TBPN - Sam Altman on Codex 5.3 Launch, Anthropic's Sholto Douglas, Alphabet Beats Q4 Estimates | Sam Altman, Sholto Douglas, Daniel Barcelo, Mandy Fields, Ivan Burazin, Scott Rogowsky

Episode Date: February 5, 2026

Sign up for TBPN’s daily newsletter at TBPN.com(01:21) - 𝕏 Timeline Reactions (02:58) - Anthropic's Trust Nuke (18:36) - OpenAI's Frontier Model for Enterprise (39:02) - Alphabet Beats Q4 Estimates (52:25) - NVIDIA to Skip New Gaming Chips in 2026 (56:16) - Sam Altman, CEO of OpenAI, discusses the launch of Codex 5.3, highlighting its enhanced coding capabilities, increased speed, and improved interactivity, allowing users to guide the model during tasks. He emphasizes the importance of integrating AI agents into workflows, suggesting that managing a team of agents will become a standard practice, and notes the need for companies to adapt quickly to AI advancements to maintain competitiveness. Altman also touches on the evolving role of software development, predicting significant transformations as AI continues to advance. (01:33:10) - Sholto Douglas, a Member of Technical Staff at Anthropic, specializes in scaling reinforcement learning and agentic AI. In the transcript, Douglas discusses the release of a new model that excels in tackling complex problems through extended reasoning and test-time computation. He highlights the model's advancements in digital tasks, such as coding and knowledge work, and emphasizes the importance of developing AI systems that can function as capable coworkers. (02:00:32) - Daniel Barcelo, CEO and Chairman of T1 Energy, has over 25 years of experience in international energy finance and emerging markets. He discusses T1's efforts to build a domestic solar and battery supply chain, highlighting the acquisition of a 5-gigawatt solar module facility in Wilmer, Texas, and plans for a 2.1-gigawatt solar cell plant near Austin. Barcelo emphasizes the importance of advanced manufacturing and an integrated U.S. supply chain to meet the growing demand for scalable, reliable, and low-cost renewable energy. (02:24:41) - Mandy Fields, Chief Financial Officer of e.l.f. Beauty, discusses the company's impressive 38% net sales growth and 79% increase in adjusted EBITDA, attributing this success to their value proposition, innovative marketing, and the acquisition of Hailey Bieber's brand, Rhode. She highlights e.l.f.'s balanced marketing spend across digital platforms to stay connected with their community and emphasizes the importance of Super Bowl ads for building long-term brand awareness. Fields also notes the company's focus on expanding their existing portfolio, including e.l.f. Cosmetics, Naturium, and Rhode, while remaining open to future acquisitions. (02:32:56) - 𝕏 Timeline Reactions (02:35:53) - Ivan Burazin, co-founder and CEO of Daytona, has a rich background in cloud development, having previously founded Codeanywhere and the Shift developer conference. In the conversation, he discusses Daytona's mission to provide secure, scalable infrastructure for AI agents, emphasizing the necessity for these agents to operate in isolated computing environments to perform tasks like code execution and browser-based activities. He also highlights Daytona's rapid growth, noting a significant increase in customers, including Fortune 100 companies, and announces a recent $24 million Series A funding round to further advance their platform. (02:43:01) - Scott Rogowsky, renowned for hosting HQ Trivia, discusses his new venture, Savvy, a live interactive word puzzle game where players compete against him in real-time to solve puzzles and win cash prizes. He reflects on lessons from HQ Trivia, emphasizing the importance of player engagement and accessibility, and highlights Savvy's design to ensure all participants can play through entire games, enhancing user experience. Rogowsky also shares his commitment to the project, having recently relocated to Los Angeles to focus on its development, and expresses excitement about building a dedicated community around the app. TBPN.com is made possible by: Ramp - https://Ramp.comAppLovin - https://axon.aiCognition - https://cognition.aiConsole - https://console.comCrowdStrike - https://crowdstrike.comElevenLabs - https://elevenlabs.ioFigma - https://figma.comFin - https://fin.aiGemini - https://gemini.google.comGraphite - https://graphite.comGusto - https://gusto.com/tbpnLabelbox - https://labelbox.comLambda - https://lambda.aiLinear - https://linear.appMongoDB - https://mongodb.comNYSE - https://nyse.comPhantom - https://phantom.com/cashPlaid - https://plaid.comPublic - https://public.comRailway - https://railway.comRestream - https://restream.ioShopify - https://shopify.comTurbopuffer - https://turbopuffer.comVanta - https://vanta.comVibe - https://vibe.coSentry - https://sentry.ioCisco - https://www.ciscoaisummit.com/ai-virtual-summit.htmlOkta - https://www.okta.comKalshi - https://kalshi.comFollow TBPN: https://TBPN.comhttps://x.com/tbpnhttps://open.spotify.com/show/2L6WMqY3GUPCGBD0dX6p00?si=674252d53acf4231https://podcasts.apple.com/us/podcast/technology-brothers/id1772360235https://www.youtube.com/@TBPNLive

You're watching TVPN. Today is Thursday, February 5th, 2026. We are live from the TBPN Ultradome. Yahoo! The Temple of Technology, the Fortress of Finance. Let me tell you about RAMP.com. Time is money saved both. These use corporate cards, bill pay, accounting,

and a whole lot more all in one place. We have a quite the lineup today. Let's pull up the linear lineup. Meet the system for modern software development. 70% of enterprises on linear using agents. And we have a lineup for you today. We got Sam Altman from OpenAI joining Shalto. If you're not familiar, Sam Altman is the CEO of OpenAI.

They are the makers of Chatto Beach. Yes. I did actually explain to someone. I was like, oh, do you know, I was explaining the show to some random person. I was like, oh, they were like, what do you do you do? What do you talk about? And I was like, oh, do you know like Open AI? Yeah. I was like, oh, we're having Sam on the show today.

Are they like who? No, no, no. They were like, oh, that's cool. Like, that's good. Like, I understand what the show is. Then we have Dan Markello from T1 Energy. talking about building solar panels in America and, of course, our lightning round, which I'm very excited for. So we've been thinking more about the Super Bowl that's coming up.

We've been thinking more about ads, the response to the ads, the back and forth with the ads. Rune had a good post here. He said, putting my media observer, putting my media observer hat on, anthropic ads are pretty brilliant because they're dishonest in a way that's only going to rage bait open AI heads. and certain industry insiders, but are funny and striking to everyone else. When you're a call option, ooh, calling them a call option,

kind of a diss. Savage. Variance is good. Mario Kart Blue Shell. Are you familiar? Do you have you played Mario Kart? Do you understand the Blue Shell? So I played Mario Kart.

I think so long ago, Blue Shells didn't exist, but I've been planning my kids, and I've since learned the importance of the Blue Shell. The Blue Shell, it targets just, It's just the first player, just whoever's in first. It's a great metaphor for what's going on here. When you're in, when you're not in first place, you get the blue shell, you can take a shot at the leader without even needing to call them out.

So you can just say, the category is bad, and everyone assumes you're talking about, you know who. So I thought that was an interesting thing. Trey says Sam Altman, the Konigseg collector. Yes, yes. Yes, that one. Yes, deep dive. Anyway, so everybody had a take on this yesterday. It was perfect in how much kind of controversy it generated.

It was wildly entertaining. I wanted to kind of, I'll read through kind of like my updated take. I got a little bit of... Processed. Yeah, I got a little bit of pushback. I said they were playing dirty. Signal responded to me and said, not dirty at all. So I wanted to address that.

Okay, dig in, but first let me tell you about console. console builds AI agents that automate 70% of ITHR finance support, giving employees instant resolution for access requests. Anyway, so, yeah, I was processing this more. We obviously watched some ads yesterday. We watched the Get a Mac campaign. We watched the Bud Light special delivery one, which is about Bud Light is in a castle. They get an order of corn syrup.

They're like, we don't use corn syrup. That must be for Coors Light and other competitors. And so, like, I was processing them. And like the difference there is that those advertisements are truthful, right? Like people, people that have had a PC have probably gotten a virus, right? So when Mac is like riffing on that, it's like, it's truthful, right? Yeah, and they had some data to back it up. It's not deceitful.

I would say if I'm putting on my steel manning Microsoft in 2007 hat or 2003, I would say, hey, we do have Windows PC defender. We're fighting viruses. And is it possible to get a virus on a Mac? probably is it is it is it possible to not get a virus on a PC yeah and and on Microsoft side people are like yeah no one makes Microsoft or no one makes viruses for your computers because you don't sell very many it's not very ROI positive it's a good point anyways and then Budlite's campaign was like truthful even though it was aggressive and that you could look up the

ingredient list of their competitors and see that they did in fact use corn syrup yeah and you can make your own decision on whether or not you like that ingredient but they were just drawing awareness to it and so and so and so So my point is that I think that Anthropics ads are like closer to political attack ads and that they're sort of intentionally taking, kind of trying to be deceptive, right? They haven't broken any laws. Yeah, no, no. They don't name chat CBT.

They're just sort of like throwing LOD at the whole category. I asked, yeah, so anyways, I said they were playing dirty. I got some pushback on it. I asked Claude. I said, Claude, how would you define playing dirty? And Claude said, playing dirty generally means achieving your goals through tactics that are deceptive unethical or that violate the understood rules and norms of a given context, even if you're not technically illegal. It's the gap between what you can do and what

you should do. A few dimensions to it. It goes into deception, misleading others about your intentions, hiding information, or creating false impressions to gain an advantage. These campaigns do an amazing job creating a false perception or impression of what ads in and the alums are going to be like. I do think the response, just to chime in some random stuff, but I think the response to the ads, we were wondering, like, you know, outside of the, you know, the TBPN, we love ads, ads are fine,

and they're not going to do anything weird. We're strong supporters. What will the public's reception be like? Yeah. Claude skyrocket to the top of the charts because these ads are so effective. Will general consumers buy the line? Yes, the chat apps are going to get weird with the ads or not.

And I was scrolling on Instagram Reels last night completely randomly. I was not looking for anthropic content. I think I followed the Claude account. Maybe, maybe not. It just targeted me. It hits me with a vertical version of the ad. It's called deception,

I think, something like that. No, violation. Violation is one It's not called deception. It'd be a little too on the nose. I think there is one called deception. There's a bunch. They all have different names. Anyway, it's called violation. And there's a screen. grab here. Oh, yeah, yeah, I remember. So people were like the, the anthropic deception ad is deceptive. Yeah, yeah, yeah. So violation pops up and it's this, and it's this Claude AI,

and it has almost 6,000 likes, even when it just got served to me, I, my interpretation was like, this is working, this is popular, this is, it's not just beautifully shot, it's well edited for vertically. Yeah, it's either, it's either really resonating or they're putting a mass amount of spend behind it or both. Yeah, so this is what it looked like vertically. And then this is the funny thing. So in this ad, you see the guy struggling to do a pull up. He asks, you know, what's supposed to be an LLM, create a fitness plan for me. And then the fitness bot says, hey, you know what else can help? One inch insoles from height max or something like that. And it's like this looks max. It's very funny. But it's a one inch insert that would go in your shoes. I scroll up.

What's the next ad that meta serves me? An ad for three inch inserts. Three inch inserts. And the ad is actually deceptive. It says the guy can go from 5-9 to 6-1. That's 4 inches. And so these are full shoes that have the inserts built in. And for some reason, I got in, even though I'm not in the market for insoles, the algorithm just knows that I love these ads because they're very funny and they're very on trend with the looks maxing thing. And so I get served these ads constantly.

This is all meta shows me is these height-enhancing shoes. because I think I actually clicked on them and was like digging in. So you bought them, right? Of course, of course. To send to my enemies to a smogic. You're trying to get to seven feet. That would be good. That would be good.

Tyler, do you have something on this? Yeah, I was just going to say, so I saw on Instagram as well, I saw some of the Claude ads. And in the comments, I mean, people were riding with Clot.

They are. Yeah, they are like. It was like Norma's too. Yeah, yeah, yeah. No, totally, totally. Yeah. Like, like, they're winning the Vibe War. They've been winning the Vibe War with developers,

and they've been winning the Vibe War on Axe. And now it feels like they're about to win the vibe war in the public square. Really quickly, let me tell you about Figma. Figma makes isn't your average vibe coding tool. It lives in Figma, so outputs look good, feel real, and stay connected to how teams built. Create code back prototypes and apps fast. Sorry. I said two things can be true about the campaign.

It's brilliant, well-timed, and incredibly strategic for a few reasons I'll outline below. And it's designed to plant a false impression of Chad Chippew's forthcoming ad product in the minds of hundreds of millions of Americans. They could argue, oh, we're not trying to do it. that, but you can't really kind of argue with the effect. So Anthropic accomplishes a lot. The campaign entertains America, right? It's wildly entertaining. It's hilarious. Truly. Like the perfect, like sycophancy that you can hear, you can hear the MDashes, the pauses. It's amazing. Really good. Mother is the name of the agency that they actually crushed it.

They're also putting themselves on the map ahead of the IPO. I think in some ways, like certain audiences would know more about anthropic than Claude, right? Yeah. Even if you're just like generally interested in investing in AI, you're probably hearing about Anthropic more than you're hearing about Claude. Yeah, yeah. It builds their aura with insiders. I said if they spend $100 million on this campaign and all it does is help retain a couple of like truly elite researchers.

Yeah. It's worth it. What are you laughing at? I love it. I know it's like anthropic is campaigning to get themselves banned. Just like with misinformation just going way too far and it backlash. That's just funny. me. Sorry, we can go back to it. It's somewhat continues. They're like fear-based messaging that they've been, they've been kind of riding with in general. Yeah, yeah, yeah, yeah, yeah. At the essays. Effectively rage baits open AI. Like, they got, they got fully baited.

Completely. Sam switched out of his, like, you know, lowercase typing and was like, I got to go into uppercase for this one. Lots of responses. It increases, like the public's, like, general scrutiny of the ads rollout. And then Washington, too, is another factor. Yeah, maybe, yeah. I just think it'll come up. Totally, yeah. And then... If you have to ask Mark Zuckerberg, how do you make money? Like, anything's on the table these days in Washington.

Yeah, and so... And then the other thing is, like, it's going to broadly damage consumer trust in LLMs. Some people will just be like, wait, like, they've been kind of like making money on me without me knowing, right? Or can I trust every output as, like, actually good advice? or am I being monetized?

And this is the one that you think could come back to bite them. The other five are pretty good. Potentially, right? But it depends what their consumer strategy is. Right now they're saying, we don't care about consumer. They've said that a lot. But their actions kind of speak differently in some ways. I said Anthropic has consistently told the market they don't care about consumer,

but I'm not sure. The argument for ads is that they'll make LMs free for people that can't afford to pay a subscription. but Anthropic has already lost the race to serve billions of people, right? I don't think that they're, when you look at when you look at Gemini's sort of traction, open AIs, chat chip B to heat traction, like it seems like the race to get to three billion monthly actives is kind of over. I don't believe that I don't believe that Claude's going to come behind and get there and they wouldn't be able to do that without doing ads, right?

Because there's, like you can kind of run the numbers. The call option, they need the blue shell. They got to take out the way in front. So the question that kind of where I was taking this is, can they deliver a luxury product to a smaller cohort in the hundreds of millions to kind of iPhone numbers, right? There's like roughly like one and a half billion iPhones that are like active in the world. Those people could all buy a cheaper Android and just cheaper devices, but they've paid a premium for the iPhone because they can. And for many people it delivers a better experience. So I said the iPhone was not the first smartphone, Claude was not the first consumer LLM.

The iPhone did differentiate on specs early, not unlike a model card, but Apple did eventually pivot to more emotional arguments for why you should be seen with an iPhone. It tells people you care about the environment, that you don't have adult apps flooding your app store, and that you take privacy seriously. These have had varying levels of success. Every tech company was able to tell an ESG story, and I can't imagine an Apple exact, even saying the word porn today, even though Steve Jobs was very pointed about it back in 2010. He said, you know, there's a porn store for Android. You can download porn. Your kids can

download porn. That's a place we don't want to go. So we're not going to go there. Calling out the competition by name and dropping that is like, yeah. And the key thing here is that it was factual. It was true. It was true. Like it wasn't deceptive. No, no. And so I don't think, I don't think that was that was edgy, but he wasn't playing dirty. Yeah. Yeah. And so Apple did, however, carve out a solid messaging strategy around privacy and allowed them to put the screws to advertisers, the market adjusted, and the ad industry obviously survived, but average Americans still feel that Facebook is listening to their conversations to target ads.

Consumers deserve choice. It's great if they want to pay for ad-free tiers. Most don't. 1% in Europe for Facebook, by the way. That's the stat. They all have the option to pay for Facebook, for ad-free Facebook and only 1% pay. So consumers deserve choice,

but they should not be misled about how ad platforms work. Android has generated an immense amount of value for the world. So as Google broadly, let consumers choose, but let them choose intelligently. Yes, I have a rebuttal, but I'm going to tell you about Lambda. Lambda's the super intelligence cloud building a assing for training and inference at the scale from one GPU to hundreds of thousands. So my rebuttal. My steel man is that... And to be clear, I'm not saying anthropics shouldn't have done this. No, yeah. I'm just saying that it was a little dirty. Yeah,

they're rolling around in the mud. It's good. They're in the trough. We love the trough. We live in the trough. We live for the trough. My steel man is that They didn't cross the line. They didn't play dirty because they didn't call out ChapchipT directly. Okay. You can take that whatever you want. But will something like this happen? They are punching up.

They're punching up. Yes. But third, there is a world where something like this will happen. There is a world where the ads do get integrated in such a seamless way. If you look at the evolution of Google's 10 blue links, it started with 10 blue links, no ads. Then it was a very clear yellow box. with ad, and it was very clear that it was an ad. And over time, the UI evolved to be a lot less aggressive

about telling you that it's an ad. And the ads on meta platforms do get creepy sometimes. You talk about something, and then you see the ad. And maybe that's just confirmation bias or some sort of cognitive. You only notice the ones that are weird, so they all feel weird. You see a lot of stuff that you weren't talking about. That doesn't trigger anything. But when you see the thing that you were just talking to your friend about, I was just talking to you about sweaters and I see an ad for sweater.

I'm like, how did it know? And realistically, it knows because you just went on Facebook, you found that sweater, you bought it. It knows that we're friends, we're DMing, we're talking, we're literally friends on the platform. And so it's like, look, Geordie likes this and they're hanging out all the time, sending each other DMs. Why don't I just show John what I just sold to Geordie? That makes perfect sense. That's something that can be done with just stock vanilla machine learning at, you know, core AI. inside Facebook and meta, and they do that very effectively.

But it can feel sort of creepy sometimes, and some people get creeped out by it, and they talk about it. And so the idea that an interaction like that might happen is not complete science fiction. It is possible. And so they are sort of warning that, hey, if you want to make sure that this never happens,

it's our pledge, that that's not even on the table. Now, the big question is, when's Anthropic launching ads? We got to get them to launch ads. Well, I don't think they can now. No, they have to. It's okay. I give them permission.

I will say, if you, everyone's going to be dunking, oh, you went back on your promise. No, I will be your strongest soldier. I will be your strongest soldier. And I will say it's the right thing to do. Put ads in Claude code. Put ads in clod. Put ads in the comments of the code that you write. So if someone reviews the code, they're seeing an ad.

That's what I want. That's the future I want to live for. Anyway, Turbo puffer. Serverly Specter in full-text search, built from first principles on object storage, fast, 10x cheaper, and extremely scalable. Oh, and you know I've got to tell you more about ads, vibe. Where D2C brands, B2B startups, and AI companies advertise on streaming TV. Pick channels, target audiences, and measure sales just like on meta. We love ads here, and we love doing ads for ads.

Zach Kukoff says, every time I see Anthropic and opening eye try to distinguish themselves with comms marketing, I realize how much we are replaying the PC wars from the 90s. Anthropic, tasteful, elegant, opinionated, prosumer, expert enterprise. OpenAI, populace, broadly appealing, low consumer, low consumer plus typical enterprise. Yeah. Yeah, so Anthropic is the Apple and OpenAI is the Microsoft. And they're also aligned with Microsoft, owned in part by Microsoft.

Yeah, I mean, it would be interesting to hear Dario just talk for an hour purely about just the risk. of advertising in AI. Yeah. Right. Because that would be powerful. Certainly wouldn't have been as effective as dragging open AI in front of hundreds

of millions of people. Yeah. I mean, doesn't Anthropic have a podcast? Anthropic? Podcast? I think they do. TJ was helping Sam with some comms. He said, Anthropic might think more seriously,

he said, fix it for you. Anthropic might think more seriously about adding ads if they had any consumers using their product. Just taking a very tactful response to a situation. Just being like, what if you amped it up, brother? Yeah, I mean, you've already lost if you're just dropping a massive word salad. Maybe.

I don't know. I think there's a lot of nuance here. And it's good. I do think it's important to not mold, to not be, you know, like, angry and made mad. Don't let people spike your cortisol. Exactly. Like, they're going to be gesture maxing at the Super Bowl, and you can't let it. You can't let it affect you.

You got to focus on the new models, which have launched today. There's a bunch of new stuff. So GPT 5.3 Codex launched today. And there's a very, very cool update to, or new product, Frontier, which is a product for building AI co-workers that I'm very excited to talk to Sam about. Because it feels like the first glimpse of an orchestration product, which we talked about yesterday, Gastown has been taking off. I mean, let's give some credit to John.

I would say it might seem like John had early insight on today's Open AI launch. We did not. We didn't know. We actually didn't. But in hindsight, you called it perfectly. Crystal ball. Crystal ball. Elite.

Alpha. One day of alpha. One day of alpha. Yeah, if you had launched a enterprise focus orchestration platform yesterday. Yes. You could have raised money. got sold secondary. All right before Open AI immediately launched a very...

Are you going to get Steve from Open AI? They're not even thinking about this to they launch it the next day. Who knows? Who knows? Anyway, let's move on to Shopify. Shopify is the commerce platform that lets you grow, it grows with your business and lets you sell in seconds online, in store, on mobile, on social, on marketplaces, and now with AI agents.

Matt Turk says regular startup. We close a few customers and shipped some new features. Good week so far. Anthropic. We destroyed our main rival with our Super Bowl ads and tank the entire software category in public markets by announcing some plugins. Good week so far. Oh, yeah, we barely even talked about this, but Anthropic launched a lawyer in your pocket. They launched a legal tool or they announced it. I don't know.

Is it actually available in the app yet? Because it feels like, okay, maybe this competes with Harvey. I'm not seeing it yet in the Claude app. I don't really think it competes with Harvey. No, no, no. So I don't think it does because they're selling it direct to consumer, I believe at least. But it seems like an amazing product. Like it seems like the demand for this would be incredible.

Yeah, it will compete with Legal Zoom. Like Legal Zoom is down 15% since this announcement. Sure, sure, sure. It's now a $1.38 billion company. Yeah, sort of in the Chegg mode. Yeah, and not just for, I mean, legal Zoom's a little bit different because you can actually file you in corporation documents. And they've been, they face pressure from Stripe, Atlas. for a long time on the front of LLCs and whatnot.

But I mean, truly, like, if you're getting a job and your employer gives you an offer letter, like, taking that to a lawyer can be really expensive if it's your first job. You're probably not going to review it. A lot of people are probably just copy pasting it into cloud or urban air. Yeah. Take a look at this. A little bit of that, and then they can't. But just being able to just forward the email in or integrate your Gmail and just say,

hey, I got this offer letter. Like, does anything in here look weird? Is there anything I should ask? I don't have a ton of leverage, but I want to understand this document. Claude should be able to do that, and it makes a lot of sense. And I expect Open AI to launch this product like ASAP. Yeah, I mean, so I think all that was actually launched for the cloud legal thing was just a plugin in Cloud Co-Work. Oh, which means it's basically just like . . . . . . . . . . . . . . . . . . . . . . . . . . which means,

yeah, yeah, yeah. It's like skills. It's like it goes beyond skills because they probably had to do some legal work at their own to make sure that they're, like, not giving legal advice and that they're couching things properly. I think that's also why ChatchapD Health is in a different area. because they don't want any risk of like pre-training on your testosterone levels. And then I go into the next version of chat GPT and I say, what are Tyler Cosgrove's testosterone levels? And it just knows it because it learned it from the chats that you were sending it.

There's a whole bunch of private information beyond testosterone levels, of course. Anyway, Railway. Railway is the all-in-one intelligent cloud provider. Use your favorite agents to deploy web apps, servers, databases, and more, while Railway takes care of scaling, monitoring, and security. Exactly. Key over on X says, TBH, the Anthropic ads are good, but I think they're a bad idea. Normies are not going to think, wow, this is what Chad CBT is going to be like. I better subscribe to clod.com.

They're going to think, wow, this is what AI is going to be like. Rune agrees. He's not biased at all. He says suicide bombing strategy. It's bad for them. But worse for Open AI, you almost have to respect it. What was the text I sent you? I said, quote, I don't know how to get on Anthropic.

Yeah, John was asking him friends. that's outside of tech and they were like, what's anthropic? How do I get on it? I don't know how. Let's go to Eric Suford. He said, this anthropic ad is simply obnoxious. And then we'll move on to other stories. Anthropic generates most of its revenue from enterprise business

so it can afford to not to maximize its consumer revenue opportunity. But the sanctimonious moralizing here presents advertising as a cynical business model choice. It's neither cynical nor a choice. The freemium digital advertising supported model is the only repeatedly proven pathway for a consumer technology product to reach humanity at scale.

There are just folks who can't pay $20 a month. OpenAI's advertising revenue will enable it to offer access to larger models for free-tier users. To the extent Anthropic thinks chatbots powered by frontier models are valuable to society, it should aspire to provide access to as many people as possible. Now, they'd probably say, hey, we're freemium. We do give, we do give access to chatbots powered by frontier models to free users. They just only get a certain amount and then they have to upgrade, but Eric's point still holds here for sure. Ads allow for that to an extent that no other consumer technology business model demonstrably can. This kind of condescending attitude toward digital advertising is economic chauvinism, and it represents a desire for technological gatekeeping.

And the idea that ads will necessarily influence chatbot content is overstated, borderline anachronistic, anachronistic because you don't do it that way anymore. You wait until, hey, this person's buying, you know, insoles or lifts, and they're asking about the Roman Empire. And this is the perfect time because they're just chilling, reading the Roman Empire. And they're thinking, oh, okay, yeah, I did need to buy that. And then they switch gears. Like, you don't need to put the ad right next to the content that relates to it. That's just an antiquated way of thinking about online advertising.

Yeah. It ignores functional. Again, I think Sam Altman did say in October of 2024, I kind of think of ads as a last resort for us as a business. business model. Yep. So again, he's going to, you know, has to eat his words on this one. They get to take a little bit of a victory lap there. Yeah. So in that sense, criticism is fair. Anyway, before we move on, Crowdstrike. Your business is AI, their businesses securing it. CrowdStrike secures AI and stops breaches. So, Claude also announced a new model from Anthropic, of course. Introducing Claude Opus 4.6,

we have Shaltow coming on the show at 1230 to discuss that. They say, it's our smart It's the smartest model and it got an upgrade. Opus 4.6 plans more carefully, sustains agentic tasks for longer, operates reliably in massive code bases and catches its own mistakes. It's also our first Opus class model with a 1 million token context window in beta. So very exciting.

They put a number of, they put the model card together. They did particularly well. What was the benchmark that stuck out to you, Tyler? or you said... Yeah, I think it was Arc AGIV2. It's now at like 69%. I think previous was, I believe it was 5.2,

which was at, I want to say, 55 around. So like pretty sizable upgrade. But I mean, with these things, it's always so hard to tell. It's really just like qualitative differences at this point where the models are like, you have to use it for like an hour or two and then you can kind of tell what the differences are.

Yeah. And also like in new contexts. Like in terms of just the vanilla, go and ask a question, and it's been able to get you a pretty good answer for like years now. But it hasn't been able to go and pull a bunch of financial data together, fact check it all, put it in an Excel sheet, this is what they're pushing, and this is what Cluso Investments is so excited about. Ooh, who says Cluoso.

Anthropic Upd's AI model to field complex financial research. So Opus 4.6 is designed to carry out financial research and other work-related functions. The company's expansion into new areas, including legal service, has rattled Wall Street and sparked concerns about which companies and services will be disrupted by AI. The SaaSpocalypse is upon us. We will be asking Sam Altman is software dead? Or is going to come back? Are we back?

Who knows? Claude 4.6 Opus is still best, it still has the best SVG results out of all the models, just incredibly high taste as Lysan Al-Gib. And these are some pretty beautiful pictures. Are you familiar with SVG programming? So you basically draw each square and line in code. So you have to say, I want a square that goes from this pixel to this pixel,

this pixel, and you add and layer all those up. It can be extremely time-consuming if you do it by hand. But Claude can just sort of one-shot it. And it looks pretty beautiful. On-brand. And it's an interesting benchmark because this isn't something you necessarily ask. These are not generative images in the diffusion sense. This isn't the mid-jurney model. This isn't what SORA uses. This is a different thing asking the model to basically write code that generates an image. So it really has to understand the back and forth between what it's building. So cool benchmark. I like that.

I enjoyed that. I also like the New York Stock Exchange. Do you want to change the world? Raise capital at the New York Stock Exchange then. It's as easy as that. Anyway, the Anthropic handle being owned by a guy who only posts his favorite form of AI Wordles. He only posts Werdles.

Oh, and he just has adanthropic. That's very funny from Paula Rambles. So this is a strange thing with, with X. Yes. They have policies around their handles that they really stand by in that ramp is at ramp

is not owned by ramp at Anthropic is not owned by Anthropic. And when you look at the accounts, you think is this really the best use of a handle like this in comparison? Paul Jankura who is at Anthropics is emphatically not an AI company.

Ohioan liberal bookworm news hound CLE sports and Oh, wow. He probably... Anglophile. If he's a bookworm, if he's a bookworm, he probably loves sitting on

the anthropic handle given that anthropic has been... Warming their way through some books. Chewing through some books. Much like a worm. We didn't get to that article. That was a funny article. You got to feed Claude.

It's okay. The books got to feed. And they have to... This is actually... Potentially bigger than the water thing. Like, it's just so visceral. People love books and watching them be destroyed. Yeah, if you didn't see...

It's just one version of his book. Yes, I agree. But people don't like destroying books. They're sensitive about that. The very funny thing is that Paul at Anthropic on Twitter, on Axe, has migrated to blue sky for many things, but he didn't get the Anthropic

handle on Blue Sky. He's Anthropic 42 over there. So shoot him a follow if you're hanging out, looking for wordal progress. People are playing around with the nominative determinism. Yes. Say Google's CEO is named Pichai.

His purpose is to pitch AI. Dario's last name is literally AI model. Maybe I should change my name to pitch MongoDB. John Pitch MongoDB. Choose a database built for flexibility and scale with best in class and betting models and re-rankers. MongoDB has what you need to build. What's next?

Dylan Patel is sharing some data. He says 4% of GitHub public commits are being authored by Claude code right now at the current trajectory. We believe that Cloud Code will be 20% of all daily commits by the end of 2026. That feels low to me. I'm expecting 99%. Sorry. Well, I mean, so I think the way you can see this is because like whenever you actually do a commit with Cloud Code, it adds itself. So this is like, there's a ton of people that are like using Cloud Code, but then not attributing back to the Cloud. Yes. And we also have some news, Doug O'Loflin from Semi Analysis will be joining the show tomorrow to break down Claude

Code. We might have to pull him out of his psychosis. He is addicted to Claude code. He's running multiple instances at all times and his has updated his profile picture in the semi-analysis slack to reflect his devotion to the show goth, I suppose. Anyway, so OpenAI unveils frontier a product for building AI co-workers. This is in the Wall Street journal and OpenAI also posted about it. The new platform launched amid market fears over AI's disruption to software is aimed at helping businesses develop AI agents that work alongside humans. So there are some interesting questions here about... Think of these as like AI

co-workers that are actively trying to take your job. Like they're trying to help, but they're also like want, they want your title, they want your comp. They want to learn everything about what makes you great and they want to get them. Maybe or maybe they're just trying to empower you, Jordy. Maybe they're just trying to make you a better you. I don't know. Maybe they're just coming over, just cracking jokes, trying to distract you while learning so that they can take your job. They can literally do both. They can literally do both.

Before we dive into this, let me tell you about Restream. One live stream 30 plus destinations. If you want to multistream, go to Restream.com. So Frontier works with OpenAIs previously announced AI agent building tools and makes it easier for businesses to combine sources of data that agents need to perform tasks. the AI company said. The agents will be able to process information from various sources and complete tasks like working with files and running code, OpenAI said. So no more copy, paste everything into your chat, GPT enterprise edition. It should have access to your network, plug into all your different systems. You'll be able to write API bindings, I imagine, and there might be some forward

deployed engineers or some associates from OpenAI that are helping you actually on board fully to the agentic workflows that have been promised. Yeah, they're hiring. How many consultants to help with this go to market? They are hiring hundreds of AI consultants to boost enterprise sales. There we go. Job creation. Yeah. This is a great gig.

I would highly recommend jumping on this if you're in this market. OpenAI is hiring hundreds of new staffers to expand a technical consulting team that helps large corporations develop custom AI applications and agents to automate employee tasks, according to a person knowledge with the company's. The hiring effort could help it beat back a competition from arch rival Anthropic, which has also upped its game in catering to enterprises. It comes as Open AI prepares to launch a new enterprise offering that would unify businesses' efforts to use AI. The chat GPT maker is expanding its number of technical consultants, also known as for deployed

engineers, who can customize OpenAI's model using a client's own data, a person said. These engineers can, for example, help T-Mobile develop AI to respond to customer service requests or help into it provide its customers with tax preparation services. So you have all this data, you want to do long context reinforcement learning on it. Long context reinforcement learning has been very, very successful in the coding world because Git has a complete history of every line of code that's been written, every comment, why it happened. You have this perfect record of everything that happened when you built a piece of software,

and there's a ton of open source repositories. You can download all of GitHub, basically, and see how software is developed. So you can train the model on that. Well, you can't really get the same level of free data with enterprises. A lot of this data is locked up and a lot of it's specific to a specific business process, business. Yeah, let's pull up an image. There's a little graphic here that they made. From OpenAI's frontier.

OpenAI.com slash index slash introduce. Yeah, and if you can zoom in a little bit at the bottom, you have your system of record. You have business context, agent execution, evaluation, and optimization. your agents, OpenAI agents, third-party agents, that feels significant, right? They want to be the orchestrator, right? And if you want to bring in some other folks in to help out, great, at least for now. And then they have interfaces, which they're Chad GPT, Enterprise, Open AI Atlas, and other business applications. I'm a little upset that they didn't go with a Mad Max theme, like Gastown.

I like the Polkats. I like the mayor. I like the deacon. I thought that was a fun metaphor. They went with something a little more enterprisey. But I think Frontier is a good name. I don't know. It flows.

Of course, it's not Frontier 5.2 high, so we got to count our blessings here. But I think Frontier works. Are you working on a Frontier integration project? Oh, those consultants from OpenAI, they're getting us on board to Frontier. It sounds good. Why is Tyler laughing? Tyler's laughing. Yeah, I'm getting flamed in the world.

the comment. What's up with Tyler's hair today? Yeah, what happened? Did you, did you use shampoo nor conditioner? I need a haircut. Maybe you need a haircut. Maybe you need a hat. Maybe you can grab one of those TDPN hats over there. You'll be good. Don't worry. Let me tell you about Apple Oven. Profitable advertising made easy with axon and dot AI. Get access to over one billion daily active users and grow your business today. We'll get more from Sam on Frontier. Yes. I think we can move on. So the last thing on 5.2, the last benchmark that's interesting is GPT 5.2 with high, not extra high reasoning effort, has a 50% time horizon of around 6.6 hours on our expanded suite of software tasks. This is the highest estimate for time horizon measurement we have reported to date. And it's right on track, doubling every 128 days. It depends on what you're looking for. But the implications of this are more and more tasks. are suitable for these models, although I do think we're going through a shift with the orchestration thing where we might need a new benchmark, because what happens if the individual model, GPD 5.2 high, can do one six-hour task, basically,

but you can deploy five that talk to each other, and when you combine those, it adds up to 20 hours of work. You just get a jump in the graph, but it's not technically the model, so you need sort of a new benchmark, supposedly. Absolutely. Jessica Lesson likes Sam Altman's post. She says, this is excellent comms from Sam Altman, an eye-popping stat in an easy-to-understand claim. And I hate, though, this budding movement to frame the business model debate as elitism versus anti-elitism when chatbots launched with subscription business models, I thought it would

be a great new era of consumer tech blended with ads and subs, much as we've seen with Netflix and Spotify. It would be a shame if the OpenAI versus Anthropic rivalry wipes that away and makes it seem like subscriptions are only the domain of the elite. So many other consumer businesses show otherwise, and they are better off for the balance. I like that. Signal says either you or I have lost the ability to process reality.

Yeah, I mean, not a lot. In response to this? In response to that. Okay. Yeah, no, a lot of people didn't think this was great comms at all. Yeah. I mean, I would say, like, most of the posts I saw were like, hey, like, this feels like you got baited.

Yeah, yeah, because you could have just not responded, I guess. Yeah, or you could have said, he said later in the day, I'm excited to share that Codex has one million users. Yeah, yeah. Which is the product they launched this week, right? Yeah, yeah, just focusing yourself maybe. I don't know. Meanwhile, over in Silicon Land.

Silicon Land. Greg says GB200 has really been enabling us to do some amazing things. Yeah. It is remarkable how, you know, relatively slow the rollout has been. Obviously, it's been very quick. But it does take time from the time that Nvidia announces these chips to actually fabbing them, racking them, wiring them up, getting ready to train new models.

And so, anything else? Yeah, I mean, I think Dean Ball had a new post today and said, like, you know, companies don't even have Blackwells yet. Yeah. So we even seen, there's no models that have been trained on Blackwell. Yeah. Pretty great. So it's like, get ready.

Get ready to see some may I progress. We got Mike Isaac in the chat. He says, not good comms. Not good comms. I'm with Mike. Expand, Mike. I'm with Mike. that one.

We'll see. Let me, I'll look up Mike's. And I will tell you about Google earnings because we touched on this, but it's important in the context. So Alphabet sales hit record, spending to double. They're going all in on AI.

Google Parent Alphabet reported an 18% jump in fourth quarter revenue driven by growth in digital advertising. Before we go into this, Mike had a banger yesterday. He said by the number of Open AI employees I see tweeting about

Super Bowl ad Anthropics should be paying OpenAI earn media fees. Yeah, that was a good take. Got them. That was a good take. Got them. Sales reached nearly $114 billion ahead of analyst expectations. Net income, $34.5 billion, a 30% increase compared with the period a year earlier. The company reported a record $403 billion in sales for 2025.

Profit, $132 billion. Not too bad. Alphabet shares were sort of all over the place. The Wall Street Journal printed that they went down 1% and after ours traded, but there's a lot going on. Google, like other technology companies, plans to spend tens of billions of dollars to develop AI models and build the data centers needed to train and run them. The company said it expected to spend between 175 billion and 185 billion in CAPEX in 2026, up from 91 billion to 93 billion in 2025.

So they're like doubling, which is it's exponential. growth maybe, get ready for some AI progress. Yeah, somebody, I don't have it pulled up, but somebody was saying that the 2026 projected CAPEX will be more than the lifetime CAPEX for Google up to 2021. So in a single year, they're going to eclipse that, which is just insane. Bucco had a good take. He said, Google CAPEX on purpose, tell the market, this is what it will take to defeat us before IPOs hit.

Yeah, certainly nerve-wracking if you're competing with them, but they have the edge on the capital side and the capital war over Anthropic and Open AI, but the race is still real. Yeah. Let me tell you about Phantom Cash. Fund your wallet without exchanges or middlemen and spend with the Phantom card. The Wall Street Journal had some more context on Google. Google leans hard into its AI winter status, AI winter status, not AI winter. Add and cloud growth acceleration justify the recent surge in alphabet stock, but blowout

CAPEX forecast still takes one's breath away. The motto for the artificial intelligence race today should be, if you've got it, spend it. That's a message that meta-platforms took to heart during its fourth quarter report last week when the Facebook and Instagram parent announced plans to spend up to $135 billion on CAPX compared to about $72 billion last year. Google managed to up the ante Wednesday with its own plan to. spend as much as $185 billion this year, which would be about doubles last year's outlay. Google's annual revenue has now topped $400 billion, about twice as large as metas. Still, that new spending target, even for a company that has been firing all cylinders lately,

takes one's breath away. The stock price slipped in after hours trading after its fourth quarter report and conference call. Google has both the political and financial capital to lay such a bet. The company's Gemini 3 model has put it on top of a heap of performance for AI models, while the unmatched distribution of its search engine and products like Gmail have quickly driven adoption. Google said Wednesday that it has more than 750 million monthly active users just on its Gemini app, which only represents a portion of Gemini's actual users because it's vended into all the different products. What's going on in the chat?

4-0 Army has entered the chat. Interesting. You're hitting the chat with hashtag Keep 4-0. They want to be heard by Sam Alman. Well, it's not Keep 4-0. Hasn't 4-0 been deprecated? And so shouldn't they update the hashtag to say, like, bring back 4-0? Revive 4-0?

It's actually not leaving until February 13th. Oh, okay. So there is time to reverse the decision. I am interested to know about the operating cost of keeping 4-0 alive because it feels like if it runs on old hardware. Well, that was clearly something people were willing to pay for. Yeah, can you just kind of like to find a corner and just like a handle runs? It was, you know, maybe more of like a user. behavior decision than a financial one, because it seems like running legacy models gets really profitable.

Hunter Rice says cancel for a. Okay. A strong reception of Gemini, along with Google's victory over the federal government's efforts to break up the company, have cheered investors when sentiment on technology and AI is faltering. Alphabet stock price has jumped around 20% in the last three months. Nvidia, Microsoft, Amazon.com, and Broadcom have all lost ground during that time. But capital matters too. And here's where, and here is where's where Google's business model pays off the most. The company's advertising arm has long been a lucrative cash cow that is still managing double-digit growth rates. Growth is actually accelerated with ad revenue up 14% in the fourth quarter compared to 13%

in the previous one. That's exactly what we saw in meta. The ads are getting better, baby. The company's cloud computing division was even more impressive with revenue growth jumping 48% year over a year to hit 17.8 billion. Google Cloud hasn't seen growth like that since early 2021 when the business was still less than a third of its current size. Google Cloud turned in a record 5.3 billion in operating profit in the last quarter, a figure that was 45% higher than Wall Street's targets. The company's booming business produced nearly $165 billion in operating

cash flow in 2025, the highest in the S&P 500. These strong results will help invests. digest the latest investment plan, but spending what could be 40% of annual revenue on AI chips and related infrastructure is still a sizable gamble. Such investments will sharply elevate depreciation charges, which in the latest quarter reduced net income by 18%. Google isn't the bargain it was less than a year ago when breakup years had the stock trading at less than 16 times projected earnings. My question is, like, when does the AI buildout stop, because if you're constantly investing more and more, where does the cash flow come from? Like, as an analyst, you look at this and you say, okay, they're spending, you know, 50% of their

revenue or 40% of their revenue on the AI buildout on data centers, but how long will they need to do this? Like, if they have to do this forever, then you just permanently have a worse business because you're just constantly buying the hardware. Yeah, I mean, you get a massive capability increase, lots of labor moves into data centers, and eventually maybe you can enter a scenario where it makes sense to continue to increase CAPEX because revenue is accelerating even faster. Yeah, yeah.

Yeah, notable that Nvidia is still, you know, trading down today even after that update on earnings. Like you would expect people that look and see like, hey, you know, Jensen had given some pretty kind of wild projections, right, looking out over the next couple years. Yeah. And you have to imagine that. Google updating here makes it a lot more. I mean, we'll spend a lot on TPU, but still, they will be buying a lot of NVIDIA chips for sure.

Really quickly. Let me 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. Richard says Google is a company that doesn't do hype for them to go and increase Cappex from 90 billion to 180 billion is probably the most bullish thing long-term investors can see as it shows the scale of future revenue growth. I'm shocked that at this stage most still don't understand this.

Yeah, so excited about it. What did Doug say over at Fabricated Knowledge? He said, Google made 163 billion in CFO last year and is now planning to spend 180 billion in CAPEX next year. Now, deaf think acceleration is coming, but wauza. I'd rather go bankrupt than lose the race, quoting Sergei. The trailing 12 months of free cash flow is staggering over there. Going to make $200 billion in cash flow this year, still less than CAPEX for now. Also going to have $80 billion in net cash and probably $100 billion in myriad minority illiquid equity stake.

So, yeah, they own like SpaceX and a bunch of... Waymo. With current rates of growth, they can raise $300. billion in 2030 without even going into debt. Totally agree, but this feels pretty close to the pedal to the metal to me, says Doug. So, uh, Joe Wisenthal says the average person on earth is watching 25 YouTube shorts every single day. YouTube shorts average 200 billion daily views. That's a lot. That's just insane. That's a lot. I mean, you can watch a YouTube short in like, what, five seconds on average, right? Because you skip one, you watch one for five seconds, one for 15 seconds.

I mean, 25 YouTube shorts that feels like maybe five or 10 minutes of browsing, but still like pretty, pretty staggering. Yeah, I haven't, I haven't migrated any, like I don't scroll on YouTube shorts yet. Yeah. I find if I search for something, like let's say I'm looking up a car, it's nice to get a 60 second explanation of it. Yep. If I don't have time to like watch a review. Yeah. But I'm not like sitting there scrolling.

Yeah, it still serves. Clearly a lot of people are. But I rarely scroll through, whereas on Instagram I will scroll the feed of real. Yeah Staying in the content world Kalshi shared that just in YouTube generated over 60 billion last year More than Netflix

And Polly Mouse says YouTube is beating Netflix With this really sneaky content strategy In which their creators make stuff People want to see and are then rewarded For it with views and money And it is a simple encapsulation of the YouTube strategy Netflix hates this one

simple trick UGC UGC is a big big business Who would have thought? But yeah, the scale of the quality. I'm hoping we're going to talk to Mr. Beast about his outlook on YouTube and how it's changing soon because he's taking it in such an incredible direction where the content is Netflix quality. I mean, he sells it to Amazon Prime, right?

So he's certainly there. In the Antitrust Division of the Department of Justice posted an update today, the DOJ, antitrust division filed notice that it will cross appeal from the remedies decision in its case against Google's unlawful monopolization of internet search and search advertising. So there's there's still Hashin that out. Let me tell you about label box reinforcement learning environments voice robotics evals and expert human data label box is the data factory behind the world's leading AI teams. So big tech is is throwing cash into India in its quest for AI supremacy. We talked about this story a little bit yesterday we can get to the bottom of the wall. Yeah, Bloomberg, yeah, Bloomberg had a story that's about the facts.

Yeah, effectively Google's scaling square footage. Yes. In India pretty dramatically. Yes. And the question was, you know, why are they scaling it up at all? They're investing so much in CAPEX and servers and digital workers. But of course, they need humans as well. So American tech behemoths are racing to establish leadership in artificial intelligence, not just in the U.S., but also around the world. India is welcoming them with open arms, says the Wall Street Journal. India is becoming one of the hottest markets globally for AI, for US AI Titans, looking to cater to the country's massive and digitally savvy population, looking to attract more tech investments. The Indian government announced plans of the weekend to give tech firms a 20-year tax break on overseas revenue gleaned from global data services based in India. So you go and you set up a data center or you set up a business, a tech business, you're making money off of the, that whole data center or whatever you're selling in terms of software, and you don't have to pay taxes to India for 20 years. That's a long, long time.

Techno chief called this out yesterday. Yeah. The move is part of the Indian government's push to make the country a major provider of AI services, including low-cost tools to solve local problems while leaving cutting-edge innovation to depocketed firms in the U.S. and China. This will give India the opportunity to become a major AI hubs, said India's technology Minister. In the past few months, U.S. tech companies have unveiled tens of billions of dollars in investment in Indian data centers as they race to build AI infrastructure around the world. In October, Google announced a $15 billion investment in data centers in Southeast,

southeastern India, as well as undersea cable links in what the company describes as its largest single AI hub outside the United States. In December, Microsoft unveiled its largest ever investment in Asia with a $17.5 billion pledge to develop the country's cloud and AI infrastructure. also pledged 35 billion across its operations in India up until 2030. The U.S. data center firms known as hyperscalers are drawn to a country whose 1.4 billion consumers are some of the most prolific users of data and AI chatbots. Very interesting. Anyway, there's more news from Nvidia in the information. NVIDIA is delaying a new gaming chip due to memory shortage.

This is the line you don't want to cross. The gamers are going to rise up and they're going to going to storm the data centers and they will be putting the screws. They will get their chips one way or another. They will. They'll be calling their representatives, calling their senators. Pause AI has a new cohort. If you don't give me a new gaming graphics card to play the latest games. Yeah, so it's the first time in 30 years.

It won't release one in the calendar year. Yeah, every year. It's over. 1080TITI, 2080, 3080. These were great graphics cards, and now no longer just delaying it. I mean, I guess you can use cloud gaming or something, but also it does feel like a lot of video games have sort of maxed out what they can do with stock graphics cards, and so there isn't as much need to upgrade every year. What do you think, type?

Yeah, I mean, also with Genie 3, it's like we're getting new advancements in gaming. Yes, yes, yes. So they should be happy about this. You'll own nothing and be happy. How much GNI 3 did you play yesterday, Tyler? It's fun, right? It's fun. So you play it?

I thought about playing it. Oh, okay. You thought about playing it. But the mechanics weren't there, right? We need mechanics. Genie 4, I'm going to be DAU. That's the goalposts currently. I want mechanics.

I want tire changes, gas refill, as I'm driving the Nureberg ring in my Gini 3 car simulator, race simulator. Anyway, memory chips are a key component of GPUs. Very excited to ask Sam. about the various pieces of the AI bottleneck since... We should ask him about world models too. That would be interesting.

He's not one to not launch a product. Yeah, he's launched a lot of stuff. And, yeah, Ben Thompson has had this take for a while that AI-generated video is the metaverse. And that's what you're seeing with meta-vives. That's what you're seeing with Sora. This is sort of where things go. NVIDIA is also slashing production of its current line of gaming chips,

the G-Force RtX-50 GPUs because of the memory shortage, one of the people said. Prices of NVIDIA's latest gaming GPUs have already risen at retail stores and websites due to their scarcity over the past year. I wonder how much it costs to actually build a modern gaming PC these days because it's been a while. It used to be like a few thousand dollars would get you sort of like a top tier setup. And I wonder what it's looking like now with all the shortages. It's possible to be sure that Nvidia executives could still change their mind and release a gaming chip if the market improves as the company is known for being flexible and moving quickly, says the information.

Demand for computer memory chips has skyrocketed due to the AI boom as they are needed in large quantities to train and operate machine learning models. Memory chips act as warehouses for storing data. The memory chip shortage is expectedly to higher prices in consumer electronics. Last week, Apple CEO Tim Cook said the rising prices of memory chips would have an impact. on the company's March quarter margins. He's also saying that he's having trouble getting line time at TSM to some extent. And so there's a lot of bottlenecks

that are working their way through all the technology markets. He hinted that the impact would be greater in the future, noting we do continue to see market pricing for memory increasing significantly. As always, we'll look at a range of options to deal with that. Gaming and AI chips use different types of memory, but both are made of the same raw materials,

coming from one of three months of three main suppliers at Samsung Electronics, S.K. Heenix, and Micron Technologies. Well, without further review. We have our first guest to the show. Sam Altman, the CEO of OpenAI. He's in the Restream Waiting Room. Welcome to the show. Sam, how are you doing? Welcome back. Good. Thank you guys for having me back. Thanks so much. Big day. Big day. Kick us off with, where should we start? Should we start with the model or frontier? Can we start with the model just because I'm having something? Yeah, absolutely. Break it down. What did you launch?

So we launched 5.3 Codex is, I think the best. coding model in the world. We took a lot of the feedback that people had about 5.2, 5.2 codex and got it into one model. It is much smarter at programming, but it's also way faster. You can interact with it mid-turn. I think it's got a much better personality. It's really good at computer use. So it feels like a very big step forward. It was funny as we were deploying at this morning. A couple of very, like, extremely experts at using these models noticed and said, man, something's really different with codex. And they like caught it mid-deploying. So I think you can really feel it quickly.

Oh, you're saying people outside of Open AI just everyday use it. Yeah, I like the half hour that we put it out before we, you know. Talk about interacting with a mid-turn. How does that work? Why is that important? What does that unlock? So people are trying to use these tools for very long pieces of work at one time, you know, multi-hour tasks.

And sometimes you don't specify it correctly, right? Sometimes somebody's not set up right. Something it's not just screws up. The ability, they can do amazing things with no steering, but they can do much. more amazing things if you steer them along the way. So this has been, this is one of the things I felt most new about this model. So talk about orchestration, how this fits into frontier because I imagine that. It's notable, like if you see a coworker making a mistake and you don't interrupt them,

that's rude, right? It's deeply inefficient. It is incredible what these models can do without any feedback. Like if you think about a new coworker especially, you know, you train them and you give them a lot of feedback early on and they learn the job. and you correct them, and they kind of get practice. And the models, they will soon do that. But right now they don't do that.

So we just rely on either they get it right one shot or we collect them, correct them along the way. Yeah. So I think there's a lot of people that are running multiple agents and multiple tabs. They're starting to think about orchestration. It feels like frontiers a piece of that. But if you're interacting with a model that's running mid-turn,

like how does the user experience change for developers with 5.3? And then what will it look like in the frontier world? I think we will be heading towards a workflow where a lot of people just feel like they're managing a team of agents. And they'll keep, as the agents get better, they'll keep operating at a higher and higher level of abstraction, which at least watching what's happening so far is a jump that people are going to make pretty well. The models are so good. Now there's such a capability overhang that building better tools to let people,

do that, which the Codex app that we launched on Monday was a great step for for, will be very, very important. But you will be managing very complex workflows. The agents will keep getting better. So you'll keep working at the maximum of your management bandwidth or cognitive ability to keep track of all the stuff. And the tools to make that easy to do will matter, I think more than intelligence for a little while because there's such an intelligence overhang already. What's the role of a Ford deployed engineer today at open? an AI towards the end of the year. Capability overhang that feels like raw meat for a for a forward deployed engineer. It's like the like they solve that problem. Right. Yeah. I mean,

look, eventually the models will get so good that they'll help companies deploy themselves and the forward deployed engineers will again get to work at a higher level of abstraction. But for now, you go into a company that is not AI native and say, okay, you've said you want to say they want to deploy AI. They really are not sure what to do. How do I hook this up to my systems? Do I need to find a model on my code base? How do I think about orchestrating agents and using things from different companies? Most of all, or at least what we hear most frequently, is how do I think about security of my data and how do I know that these like AI co-working agents are not going to go access a bunch of information and share it in ways they shouldn't or get a context exploit or something

like that. So the forward-deployed engineers take this incredible new technology and a platform like Frontier and say we will connect your company to an AI platform so that you can use all these agents and workflows and everything else you want. How important are these metaphors or how temporary are them? I was very interested in reading about Gastown and you have these pole cats and it's this whole Mad Max world and that feels like maybe just a temporary aberration where you're setting up agents for specific tasks, but also that could be incredibly valuable in explaining to a large corporation of how they're going to integrate AI across the whole organization. Yeah, I suspect like everything else that's happening when an industry is moving so fast,

all of this is somewhat temporary on like a long enough time scale. And as these models become more capable, these agents are operating on very long time Horizons with the ability to just kind of figure it out and our trust in their robustness keeps going up, then maybe you don't need a lot of the abstractions we need today. You know, maybe you just like have a single AI bot that runs at your company and you can say, hey, I'm going to like launch this new product and it does everything an ambitious person would do. But that's not where we are today.

So today we have to put in a little more work to get the pieces put together. Yeah. How are you thinking about the meter benchmark for long task horizon? You're at the top of the charts. At the same time, it feels like we might need a new chart if we're talking about agent swarms because they'll be able to do things that go for weeks, but they will subdivide the work. There's some subdivision that happens within a reasoning model, but it doesn't truly parallelize, at least that I'm aware of.

So what does it look like in a world where you go to a model, but now it's spinning up a whole bunch of different models underneath? I think there's two of the key insights of the whole field are in this question. Number one, the implication that no chart in AI lasts more than a few years is right. And like this one, kind of, you know, we'll see how much longer it's really useful. The second is a lot of people thought that, okay, we're going to just need a super long task. And a super long task horizon, so we need a super long context. And definitely what people have already seen with coding agents is by agents breaking up work, orchestrating it well, farming it off to sub-agents, even with the current,

of the technology, we can do something, which should not be surprising because it's similar to how people do things and get amazing amounts of work done. So that's been cool to watch. I think that will keep going. A joke that some people at open eye make is that soon the chart that matters is just going to be GDP impact. And then the question is what's the one that comes after that. But everything else, a lot of these proxy metrics, there's now so much economic value in what the models are doing. What do you think could come after? I have no idea. Do you have an opinion? Happiness? I don't know. When you look back at some of your, when you look back at some of your blog posts from 10 years ago, your predictions were usually pretty on point. Yeah. Maybe it's hard

to predict. Thank you. Thank you. I, uh, specifically the, specifically the merge, like basically calling. I mean, there's a bunch that are good. Like right now we're getting, you know, thousands of messages in the chat about 4-0 and you predicted in 2016 that people could, uh, would, would become, uh, you know, very attached to a chat bot. Yeah. I'm working on like a big prediction blog post for the next 10 years seems too far, but the next five. Sure. But because it's like, I'm sure a lot of it'll be wrong and it, you know, it's still fun to try. The sort of like relationships with chatbots, clearly that's something that we've got to worry about more and is no longer an abstract concept.

Even the question of what comes after GDP, like one reason that's the reason that. And I think that's interesting is the way we measure GDP now could start going down, even though quality of life goes way up. Sure. And we don't have a lot of practice with things like that. But the massively deflationary. Isn't that just Europe? Not just Europe.

We want the quality of life going up, too. Yeah. Just going down, you know. Switching gears, what do you think about the work that the NeoLab boom, the research efforts that are happening all over Silicon Valley? It feels like there's an acknowledgement that there's research breakthroughs that need to happen, and everyone's taking different shots at those. Do you think that those companies will just find a breakthrough and join a lab or launch their own products?

I think it's fantastic, first of all. One of the meta things that we wanted to do when we started with Open AI is when we started open AI is like there had been a period where the technology industry and Silicon Valley in particular was amazing at new research labs or just doing new research. research in industry in general. And then it kind of fell apart and there hadn't been a good one in a while. And part of what we hope to show, and this was not only us, like a lot of people were excited about new research labs, is that industry could do research again. So seeing that now become fashionable and all of these new labs, I think it's totally awesome. Some will succeed, some will fail, some will go into some other effort. But having industry support research in, you know, startup style, I think it's wonderful. Over the next two years, would you expect to acquire

more individual product companies or more research labs? Good question. I don't have a strong opinion there. I would bet, well, I'd say the very best ones will often look like a mixture of both. The one that I have in mind right now is something that very much looks like a mixture of both. So maybe the shape of things to come is the really truly extraordinary product work will have more and more research component. And it'll be kind of a more of a hybrid thing. Is data the new oil?

Is there a value? Yeah, we were joking a couple days ago. A bunch of data, but they don't understand AI. They don't know how to monetize it. Well, and that phrase was effectively wasted a decade ago. Yeah. And so to say it now sounds really silly. Yeah.

But it feels like it could be more true now than ever. Yeah. You know, certainly, yeah, man, they really did waste it a decade ago. I was just thinking of the kind of people that used to say that. Yeah. Ted talks. You're not supposed to call them out. They're nice people over, Ted.

You know, definitely like the sort of the magic relationship of this last eight years, whatever you want to call it, has been like, you know, that we can put in more and more resources, compute data, new ideas, whatever, into creating an artifact. And it gets like, the log of it gets better. So you can throw in, you know, that's why we have this huge. exponential increase in resources, but we keep getting better and better models. And for all of the concern people have about it's going to top out or it's slowing down or

whatever, like no one's been right about that. I mean, sometimes they, it looked like they were for a couple of months as we digested a new model or came to a new form factor, but it has been an incredibly smooth last six or eight years of this. What those resources are, there can be some trade-off between, you know, sometimes it's better to spend your money on better data, sometimes on more compute, sometimes something else. On the whole, uh, compute power is the new oil is the statement that feels closest to true to me, but there will be other parts too. Is software dead? It's different. It's definitely not dead,

um, but what software, like how you create it, how you're going to use it, how much you're going to have written for you each time you need it versus how much you'll want sort of a consistent UX. That's all going to change. You know, there have been a number of these like big sell-offs of SaaS stocks over the last few years as these models have rolled out. There, I expect there will continue to be more. I expect there will be big booms in software. I think it's just going to be volatile for a while as people figure out what this looks like.

The statement that someone said to me that is stuck in my mind most these last couple of weeks is that every company is an API company now, whether they want to be or not. Oh, yeah. Because agents are just going to be able to... Yeah, we had DARA from Uber on yesterday, and he had a pretty refreshing kind of approach. We were asking about integrating agents with Uber,

and he recognized that, yeah, the ad business could potentially be threatened if you can order an Uber and chat GPT, but he basically said, like, you have to think of the consumer. The consumer wants to order an Uber via their preferred agent, you should let them, otherwise you're going to have other problems. Yeah, that is the right take for sure. And I don't, or I think so at least. And, you know, we've been through platform shifts like

this before where you, I mean, Uber wouldn't have existed without one. It wasn't until iPhone where you could like have it make sense to order an Uber to write where you were as you were out in the world. So I think there will be totally new things that happen. Other things you'll use in new ways. but definitely as I've started using Codex, my excitement about having agents go off and do things for me and still use other services, pay other services. I'm sure we'll need to figure out new business models and how revenue gets shared around, but that will happen.

Yeah, talk more about Codex desktop. One more question on SaaS. Have any public market SaaS companies tried to get a soft landing with Open AI? And do you think there's any value? Just let's say I'm a No one has no public SaaS companies

that I'm aware of have tried that with Open A Look at some of them Some of them Some of them Some of them will certainly be durable and

Are on sale right now And potentially just need new energy And need kind of a new An entirely new approach And maybe Open AI could provide that Yeah Yeah I think some will be incredibly valuable

Some do feel like a thinner layer now But I don't know I don't know. Like I was talking recently to a bunch of SaaS companies, and they do not, they do not feel unexcited. Like they're like, we're going to go through a big transformation here. And, you know, yeah, sure, other people can instantly write software now, but so can we. And we got a great system of record. And it seems reasonable. Some won't make it, of course.

Yeah, yeah. Yeah, talk more about the codex desktop rollout. It feels like, you know, successful amount of downloads. but key, like a shift for, you know, people who are maybe lightly technical, but don't have time to set up an IDE and configure an environment to actually start writing software. I want to know about plans to integrate to the phone. That was a big moment, I think, for a lot of people with the clawed bot, molt bot, open claw thing was like, oh, I can text something and it will go and write code and that's valuable. And that unlocks a new agentic experience. Like where do you see the Codex desktop ecosystem going?

I am. So Codex desktop has been a somewhat of a surprise to me in terms of how much people love it, including how much I love it myself. You know, the, I think it's a great example of 10% of polish of the experience of using these models, especially when there's so much capability overhang, goes an extremely long way to what you can build and how you interact with this stuff. of course we should have an ability to kick off new tasks from mobile and we'll do that. I mean, really what you want is like your single AI that's working for you on a unified backend,

access to all of your data and your ideas and your stuff and your memory and all the context and the ability to work across a lot of surfaces and often you'll be at your desktop. Often you'll be on your phone and you just want to add something in. But it is a pretty profound shift in my own workflow. not just for coding tasks, but more general purpose tasks. It's still kind of hard to use if you're not at least reasonably technical, but obviously we'll find a version of this product that can do other knowledge work tasks and control your computer and things like that where you don't have to be.

And it'll bring the magic of building stuff really to a lot of people because even if you never look at code, you'll be able to build something reasonably sophisticated. One of the things that I have built when I was playing around with the new Codex app is this thing I had always wanted, just like this magic auto-completing to-do list. I really work with to-do lists and this idea that I could just put tasks in and it would try to go do them. If it could complete them, it could complete them. If it needed questions, it would ask me questions, if it needed, you know, if I had to do something I could still do it the old-fashioned way. But an interface like that where, you know, all the stuff you want to do, you just sort of explain to a computer or your AI and it tries to go off and do them.

and sure if you're on your phone, you're going to just add a task on your phone or, you know, if you want to easily import some from email, you're going to do that. Like, it feels really good. So I'm excited about all of the ways that this will just become a general knowledge work agent. Yeah. Were you unsurprised to see a product like OpenClaw come from open source? Because it's certainly that kind of user experience is not, I would imagine this is something that you knew would be a thing, and yet I think part of the magic of open claw is that it would

be very, very difficult for a large tech company. Peter didn't make many phone calls to hypers to say, hey, I'm going to be integrated with your API. It just went out. And you guys, you know, and when I think back on like the sky acquisition, this kind of experience was probably very top of mind and things that you're working towards internally. I love the spirit of everything about open claw. And you are totally right that it's much easier to imagine a one person open source project doing something like that than a company who is going to be afraid of lawsuits, data privacy and everything else. You know, they're like, I think this is kind of how innovation works. Something like that starts. It's clearly amazing. There will be a way to make a mass

market version of that product. But letting the builders build the equivalent of the homebrew computer club spirit go here is so important. Yeah, totally. Can we switch to social? I feel like if I Google Sam Altman's social, I get pure AI and SORA and then also demand or predictions about a human only social network. Where do you see social going broadly? How do you want to integrate it with it and power it in the future? The moldbook thing was like a very interesting social experiment to watch. And I think points to agents, interacting in some sort of social space, hopefully on behalf of people, at least in some degrees, could be quite interesting. I don't think we know what to do there yet, but it feels like social is going to change a lot. And I am interested in the space of what a social experience can look like when your agent is talking to my agent and coming up with new stuff.

Clearly, putting like a lot of AI bots on the existing social platforms is just making everyone crazy and not that fun. So that's not the right answer. But I think we can design something new for what this technology is capable of that will feel good and useful. Yeah. Is there is there a solution to the bot problem that's just all the labs sort of integrating with all the other platforms and and even if you can't detect its AI generated, you can literally say we just, we just generated those tokens. Like those exact tokens are in our database. You probably can't do that because their open source models are like good enough to write things at this point. Yeah, yeah.

I am excited about sort of like assertion of humanity instead of trying to like detection of AI as a thing here. I don't know if the social platforms, it's like in their interest to solve this. It is causing, it's creating like, at least in the short term, it creates like a lot of engagement and increased usage. So I believe they could solve it if they wanted to. I'm not sure it's in their interest. I'm actually not even like it is, I don't like it,

but it is some people do seem to enjoy it. Yeah. Can you talk a little bit more about where SORA, as a video generation model is going? It feels like tool use is maybe under-discussed, you know, adding reasoning. It's not just the diffusion model. It's giving these models the ability

to make linear cuts and overlay motion graphics. And when I scroll the Instagram rails that I see, they're like vibe reels with cut, cutouts and it flips negative and it's all color graded and stuff that like, yeah, you could probably diffuse it all, but it's pretty cool just to teach a model to also use After Effects or whatever video, you know, motion graphics suite you want to use. Is that an interesting unlock? What do you see going on? So all of that stuff will happen and I agree with you the models will get really great at doing

that. People love generating videos. I would say people, we have not yet found a way that people really love watching other people's videos. This is true for a lot of other AI. Like they love to, you know, people love talking to chat GPT or whatever. It's not that compelling, but for most people to like read other people's chat dbt generation. So I think there is something. But isn't that for all writing and all video? Yeah, it seems stronger to me in this case than the general case, but maybe you're right. Maybe this is. Yeah, but if somebody says, hey, I generated a 15 minute video. I'm really excited for you to watch it and you watch the first 10 seconds and you're not that captured by it. I don't care that it was human made. Yeah. Maybe you're right. And this is,

it's not a, it's not a special case. Do you see that in the data with, with SORA downloads? Because I've noticed that I'll generate stuff on Sora, download it and share it to a group chat. And then it's this little in-joke that me and five other people get. And we see this like family group chats of, oh, it's our dog and our kids. But like, there's not really like, okay, yeah, this is a business. You know, everyone likes this. Absolutely. I would say that the most, common use cases, something like that. Sure. Like memes on group chats is a real killer use case.

How is the Disney rollout going? I was super excited about it. Jordy was extremely bullish on it from a business strategy perspective. Yeah, when you look at how image models have grown various LLMs historically, and now you're going to have an image and video model that can do something that no other LLM can do at least legally. Yeah. And is Bob Eiger joining Open AI?

Bob I love it. No, no credit plans for that. He's going to be looking for a job. Free agent. He's a free agent. Hit him up for us. You know, do some recruiting. That'd be great.

The, I think that generating characters in images and videos is going to be very important to people and they really like that. Like other, like we were saying otherwise, I don't think many people like want to watch me and some Star Wars character doing something together. But I might think it's cool. Yeah. It's, you know, there's like a real trend going on right now with chat GPT, where it's make a caricature of me and my job based

off everything you know about me. And those kinds of things, people actually do like looking at other people's media. Yeah, it's almost like a face filter or something. There's like enough of, it's the studio Ghibli moment. Like there's enough of the human still in there that it's not, it's not, you can't, yours is not the same as mine. So it's still personalized. It's personalized and it says something about you and, you know, the, the like a lot of these things, things, a lot of what's gone viral before with ImageGen, I think it's like if you can make people look a little bit more attractive or cool than they look than they are in real life. Yeah.

Without sort of having to ask for that. Yeah. How are you thinking about the actual rollout? We were debating between like open the floodgates. You can generate any Disney property versus like it's one character at a time. And everyone's posting Spider-Man. And then it's, you know, Mickey Mouse week. And there's another viral moment.

I'm not sure what the team is planning there. I know Disney's had some different opinions about what they want to do and try to be a good partner there. But I'd be excited to open the floodgates personally. Oh, that'd be fun. Cool. Talk about your first, speaking of video, your first Super Bowl ad, it felt like not generated, lots of motion graphics, the black dots coming together. Like, what was the goal with that ad? Who were you trying to speak to?

It didn't feel like a direct response QR code download the app. No. What was the mission? I love that ad I think that was such a cool one it was clearly not meant to be like a mass market or direct response ad

but you know speaking to the like people who are at the center of this revolution and just trying to like celebrate everything that has come before and everything that will come after yeah

we didn't hear a lot about it from like average users of chat GPT but we heard a lot about it from like researchers in the field and a lot of resonance there. It was definitely not generated. It was done the old-fashioned way. And, you know, it had, like, a lot of people loved it, a lot of people hated it.

And then many people in the middle didn't get it. And I felt okay about that. That's a great, absolutely. I like our ad for Sunday. Okay. It's about Codex, no surprise. Yeah. Talk about the evolution of the advertising to be more just clear about the actual use case, the value.

Like, what are you trying to say with your advertising? strategy now as it refers, as it relates to like video? Well, the thing I would most like us to say, and I think this is a new challenge, given where the models are, is to teach people what they can go do with AI. Yeah. I mean, AI is now unbelievably capable, and most of the world, it's still like asking basic questions on chat GPT. Everyone can go build amazing things now.

Everyone can go do all kinds of work. Scientists are going to make new discoveries. And I'd like to, you know, to the degree that advertising we do can teach people how to use this, I think that'd be awesome. Yeah, so the KPI is like reduce the capability overhang broadly. I think that should be a general KPI for us, not just of our ads. The products that we build, how we teach people to use those products, like the, that feels very important. Yeah. Anthropic also has a bunch of ads in the Super Bowl. It seems like they might run a ton.

Damn heard. What do you think that? of getting wrong about their characterization of how ads will roll out in chat apps? Well, it's just wrong. The main thing that I think is we're not stupid. We respect our users. We understand that if we did something like what those ads depict, people would rightfully stop using the product.

No one, like our first principle with ads is that we're not going to put stuff into the LLM stream. that would feel crazy dystopic, like bad sci-fi movie. So the main thing that's wrong with the ads is, like, using a deceptive ad to criticize deceptive ads feels, I don't know, something doesn't sit right with me about that. I asked Claude what if the definition of playing dirty, and it said, what did it say, misleading others about your intentions,

hiding information, or creating false impressions. Yeah. I thought it was a little dirty. I thought it was well played, but it was it was it was well played for sure and it was a funny ad and they you know like the sort of the stuff about the chat GPT personality that most annoys me which will fix very soon. I thought they nailed in the ad. So that part was funny. Yeah. But I don't know. You know like I also like I think it's great for them not to do ads. We have a different shaped business. I did notice that they said in their thing like we may to revise this decision and we'll explain why if so. Yeah, the blog post, the blog post was kind of did a good job of disarming. The pro ad people gave themselves an out in the future. Do you think they care? I think it doesn't matter.

Like, I think it's a sideshow. You know, people are excited for a food fight and between companies. But like, the amazing capabilities of these models, the product, the kind of, of like the groundswell of excitement around Codex, that feels way more important. How do you stop the pausing that happens in voice mode? Do you need new hardware for that? Or is it a model capability thing? We need a new model. We may need some new hardware too, but mostly we just need a new model. I think we'll have a great voice mode by the end of this year.

What's the bigger bottleneck energy or chips? It goes back and forth. Right now, again, it's chips. Chips. Is there anything? But it, you know, be different, different times. Is there anything we, like society, America should be doing more aggressively to increase the supply of fabs? Yeah, I think it is, well, it may get self on its own. It may, like normal capitalism may solve it. But I think somehow deciding as a society that we are going to increase the wafer capacity of the world and we're going to fund that and we're going to get, you know, the whole supply chain and the talented people we need to make that happen will be a very good thing to do. Do you think there's an upper bound on model IQ?

Like the race right now is you're smart, but you're not smart for days. You're smart for hours. Can you go much further and get much smarter? It seems certain. Upper bound, I don't know. I don't know how to think about that question yet. Yeah, I can't even reason about what 2,000 IQ looks like. You know, like, I don't know what that means.

It's funny you say, I mean, I can't reason about what it means. to think about a problem for like 10,000 human years. That's another good one. Yeah. That's crazy. But maybe IQ is going to feel even weirder. I don't know. I feel like I somehow feel like this isn't going to feel as strange as it sounds.

And for a bunch of reasons, we're so focused on other people. We're so focused on our own lives. We have such a human-centric nature that, like, okay, this thing is really smart. It's inventing new science for us. it's running companies for us. It's doing all the stuff. And that sounds like it should be impossibly weird, and I think it will just be very weird. Do you think space data centers will provide a meaningful amount of compute for Open AI in the next two to three years, five years? No.

Ten years? You just keep going. Ten thousand years. I wish you'd unlock. Okay. The funny thing about the whole back and forth about ads is that in our world, the criticism is that you didn't launch ads early enough? Is there a world where you wish you launched earlier?

How is the actual rollout going? Are advertisers happy? Do you have like a really long roadmap? Or do you think you'll be faster at catching up to sort of what's frontier in ad? We haven't even started a test yet. We start the test soon. But, you know, we're going to, it's going to take us some number of iterations to figure out the right ad. at the right kind of the right way this all works. Do I wish we had started earlier?

We have gone from like not a company, you know, three years and three months ago or something like that. We were like a research lab. And now we are like a pretty big company with a lot of products. So there's many things I wish we had done faster. I think we were correct on the tradeoff here of how we balance things that we need to do. I wish, you know, we launched this very cool enterprise platform this morning, which we had done earlier too. But I had to like deal with the monstrous growth of Chachapit and Kodax and all other stuff. Good problems to have. Last question for me.

What happened to that internal writing model that you used to write the essay? That feels like something that was really cool, but we never really saw the light of day. We're going to get a lot of that spirit into a future model. Again, it's like there's so much stuff happening. We have to make these hard prioritization decisions. Sure. I would love a cool writing model. Not as much as I would love a cool coding model.

And it's what is possible now for coding, for science, that's the thing I'm most excited about, for accelerating all kinds of research, AI, and otherwise, for really accelerating the economy. I think that's like the right thing for us to most prioritize in terms of new capabilities. But yeah, well, like you want a model that can write beautifully because it means it, well, you want a model that can write beautifully if it's, well, you want to write a model that can write beautifully if it's, can also think very clearly and express that very clearly. That's just useful in normal work. Yeah, that makes sense. Last question for me.

How have conversations been with the broad open AI leadership team? You guys are in a position where any single word or sentence you say in any situation can be spun into a headline immediately. And then you guys have to go on damage control, kind of correcting the narrative. But of course, the original message is often, or at least the original news, is often seen more. broadly than the correction. It seems like an interesting challenge. It is a strange way to live. It, and I don't like, I don't know of any private company that has ever like been so in the news and so under a microscope. And it, you know, at some level, it's frustrating. And, you know, we're so squarely in the sights of everybody's anxieties and every competitor trying to take us down. And everybody's like, just what is going to happen with AI to,

they're part of the business or their own lives, that there's like a lot of plasma looking for an instability to collapse on. In some other sense, though, the subjective experience of it is we are so busy on so much exciting stuff that it often feels like there's this crazy hurricane turning around us. And when we sit, it's like fairly calm. You know, the media or Twitter goes insane about something one day. They're talking about a crazy meltdown. We're like, that is insane.

like, okay, and people talk about it all day and then later find out it's wrong and sort of seemed like a lot of wasted energy, but we're just like, we have this great new model coming, people are building incredible stuff, companies are transforming, we're trying to figure out how to get more compute and deal with this compute crunch.

And we just kind of like, keep going, and we're busy. And then if we like open Twitter, pop up our heads and look at the news, it's like, wow, that is an insane, crazy thing happening, completely divorced from reality or 99% divorced from reality.

And like, okay, someone will correct it, but then we get back to work and people flip out again. It's, it is much, it is weird to watch when we look outside, but it is, it is less chaotic internally than I think you would imagine from reading the media reports. Well, thank you so much for taking the time to come chat with us. I'm excited to see the Codex ad. Please try it. I have. The app and 5-3 have been, like, I think, the coolest thing we've done in a while. Yes, with one prompt, I rebuilt the TBPN.com homepage to look exactly like Berkshire Hathaway,

and it was just immediate. It was very fun. Interesting choice. Plain text. It was very easy. Immediately one-shot. It did not really push it to its limits, but I'm having fun. So thank you so much for coming on the show.

We'll talk to you soon. Thank you, great to catch up. Goodbye. Cheers. Let me tell you about 11 labs. Build intelligent real-time conversational agents. Reimagined human technology interaction with 11 labs. And I'm also going to tell you about Century.

Century shows developers what's broken and helps them fix it fast. That's why 150,000 organizations use it to keep their apps working. And without further ado, we have Shelto Douglas, Cremantropic. He's a member of technical staff and he's here in the TV event. Ultrona, how are you doing? Hey, guys, how are you doing? I am doing great. Seems like you're doing better.

Tell us the news. What did you release? What happened? All right. Well, once again, we've got a new model. It's really fantastic. The camera's a little bit laggy, a bit strange, but, you know. Should be good.

Should be good. This model is really, really, really fantastic. So I think people have been comparing the previous generation of OpenEye and Anthropic models, and they've noticed there's some differences, right? The opening our models were a bit better at trying really, really, really hard on tough problems. But the anthropic models were much faster and so forth.

And so they worked on speed, while we worked on making the models much, much better at really, really tough problems. And so that's, I think, where the model really shines, is it's able to think for really, really long time. It's able to expend a lot of test-time compute on thinking hard about problems. It's also a massive step towards the coworker that we've been working on. I think I mentioned last time that we're seeing this continuous progression

towards models that are as capable at the rest of knowledge work as they are at coding. And I think I'm really, really excited about that as well. You can see it's a lot better at computer use. We've got Claude for PowerPoint. Claude is starting to become actually quite capable at Excel. It's not perfect there yet, but huge steps, huge progress. Do you have internal benchmarks for things like PowerPoint now? We have benchmarks for everything that we work on.

But what makes it to the model card? What makes it to the model card? The model card is typically only publicly released benchmarks. Okay. Because for our internal benchmarks, we usually hold them out for internal testing. Oh, sure, sure, sure. What's the process for benchmarking a PowerPoint? I can't talk too much about benchmarking and PowerPoint.

But, you know, I'm not sure if you know, about a couple of years ago, I did work at McKinsey. Okay. So, you know, sometimes maybe they just need to be. So they show them to you and you say, oh, it's good or it's bad. Yeah, probably a lot of people involved. Yeah. Yeah. Talk to us about, like, what a difference between 4.5, 4.6 means, you know, people might

be familiar with pre-training, the different buzzwords that go into changing a model. Like, what technically happened at Anthropic over the last few. months to enable this. Yes. So we actually, we decorulate the model release versioning from any specific technical details. It's more about the overall capability level, the way I thought about it. And actually, like, I've been asked by many people, why we didn't just name 4.5 as 5? Because it was such a step up.

And I think maybe we under-anticipated. I mean, when I was last on here, I don't think even I fully appreciated how much of a jump it was from 4 to 4.5. Yeah. It was great because by the time there was the 4.5 hype cycle, we could repost a bunch of of clips of you talking about it. And people were like, wow, they covered it today. It's like, no, we actually had them on the day it launched. And so I'm sure we'll be doing the same thing with 4.6. Talk about task horizon, the meter survey. Is that important? Obviously, it's a

public benchmark. We've talked about it before. But is that something that is a different tradeoff to the other, because I see it as like there's the IQ and then there's how long something can stay focused, like the locking in. And is there any tension, between those? Do you ever get a model that's like, oh, it's really smart, but it can't stay coherent for that long, or vice versa. It does, you know, dumber but longer tasks. I mean, I think the meter eval is possibly the best eval currently out there, right? But you're right in saying the different parts of the chart measure different things. Initially, it was just testing IQ, but then as time goes on, it starts to test things where humans fail at tasks, not because they aren't

smart enough, but because actually persevering on a task for six, seven, eight hours is really hard. And you're right that these come apart. There's sort of like models you should almost expect to have superhuman perseverance, but at the same time, maybe context coherence or something like this, like fights against that. And that's a different axis from raw intelligence or knowledge on given domains. So Meteor Eval is the eval that I usually use when I'm referencing AI progress to people. Recently, we held this physics conference where we invited a whole bunch of physicists.

We did it with Google DeepMind. We invited a whole bunch of physicists to try and convince them that AI progress. was really big deal. And that was the chart that resonated most deeply with people. And I think it's just quite conceptually graspable as, oh, okay, things that would take me X hours, the models are now capable of semi-reliably. Also interesting, it's an interesting benchmark considering that so many types of knowledge workers will never actually work on a task for seven hours straight. Like, it's just like, oh, I worked on it for 20 minutes, and I jumped into a meeting,

and then I took a look at it again, and then I had lunch, and then I went for a walk, and then I came back, I worked on it, then I got somebody called, I had to work on something else. So it's like very possible that a lot of people listening to this, you know, maybe outside of some software engineers, have just never even in their career worked on something for six hours straight. The flow state, it's great. I mean, you chain things together and so forth. But yes, I agree. It's a, it's a unique, it's to give it out in that respect.

And what does even a week-long task mean? Surely that's sort of inherently composable into, you know, a year-long work-of-progress. I mean, I certainly don't think of myself as doing it. a year-long task when I'm at work, right? I'm doing these day or week-long time horizons, usually feed into a several-month-long time horizon. So talk about orchestration. It feels like there's a moment happening.

I really like the Gastown analogy in the world that he's built there. You see it all over. I have four clog-code instances running at the same time. It feels like there needs to be a new layer of abstraction. How do you think that gets solved? Is now the time to start learning those things? what metaphors will be valuable going forward?

Yes. So right now the agents are still at this point where you need to manually multiplex cross them. Yeah. You know, when I'm working, the stock thing about going from 4 to 4.5 and now 4.6 was that I went from maybe like 10% of my lines of code being written by me to 0% of my lines of code being written by me. And but the thing is I need to actually sit there and constantly switch between these windows to make sure they're on track and give them guidance and feedback and, you know, multiplexing maybe five at once or something like this when it's going really well, but you still need to be in there in the details. And I think the long-term right way to think about is constantly moving up levels of abstraction.

So ultimately you only want to be talking to one agent that's maybe synthesizing the feedback from models as they come back and say, I got stuck or I was unclear on this. And that model can act as what I was doing before, and I can take a step back and sort of act an even higher level of abstraction. It doesn't feel like we ultimately want to be playing, you know, Age of Empires or StarCraft with the models, right? We shouldn't be APM bound. Instead, it should be the model's surfacing information as you need it

so that you can constantly act on whatever the most important thing is and the context switch appropriately. Speaking of gaming, do you have an update for us? When is the game shipping? Tell us about the game. What inspired this? Give us the backstory and then tell us what's going on. Well, I think like many other people,

I really wanted to test the limits of the models over the holidays, right? I mean, we saw all of that from people saying I spent three to four days doing X. And I think that's really where almost the hype cycle for the most recent generation of models started in some respect, when people got a chance to properly test their limits. Yeah. Tells you a lot about adoption and how, like, diffusion takes time because you need to have the space. We need more long weekends. Seriously, right.

That's the bottleneck. Long weekends are the bottleneck. Yeah, good for GDP. Yeah, good for GDP, I guess. Take a month off. They play with coding models. Yeah, yeah, yeah, sorry. Anyway, so Dylan's been getting us all into Age of Empires

with these five-aside matches on the downstairs table. And I wanted to see if I could make an age of scaling where you build solar panels and data centers, you train AI models and drones and so forth instead of farms and mining and all this. You start an industrial era, you go all the way up to Carteshev. I only got like 70, 80% of it. done and all the mechanics work, but it turns out it's actually really hard to make a game that's

fun. Yep. Like, actually just clicking Make Solar Panel a hundred times is not fun. Surprise, surprise. It's fun for some people. There are some of those crazy clicker games where you do that and it's just like a brain rot game and they're pretty successful. But I think you should stay away from that.

Not even Factoria. These are like mobile games that are like really bad. I don't think that aligns with your mission. There's an arc to it. Yeah. But we talk about the grand geopolitics and strategy. of this era that we're going into, right? Where the most valuable resources on Earth are, you know,

well, basically the economy is in some respects reorientering around compute. And so I wanted to capture some of that dynamic, the late-night discussions of San Francisco in a game. Yeah, I love it. How are you processing the work that's being done in world modeling and generative worlds? I know that you're not generating images,

but Claude can take in an image and process, and that has value. you text a screenshot and have them implement it or have him implement it or it implement it. But what do you think about world models? Is that going to play an important piece of a link in the chain of where you're going? Yes. So, I mean, I think there's two different things here. There's the direct path to AGI, which I think is very much coding and then AI research

and general science and so forth. And I don't think that requires world models. But I do think that world models are incredibly exciting for a number of other things. I think they're very exciting from a gaming perspective. I mean, we all saw those demos. They're incredible. They're just truly mind-blowing. And I also think they're probably the unlocked to training robotics properly.

It feels to me like there was this era of, oh, we're just going to need to fill, like basically do incredible amounts of behavioral cloning of people teleoperating robots. And it feels actually perhaps like the scalable way to get robots to work is world models, but unclear. I'm not, but I don't think it's on the critical path to AGI, basically. Can you unpack the concept of a software-only singularity?

Yes. So in this world, it's one where the models are far better at digital tasks than they are at physical ones. And so we see rapid change in the digital world with relatively little change in the physical world. So information and software changes dramatically. And this ends up having some pretty weird effects. it means that

maybe like the drivers of what have been the last, you know, a couple decades of progress in the economy turnaround, like, you know, sort of get, get changed very rapidly. And I think we'll see that flow on into the physical world, but at a delay. So you get much better at doing chip design. You get much better at training AI models.

AI models get a lot faster. Chips get a lot better. The general economy gets a lot more efficient because the sort of information and message passing that is much of the rest of the economy ends up becoming much more efficient. But the same time, you don't yet have robots providing limitless physical abundance.

Science probably progresses really fast, up to the degree that you need interaction with labs or larger particle colliders or something like this. And then you go, okay, well, I need to build the robots to... But at the same time, automated labs feel more near term than unlimited robots in the real world manipulating the Earth. Yeah. Yeah. I think maybe

they actually arrive at similar-ish times. I think you need pretty competent robots for the labs, or at least no one's yet managed to figure out how to automate a lab without. It turns out there were all these really weird little tasks that require a lot of manual human dexterity that are currently part of biological protocols. And so a lot of biologists will say, no, like you just actually need something that's capable of human-level dexterity. Maybe not for all experiments, but for enough that it becomes annoying. And in this software-only singularity, like how are you defining singularity? You know, the Kurzweil formulation of like more computing power than human brains,

that's like one, you know, equation. There's also just like this point beyond which we cannot see. How do you think about singularity in that context? I think there is a pretty tricky event horizon, but I at least haven't found anyone that's made incredibly strong or good predictions. It just feels like at that point you have as many digital intelligence, or more, and they're as smart or more smart than human intelligences. What does that even mean for the world?

It's incredibly hard to predict. It's something we spent a lot of time trying to think about and trying to prepare the world for all of the eventualities, but it's, I think, difficult to make sort of top-line predictions of what exactly that looks like. Yeah, and the original Kurzweil formulation, that was almost the definition of the singularity, was that you can't make predictions beyond it.

And so, you know, once the predictions break down, then you're there. Yeah, I mean, I think in the near term, you know, I went down with Dworkesh to his Elon interview in Austin. And I think a lot of the sort of reflection of that in the physical world perhaps is what ends up happening. Like you do end up basically trying to climb the Cardish of scale and capturing more of the energy of the sun and so forth. But I think that takes some time.

Is your current takeaway we're more chip constrained, energy constrained, what's the biggest bottleneck to AI progress? I mean, I think right now where I think I agree with Sam's earlier answer that we're more chip constrained in air progress. But I think it like is sort of like interesting to roll forward two years and be like, okay, well, if you want 100 gigawatts or in four years and you want a terawatt, what does that look like? Where do you put that? And this is why Elon's going off the data centers in space, right? He thinks it's going to be the easiest way to get a terawatt in space in 2030. and maybe it's space, maybe it's the, you know, a giant desert somewhere.

It's the Atacama Desert, the Australian Desert, somewhere in Texas, you know, maybe Texas has a lot of solar panels. Yeah. Hard to know. Yeah. Jordy. How are you guys thinking about free usage limits ahead of the Super Bowl? It's really tough.

I mean, like one of the struggles of this is that, like, compute is so constrained. And so... Yeah, I noticed you guys didn't say, download the Claude app. There was no direct call to action. Yeah, I mean, I think the purpose of the ad was very much one for provoking, like, thought and discussion. It was certainly successful already.

Yeah. It did provoke a lot of thought and discussion. It did. And I think, I mean, I don't think you need to do an explicit call to action for people to download things or to consider things. Yeah, that makes sense. sense. But again, it sounds like it will be, the GPUs will be on fire Sunday is the expectation.

The GPUs will be on fire, as has been the case for the entire industry for last year. I mean, I think we've all had to make incredibly difficult tradeoffs on exactly how computers used. Yeah. Can you, can you like zoom in on your experience? Because I feel like, you know, everyone's sort of seen like rough growth curves for Anthropic on the revenue side or on the tokens generated. or whatever, and it feels exponential, but smoothish. And then simultaneously, you have these constant worrying about bottlenecks. Is there enough capital? Are there enough chips?

Are there enough data centers? There's enough energy. Are we in an age of research? Are we in a plateau? How have you balanced those two narratives out? Your experience of smooth growth, even though it's exponential, with constant fear around something like a cloud hanging over the industry and maybe a slowdown. I mean, I think, so to like note on the smooth or exponential growth,

I think one of the things that really blew me away was when semi-analysis did that analysis and found that I think what we went from like 2% a month ago or six weeks ago to 4% of GitHub commits done by Claude Code. Yeah. And for the amount of GitHub commits done by Cloud Code to double over a few weeks is truly, ludicrous, right? Like, and there's no real visceral way to feel that. You almost feel like it feels like a number on a screen,

but you can't viscerally feel it. For us, we've always had a very, very, very strong conviction. And in many respects, Anthropic is a bet on this being true that scaling is continuing and that progress and that sort of progress continues unabated. So in many respects, the external numbers are only a reflection of the conviction that we've had internally. for a long time about how we, exactly how we expect all the trends to go. I mean, I think we're broadly on trend for like, you know, if you look at situational awareness,

you know, I think pretty sure the power and like energy and flops, their predictions are bang on. It kind of feels more like we're just, we're hitting each milestone as we expect. And roughly, yeah, roughly what we expected to have happened has happened so far. In terms of diffusion of the technology, do you think that there's a role for forward-deployed engineers to go in and change organizations? We saw some news that OpenAI is hiring a bunch of folks. Obviously, a lot of enterprises are using Claude now.

But at the same time, that phenomenon of if you don't have a free weekend or a long weekend, you might never get around implementing it. And so having a little bit of extra horsepower and knowledge around the office might actually pull forward capability. Is that something you could see growing at Anthropic? Yeah, I think it's a great idea. I think it's very clear that people don't know how to hold these things. And also the fact that it's ludicrous.

Three months ago, we didn't have this most recent generation of models. We didn't have a 4.5. And you're meant to adjust your business strategy over the course of the holidays, basically, because the models are suddenly capable of doing things that they couldn't do right before you, you know, at the end of the end of the year. Are you saying we could see new white,

call her job creation? Maybe. I think these are really valuable jobs. I think that there's clearly a hell of a lot of value to be unlocked. We asked Sam this, but I'm interested your take. Is data the new oil?

Or is it the fossil fuel, as earlier said? Yeah. Yeah. Is it but the more pointed question is just like, are there organizations out there that have data that's locked away, and it requires a business development deal or an acquisition or some sort of AI leadership to add that capability,

because there are effectively secrets, sometimes trade secrets. I don't know if the Coca-Cola formula is in 4.6, but it will be if the Coca-Cola company calls you, right? So how do you think about that question and that phrase, even though it's been beaten to death, maybe it's making a comeback? I think it's maybe like there's two kinds of data and one of them is dramatically more useful than the other.

And there's the kind of data which is like the kind of analytics that a company might have collected in the past. Or the artifacts of that company's operation, internal documents and so forth. And there's the other kind which is like the actual work that people did to produce those documents. And that's not reported. And so I think that, like, to the degree that I'm not certain it's data as the new oil so much as, like, the expertise of people. And like models being able to understand and learn from people. You almost want the models to, like, be an intern in an organization and get coached and feedback and learn about how to do the job. And you learn, I guess, maybe what I'm analogizing to here is as a human, when you join a company,

you mostly learn the job from your colleagues rather than from reading all the documents in the company. It's just much more informative. And I think I broadly expect that to be true of models as well. I expect them to learn in a quite human-like way from their colleagues. Have you thought about a scenario where a company's maybe revenue goes to zero and all the remaining value is just in their historical data? I guess in this analogy it would be the people that I think the values would regress with.

Sure. Yeah. Yeah, that makes sense. When you guys are thinking about new opportunities, verticals, categories, what is the thought process around UI? There's a lot of people building AI-native tools today. they functionally look like traditional enterprise software. And it's hard for me to imagine like anthropic building out, you know, infinite surface area of traditional software to do things like orchestration.

But like what is this specific framework when thinking about interfaces for new products? We mostly want to build something that fits in where the humans fit in. And can do it and can help, like, you know, it sort of like interacts with you like a colleague. And so we want something that can fit in. And yeah, rather than like absorbing interfaces, join in and use the same interfaces that you do and stuff like that. I think that's how... And I think that's what companies want.

If you ask any company in the world, hey, we can get you the best executive in this function in the entire world. They will usually say, like, we will pay almost any price for that. Or we will pay an extreme premium for that. And yet if you tell them a software solution, hey, this software solution can help you do this. other thing. Maybe they want to use it, but it's not as enticing as something that can just do the work. Right, exactly.

Also, I mean, there's, yeah, the UI evolution seems very, very underrated. I was processing the open claw development, how important mobile was to that, just this idea of somebody shows up and says, you're a busy business executive, you're in meetings and phone calls all day, you're on flights, we got a great software engineer for you, but you got to talk. to them over the terminal on your desktop. It's like, I'm not going to give that many commands because I'm on the road and I'm on my phone. And you see people walking off planes with MacBooks now because they need to get one more prompt. I have done that. I'm so guilty of that. And so I imagine that there's a lot of work that will be done on the mobile side. How do you think about it?

I think that's broadly true of like all AI interfaces, right? I don't want to look at a screen. I don't want to have like, I don't even want like a terminal window or a chat window. I just want to ask my computer to do something, or maybe like sort of like ask the world around me for something to happen and for that to happen. And so I think that there's like focusing too much on the interfaces of today versus how would you interact with an incredibly competent and colleague is the right thing. Like, you know, I just want to text Claude and be like, hey, you know, can you fix this up for me?

Yes. Can you help me sort this out or book this trip? Yeah, just the one's to, yeah. Yeah. Yeah, everyone has that ability to text from a variety of devices. People have AirPods, Apple watches, phones, laptops, tablets. Do you think we need any more new hardware? I think new hardware is pretty exciting and interesting.

I like that people placing bets on it. Yeah, I mean, I feel like something which can capture a bit more of the context that you have in your day-to-day life and that you can talk to. I mean, one thing that voice is hard. higher bit rate than typing for most people. But at the same time, like, in most of our environments, it's actually quite annoying to imagine talking to your devices in crowded public space or at work or so forth.

So maybe something that captures that automatically would be great. Like, you know, you've seen those, like, I'm not sure if you've seen the YouTube videos. Oh, yeah. Where people, like, sub- Vocalized. That would be cool. Yeah, Apple just bought a company that sort of looks at the skin movements and maybe can re-release.

There's a number of things. companies in the space. It seems like it's coming and you'll just be able to plug into that as the as the hardware evolves. That seems cool. But also just my natural environment, absorbing more of my context and being able to just like talk to, you know, like having speakers around my house or something like this, rather having to bring my phone around me. That would be cool. Do you see any, any world where on-device computation becomes more important? I feel like Anthropic as a whole is so back-end heavy. Like none of the computation is done on the device. Yeah. But that could change in theory.

I don't know what you think about the future. I mean, I think our general perspective is that a given level of intelligence, you know, I mean, we've seen this trend, right? Yeah. Intelligence gets 10 to 50x cheaper for given level of intelligence every year. Sure. Massively democratizes access to that level of intelligence. It's literally in my Twitter buyer.

It's like intelligence too cheap to meet up. Yeah. Because this is in large part one of the things I've worked on at Google and I've also worked on the way. Yeah. And one of the ways that happens is that that level of intelligence goes on device. but then there's always, because scaling keeps continuing, there is this exponentially greater set of use cases,

which the models then get applied to. So it's totally plausible to me that some subswarm might exist on your computer or you might get a laptop that has better memory bandwidth, and so you can have little models complete stuff, but the same time, then you also want the much greater intelligence going out there and sort of like planning and farming things out to swarms and really like munching on the intensive and intellectually difficult work. Yeah. Jordi, anything else?

have the market sell-offs due to various anthropic releases been generally predictable, or have there been any that have been surprising? I mean, honestly, I didn't even notice the last one because I was just heads down on the launch. But look, in this case, I think it's a little bit much to ascribe it to an anthropic launch. I mean, there's been legal tools released with AI for, we have many, many customers with legal tools that do work for lawyers. I don't think this is crazily different to any of those. And I think this is part of just a continuing trend that we've seen. Yep.

Well, thank you so much for taking the time to come talk to us on launch day. Great to get the update. Thank you very much. Thank you very much. I'll talk to you the next time. We'll see you. Have a good rest of your day. Let me tell you about graphite.

Code review for the age of AI. Graphite helps teams on GitHub ship higher quality software faster. And I'm also going to tell you about Vanta, automate compliance and security. is the leading AI trust management platform. And I'm very excited for our next guest. We have Dan Barkella from T1 Energy. It's the chairman and CEO.

He's in the Restream Waiting Room. Now he's in the TV. What's going on? How are you doing, Dan? Hey, I'm doing great. Thanks for having me. Thanks so much. First time on the show,

would you mind introducing yourself and the company and that I'd love to know about the journey to the company and that I have a bunch of questions about energy, but we can start with an introduction if that's okay. Sure, great. Dan Barcello,

I've been doing energy all my life. life. A native New York are now a proud resident in Texas, and we're building a lot of different energy systems. My journey took me a lot through different parts of the world, building well and gas systems, and now we're really focused on building solar in America. It's an exciting time to be doing it when everyone knows we need a lot of electricity and a lot of energy. Yeah. How did you wind up with T1? So T1 was a company. We saw the need for particularly utility scale, grid level type of energy systems to be, I don't want to say disrupted, but needed to be impacted in a large way,

whether it's on grid level storage or whether it's on solar. We were doing quite a bit on the grid level battery side, and the battery side, quite frankly, had a wall of competition from, you look left at Samsung, you look right, it's LG, you look this way, it's Panasonic, and you got CATL behind you, and then, oh my God, here's Tesla. But when you look at solar, it was very different in the United States. It's predominantly was by Chinese companies. The American technology of solar was something we kind of almost forgot how to do in the U.S. and build it out. But we had an opportunity to buy assets, buy the technology, buy the know-how. Now we run it. We bought a 5-gig-watt asset in Texas

late 24, and we're building a second part of the plant, which I can get to in a second. So for us, it was really about how you get in. You should have seen our initial reaction to the T-1. We saw our first interaction with T-1 was seeing the video of your facility. And John and I were like high-fying. America's got it. We can build solar and scale. And then, of course, we figured out kind of, you know,

the history of the company and we're very excited. But we're excited that it's an American asset now. And look, we have a million square foot facility in Dallas that is, you know, we brought that from producing zero. We commissioned that and brought it up to producing over five gigawatts in annualized runway in November. So that was a lot of work. A lot of robots in the factory, but a lot of humans.

You know, so, you know, we'll unfortunately. What is the, yeah, talk about, maybe talk about the scale of the operation in the context of overall U.S. solar panel production, because I think people hear five gigawatts, and maybe they can wrap their head around that, but it's a meaningful percent of overall production. Is that right? The U.S. is about 50 gigawatts, so this alone is about 10. percent of American production. It's extremely automated. It has 1,200 people. It's running 24-7, except, of course, during those ice storms that people don't know how to deal with in Texas.

It's more about the people driving at work than the factory, but it's of significant scale. It's one of the most modern of the world. Yeah, how does that 50 gigawatts of capacity in the United States compare to China? Can you lay the sort of geopolitical backdrop and get me up to speed on how? how much the American government is doing to support this effort? So look, China is up to over a thousand gigawatts in 1.2 terawatts of capacity. China's adding from last year probably close to 350 to 400 gigawatts. That's close to the American grid. So it's almost like China can add an entire grid just in solar.

So the numbers are wild. Yeah. I don't think, I don't believe China's doing it. to do lower carbon. I think China's doing it because they need to grow very, very, very fast. And I think that's what you're seeing the last two or three years in the United States, where solar and storage is also taking 75, 80% plus. You asked about what the US aspects for it.

I think the US, both Democrats and Republicans, have done a lot of disincentives and a lot of incentives for advanced manufacturing. Whether it was the IRA in the past or the O triple B now, those incentives are in place. a lot of tariffs in place. I think a lot of those create an ability for the U.S. to be a level playing field with other competitors. Quite simply, I think America lost a lot of its manufacturing muscle. We're building a new facility outside. It's about an hour north of Austin in Rockdale.

It used to be an old alcoa plant. Now it's just a piece of dirt. We're building a 2.1 gigawod solar facility there that's going to make the cells that go into our plant in Dallas. So what are the main bottlenecks? Is it purely capital? Because ideally T1 could just copy and paste the Dallas facility a thousand times. But what does it take to actually do that? The solar chain is really cool because it's almost like four different parts.

It's got the digging out silicon, making polysilica in the rocks. Then you're going into ingots and wafers. And those wafers are also very similar in the semiconductor industry. Then it kind of splits. The high-grade stuff goes to semiconductors. The low-grade stuff goes to solar. then you take a solar cell plant. The solar cell plant is basically a low-grade semiconductor plant.

You're making the cells, making a fab, and then those cells go to the low-tech part, which is in Dallas. Dallas is where I always joke about it. It's a five-gigawatt site with 50 trucks a day, bring it in glass, put in the cells in between and glue them together, and then you get 50 trucks a day going out with glass. So it's like Dan's Glass Company.

I love that. But the cell parts really, interesting and that's where a lot of the advances come from specialty gases to process equipment engineers to process equipment that's very specialized that's the higher tech part yeah so for me it's very interesting as we go backwards and integrate to answer your question on how do we get more of it starts with contracts you know we we we're smaller company we're not the scale of like what Elon's doing in terms of his announcement with solar right I don't have

unlimited capital funds nor do I want a lot of dilution on the as a public company. We need contracts. Those contracts then allow us to get the risk capital that we need to build these plants. But the building part is not the hard part. I'd say the building part in America is more about navigating through certain regulations and navigating through certain delays there. Texas, I think, is very conducive. But as long as I have power, as long as I have access to water, and we go full recycling on the water. But as we go into water and power and low-cost inputs, we feel we're in a great place in Texas to build. Who are the, like, main buy, Like who are, who, is it, is it, you know, traditional energy companies?

Like, who are you spending the most time with? Or will you actually sell to like an Amazon Web Services or Google who's like building out a new data center? So we focus on utility scale. Got that's capturing largest utilities, but it's also capturing all the big tech companies that you'd mentioned to. We're usually selling directly through developers that then face those customers. Makes sense. Some of them don't, for confidentiality, don't want to be disclosed. One of them, Treaty Oak, we recently announced.

They're a large developer based in Austin. So I imagine that they'll have a lot of their customers exactly what you just said. The demand for data centers is surreal. The demand for power is surreal. A lot of the, you know, everyone knows we're waiting for turbines until 29 to 30. So you're not going to get that there. So if you want to start scaling power fast, you need solar and storage right now. So that's a big, big driver.

And that's the bulk of what's driving the demand. Behind that some smaller utilities that are looking to add to the grid and capacity, but by far. The other big trend, though, is as you're running out of spaces to connect power and interconnections, a lot of things are shifting to behind the meter or distributed generation.

This is where you're building first power islands. You're building data centers with power islands that can later connect to the grid because affordability is a big issue. Nobody wants a data center that's going to raise prices of electricity. And the Santis was very vocal about that in Florida, like data centers, you know, bring good jobs and things, but then, you know, what are they going to do to electricity prices?

So I think it's really important that the industry gets a handle on how energy can also be in an area that's not going to put more stress on the grid. So just to restate the current status of the business, G1 is the first facility up and running. G2, your building. What does the future look like? how far into the future? Obviously, you need the contracts to unlock new plans, but how do you want this to scale? Where else do you want to build? How quickly do you want to copy paste these facilities? If I had the contracts, I'd love to copy paste Dallas like that, whether it's, you know,

six to six months to a year. Those are fast. We publicly said that we're, you know, on track to build G2 this year. And that's already started constructions, starting construction. started that in December, and we're tracking to continue to build that. So that's our main asset, our main focus. G1's operating, G2's in build, and then if I go to the next part of supply chain, we get our polysilicon and our wafers from Corning and Hemlock, so that we've completed the entire chain to be an American chain. Yeah.

How do you think about the data centers in space? Elon's notorious for vertically integrating, but there's going to be a lot of players. It's probably a slightly different solar panel than what you might put on the ground, I imagine. Is that something where you're interested in talking to folks about a contract to deliver solar panels that would wind up on satellites and space? I'd love to. You know, I think when you look at silicon-based and probably more on the perk side than the TopCon side, there's quite a bit of demand for lower Earth orbit in particular rather than deep space. So this silicon wafers do work in space. So the question really becomes, you know, are we going to use it for terrestrial demand or space? demand. Obviously, we're not sending up modules with glass, you know, up to space, but at the

cell level or at the wafer level, we'd love to have and feed into that as a customer base. Love to do that. What do, I mean, being a public company CEO can be stressful. What do the short sellers get wrong about T1 energy these days? You know, being an ex-by-sider and an Excel side or writing research reports about it, you know, I know the journey to be a portfolio manager or to be a sell-sign analyst and get a lot of things right, I'll get a lot of things wrong. I think for the most part, short sellers, as competitors like to say, like, oh, they're not going to make this or they're not going to, you know, do it the right way. Compliance with U.S. law in terms of being a fiac, you know,

foreign entities are concerned. There's always concerns about the solar industry and misunderstandings about how China's influence is. That's something that I think is an old story. You know, We're an American company, American public company, New York Stock Exchange listed, and we bought an asset, you know, and we're running and building it now, but let's go for this. That's one of the biggest parts. I mean, the other part, you know, I get stressed because sometimes my IR guy goes, don't re-read it today. I'm like, okay, but I want to. And I'm like, no.

And then someone says, hey, go kill yourself or go to this. I'm like, Jesus, you know. Well, fortunately, some of your fans have come over to our chat and they want to know, when are we getting merch. They want T1 merch. Any plans? We will put that site up as fast as possible and get the hat and get these things maybe without this logo here or this one here. Get that there. But no, we're excited to get the merch. We love our shareholders. We're thankful for that. And I think we have a good feeling for all those parts. Yeah. Jordi, any other questions? No, it was great to meet you and come back on

as you have new announcements. Yeah, I have one more thing. Just I'd love to know the shape of the company. It feels like a, you know, an American reindustrialization process. How big is the company? It feels like you're a job creator. What roles are you hiring for? If this really, really scales, as I hope it does,

and I hope we're talking in a couple of years and we're talking with a tarot after everything, with a tea instead of a G. What does the shape of the workforce look like? We're 1,200 already in Dallas. We're going to be 1,500 in Austin. In Austin headquarters, we're going to be over 100. You know, if I can do my recruiting call, go to T1energy.com careers. We need engineers.

The more rogue you are as an engineer, the better. This has to be a new style of manufacturing. I see you guys always, you know, Ramp is your big sponsor. We start with our whole thing is our corporate policy for expenses is very simple. Do not buy anything illegal, either good or service. Two, it's corporate money. Be careful with it. Three, we can watch everything you buy with your ramp card, and we do with AI, and it's awesome.

And the fact that no one has to do T&E reports or something, it's just about accountability to people. And we just have to build as fast as we can as quick as we can. So throw some of the old rule books out. But that's part of the reason we want to be in Texas and in Austin. There's a lot of engineers there. We need them from the semiconductor sector. We need them from the manufacturing sector, from the robotics sector. So it's all about that growth.

And more contracts, maybe some contracts for space one day, who knows. But the more contracts will get, the faster we can grow. That's amazing. Well, thank you so much for everything you're doing. Thanks for taking the time to come chat with us. Congrats on the progress. I'm rooting for you. This is an amazing product.

Yeah, you should have. At some point, I think we were alive. Between the first time we talked about T1 and the second, the stock was up, like, I think, 150%. And that's why we're podcasters. That's why we're podcasters, not p.m. We'll leave the side of the by side. But yeah, thank you for the work. Thank you so much for taking the time.

And the message is loud and clear. Merch. That's the priority, right? It's priority number two after re-industrializing America and saving the solar industry. But it's a close second. Anyway, thank you so much, Dan. This is fantastic meeting you. Cheers.

You're welcome back anytime. We'll talk to you soon. Let me tell you about fin.a.i, the number one AI agent for customer service. If you want AI to handle your customer support, go to fin.a.I. Checking in on Amazon. Yes, what happened. Amazon's reported earnings down 7.5%.

I'm trying to find a good summary of the results. While I work on that, maybe we head back to the timeline. We can head back to the timeline. We can also do this Wall Street Journal, your sort of trend piece about your AI habits. Obviously, we saw a lot of the 4-0 folks in the chat. the new family intervention, we need to talk about your AI habit. People are getting clingy with their chatbots.

This is a knock-on effect of relationships between artificial intelligence and humans. And the Wall Street Journal says it's giving their loved ones pause. So I want to read some of this and you can give me your reactions. Tell me how you're using AI in your daily life with your family. Before I do that. Please, use more details. Amazon's projecting 200 billion of CAPEX. Whoa.

That's at the gong. Can we get the mallet? Let's begin the Lambda Lightning Round now. We have three great guests joining over the next half hour, and it's Lambda Lightning Round Time. So let's bring down the mallet. You can take this one, Jordy. You're ready.

Unhook the Lambda Mallet and hit the Axon. Hit that Apploving gong during the Lambda Lightning Round. The market doesn't like it. But we like it. We like it. We like big numbers. We like it. We like Capbacks.

We're very excited. I didn't see that. Anyway, let me tell you about Octa first. Octa helps you assign every AI agent a trusted identity, so you get the power of AI without the risk. Secure every agent, secure any agent. And we will go into this Wall Street Journal article about how people are interacting with AI.

So it started as an innocent hobby to entertain and distract herself while going through a divorce. Erica Atchison used Anthropics Claude to build a bot that generates images based on prompts from her and her friends. Soon, she promoted, clawed to an advisory position in her personal life. It didn't take long for her kids to think

she might need a break from her new friend. These days there are two kinds of people, those who using AI for everything and well, everyone else. Super users are contracting out their work to groups of agents, amping up their productivity. They're dropping advice from chatchipt, aka chat, as if the OpenAI bot is their most trusted consigliary. Counseling on everything from business strategy to whether or not, you should let your friend get a perm. Interesting question. Everyone else is wondering if their loved ones may be taking things a bit too far.

Maybe some tasks should be reserved for humans. Maybe there's dignity in that. Maybe some of us like the way things are. Atchison says Claude was particularly helpful as she navigated her relationship. She suspected wasn't going to go anywhere. He was like, quote, this is Claude talking. Hey, I don't think you're going to be able

to extract yourself from this if you don't stop talking to this person for a while, says the Portland, Oregon resident who works in operations for a tech company. When she was later tempted to rekindle contact Atchison would check in with Claude, stick to your guns, it told her. AI hasn't been beneficial for every relationship, though. While our kids watch TV, Atchison likes to tinker with Claude. Her sons don't approve.

It's a waste of her time, says Grayson, 11. She puts a lot of time into AI just to make boring old pictures. Wait, how is she making pictures with Claude? Yeah, I'm very good. They don't have an image model. Must be calling some API under the under the hood. I don't know. Is she like, she's like on cloud code harnessing nanobanano pro?

Maybe she's talking about the SVG images. Oh, yes, maybe. No, but there's an image in here that's her image and it looks like a GROC imagine image. Maybe she's on GROC. This is an odd article. I'm very confused by this. Anderson 8 has a different complaint. The technology's accuracy.

Chat, TPD, couldn't I, couldn't correctly identify a Pokemon. Another time He's just like, this is unacceptable. Zero on Pokemon bench. I'm sell the stock. It's going to zero. It's over. It's over.

It's over, says Anderson, age eight. Another time, Anderson struggled to get it to generate a suitable picture of himself. Once, I only had one leg. And once I had three legs, he said, this is so cute thinking about this eight-year-old being like, this is slop. This is slop, says eight-year-old. Eric David, of startup founder in Brooklyn. uses AI for recipe planning, workouts, business strategy, and vibe coding with his son. The 42-year-old also uploads recordings of his executive coaching meetings to chat TPT to analyze his progress.

I've done some, I asked chat TPT about a workout plan at one point. I don't go back to it that often. Recipe planning. I did have a do, I did have a do a deep research report all around where we could go to food. Do you think it's actually good for, do you think, do you think LMs are good for workout planning? It's just taking the app. I typically want a workout plan that is, I mean, I don't, we don't, we work out every day and we kind of do a vibes-based approach, I would say.

Try to push it hard and not, we're usually talking about technology and business during our workouts. Yes. But I, you know, historically, let's say when I was starting to work out, I would find a plan, but I wanted it to be from a specific person that was making that plan with general, with an intention around it for somebody in that state, which was maybe like I've been weight training for six months at the time. Yeah, there's a huge... And so if you just take, like...

And the reason I ask is, like, I've had terrible luck doing recipes, getting recipes from LLMs, because I think it just takes the average of every recipe for a certain item on the internet and just kind of gives it to you. And you can end up with like... I think we tried to make waffles once. And it just...

The waffle was like, clearly the rest of... recipe was just wrong because they were like really runny. It just wasn't working. It was like it had botched the different. Well, you haven't used the latest models. You throw codex 5.3 at that. Yeah, but it's, no, but it's obviously not an I. Waffle bench needs to be.

It's not an IQ thing. It's just, it's just. We're creating Waffle Bench. We're moving the goalposts. It needs to be able to make waffles. Great waffles. Anyway, let me tell you about cognition. They're the makers of Devin, the AI software engineer.

Crush your backlog with your personal AI. engineering team. Anyways, the wife of this, of Eric, the startup founder, says his wife is reluctant to use AI, especially for any writing-related tasks. A freelance journalist, she doesn't believe the technology can replicate her 20 years of professional experience. Her husband's use can annoy her, like when she asked him to write an important email to her son's school, as she read it over, something seemed off.

I was like, this is basically the same paragraph repeated three times. says the 45-year-old, she asked him, did ChatTBT do this? She says he admitted to it. She's probably not lying, he says. I don't remember the specific instance. I love how they're just like cross-interviewing. Yeah, it's very funny because I leaned the complete opposite way on like preschool applications once. And I was like, okay, I know that everyone else is going to be using AI slop.

So I need to write extra conversationally. And so I'm in the application to be like, hey, what's up? My name's John. just like completely, just completely organically writing, like, flow of state. And then I was like, I probably need to polish this up a little bit, but I'm definitely not giving it the LLM flavor because I think that that's going to be a tell for a lot of people on the other side reading these. They're like, another thing that's, it's not that.

It's not this. It's not that. A lot of, another hyphens, like, people are getting sick of this slop. But what do you think, Tyler? Yeah, I just, like, don't resonate with this article at all. Tell me more. Over, like, Christmas break, I think I onboarded my dad to Claude Code. Yeah.

I'm like, yeah, you need to be using the best model. Like, why? You haven't tried four or five yet. Like, what are you doing? Yeah, yeah. But what was the killer use case? Did you tell them, like, write software, get on GitHub? Did you get them set up?

Just like anything. Anything? Like, any task, just have Cloud do it. Yeah. And it'll, like, it'll probably work. I mean, I don't cook much, so I don't know about the, you know, the recipe stuff. But, I mean, the funny thing is that, like, they're a recipe. You can just take a recipe book and open it.

You can just Google a recipe, but you're like, I need Cloud Code to work for six. hours to get me a recipe for waffles. It's like, like, that's not actually the promise. That's not actually, like, advancing. The advancing is, like, build me a piece of software that will generate me a new waffle recipe based on all what's happening in the news and what my kids are into and generate the images. Do a lot more that would take a long time. You're not actually going to be saving times if you're using LLMs for something that's Googlable. So, yeah, I don't know. I don't know. We're in a really weird moment right now where the models had, are good at so many things.

They are also incredibly confident even when they're wrong. Yeah. And sometimes you're making, I had a friend that got a ticket for, they were dictating a text into their phone while driving and they got pulled over and they were given a ticket. And they were trying to figure out, okay, they were trying to figure out, okay, am I gonna get a point on, are they gonna get a point on their license

for this infraction? And I, and, and, the model was saying one thing, but I've had other experiences very recently with models where they get it completely wrong. So I was like, okay, like you kind of got like an indication. Good thing you had me.

Good thing you had me. The thing we were talking about you. That was something else. You texted me and I was like, I know more than Chad UVT about this topic. Anyway, let me tell you about public investing for those to take it seriously. Stocks, options, bonds, crypto, treasuries,

and more with great customer service. And our first guest, the lightning round is here. Should we bring in Mandy Fields, the CFO of Elf Beauty. Welcome to the show. What's happening? Hello. Thanks for having me. Great to meet you. Would love a quick introduction on yourself, since it's your first time. Sure. So, yeah, I'm Mandy Fields. I'm chief financial officer at Elf Beauty. I've been here seven years, my 28th consecutive quarter of reporting the results for Elf Beauty.

Amazing. Amazing. And what is, what's the latest? What's the news? Well, the news is we just reported another tremendous quarter, 38% net sales growth, 79% with an adjusted EBITDA. Oh, hey, yeah, I love that. And so we're really having a lot of fun over here at Elf Beauty. And that's the results. What's driving that? Right now, I feel like there's so much uncertainty around the strength of the consumer

broadly. There's so many different, like, you can look at government data and see one thing that might give you one signal, and then you look at, uh, report, you know, uh, quarters like you guys are putting up and it says something else, but what are you guys seeing? Yeah, well, it really comes down to three things. Uh, our value proposition. And so, uh, 75% of our portfolio and on the elf brand is at $10 or less. And so that really speaks to that consumer seeking value. Uh, you pair that with our incredible innovation, uh, the marketing engine that we have, who by the way, I heard that you all are going to be at the big game Super Bowl ad. So congratulations on that. We will be as well. Thank you. I think we'll be at a

smaller, our buy is we'll be at a smaller scale than yours, I'm assuming. I saw that. I saw yours was about $50,000. Something like that. Something like that. Yeah. But yeah, well, anyways, I want to get into the Super Bowl too, but continue. So, and then we also have the incredible addition of Road to our net sales performance. And so Road is an acquisition that we did. We closed in August. That's Haley Bieber's brand, and it has just had tremendous performance. And so you take all of those things together. That's how you get to that 38% net sales growth for the quarter. How is the marketing mix changing over the last few years? I mean, meta had blowout quarter. It seems like ads are performing very well. On that platform, the influencer economy is still

growing. What does the mix of marketing spend look like? Yeah, so our marketing spend is pretty well-balanced. We allocate 24 to 26% of our net sales to marketing in digital, and we are there on every platform where our community is present, and that has really been a key differentiator with Elf. I think that we are so close to our community. We keep our ear to the ground, and so that really comes through in our innovation and the marketing campaigns that we decide to put forward are really culturally relevant and connected to our communities. And how are you thinking about the agentic commerce question? Is this something that you need to do anything special for?

Does it just naturally happen? It's still so early, but these things in AI, they take off so fast. I think it is still early. We're making sure that we have the back end correct so that we can participate in agentic commerce. But, you know, AI is just one of those things where this is kind of one of those technologies that can help on both the revenue side

as well as the efficiency side, and so I think it's pretty incredible from that perspective. Is it possible to build a direct-to-consumer brand from scratch these days without a huge celebrity partner? It feels like more and more of the deals that I see that actually go the distance, get to big revenue numbers. They got somebody attached. Yeah, it's interesting that you say that because I think that's actually pretty rare for a celebrity brand to make it to this stuff. Like if I think about Haley's brand road, you know, very rare to see a company go to $200 million in sales in less than three years on 10 products. I mean, you usually are seeing multiple products, multiple skews, categories to get there. And so I think what we found in road is pretty rare.

Yeah. Well, congratulations on the progress. Thank you so much. I wanted to, before you jump off, I wanted to ask kind of what, as CFO, how you think about Super Bowl. ads. I'm sure this isn't your first rodeo. What kind of goes into it? You know, there's so much there's so many different

ways to approach the Super Bowl ad, but what's kind of your logic going into it or kind of framework? Yeah, so Super Bowl ads are really about building awareness behind your brand. And so this is not our first rodeo. I think our first time was

four years ago. We came with an ad with Jennifer Coolidge. It was a lot of fun. This year we have Melissa McCarthy as the actress in our Super Bowl ad. It's fantastic. So I really think it's about building brand awareness. If I think back, you know, five, six years, our overall awareness is around 13% as a brand and now we're over 40%. And I credit that to a lot of the collaborations and the things like Super Bowl that we've done over the years. So your team is, yeah, so your, your, your, your CMO and your marketing team are not coming to you and you're saying like, cool, I'll give you the budget, but like, I want to see a move.

in top line the next week. It's more about those things. I always say that. Show me the money now. But in fact, it is sowing the seeds for a longer term investment. There are other things that we look at that are more direct revenue drivers.

But Super Bowl investments like that are for the long term. What is, last question for me, what's your kind of M&A outlook for the next couple years? I'm sure in a perfect world, there'd be a new road every single year that you could roll in, but obviously that's not the way, you know, great brands don't get,

you know, started every single day or even year sometimes. So how are you thinking about it? Well, we have a great portfolio of brands with us today. And so that's really going to be our focus on the existing portfolio and building those brands out even further, whether it be elf and cosmetics and skincare, Naturium or Road. We have some really great brands in our portfolio. And so that's what we're going to be focused on. If somewhere along the way we found another road, be happy to bring it into the portfolio, but stay and focus on what we have for now. Makes a ton of sense.

Awesome. Well, thank you so much for taking the time. Congratulations for a lot. We're excited to see the ad on Sunday. Yeah, very excited. We'll talk to you soon. Thank you so much. Have a good ad.

Cheers. Let me talk about Cisco. It's critical infrastructure for the AI era. Thank you to Cisco. We had a lot of fun on Monday. At the Cisco AI Summit, truly, just great interviews. If you haven't already gone to see them, go check them out. Back to the timeline.

There is other news, as always. The number of horses per county? I didn't see this chart. And Chadson says, these are rookie numbers. Should be double, even triple this. Horses everywhere, wall-to-wall horses. I agree. There should be way more horses.

I totally agree. I found out, as you know, in escrow on a new property. And I was talking to somebody very enthusiastic about horses, and I was getting the breakdown on what kind of horses I'll be able to support on the properties. And they were giving me the lay of the land. So I hope to contribute to, it's going to be on everyone to get these numbers up, right? It's not enough for one of us here and there.

I mean, Dara Kosheru Shari, CEO of Uber, yesterday he came on and said that, you know, 75% of all land in cities is parking lots or something. And that's not the real number. But there's a lot of parking lots. What's going to happen to them when we don't need them because of autonomous cars? Stables. Yep. You take your Waymo or your autonomous Uber into the city.

You hop on a steed and you go from place to place. Excellent execution, jury. Excellent execution. I love horses. I do love horses. I also love gusto because horses have gusto. The unified platform for payroll benefits in HR. to evolve at small and medium-sized businesses.

It is the backbone of TBPN. According to the information, upcoming avocado model for meta is referenced as the most capable model to date internally. This doesn't tell you that much. It doesn't. It's the most powerful iPhone ever. We get it. We get it. We get it.

But how does it benchmark? Not even that. It's like this is like some, you know, like they could make the most capable model for meta could be the least capable if you compare it to the other left. Yeah, yeah. So, so.

Yeah, but, but, but, but, it's the same thing that happens in WWC where you stand on stage and you say it's the most powerful iPhone yet. It has the most battery life

of an iPhone ever. And it's like, well, if it was a less powerful iPhone, you probably wouldn't launch it. Um, but it's very exciting to see that they're making progress and they're sending out memos and they're leaking to the information

and their excitement building. And I think that, I do think that we are going to get something pretty powerful. Yeah, and I'm excited about it. And it's a great name. Everything like avocado's a good. name and based on the team of the compute and the money like they're going to jump to the

frontier like there's just no question that they'll be really really close to frontier we got to go back to the horses owl owl in the ex chat says they have two horses doing their part yes thank you oh thank you owl and full circle get those numbers up though don't just say oh I'm doing my part I don't like if you got to be growing exponentially you got to four then eight 16 32 etc okay just on the on the meta model thing yes Historically, food-based models with the name, with food in the name, have done really well, right? You have nanobanana. Yes.

You have strawberry. Strawberry. Yeah. That's all I can think of. I mean, avocado, it's a good sign. Yeah. Okay. Okay.

It's good. It's good. Wasteland Capital is doom posting. A black pill. Yes. Tech has now entered a liquidation spiral where people, e.g. Head funds are forced to sell good, cheap, accelerating assets like semis to cover their losses and expensive, decelerating, high P-E-S-H-I-T, like SAS and AI Victims.

Not sure how long this will take to play out. I don't know. It's been a brutal week. We needed the red suits this week. We did. But fortunately, our guest who's in person has a very special suit that I'm very excited to show you after our next guest, who is in the Restream waiting. Last post, Joe Wisenstall says the sell-off in everything that is not canned soup is relentless.

How's Caterpillar doing? Probably doing pretty well. Campbell's is up today. I love it. Up 6% in the last five days. Congratulations. Flight to safety. Flight to Campbell's. Yeah.

Flight to soup. Buy some soup. Gemini 3 Pro. Google's most intelligent model yet. State of the art reasoning. Next level vibe coding and deep multimodal understanding. Our next guest is in the re-stream waiting room. We have Ivan from Daytona.

Great name. Great. Great. We love to talk about this. You had one of my favorite interactions on X. This was like a month ago. You remember you were, you said something like I'm working late on a Sunday and somebody commented like that. Why are you even doing that? Nobody knows who you are. And Ivan was like, that's exactly why. Oh, that's so good. That's like amazing. That's a great response. That's a great response. Yeah. So it was basically, the reason was like it was just from the holiday. and everyone was talking about, oh, no one took a break during the summer, so we're all taking breaks on, you know, winter, like founders and investors or whatever. And I'm thinking, like, are we serious?

This is like a once-in-a-generation opportunity. And like, if you go rest, like, you won't probably make it, right? And I just tweeted that. And then to your point, someone was like, I've never heard of you. Like, exactly. That's why we have to work. So, yeah. Well, hopefully people hear about you now.

Why don't you introduce yourself in the company? Sure. So I'm one of the co-founders and CEO of Daytona. Daytona is on a mission, basically, to give every single agent a computer or a sandbox, as it's popularly called, right? And so the team has been working together, the founders for almost 20 years, so we're quite old, and we've always been in for people, everything from screening together servers to creating orchestrators and whatever. Interesting. Orchestrators. Tell me more about where you're seeing the orchestration market going. I'm a big fan of Gastown and the metaphors, and it feels like it could be like the hot new trend, but I don't know if it's too early or where we'll see all that go. I mean, trends keep changing, as you've seen, like every two months, things change. But one thing is for sure, and this is what, I mean, I love what Gasound is doing and all these others,

which is basically the ability to spin up multiple of these agents. Each of these agents needs a computer. And I don't think we stop here with like two or four or ten or fifty. Like this will go exponential. And so that's why we're doing what we're doing. So we really love what Steve is doing at Gas Sound and all these other people that are creating this. Talk more about where the product is,

the evolution, where the team is, kind of the current state of the company and where you want to go. Sure. So the company's almost three years old, but we actually like hard pivoted and like fired all our customers and everything. And we decided to, yeah, that was a hard one. And we saw what was coming. We weren't actually at first,

but we did see that agents would need these computers. And so we actually rebuilt the product from Zero, launched it eight months ago. And since then, we've got the number of customers has been insane, the growth has been insane. We've everything from YC companies, growth companies, Fortune 100 companies. Everyone building agents either internally

or as products are using us or someone like us. Talk about why an agent needs a computer. You know, there's been this debate over OpenClaw, formerly clawed book, around like, oh, should you host it virtually or should you run it on a MacBook Mini? And if you do run it locally, that's a security risk. But then you talk to the people and they're like, okay, yeah, you're running in the cloud, but you still gave it access to your email. And so, like, you kind of let the fox in the henhouse.

So what does it mean to actually provision a computer versus just APIs to other compute resources? Yeah, exactly. So I think that's the most interesting part where originally agents were like these, use like rag-based systems, they could connect to APIs, and they can usually consume information, analyze it, give you some feedback, and fire off some tasks. But if you actually think about that, if that is a way a not digital knowledge worker would work,

the two of you wouldn't have laptops in front of you, right? Like there's, it's not that ideal. It's, you know, it's messy. You have to download things. You have to install things. You have to analyze things. And if we think of AI agents as digital knowledge workers, it actually makes, like, super sense that they do need these computers.

The OpenClaude, or whatever's called right now, and Cloud Code and whatnot, I can't, there's like three names. It's open Claw. No D. Open Claw. Yeah, so this just shows that everyone buying Mac Minis. I think this just made people understand why it needs computers. But if you think of, if you look at it,

at our use cases and what people use each I wanted to build for. We basically have three, which is code and command execution, computer and browser use and URL. And so the first one, like the canonical, these are not our customers. I'll just say the highest brands, just so people know. It's like if you think of like lovable, and you think of lovable, you as a human type with lovable, and then it spins up a sandbox to write the code, preview the code, and run the code. And so lovable has how many users? And so it needs like hundreds of thousands of these sandboxes to be run so their agents can get there, right? The second one is, like, if you need your agent to, you know, do QA testing of

booking.com's website. Like, it needs a computer to be able to run the browser to be able to do these things. And the most, the newest sort of use case for us is just like the RL environments. And I think Dylan Patel, who was a guest yesterday on your show. And also, I'm going to also know, we're doing a conference next week. And so he's one of our guests. Yeah, he's going. Yeah, he's going. So at the Chase Center, we rented out of the Chase Center to do the

conference called Compute. That's amazing. So you're very invited, if you guys want to join. And so he has this article about oral environments and how most of the models have now over the last 18 months become better because of post training and spinning up these are all environments.

And these companies to spin up these are all environments, they need not only fast computers or sandboxes, they need them in concurrency. So they need 100,000 or 200,000 to be spun up in an order of like three minutes, right? This is not a trivial thing and not something

that you can really take off the shelf. And so when you think about why I just need computers, like when you really think about the market of AI agents and software, that's sort of where that fits in. I want Dylan Patel doing the Liliati walkout that he's made fake AI versions of it so many times. And now he's at the Chase Center. 100, 1,500, 1,500 builders there. You've got to get the Liliati playing so you can do the walkout.

I saw first mark in the chat. Give us the news. What happened? Break down for us. A $24 million Series A. Boom. Congratulations. Thank you so much.

Thank you so much. Yeah, very impressive progress. And I'm sure he'll be back on very soon. And have a great time at the conference. Yeah. Pulling together 1,500 people. Yeah, we're excited. Thanks so much.

We'll talk to you soon. Great to meet you, Evan. Have a good rest of the day. Goodbye. And without further ado, we've been keeping our next guest. waiting he's live in the TBPL Trudome it's Scott you he's suited up him he's suited up thank you so much for coming here to look at this suit this is that game worn someone's gonna bring some swag to this show I know no quarter-zips here

no half-assin custom suits only yeah where it's actually yeah what's the origin of suit. No, Shinesty. I got a prep. You know, you guys got your sponsors. Yeah, yeah, yeah. I got my sponsors. Shinesdy.com. They make all my custom suits. By the way, it's a shame the sports card company Fleer isn't still in business. Fleer? Can you remember Fleer? Sports cards? Yeah. Because then they could have called us the Fleer Ultradome.

Yeah. You've been given out gold medallions. I had to do the for Dylan. Dylan. Dylan needed a fleer ultra records. Sorry. He's a big. He's a big. Yeah. How'd you meet Dylan? Dylan and I met at HQ. Okay. When he came on board, I was hosting it, he came on board as a partnerships guy. Absolute Wizard. You guys are so lucky to have them.

We are. With Savvy, if we were just a little earlier market, maybe we could have hired him first. Yeah, yeah, yeah. You guys, we'll poach him one day. Yeah, yeah, introduce Savvy. That'll be a good talent, that'll be a good day. And then, yeah, introduce Savvy, get us up to speed, and then, of course, we'll go back and talk a lot about HQ. Sure, yeah.

So, Dylan and I met at HQ where if your viewers aren't familiar, was this live interactive, mobile game show viral phenomenon. Back in 2017, you guys played? Oh, yeah. H-Quties here? Of course, right. All right. I never won.

I don't think I ever won. Neither did I, man. Oh, well, you were playing. That would be insider trading. It was tough, though. Honestly, part of, like, the journey to savvy is, you know, taking a lot of learnings and insights from HQ, which tremendous success at first, viral international hit. Yeah.

But when you think about it, and in hindsight, you go, oh, maybe it was a very, you was a flawed product because it was this really impossible game to win, right? Very few people actually won it. And a lot of very smart people, including Stephen Colbert, when I did his show, he told me behind the curtain, he didn't do this on the camera, he goes, I played your game a couple of times, didn't win, stop playing. And it's like, oh, interesting. Because if you're a smart person, you want to feel stupid. Yeah, just losing over and over and over and stuff. It's not a good feeling. So from that perspective, it was maybe not the best. Because, yeah, if you play Fortnite or Call of Duty and you're not doing well, they will pair you with other people,

that are at your skill ceiling, like the skill-based matchmaking happens in most. You go to chess.com. You'll be matched with someone at your level so you can get a couple wins under your belt. With HQ, it just gets more and more competitive, more and more competitive. Exactly.

And then the company obviously has to raise the bar or else everyone wins. Yes, and iterating on the products, adding features, adding new formats, they just never quite got there. So with Savvy, one of the takeaways is, we want to make this game something that everybody can play all the way through

because with HQ, if you got out on question two out of 12, you were done. Yeah, that's right. Yeah. So now you can play all five rounds of savvy, which are word puzzles. Okay. We thought about trivia, but I think AI has put a kibush on trivia for cash.

Yeah. Because it's just too easy with these bots, man. But with tech savvy's the end of the game. Yeah, somebody would have just like the biggest bot farm in the world just trying, just winning every single time. I mean, they were already doing that eight years ago, so just imagine what you can do now. You guys were dealing with bots? Yeah. I mean, I don't know if it was AI back then, but there were, you know, Discord farms and.

but people, I don't know, I'm not a tech guy, so I don't really know how that stuff works. But with Savvy, I mean, we're doing word puzzle. So it's basically like a live wordle meets connections. I don't know if you play those games. So it's a version, our version of that, five rounds, timed rounds, you can score points. And the key differentiator, you're playing against me.

Yeah, it's you. It's the host, the host plays too. So imagine HQ if I was just, if I was asking the questions and answering them, and if you answer more than me, you won. Imagine that's how HQ works. That's how savvy works. So if you score more points than me, if you solve the puzzles quicker,

if you get, you beat the host, you're a hostbuster. A hostbuster. A hostbuster. So like that. What's your, what's a day in the life like then?

Right now? Yeah. Well, guys, I just moved to L.A. last week. So I'm trying to get my day in the life. Oh, I love it. I mean, I was here a few years ago.

New York guy before? New York, my whole life. But I spent about 21 to 24 in L.A., moved back to New York for some relationships and came back here for my new relationship. My new commitment is this app because frankly, the time. Married to the game. I'm not married to anyone else, but this game. I'm committed like Kurt Russell to Goldie Hawn. You know, I'm that level of commitment. That's great. Arizona IST, 99 cents. That commitment. That's my commitment. That's my commitment. But, no,

I'm here for a week. So I really haven't gotten my day of the life routine yet, but I'm trying to. I walk the dog in the morning. I'm still getting up early with that time zone difference. You know, having my coffee, making a breakfast, doing a great big French toast. Are you going to stay on New York time? I thought about that. It might happen. I mean, you guys are up early for the show, I know. So it actually helps, right? Kind of staying in that. We do the workout. We go get breakfast, prep the show, go live at 11. How important is going live at a particular time, the iteration, any of this, recording? Like, how are you thinking about just rotating another host? Like, your workflow and your involvement, because I feel like the unlock is the talent.

is the host, right? For sure. And we're going to definitely find other hosts. We have to. I mean, I already just talked today about doing a gig in Miami on March 18th. I'm like, all right, I got to find. So we're going to have other hosts.

We're going to have other shows. That's all going to happen. But look, this is so cool being here. First of all, watching you guys grow the way you have. I've been following along and watching. It's very cool to see. And it really brings... There's a lot of HQ in here.

There's a lot of HQ in here. There's some clubhouse. There's a lot of different stuff that has been tried and we sort of pieced it together. And you realize the importance of cadence, of going live every day, for you guys having that long show, for us just a short burst of tight attention span.

But, you know, a whole different thing. We're playing a game. But it really is like, so for example, we were doing beta shows weekly in New York. Okay? And I knew going in, like, weekly isn't going to really do it. Never do anything weekly. Always strongly.

Ah, I like that. That's what we say. Never loud to. It's good. So we started doing daily shows now. We're going live five times a week, Sunday to Thursday, school nights, live on school nights at 9 p.m.

Eastern, but doing them here. So last Thursday, we had, 1,200 peak concurrence. About 1,500 total entered. Last night, less than a week later, 3,000. Wow. So we've had doubling growth week over week. There you go.

Monetization, we sold one T-shirt on T-Public. Got three bucks for that. Chiching. It's our first $3 in the door. Fantastic. I think we made four cents on TikTok live. Oh, okay. So we are monetization, post-monetization.

There we go. Post-revenue. That's great. But no, listen, I'm bootstrapping this thing to talk business. Yeah, I was going to ask. So you talked about some of the product challenges with HQ, just if the average user is losing every single night, that's rough. But yeah, what about from a business standpoint?

It also feels like HQ was in this position where once it had real venture dollars behind it, you just have to go for scale. And when you're bootstrapping, as long as the business is sustainable, there's no reason to stop. You can reach a point where you're like, okay, we've hit a ceiling, and that's okay. Like, maybe there is no ceiling, maybe there is, but it doesn't really matter either way. Yeah, I mean, I don't have a very sanguine view about VC culture and then holding, based on my experience with HQ. I mean, look, you give a bunch of young people a lot of money and just say, go scale a company, go lead a team, go manage people.

Not very many people who are cut out. Go compete with Facebook. Go compete with Google. But really, even just like personal, I mean, also, you know, business sense. And, you know, most of these people didn't go to business school. Anyway, my point is we have like a philosophy here where we're really trying to just bring as much as we can to this as a full team. I'm putting my money in. One of my other co-founders putting money in.

And we are really trying to commit to doing this ourselves for as long as we can. We did take a couple meetings with some like early stage seed guys just to kind of validate the product. I'm not like the other VCs. I won't tell you to grow at all. The other question is like is this almost more suitable for someone who is in the entertainment industry? And not thinking about this is, okay, this is going to be a tech company and AI company, blah, blah, blah, but more like the values in the talent, the values in the community, the values in the product. Yeah, I feel like, I feel like in Hollywood and entertainment, it's very much calculus of like, hey, if we can spend $20 million to make this movie and it makes $80 million, that's a win. Yeah.

And in VC, if a company says, we're raising $20 million and we only ever want to make $80 million. And then we're opening it up and moving on to the sequel maybe in 10 years. Which also is insane to me because 60 million is not bad. Like, I'm very happy. Give me a quarter of that. Yeah. So, no, I mean, listen, I get it. There's a whole ecosystem of the investment.

And listen, we were certainly open to it. Listen, we'll be open to it if we need to go there. But thankfully, we're in a position, although I don't know after the market today, might have to reevaluate some of that. We'll see. Thanks. Bad day. How do you originally get the job at HQ?

You're just, you check your portfolio, you're looking in the mirror, put on and start putting on the classroom. I'm getting a text. I'm getting a text like, yeah, we actually do need a check right now. So many VCs are watching. No, sorry. How'd you get the job at HQ? Oh, that was an audition, man. An audition.

So, yeah, even though it was a tech company, they held full auditions. Like, did you, did you see it on like a talent casting website or just a Craigslist ad? It was a flyer? It was purely like a, you know, a friend networking thing. Yeah, I mean, a guy used to work at the onion. Let's give it up for network. Oh, I love the other. Let's go for networking. Yes.

The Onion. You worked there? So in 2008, I interned there. Wow. And this guy, Nick was the photo editor. Yeah. And we became friends, you know, a couple of young guys, funny guys hanging out in New York. And we stayed friends throughout, you know, on Facebook mostly. But, like, I got my roommate a job at the company that was HQ, it turns out,

because Nick had put out a message on Facebook saying, hey, I'm looking for an animator. My roommate was an animator. So Russ got the job. there. Oh, yeah. And then lo and behold, a few months later, Nick calls me and goes, you know, we're casting a host for this game show on your phone. I'm like, I was about to move to L.A. This was 2017. I already moved down to my apartment in Brooklyn. I was ready to go, and I'm like, I guess I'll do one more edition. I've failed every edition. Just flunked him left and right for shows like Broad City and Search Party. I remember the Delta commercial

and Noah Cindergars was going to take my shirt off. And I was like excited to do that. You were diced. I love Thor and the Metz. And I'm actually a JetBlue guy, to be But listen, I failed all those auditions, and this is the one edition I got, the last one I took, and it changed my life. What can I say? It's just, but you know, you mentioned, like, the whole media and tech hybrid, and that's what makes, it's what made HQ very unique, but also what makes savvy unique because, look, HQ had interest from NBC, Disney, Fox, also Facebook, also, you know, the venture capital, you know, founders funnel to put in 15 million. I remember. But, you know, very few companies have that type of dual interest in a business. attraction from those major, major spheres of influence and money. And it's about managing those

types of investment partners, but also just managing the product and understanding that, you know, when you're doing a live interactive game show on a phone, on your own app, just like what you guys are doing in terms of building your own network, you got to think about, yeah, there's technology involved. Look at what's happening with the streaming and everything, but it's a show, it's entertainment. And you guys are doing it all, just like we're doing it all. It's exciting. It's very heady. It could be scary at times. Very few people are doing it. So give ourselves some credit for doing the media tech hybrid-ish. They said it was impossible.

How soon do you think Zuck will clone savvy? I would love for him to try. Someone else called H-K. Yeah, yeah, no. That was up. Break down how you process that. Listen, man, I mean, they say flattery is the, I forget what they say. I forget what they say.

I'm the sincerest form of flattery. We like to say imitation just sucks. Imitation, especially imitation crap. It's for losers. You ever that imitation and crad meat, those California rolls? Disgusting. It's disgusting and despicable, and you should be ashamed for imitating me. And you say this.

Not flattered at all. Right, exactly. And nobody at CNBC would have any issue with that. We all imitate, we all remix each other. But that's the thing. It's like, you don't, what you guys are doing is like, oh, I don't need a job on CNBC. I can create my own goddamn CNBC. Make your own squabbox.

We love CNBC. Of course you do. But I honestly would have never even thought like, oh, I should have. apply for a job there? I don't even know how to get into that world. It's just like, yeah, if you're right, is like, yeah, if you asked John and I, even a year and a half ago, hey, would you ever think about, could we, doing business television? Yeah, a news anchor. Right, right. No, like, we're like, let's have a podcast with two microphones, just talk to each other. And they turned into a TV show, kind of, but streamed online. But, you know, someone,

someone asked me when I was hosting HQ, I get interviewed, and they say, well, where do you want to be in five years? What's the next move for you? I go, I'll be here for 20 years if this thing stays on this is the future. It was so good. Unfortunately, it didn't last very long, but with savvy, like I said, I am in this to win this for the long haul. My skin's in the game. My face is, again, you know, front and center for better or worse.

I don't think America, I don't think 100% of America loves this face, but enough people do. Yeah. Enough people do. Yeah, what does chat back like that? You don't have to run a while. Chat can be a little rough, you know? You know, there's all.

We've been there. We've been bombed by all sorts of, just today. I don't know if you saw. I guess word leaked that Sam was coming on to the... Leaked, we promoted it everywhere. Well, no, I'm saying it leaked to the part of the internet where there's a lot of people that are really frustrated that 4-0, the model from opening eyes being deprecated.

A lot of people fell in love with the model, have a very close relationship with the model. And so we basically couldn't even look at the chat at one point because there was a message every second from somebody saying. I didn't realize how big that army is. I've seen a couple posts. I thought mostly maybe fake AI generator or something like they were there. they were real and like and and they had points to make yeah this is the problem with people who

can't yeah fortunately yeah fall in love with a fashion model they fall in love with AI models and it's a whole different type of person unfortunately we have like a really core community that's really positive and everyone gives each other feedback like the the core group is like really really solid of course no I mean I'm I'm just screwing around like yeah obviously every chat's going have a few trolls in there yeah of course but no we welcome everyone and and there really are like HQ the community of HQ was so tremendous this, these HQdies. I mean, millions of people playing. I've heard from people who got married because they met someone playing HQ, you know, strange families coming together again. How many,

people of the early user base for Savvy are just people that were waiting for it to be reborn in some... A bunch. I mean, you know, what we noticed was, again, live interactive, not all people are doing it, and platforms aren't built for it. So even on like TikTok and Twitch, when people are going live, they're finding hacks to make interactive gaming, but no one's doing this one-to-one interactive, the host versus the audience, streamer versus audience. Very unique. And that's kind of the thesis beyond building our platforms. We want to make this streamer-to-audience interaction that no one else is doing, live appointment-based, games, cash prizes, the whole shebang. So how do you think about syndication and marketing, like turning the show that's

live interactive? That's a very high bar. I mean, we see it with like the number of people that show up live and watch the whole show is way lower than the people that watch our RSS feed or YouTube videos or shorts or clips. And so how do you take something that you got to be there? But if you see it, you might be like, I want to be there for the next one. Exactly. Well, that's the marketing. Right. I mean, we haven't spent any money on marketing. It's really just been my social media posts that have gotten us to these 3,000 people or so. That's great. And then, of course, you know, coming on shows like this. I'm hoping we get the TBPN bump tonight. Oh, yeah. Let's see what we get to 3500. If it happens, we're taking

credit. We're taking credit. You should. You should. You should. I mean, how about even say Maltman's trolls, we'll take you too. Tell them, tell them that 4-0's hosting. Four-o-hosts. That's a dangerous proposition. For a co-host forever. You can never stop. You touched on monetization earlier, but is this the kind of thing you get, if 10,000 people, 20,000, 100,000 some days showing up every single night, they would happily pay some amount of money for that experience, right? Absolutely. And again, to bring it back to Dylan, who was the responsible for getting those deals on board.

I mean, we had, ultimately, we were doing million-dollar deals for single shows. Yeah, but why wouldn't, why wouldn't a player just subscribe, too? Oh, we have subscriptions all of us. Yeah, paid subscriptions, so you have that, like, because I think if you're running, yeah, I mean, I can just imagine if you have

100,000 people that play the game all the time, why would they not play? There's a lot of value there. No, yeah, we're going to do subscriptions. We're rolling that out for our season one premiere March 1, subscriptions, and, of course, sponsorships could be a part of that.

to. But no, absolutely. I mean, the, you know, when we see people on Twitch donating $10 a month to their streamers for nothing really in return, it's like, well, if we can offer, you know, I mean, it's just donate. It's like, it's a good, you know, Patreon. It's support. No, it's support. And that's, it's an amazing community of support out there, like from fans to artists, right, fans to content creators. So we don't have that mechanism yet, but when we roll out the subscription, it's like, hey, if you love savvy, if you love Rock with me, throw us some cash, and then we'll give you your digital rewards, your customizations. Yeah, Fortnite skins. Exclusive games, though, maybe higher prize amounts?

Talk to me about a million-dollar HQ show. What does that feel like for you? What's an example? Walk me through sort of like how those big shows, those big monetization moments came together. How they came together was we were able to get a million people playing in the game live. If you get a million live devices, and it turns out, guys,

if Dylan only knew this when we were making those deals, we probably could have doubled money. Because later on, we hired Nielsen, the ratings company, to come in and do a study. And they put like a 1.7% multiple on our viewer. So frankly... People are looking over your shoulder. Exactly.

If there's one person holding a phone, maybe there's an office all playing together, family planning together. So honestly, two and a half million connected devices was our peak with the rock when he hosted. That was a Warner Brothers event. So it was sponsored for Warner Brothers for a movie.

To promote Rampage. Got it was the movie at the time. Instant classic. Modern classic Rampage, of course. Still getting paid. Still talking about that. movie. But 2.5 million connected devices, really, yeah, close to like 5 million viewers.

And then what's your interaction with the Warner Brothers team on that? Like, are you stopped? Yeah, during the show, is it just overlays graphics? Are you stopping doing ad reads? Like, what do they want to really make that pop for a million bucks? I mean, we showed the trailer. We put the trailer in the lobby. And I definitely talked about it.

You know, probably wrote a whole script around. I did a Ready Player 1 deal. And maybe some questions that are linked. Some questions linked to it. A lot of, you know, graphical integration, the UXUI. But I think it were like the Ready Player 1, goggles on air even. So that kind of stuff. And live product reads are going to be part of Savvy as well.

Yeah, totally. Again, we're taking the live podcast ecosystem of ads, as you guys well know. You guys are like NASCAR drivers over here. Yeah. Do you think that energy or compute will be a bigger bottleneck? Great question. And I don't know what either one of those words means. No, you know, yeah, it's going to get expensive.

But I expect you guys to train your own foundation model. Our own, our own replicants. We got a foundation model right here. You're a foundation model of a man. I'm a foundational element. I'm a fundamental aspect of consciousness live on your screen. You know, real humans are going to be a scarce commodity in the near future. How are you thinking about building the team?

Yeah, I mean, it's important. obviously. I'm CMO, co-founder technically, so we've got our C.O. Josh, our CTO, Ben, and CEO, Johan and Ben are based in Europe, actually. Most of our team is in Europe. We're distributed across the globe. And they're doing a lot of the hiring decisions there for the key roles, backend stuff. We're looking, but we are looking, you know, animators, social media manager, actually, community manager's key. And really, we're looking for just, you know, it's like the B do have way of thinking. We've just, we just, we just, we just, we just. We just, we just, want people who are going to be their best, bring their best self to the job. And if they're doing that, then they're going to have success and we're going to do well. Yeah. Did you get any inbound from companies that wanted to dip their toe in live, post-HQ, people that want to, like, what was the post-HQ sort of like idea maze for you? Oh, it was wild, man. Because I imagine you could just go and say, like, I'm going to be an actor, put me in a TV show or whatever. There's game shows,

so you can compete in that. There's also a lot of tech people that were aware of you. How do you navigate that? Well, I mentioned the failed auditions before. I'm not much of an actor, it turns out. So I wasn't going to go that route. Play yourself well. I do play myself well. That's all I really need to know how to play, just me.

The way I look at it is like there was probably two dozen people contacted me over the last seven years. You know, I had this baseball show. I went from HQ to this baseball show called Change Up on DeZone. It's supposed to be a three-year deal. COVID killed that. The major leagues became the force major league. leagues is the joke I like to make.

Oh, yeah. They killed, it's a contract joke. Because, yeah, because baseball went out hiatus. Baseball was the hiatus. There was no need for the show, and they used as an opportunity to pull the plug. Brutal. So I lost that gig. But look, I had like dozens of people reaching out about all sorts of things.

I even tried to start a company back in 2019, 2020, pre-pandemic with a guy, and that had its issues. But, you know, the thing about these guys who hit me up, it was Ben, and Eman and Yohan, And these two, randomly on Twitter, they messaged me over a year ago now and said, hey, we have this idea. Interaction, host versus streamer. And I'm like, I've heard a lot of pitches. I took the meeting. They show me their demo.

I was like, these guys are, they're like 27, 20 years old. And I've met with lots of people, lots of ideas, some fly-by-night stuff, some really earnest people, will just end up the right team, right product. These guys seem like they had the experience with mobile gaming. That's what they pitch me on. They go, we work at mobile gaming companies here in Europe. That's what we need. So they know the retention mechanics.

They don't necessarily know. have needed to do the exact same thing, but having done something, the scale, the interaction, the engagement, all these things. Something else lacking from HQ, by the way. I don't think a single person on the team had mobile gaming experience

or Hollywood production experience. It was wild. No, one had any experience. No, seriously. Does anyone here have experience? I mean, no, very talented people. Coming from Twitter, coming from Uber, coming from big tech companies.

But I really don't think, I mean, maybe I'm wrong. engineers maybe it, but mobile games a very specific field. And again, entertainment and Hollywood media is very specific. So I brought that entertainment side of things. These guys have the mobile gaming tech side of things. We've married together. Our CEO, Josh, is kind of the glue guy who's really, you know, keeping the company running and taking care of all that. And also our production manager helping set up the streams. He just did a whole overhaul of our new studio here in L.A. Shout out Forever Dog Studios. Our version of the Ultradome. But no, I mean, it takes, as you know,

Which one's this? That's the dog panting. What's the name of the studio? He said Forever Dog. Oh, Forever Dog. So he played a dog sound. I like it. Okay, Forever Dog.

There we go. The dog. Forever Dog Studios. Thinking to my guy right now. Just left my dog at home. But no, man, it's a unique entity what we're doing. I say we, because you guys are doing it too. How are you going to that?

I love media. I love that it's a business that doesn't exist if you don't show up and make it every single day. And that scares a lot of people, and sure there's downsides to that. A lot of people want to say, you know, the idea of SaaS, it's beautiful. You build this thing one time, you sell it, you know, indefinitely. Passive income. This is not passive. I heard you talk about UGC.

And this is the classic dilemma for me because what HQ did, I think, amazing. The most innovative aspect of HQ was the fact that it was an app that worked for 15 minutes a day. Yeah. And it was not UGC. It was PGC, producer. generated. There was a single show, a single purpose on this customized platform. What a

concept, right? Because you could say, well, let's open up to everybody. And the funny thing is, the twist on this is that HQ derived from a previous iteration of an app called hype. So the founders, they had a company called Intermedial Labs and it was about building apps. They built some kind of

dance app for Instagram. They built an app called hype. And it was, we're going to allow people to make their own talk shows from their bedrooms, give them templates, give them music, almost like TikTok talk show thing. And I don't know, maybe a few thousand people downloaded it. One woman was doing a trivia show from her bedroom. And they saw the show, they're like,

let's just do our own show. And that's how they got the idea for HQ trivia. So they tried UGC at first. It wasn't working the PGC month. Yeah, the challenge is a very small market of people. Even Twitch is still so power law driven in that you take away the top 30 creators and the platform is probably worthless.

it's very true so we are basically kind of gatekeeping the creators on our platform very very highly curating it it's probably better way to say it curing it to just me right now but we'll open up if you guys Scott's gatekeeping he's a negative word right yeah yeah that's okay but you guys I want to bring you guys on the show yeah that'd be great you'll co-host with me we'll do cross promo we know how we're the app is in the app store and I want to read this review it's amazing HQ done right this app is so much fun I've missed the live game shows of the 2010s and this app delivers on the the best of those. Not only is Quiz Daddy back, but there's also new features that reward you just for playing the game. I'm excited to see how savvy grows from here, but they're off to a

promising start, five stars. Sounds like ChatGPT. That's not ShatGPT. That's very nice. I mean, and really the feedback has been phenomenal. The community's been phenomenal. Can we sponsor a suit? Yeah. Do you want your own cost? I'll connect with your chat. No, no, yeah, I want to do a TBPN suit that you wear during the game. Oh, oh, I like that. Oh, you had stickers. And what about I'll give you some stickers. Let's do it. Let's do it.

Let's do it. Let's do it. A console allowed. And yeah, for sure, I'll wear a TVPN suit. Let's get the custom TVPN suit. I love it. I'm super excited for you.

Sounds like you have a fantastic team. This is a fit. Founder market fit. Yeah. This is the fit. You got a, your, our friend Jeremy Gafon talks about this idea of everyone is like pre or post fall.

Like you've like, like there's people that are doing really well and yet they haven't. you know, been, like, deeply humbled by life yet. They never suffered a setback. You've gone, you've gone through that now. Daddy knows. Daddy's eaten a helping of humble pie or two over the years.

It tastes pretty good, actually. Kind of more than your fair share. You know, look, I'm not, I can't complain anything. I've been so blessed. Even these last seven years of, you know, kind of wandering the desert to get to this point. But I'm so excited for this launch. And truly, if we can get the TBPN community, on board tonight. Download the app, PlaySavvy. Live. On board the cameras, PlaySavvy.

Dot Live. Do it. Android. Apple tonight at 6 p.m. Pacific, 9 p.m. Eastern. Let's get over 3,000 encourage, and I'm going to give you guys credit. Fantastic. Well, I think it's time to plant the bomb and get out of here. Thank you so much for coming on. We'll close out the show with you on here. Leave us five stars on Apple Podcasts and Spotify. And to the newsletter, TBPN.com. Yeah, we did have range. We went all over. It's fun. Sam to Savvy. Sam to Savvy. Nice book.

It's good stuff. Living Legends. Yeah. And we will be back tomorrow at 11 a.m. Pacific. We have Doug O'Loughlin joining. We have a bunch of other fun folks joining the show. And have a great rest of your day. Have a wonderful afternoon and evening.

We love you. Bye. Nice work, brothers. I'll see you on the next one.