole piece about how obscure billboards are in San Francisco. It's like a few people talking to each other, very in-group. but I love when people do fun stuff with billboards. Got to stand up. It's always great. Got to stand up. Thank you so much for taking the time. So great to have you back on. I'm sure you will be back on this year with the amount of momentum you guys have and looking forward to it. And, yeah, thanks for this protein, Max Bowl. We'll split it. We can each still get, like, 50 grams of protein. That's plenty. That's plenty.

2:56:59Plenty. Thank you. We'll talk to you soon. Great to see you. Have a good rest of your day. Goodbye. Cheers. Railway. railway simplifies software deployment web apps servers and databases run in one place with scaling monitoring and security built in uh tj parker asked a question he said why are there not a hundred times more weird consumer ai things going on nikita beer answered he said hold that thought until dom hoffman launches and dom quote tweeted and just said sup of course uh dom is the creator of Vine. And his company is called Sup. Oh, I didn't know that. But it was a good response. Dom created Vine.

2:57:43Yeah. And a number of other things. A fantastic creator on the internet has created a bunch of fun moments on the internet. I've been following his career for a long time. I enjoyed Vine. Were you ever on Vine? I was on it, but I never fell in love with it. I created the schizo added on Vine. No, not really. But I was actually experimenting with the Vine format very seriously. And I would do these like hyperlapses and infinite loops. You could do this fun format where you would take a little, like basically an image as you went around your house. And if you ended at the start, it would be an endless loop.

2:58:21And Vine was very good about endlessly looping videos. So you could do all these cool, fun things with it. I had a lot of fun with that format. It was great. Did you see that Havana syndrome is real? Apparently. This is crazy. Noah Smith says Havana syndrome was real and we got the device that causes it. Inspiracy theory that turned out to be true. Apparently, a number of months ago, the U.S. captured a weapon that has been associated with Havana syndrome. Yeah. Havana syndrome. It's a disputed medical condition starting in 2016 in about a dozen overseas locations. U.S. and Canadian government officials and their families reported symptoms associated with a perceived localized loud sound.

2:59:01The symptoms lasted for months, including disabling cognitive problems, balance problems, dizziness, insomnia, and headaches. Havana syndrome is not officially recognized as a disease by the medical community. But apparently there is reporting in CNN now that there's a device. Well, there have been a bunch of real investigations in this because it's happening all over the world. People are like, I feel terrible. Why do I feel terrible? I'm hearing things. I have a headache. I can't balance. And so this was just a great mystery. CNN now reports the device linked to it was purchased by Homeland Security in the waning days of the Biden administration.

2:59:40And DOD has spent a year testing it. It has Russian components and fits into just a backpack, which is scary. Will a defense tech startup build one and go through Y Combinator? Who knows? Stay tuned. We might see someone commercialize this technology any day now. Tyler, do you have something? I want to see someone, if they're building it, I want them to use it on themselves, like Jake Adler at Pilgrim. Oh, yeah. He cuts his leg open at the pitch meetings. Prove that it works. He shoots the VC with the... Just kidding. I'm so dizzy, I've got to write you a term sheet. Owls says that jobs exist because important people want to have large pyramids of people reporting to them, even if the pyramids do nothing useful.

3:00:25AI will not change this. Oh, interesting. Fake jobs. In other jobs news, there's an opinion here. Our brightest minds are disappearing into finance, management, consulting, and corporate law. Is this one of the oldest days? He says, no, we are using our best minds to allocate resources optimally. How terrible. Of course, we love finance, management, consulting, and corporate law. These are all valuable pieces of the economy. They are important. New research. Salesforce, who we have the CEO of joining us right now. Mark Benioff is in the Restream waiting room, and we will bring him in to the CBP on UltraDome.

3:01:04Mark, how are you doing? He's back. Great to see you guys. Back for round two. I prepared for this interview. I prepared for this interview. Wasn't that bad? No, we loved it. We loved it. I will say it was one of the most fun conversations we've had all last year. It was amazing. Wow. Really? No, yeah. Well, then you guys are just not having enough fun in your life. I'm your biggest fun. Here it comes. Here it comes. Don't sell yourself short. The bogging starts. But I prepared for this interview by going to Slackbot, now powered by Anthropic. and I asked, what is the single best Metallica song?

3:01:42It said Master of Puppets. Agree or disagree? Okay, guys, I know that you are, this isn't at the tip of your tongue, so I was ready. And I was going to flip it for your partner so he can get all the words because we had some issues last time. We did have some issues. I'm going to go through the details. Are you going to, are you about to say, are you going to give us a... Metallica, name your tune. Yes. It was not a great moment for your show. It was a rough moment. It was a rough moment. It was rough. Are you going to sign that and send it to us so we can hang it in the rafters? No, I'm not. It's mine.

3:02:16You can get your own. Metallica.com. We run it, actually, on Salesforce. Oh, you do? We run Metallica.com. No way. You can get to your Metallica album on there and all the other good stuff that you need to kind of... Okay, so I'm going to give my recap of John using Slackbot this morning. because he, this was an ASI moment. Okay. He was like, John, we do the show all day long. There's a lot going on in Slack. And so catching up is, is tough. Yeah. This absolutely nailed the summary. Immediately found something that was like very actually interesting and important. And we're like, well, we missed that.

3:02:57I probably truthfully had about like 200 notifications across 15 different channels as they build up. And it basically put together a deep research report of every little item, picked it out, when am I mentioned, and it was exactly what I wanted to catch up. And then I went to it and I asked, okay, clear all the notifications. It couldn't do it, but it gave me the shortcut shift escape, and it did it in two seconds. So I was very satisfied. How has the response been from your customers? Well, first of all, well, there's the response of my customer right there. I would say I'm so excited that we could finally get this out of the shop and into people's hands.

3:03:33I've been using it myself for months. It's completely changed my life. I think it will change your life. I think it will change your show. I think it's going to make everything better. Because, you know, we've talked a long time about prompt engineering and AI. You know, when you're writing your prompt, you're kind of coming up with your question. Now you really see the power of context engineering. That is, it's looking at all the data about your show and everything you guys have ever said. It can even read your DMs. And then it's coming in and it's saying, hey, right here, this is what you need to ask Mark Benioff, this question.

3:04:05This is the zinger for him. And that's the prompt you could give it, which is, hey, based on everything we know about Mark, everything that's ever happened in the show, everything that's happening inside Slack now, plus everything that Anthropic knows, everything that everyone knows, what's the zinger for Mark? and then it's going to come up with it. Yeah. Talk to me about the tone and the speed. I noticed that it responds like an LLM. It's giving me a few paragraphs. It's not acting like a person who might respond with on it and then, okay, here's what I found. And then it's like six different messages.

3:04:43Is that intentional? Do you like the tune or do you see it evolving to your style over time? Where does all this go in terms of the flavor of Slackbot's personality powered by and product? This is really different than our last interview. We're going deep in product. Let's do it. We love it. Number one, every company is going to have a customer agent and an employee agent. And that's why we've been working on AgentForce, to build the orchestration layer, the observability layer, so that companies can get out there with their agents. And that's what I've been so excited about. Then all the apps, like Slack is only one app we make.

3:05:21We have a whole bunch of apps like Tableau, our sales cloud, our service cloud, marketing, commerce, and on and on and on. We have a whole family of apps that help you run your business. And then a huge set of data capabilities, including Informatica, MuleSoft, our data 360 layer that let you federate and integrate your data, which means that it's going to find all the data in your company and bring it together. Now, what you're not seeing yet in Slackbot, but what I have is that you just hit a button and it connects to all your Google sources also. And it connects to all your Salesforce sources.

3:05:57You're keeping the good stuff for yourself. This is a fast takeoff scenario. And you're like, we're keeping this for ourselves. We can't let it out. Well, I was at a conference for the last two days, really cool conference in Honolulu, Hawaii. I live in Hawaii, so I popped over to Honolulu, and it was called the Honolulu Defense Forum, and there's so many of my customers there. It was really cool, but I didn't have time to get briefed on all these customers, so I'm doing the briefing myself. I just say to Slackbot, in real time, I'm meeting with this customer. Tell me everything about the customer, and it's not just getting the experience from the public web, but it's also then looking at all my proprietary databases and saying, boom, here you go, Mark, you're ready.

3:06:43And the prioritization, I think you've got to have a great experience. I want to hear your exact feedback after you kind of get into it for a few days. Because it's changed how I work. It's awesome. Yeah, I had a great experience. I didn't feel like I needed a smarter model. It did seem like I was on Claude 3, not 4.5, which I've used and is great for some things. Do you want to know why you feel that way? Yeah, why? because you're getting the data from Slack. You have so much data about your business or your show, everyone. That's what you've been missing in all of your prompts. When you go on Anthropic or you go on Gemini or OpenAI or whatever you use, it doesn't know anything about you.

3:07:27It doesn't have your context. It doesn't have your data. And that's what I mean, context engineering. It knows you now. It really does. It knows so much about you, your company, everything you guys have done, what you've built, this great show you have. And now it's able to really tell you. And you should ask it like, hey, what are the three things we should do to increase viewership? What should we be doing to increase our revenue, make us more profitable? What is my prioritization for the week? Try it like because it knows so much about your company now because of the huge investment you've made in Slack, you're going to get a great outcome.

3:08:03I'm really confident. And, you know, I don't know if you've seen some of the videos that Mr. Beast has been doing, but he is really in it to win it on this. And he runs, you know, his whole company. People talk about Mr. Beast like he's this consumer guy, but this guy loves enterprise software. He is the entrepreneur's entrepreneur. Yeah, it's true. Oh, he's amazing. The first time I ever talked to him was a couple years ago. He goes, I am going to be like Steve Jobs. I am going to be like Larry Ellison. I'm like, what? He's like, yeah, I want to be one of the great entrepreneurs of all time. So he really has a mental model that's super unusual.

3:08:46Well, he runs his whole companies on Slack. So now he has the ability to kind of get to a new level by bringing AI in. Because when you take AI, data, and the app together, as you see, you get a tremendous outcome. AI by itself, the models, great, fantastic. But they're commodities at this point. We all know that. The data sets, those are commodities on the consumer side, but not your data set. The only one who has your data set is you. And then three of the app, you've chosen Slack. Thank you for that. And so now you put one, two, and three together, and bam, you are ready to roll. You can now sleep with one eye open.

3:09:26There we go. Give us a Super Bowl preview. you what what's what's been exciting in the past that you've done what do you want to do in the future with the what's your most memorable Super Bowl yeah let's start there I think there is going to be some pretty cool stuff coming from us uh for the Super Bowl I'm not going to give you all the secrets okay I may have already given you some of the secrets okay but I don't think I can give you any more of the secrets because it is going to be astonishing what we're doing and everything that we just talked about somehow serendipitously maybe you guys already know yeah you know but Everything that we just talked about, take all of that and make it into a Super Bowl ad over the last 10 minutes.

3:10:03And I think you're going to have something astonishing. Yeah. Well, what do you think about model pickers selecting the model that I want to use? How much of that should live with the end user? I go to Slack bot and say, hey, I want to use Opus or I want to use 4.5 versus your team deciding what makes sense for that overall infrastructure that you've built. And then it's below the surface and I don't really care as a user. Well, you're already using a whole synthesis of models. Okay. Because Salesforce, we make a lot of our own models. Sure, sure. We use all the major vendors. Yeah. And we bring it all together to get what you need to be successful.

3:10:46We don't want you to have to worry about that. We want to just deliver a great, I want, you hit Slack bot and it just says, boom, here we go. And I don't want you to have to think about those things. I want to make it just great, a great experience. And I want to help you grow your business. You know, I want to help you grow your customers. I want to help you service them. I want to help you make your employees more productive. That's what I'm excited about. You guys have insane scale as you look across the, yeah, to put it lightly. Yeah, it is crazy. But specifically, have you been surprised by any industries that are resistant to AI?

3:11:27like just blanket saying like, yeah, we're, there's so much dark things to AI that it freaks everybody out. I mean, I mean, how to know how dark do you want to go? Like we can go really dark. I mean, we were talking, there was a, the journal had a statistic earlier. They said only 10 % of adults are excited about AI. I don't think that's as dark as I would go. I would say the rate of suicide that I saw with kids this year because of AI, that was the, one of the worst things I've seen in my life. It was very reminiscent of kind of what we saw with social media, kind of this unregulated technology dramatically impacting the families in the most horrible way possible.

3:12:07I watched an episode on 60 Minutes a few weeks ago on character AI, and I could not believe what I was watching. And I think that that dark part of AI, the unregulated AI, because we know AI is kind of, you know, it's inaccurate. It's, you know, it's very kind of unwieldy. We don't know how these models work. And to see how it was working with these children and then the kids ended up taking their lives, that's the worst thing I've ever seen in my life. So that's the darkest part of AI. And I think we have to be careful. You know, tech companies hate regulation. They hate it. You hear that all the time, except for one regulation they love, Section 230, which means that those companies are not held accountable for those suicides.

3:12:56So those social media companies, or in this case now, the AI companies, that's not, they basically say, oh, we're protected by Section 230. We're a platform company. They will quickly run to that regulation. but the reality is is highly unregulated in every other area so this is kind of a moment i'm sorry am i going too dark no no i'm sorry you said no we we have these kind of conversations we've talked about all this stuff yeah and uh in general i i wrote i wrote about this in our newsletter last week which is that why would the why would the average america the the the all the interviews that all the lab CEOs have done have just resurfaced over the last couple years.

3:13:40Maybe they did the interview five, six years ago talking about how the classic line from Sam is, you know, AI will lead to the end of the world, but there's going to be a lot of great companies created in the meantime. And so when people, I saw that exact video yesterday had 100 ,000 likes on Instagram, right? So like that is everyday people just seeing this and being like, okay, I like, they maybe like ChatGPT, but they're like, I like my job more, or I like living in this world and being a human. So I think there's a real narrative problem with the industry right now, and it's going to be tough to turn around.

3:14:20And look, and if we're not about protecting our children, what are we about? So let's just start there. And because we now have evidence of what happens when it's fully unregulated, and because we have the evidence of this huge horrible situation we have to take action and i think some countries you know have taken a lot of action especially when it comes to things like social media you know if you're in singapore you can't use social media if you're under 16 years old so it's like it it will eventually get taken care of but we live in you know in our world here in the United States, we're very much like, oh, don't regulate technology because we got to keep the innovation going at all costs.

3:15:04We got to keep the growth happening at all costs. Yeah. So what's the real solution? What's the real solution? I get it. Yeah. Yeah. What's a pragmatic approach here? Because at the same time, LLMs can have some negative externalities. We've talked about the example of like, you know, if grocery stores were invented last year, all the headlines would be like, this guy just went insane in a grocery store and that would be a huge story right or like this person was murdered in a grocery store that stuff happens but like we don't it doesn't get written about because of every day it's not because of the grocery stores just because um and at the same time ai from an education standpoint all the focus is on people like not writing essays anymore but there and there's very little writing on like hey this you know uh you know a 15 year old might like teach themselves like physics through using an LLM, and maybe that's amazing, right?

3:15:59So it's like, it cuts both ways. Well, I own a media company too, not as exciting as yours. It's called Time. And in my little media company, we're held responsible for the content that we write and what we say. Oh, by the way, you are too. But in the case of this technology providers, they are not. So I think step one is let's just hold people accountable. Let's reshape, reform, revise Section 230, and let's try to save as many lives as we can by doing that. Who are you looking to in terms of leading the messaging around AI safety? Do you like Dario Amadeus? Oh, I'm turning it over to you right now.

3:16:41Is it us or is it Dario? I'm passing this over to you guys. You guys can make this part of your narrative and take it forward. Yeah, yeah, it makes sense. What about in the workplace? Have you had to contend internally with the idea of someone confiding in, as silly as it sounds, confiding in Slackbot about a problem that they're having in an HR context and how that might be a new territory, a new surface area for you to deal with? Yeah, we need to talk about Slackbot safety. I mean, it sounds silly, but really, if somebody goes and they're like, you know, my boss said is going to fire me next week.

3:17:18I'm overstressed. I'm thinking of doing something crazy. You're in this bizarre scenario for the first time in perhaps your entire career. We are. And I think in two areas, first of all, as I mentioned, every company is going to have a customer agent. And you've seen like there's 10, since we've spoken like 10, 20, 30 ,000 companies, I don't even know what the exact number is today of companies deploying agent force for their customers. and we've seen these amazing agents emerge like Olive, you know, William Sonoma site, or we're going to see one emerge next week at the World Economic Forum in Davos called Eva.

3:17:52And every company has an agent to work with their customers to make them more successful. And when we deploy those agents, we have a tremendous architecture to create the guardrails to help prevent crazy things happening. I mean, we've seen this where customers try to come along and have kind of cyber sex experience with these agents on these websites. Like, no, that's not going to happen on the Williams-Sonoma site. That's not what you're going to use the turkey baster here for. So we have a whole different approach. And so that level of guardrails is super critical. And on the employee side, we're doing exactly the same thing.

3:18:38We have to kind of put those tools together and work with our partners, but also do a lot of deep engineering work to make it as great as we can. And that core platform, AgentForce, has all of those capabilities built in. Okay. On the employment side, has AI rethought your thinking about employment planning, how many people you need to hire, the scope of the organization? Has it been a roller coaster, or do you feel like you've been on one trajectory, or the trajectory has changed? How have you been processing just AI making your employees more effective potentially? Yeah, what a great question.

3:19:13I would say that we have about 80 ,000 employees now, and we'll do a little over 41 billion this year in revenue. We're now the number one enterprise software company. In AI, we just passed 11.1 trillion tokens. Let's go. It's like crazy.

3:19:38Are you surprised that companies are not getting valued on token generation yet? There's such big numbers. I am. I think, I mean, we're probably number one or number two on the token side. That probably has to do with our relationship with Bilbo Baggins. But otherwise, we've got this figured out. I would say what we have to do when we think about employment to get back to your core question. I've changed the employee mix. This year, we hired more than 20 % more account executives. There still needs to be people like me and you who go meet with our customers to really explain all of the intricacies and nuances of the technology.

3:20:23There's a lot of false narratives and people playing the market and saying things about AI that are not true, especially in regards to the enterprise. and we have to go explain it and talk to it and build the relationship and show them and prototype and deploy. And so those people are more valuable than ever and more important than ever to us. On the engineering side, I've held my engineering headcount mostly flat this year because I've gotten so much productivity increase. I probably have about 15 ,000 engineers and they're more productive than ever. I'm so proud of them. Look at what we just delivered for you guys.

3:21:00They're doing great. And we need to go even farther. But you mentioned Anthropic. We own about 1 % of Anthropic. I would say that those coding tools, you mentioned CloudCode, other things, Cursor, we use that inside. We do things to have a more efficient, more productive engineering organization. And it turns out we've reduced the number of CEOs to one because of Slack bot. We can do more because we can deploy that AI technology to the CEO also. I mean, you've spent your entire career in technology. Did this, did the chat GPT moment, did the AI boom, were you expecting it or did it hit you like a flashbang?

3:21:43Was it just a massive flashbang that just hit you like a ton of bricks? Hit me like a ton of bricks. I would say. We have a flashback. In fact, I had to use it. Ten years ago, I had the nastiest dream about AI. And so I started a whole AI initiative at Salesforce called Einstein. Okay. And we became the number one enterprise AI. And so we were building models, doing these things, making all that happen. I always believed that this time would happen. But it was more about predictive AI and machine learning and machine intelligence and these kinds of things. And we still use all that, by the way, to make this work.

3:22:23And then when the large language model emerged and prompt engineering emerged, and by the way, you may or may not know, but I'll give a little credit to my engineering team because they deserve it. But prompt engineering actually emerged out of Salesforce research. and then all of these companies you know have new companies came along to say we're just going to focus on the large language model and it's been a huge incredible success story so um i am just so impressed with what's happened in the last few years but it's just another core part of our infrastructure now and i think you see that in slackbot so getting back to slackbot it was just a seamless upgrade for you today.

3:23:05You have Slack. We added Slackbot. When we added Slackbot, we added all the AI and all the data capability. And now, actually, in the new version of Slack, you're going to see a lot of new CRM capabilities as well. You'll be able to manage your customers' information without even having to go to Salesforce. You'll be able to manage your service organization. But here's the secret. Salesforce is behind the scene doing all of it. And at some point, if you want to upgrade to the bigger, more advanced adult versions of all those things, you'll be able to do that too. I want to talk about that because we've talked to a lot of companies where they have a single product, they're a point solution.

3:23:42But because of AI, they're able to add on other surface area. Everyone's bleeding into everyone else's territory. Everyone's expanding right now. What is it like running an organization where you might have two products that are now bleeding into each other? and do you have to go and play tiebreaker a lot? Do you have a pattern for this? How do you work through all that? I love it. I think everyone should innovate and create and compete with each other, especially inside the company. Some people stay in their lane, some people don't. I'm not going to interfere with that. We have these tremendous leaders like running Slack or our sales product.

3:24:21Our sales and service product are$10 billion products each. They are some of the largest products in the software industry. And the new versions are incredible, but the new versions are all Slackbot first. And you haven't seen them yet, but if you want to get a demo later after the show, you'll see Slackbot and Slack sitting on top of Salesforce. And then, boom, you can have all this incredible productivity. But it's the lineage of Salesforce coming forward into the present moment reality of these unbelievable new advancements in AI. And that is what I've been working on. So a Slack bot first world, the ability for every company to have a customer agent, an employee agent, the agent force layer, all these incredible family applications we have.

3:25:04We just bought a company called Informatica, by the way, for like$9 billion. And it was kind of an incredible acquisition. All right. Well, there we go. I love, I always love the company, but they've done things in their new technology that for AI, if you don't have your data right, you don't have your AI right. Remember, the reason why Slackbot is working is we have all your data in there normalized. And so it's able to take all your data and give you really cool answers. Well, every company needs to normalize their data. So that's why we bought Informatica. Everybody needs to integrate it together.

3:25:43And then you have to go and connect all the other data sources in the enterprise, like Snowflake or Databricks or BigQuery or Redshift or even IBM mainframes. We call that zero copy. So our system goes out, grabs all that data in all these other places, brings it together like it is for you already in Slack. And then everything sits on top of it and you get this really powerful AI experience. So that's what's going on. That's what's exciting for me every single day. Let's talk about M &A. Hey, you talked about Informatica. There's, you know, the SaaSpocalypse has been widely reported. I'm sure you look at some charts and get a little excited.

3:26:21I'd rather not look at the charts. I can't figure them out. I mean, our revenue and profitability and cash flow go up, and then the stock is all over the place. Because people think there's some weird end of SaaS. But I would just ask you guys, did Slack become a lot more valuable for you today or less? Way more. Way more. Of course. Way more. I'm talking about you looking for value. You've done a ton of M &A. You must look around at companies that you feel like are oversold. I think there's a lot of incredible opportunities right now in the market. And if you were not looking at, look, we do about$15 or$16 billion in positive cash flow a year.

3:27:04We're going to use that. We're going to buy back some stock. We're going to buy back some stock. We're going to do some dividend action, too. But what else are we going to do with our cash? I think we should buy some companies also. Because I love organic innovation, but I love inorganic innovation, too. I bought about 100 companies. Give us a three-minute M &A masterclass. There's a lot of companies whose market caps have inflated. They've got a bunch of cash on their balance sheet. They're 1 % of the size of Salesforce, but maybe they're starting to think about M &A. what have you learned about how to do it well in a few minutes?

3:27:42And this is the last question. You have to innovate. Innovation is one of our core values. What are our core values? Trust, customer success, innovation, and quality sustainability. Those are our five core values. We can talk about any of those. But when we talk about innovation, we talk about organic and inorganic innovation, right? Organic is those 15 ,000 engineers making Slackbot. It's 100 % organic. But then, well, actually, but it's not because we bought Slack. Remember five years ago when I bought Slack? Because I love Slack. I thought they did such a great job. I thought it was a one-time trade.

3:28:16I don't think anything was going to replace it. And I thought I could make it a lot better. I thought I could put it on top of all of our products. I thought it could be a user interface for the future. I thought AI would be incredibly benefited by the data inside Slack. We just proved that to you with SlackBot. So we bought Slack. Well, it was a huge purchase at the time, one of the biggest ever in the software industry. I think it was like$25 billion, our biggest. But guess what? The value that it's provided now, it's not a sub-billion-dollar company like when we bought it. It's a multi-billion-dollar company.

3:28:54I think it's like almost$3 billion or more. I don't have the exact number in front of me. But you can just go, wow, this is like incredible, the value that Salesforce and Slack together provide. And that's what I'm looking for. Or Informatica plus Salesforce together. Or MuleSoft, I mentioned. We bought it when it was a$300 million revenue company. And now it's a huge multi-billion dollar software engine as well. So we have so many of those. And our job is to buy them and make them work. Not all of them are going to work. Acquisitions are very risky. You got to be really careful. And then you can have it all worked out and it still doesn't work out.

3:29:35But it's worth it to take the risk because innovation is risky. And organic and inorganic innovation, you have to do both. What is a good hit rate at scale? You said you bought something around 100 companies. But when you're talking with your M &A team, how often are you okay with them? and the company just swinging and missing. I have such a cool team there because not only do they do that, they also run a$5 billion investment fund. We have invested in so many amazing companies and taken so many amazing companies public. We just sold Wiz that we helped start to Google. We helped grow Snowflake and took it public and made one, I think we made a billion and a half dollars on that.

3:30:25I think we made a billion and a half dollars on Wiz. Because I think we have now about a point of Anthropic, which is a few billion dollars or so. Fantastic. And it was started when it was a tiny company. So we love investing, number one. We love buying things. And we think all of that is innovation. I love it. And it's all in one big innovation bucket. But we're trying to do something that others have not done is we're trying to innovate in the enterprise at scale. And that is the hard thing because, look, as I said, we have 80 ,000 people and more than 41 billion. And we're going to start our next fiscal year out on February 1.

3:31:08And it's going to be a massive year bigger than that. So I'm like going, whoa, these numbers are huge. But I've got a huge dream. I'm trying to go as fast as possible to$100 billion in revenue. But while maintaining this incredible profit, oh, we're one of the most profitable enterprise software companies, too. We're delivering more than 34 % margins this year. I have one last question. Thank you so much for coming on the show. Last question. Are dreams an underrated source of business alpha? You said you had almost a nightmare about AI, and that started Einstein. Do you look for signal in your dreams?

3:31:47I pay attention to my dreams. I pay attention to my visions, my insights. And Salesforce started because I was swimming with a pod of dolphins outside the coast of Hawaii where I live. And I was in about 100 dolphins and I was one with the pod. And when I was one with the pod, all of a sudden in my mind, I saw this vision of what Salesforce could look like. And I came into Larry Ellison and I said, hey, I think I'm going to quit my job. And he gave me$2 million and we started Salesforce. And that was that. That was 26 years ago. That's amazing. Well, thank you so much for sharing the story with us.

3:32:24Thank you, Dolphins. Thank you, Dolphins. Thank you to the Dolphins. The sound is amazing. Look at that. Thank you, Dolphins. Thank you for coming on the show. Take care of the Dolphins. We will talk to you soon. Have a great day. Love you guys. Goodbye. Super fun. The chat's calling him the Steve Jobs of AI. Indeed. He's a strong candidate. Strong candidate. Always great to chat with Mark. What else we got? There's a few posts. There's some breaking news. Andre Carpathy has been on a following spree on X. This is why you come to this show. He followed Jacob Rintamaki, friend of the show. He followed Eric Gleiman, the CEO of Ramp.

3:33:09Andre Carpathie is making the rounds. And Jacob Rintamaki says, you know, I still think it's funny that the guy who ran Tesla's full self-driving program, his name is Carpathie. It's a good nominative determinism for Andre Carpathie. It's always there. Although he's working on education now. So we're going to have to rework the nominative determinism to figure out how that works. In other FSD news, Elon says Tesla will stop selling FSD after February 14th. It will only be available as a monthly subscription thereafter. So February 13th, you go in, you buy the full self-driving. It's a one-time purchase.

3:33:48Then he's going subscriber mode. Probably good for the finish. Yes. Somebody here is in Google just searching. I'm a kitty cat. Where is my owner? my bowl is empty. The AI overview pops up. It sounds like you are very hungry and miss your owner. Since you can't find your owner right now, you might want to meow loudly near their usual spots to get their attention. Check the common areas of your home like the living room or their bedroom in case they are sleeping or relaxing elsewhere. I'm really glad that the AI overview and Gemini powering it are thinking about cats. It's open to all species. It's not humanist.

3:34:30It's not focused on humans exclusively. It embraces the kitty cat inside of all of us, I suppose. I started saying yes to all the apps that can track me prompts. The internet is now noticeably better with more relevant and less spammy ads. Apple did us all a disservice and misplaced emphasis on privacy. He says, get into the targeted advertising flow. You want to see good stuff? Hit yes on this app can track me. Let them target you with the best possible ads. Well, speaking of targeting, Hermes is allegedly stalking their clients. A Glitz investigation has revealed that Hermes employees are Googling clients' home addresses to determine whether they have a prestigious enough address to be deemed worthy of a Birkin or a Kelly.

3:35:17Hermes is also allegedly scrutinizing clients' social media accounts and the content they post. After a quota bag is purchased, they continue to monitor for resale activity, which if detected results in an immediate blacklist for the client. No surprises there. I think OpenAI should do this with their device. And they sell those. They should say, oh, you know, there's a quota. There's an allocation. We'll try and get you in. But you're really not using ChatGPT all that much. And the questions you ask aren't that interesting. So, you know, I don't think we're going to be able to get one for you.

3:35:50I know, by the way, AirPods, we were debating, is the open AI device going to cost$100,$300? I think they got to do like$30K. I think it's got to be like 30Gs. Yeah. It's got to be elite. Add a zero. Why not? Add a couple zeros. Yeah. Add a couple zeros. And then make it a huge hassle. Make it, oh, well, how many ChatGPT Pro subscriptions have you bought for your friends and family? How much money have you spent with us? What's your purchase history with us before we get you one of these devices? You need to make it elite. This is the key. Have you ever turned? Winning in consumer electronics.

3:36:23Have you ever churned from ChatGPT? Yeah. Well, you're going to the back of the list. Oh, yeah. Yeah, what was going on in November? We see that you tested Claude. Yeah, yeah. You're not getting the device. You're not getting the, what's it called, Sweepy? Sweepy is a funny name. Ben Mullen at the New York Times says, what AI-generated PR pitches are doing to reporter inboxes everywhere is a catastrophe. It's brutal. You're absolutely right. It says Lulu. Yeah, this is crazy because the challenge is they can, like it used to be, you know, from our side, even maybe a year ago, like if someone was like really enjoyed this segment where you were talking about this, this, that, this, that, it like actually seemed like they watched it.

3:37:10And now you can just easily get that from Gemini or wherever. So it's a disaster. Yeah. Fortunately, Nick is there just chopping emails down, dodging PR lasers. Maybe the last piece of news, Mike Krieger, the co-founder of Instagram, who joined Anthropic two years ago as chief product officer. He's changing roles to co-lead an internal incubator dubbed the Labs Team. I could see him cooking up some really cool stuff. He, of course, in between Instagram and Anthropic, worked on a news application called Artifact that was sort of a precursor to Pulse. It was like almost at the perfect time, but AI wasn't quite there.

3:37:53And I could see him doing something really interesting. Obviously, he's good at going from zero to one in a consumer application. If there's some new surface area for an AI tool, an app, I wouldn't bet against him. So congrats to him on the new role. One final post. Freya says, so fascinated by this product. And it is in some type of booking. Looks like it's in the booking.com world. And it's a get$500 if it rains on three days of your nine-day trip for only$26. So we're bringing gambling. Now, this feels like a pretty straightforward insurance product. This is just Friedberg's first company, right?

3:38:36It's climate insurance. Not really. He was selling to farmers, but yeah, it is sort of weather futures and crop insurance has existed for a long time. And yeah, this makes sense. But are we getting out of here? The bomb has been planted. Thank you for watching TVPN today. We had a lot of fun with you. Leave us five stars on Apple Podcasts and Spotify. We actually do it. Please actually do it. Leave us, drop your email at tbpn.com. We have our daily newsletter. Thank you to our sponsors and thank you to our production team and Tyler for hanging out with us today. And Dylan, who's here today. Dylan's here in the studio.

3:39:15And we'll see you tomorrow. We really can't wait. We can't wait. 11 a.m. Pacific, we'll be here live. More great guests. Goodbye. See you guys soon.