TBPN - Dan Wang's Annual Letter, Meta Acquires Manus, Nvidia's $20B Groq Deal | Justin Mares

Episode Date: January 5, 2026

(02:25) - Dan Wang's Annual Letter (43:08) - Meta Acquires Manus (59:33) - Nvidia's $20B Groq Deal (01:10:48) - Doordash Responds to Reddit Firestorm (01:43:53) - California Wealth Tax Targets Billionaires (02:00:54) - Scott Belsky's 12 Outlooks for the Future (02:27:20) - 𝕏 Timeline Reactions (02:32:31) - Justin Mares, co-founder and CEO of Truemed, is a wellness entrepreneur dedicated to making preventive healthcare more accessible by enabling individuals to use HSA/FSA funds for health-promoting products and services. In the conversation, he discusses Truemed's recent $34 million Series A funding led by Andreessen Horowitz, the company's mission to shift healthcare spending towards prevention, and the growing adoption of HSAs and FSAs for proactive health investments. 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.comRamp - https://ramp.comRestream - https://restream.ioShopify - https://shopify.comTurbopuffer - https://turbopuffer.comVanta - https://vanta.comVibe - https://vibe.coFollow 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 Monday, January 5th, 2026. We are live from the TBPN Ultramm, the Temple of Technology, the Fortress of Finance, the Capital of Capital. We have so many updates for you. But one that's not changing is that this show is still brought to you by Ramp. Time is Money, Save Both, Easy Use Corporate Cards, Bill Pay, Accounting, and a whole lot more, all in one place.

We are very happy that Ramp is staying with us on the journey. That's right. Massive news. new graphics package. Yes. If you haven't noticed, we worked with John Palmer and Luke over at Area Technology. They updated the overlay, the logo, a whole lot of stuff. Glove in there telling you, we're taking you all over the world.

That's right. News every day. That's right. It is so good to be back. It's amazing. So we met this morning at the gym with the whole team, and I hadn't seen you for more than two weeks. It had been the longest time we've gone without.

hanging out since we started the show. I mean, last break was crazy. We were still doing, we weren't even live yet. We were doing it just as a podcast, but we still, I was in Yosemite once, and we did like a 20-minute episode that was recorded locally, and we had to sync them up

and, like, edit it together. It was a very odd show. We did a show on the Eve of Christmas Eve. We did. That felt really unnecessary, but at the same time, there was so much news over the last two weeks, there were a lot of moments

where we wanted to be live. Yeah, I feel like there were three or four, maybe five really big stories where we could have had great shows. But there were a lot of days where, you know, it probably would have been us just hanging out, which still would have been fun. And there were a lot of situations to monitor too. There were. There were. There were kind of outside of our. Our domain.

But still interesting. And we'll take you through some of those. But fortunately, we have a ton of content and a ton of stories to get through because so many things have built up. The Grok deal, the Manus deal. Dan Wong's annual letter, there's so many different things that we're going to. take you through today. We have a very light lineup today, just Justin Mayors from Trumad joining, but we have... Pull it up. Let's see it. There we go. Which explains the run of show, so we're going to be building this out for you. I'm not to keep you guys in the dark.

You exactly what we're doing when, so you know when to tune in. But we should kick it off with Dan Wong's annual letter, his 2025 letter. He skipped 2024, apparently. This sort of rocked the timeline. Dan came on his show during his book tour. for his, you know, excellent book that sort of reset the narrative around the AI competition. He uses this phrase AI, he doesn't like the phrase AI race. Yeah. He likes, he doesn't think it's something that you can win. A race has a definite ending.

There's a finish line. Whoever crosses at first wins. And that's where a lot of super. He prefers to say that the U.S. and China need to win the AI future. Yeah. Yeah. And I feel like he. And this echoes what Gurley.

has said too. True. Obviously, at the time he was coming out in defense of Manus and benchmarks investment in Manus. Totally. He was saying, I don't, he's like, I don't know what the AI race is. I mean, he looks great now that Manus is an American company. It's a meta property. So, you know, and I mean, the Manus team, of course, it was from China, but quickly

moved to Singapore and now Manlo Park presumably. No, I think they're going to stay there. But still, I mean, you have to imagine that a lot of talent migrates. and it's like a fully American controlled asset, essentially, now that it's controlled by America and Meta and Mark Zuckerberg. Let me tell you about public, still with us, investing for those who take it seriously, stocks, options, bonds, cryptos, treasuries, and more

with amazing customer service. You can see we have a new ticker. Let's take it over to Tyler. Let's check him with Tyler. Oh, yeah. What are we checking in with Tyler? The chat misses Tyler. Oh, they just missed Tyler.

How are you doing? Also, we signed for 2026. massive contract extension with Tyler. He hasn't, he's not technically a college dropout yet, but he may as well be. It's a gap year. It's a gap year. It's a gap decade. It's a gap century. Tyler, Tyler's parents, if you guys are watching, I'm just joking around. It's the age of gap years. We're in the age of gap years. We're in the age of research and we're also in the age of gap years. And so we thank Tyler for sticking around hanging out with us. We're going to have a lot of fun this year. So the Dan Wong piece, we should read through some of it.

I wrote, it inspired me to think about, you know, what am I really looking for this year? We've seen sort of benchmark saturation. AI can kind of do everything now. We'll talk about the Claude Opus 4.5 and Claude code sort of sensational takeover of the timeline. So many people clearly hadn't really tested Opus 4.5 when it came out, which was a couple weeks before the break, but they got their time off. They were, you know, with their families and probably stepped out to the laptop and fired off some Claude code prompts and had really good experiences. Yeah. Because it is really magical the first time you go to it and you say build this

website and it just does it. It's actually remarkable for small hackprivs. People are thrown around the Claude. AGI. Yeah, yeah. Fortunately, we picked up a goal post. Oh, yes, yes. We have a camera view that we can put on this. Maybe if I go over. Just pull it, pull it over. Uh, So we picked up a gold post, so, you know, as we continue to see AI progress, we can actually move this goal post ourselves physically here in the studio. Physically moving the gold post now. There you go. There you go. Because my only definition of AGI is it needs to be able to smell.

It needs to be able to smell. It has to be able to smell. If it can't smell, it can't do all white-collar work because I regard the job of a Somalié as white-collar work. Now, AI disagrees with me. I actually went to one of the models, and I said, is a Somalié considered a white collar worker? And it said, no. And I think that's all just a sci-hop.

I consider it white-collar work. You often drink wine wearing a white collar. Do Somaliers not wear white collars when they're serving wine? Absolutely, they do. They should be considered white-collar workers. And in order to fulfill the AGI promise of automating all white-collar work, it must be able to decant. I sent you that paper, though.

That's about the olfactory sense of training models to have olfactory senses. Well, as soon as it gets baked into the next version of Claude, we'll need to come up with a new benchmark, and then we will move the goalpost once again, because that is the nature of the show. We continue to move the goalposts endlessly, but you know what doesn't move the goalposts? Gusto, our latest sponsor. It's the unified platform for payroll benefits, NHR, built to evolve with modern, small, and media. V-sized businesses. We've both been using Gusto for a decade now.

I went through Y-C with them. Or they were either right before me. I think they were 2011 or 2013. And yeah, amazing, amazing platform. I've actually, for over a decade. My entire career, I have been getting paid via gusto. Used it. That's true.

At my first company, I still get paid through it. Yep. It's highly recommend. Continuous. So very, very excited to work. Serious, serious endorsement. Anyway, back to Dan Wong. the goat. He is reflecting on the AI future, what it takes to win the AI future. And the thing that

stuck out to me was that we've seen this amazing, just, you know, this remarkable march of research. The models are getting better. All the benchmarks are getting, you know, completely dominated. Then we're seeing the massive hyperscaler buildouts. We were reflecting on like, what does reindustrialization mean? What does American dynamism mean? What does this idea of like America needs to build big things. We need to be able to build big things. New bridges. We need to build high speed rail.

And we've failed at a lot of that, obviously. But we've succeeded at building multi-billion dollar data centers. No, I mean, yes. Hot-Kess, yes. I mean, truly media has been a cultural export of America for a very long time. But, but we have successfully.

So Dan saying we're very good at making data centers. haven't been so good on the energy set. Yes, and so I tried to pull some of the data on, you know, my big question was like, we've been talking a big game about energy being a bottleneck for AI for a long time. It wasn't truly the bottleneck with bottleneck was chips and then data centers and then just moving the energy around, but it feels like the last year we were just kind of shuffling energy assets around the board, and that's why we saw energy prices go up where if just supply and demand, if we'd been building a ton of energy infrastructure, well, supply would have gone up and actual prices would have fallen or at least stayed flat because all the net new data centers would just be using the new energy infrastructure.

But that's not exactly what happened. So from 2008 to 2021, America's annual growth in energy production was 0.1% annually. Really, really bad. Not good. But it is getting better. The EIA's December 2025 short-term energy outlook projects generation growth of 2.4% in 2025 and 1.7% in 2026. That's like, you know, what, 20 times higher? That's great. But also that doesn't feel like, oh, fast takeoff. We're going to be doing 10% jumps. We're going to be really, really ramping up here. So it feels like there's something that still needs to change. And then simultaneously, you have this, you know, massive political backlash to rising energy prices. And I think most importantly, like,

we've identified some of the characters that drive AI. You know, we know Dario, we know Demis, we know Sam, right? We also have identified a lot of the people who are, you know, running the neoclouds or doing, you know, build out of new data centers. We don't really know the characters, who's the Elon Musk of energy is something I keep coming back to. And we've talked to some of these people. Doug from Radiant is like a contender in the sense that he's building nuclear power plants. Isaiah is also getting in the race with data centers. Blake Scholl from Boom is sort of pivoting or expanding into energy generation.

But there's no one who's really become like the main energy guy or gal, right? Yeah, I would say like Chase at Crusoe is potentially a contender. Totally. Unclear yet if he's a Joe Rogan CEO, right? Sure, sure. And I feel like when you think about who the Elon of energy, it's somebody that can go and throw down on a three-hour episode. Yes.

But also, like Chase and like the rest of the Neocloud founders are, you know, marshalling energy resources, but not really driving the underlying infrastructure as much as I think. They're more like, let's go find where the energy is cheap. Let's build there. Let's do the construction project. So other crazy. easy stats. So while we're growing at like 2.4%, 1.7%, China is consistently putting up 6% growth. So not great by comparison. China now accounts for one-third of global electricity consumption and

contributed 54% of global demand growth in 2024. So more than half of the growth in demand came from China. So stack that up over a few decades and it feels like the AI future is basically in the bag because if energy is a bottleneck, they'll have a ton of it. Now, I guess the wild card is maybe you go to space. Dan has an interesting point where he, let me actually find this exact thing. You see semi-analysis. Dylan was sharing that Venezuela has a bunch of underutilized gas turbines. No way.

He was showing satellite imagery. Did you see their Instagram? I don't know if they have like a bug or something, but they posted the same video to Instagram like 50 times. And I think they might just have like automated it and it just like, went haywire or something. Semi analysis. Yeah, semi-analysis.

Like, the official semi-analysis Instagram. Maybe they cracked the code. Maybe like the trough wants. It's a funny video of like all these guys and this like bar. I don't know. And then it has some comment. It's just like a funny thing. But I was just like, why did they post this seven times in a row?

Anyway. So what Dan says, he says, I think that the, so he says, I believe that Chinese technological success is now the rule rather than the exception. He says that China's obviously caught up in so many different technologies, drones and EVs. There are two fields in which China is substantially behind the West. Semiconductors and aviation, the chip sector is gingerly attempting to expand under the weight of U.S. restrictions. Meanwhile, China's answer to Airbus and Boeing is on a very long runway. I grant that these are two critical technologies, but China has attained technological leadership almost anywhere else.

And I believe its technological momentum will continue rolling onwards to engulf more. of their Western competitors over the next decade. Quickly, let me tell you about console. Console builds AI agents that automate 70% of ITHR and finance support, giving employees instant resolution for access requests and password requests. And password resets. So there is one that I think he missed. And maybe you put it in aviation, but I think space travel.

They're way behind on rocketry. Yeah, he said the Boeing Airbus. They don't have that equipment. And maybe you put SpaceX and Blue Origin in the same category as Boeing and Airbus. I particularly don't. I think of them as separate technologies. And I think of this as potentially like the third category that China's way behind in. And if you, I still wait it with somewhat low probability.

Like I'm still somewhat convinced by Delian's take that it's going to be really hard to actually get a ton of data center capacity in space on a short time frame. But at the same time, it's like, you know, never bet against Elon. He's got to figure something out. It just feels like something that could happen. And I don't understand enough of the physics to really, to really figure out like what timeline it happens. That was also before people went on break and we're using Claude code.

Oh, so now. Yeah. Now you can get in space. It's going to put this up there. Yeah, yeah. But like if the end game really is like you got to do, you got to do compute in space because the math just works so much better up there over. you know, a decade or two, well, then the advantage kind of flips back to America. But, you know,

something needs to change clearly. We're not asleep at the wheel. There are a lot of companies working on this in America. Some public company stocks have already mooned. Lots of people are tracking the gas turbine market. Risky startups that could never get funded a decade ago are pulling in hundreds of millions of dollars now. And we are taking the problem seriously, but I'm interested to see how quickly things can actually change. As the big AI labs mature and probably go public this year or next year, the neoclods and hypers build ever larger clusters and energy prices become more of a political issue, I suspect discussions of what we're doing to make more energy to dominate the conversation in 2026. Let's get this growth rate up. So I'm optimistic it can happen, but I feel

like it's the new benchmark that I want to be tracking. Obviously we were tracking the performance benchmarks of the models. Then we were tracking the diffusion. What's the actual revenue? Is it sticky? Like there was a lot of risk of like, oh, this, this company comes up with some AI app. They, you know, ramp to a hundred million. Will it stick around or will they just get, you know, swept by the wayside and people will go back to doing it the old way or they'll use something else or they'll pay way less? All of those, I feel like we kind of beat all of the bubble allegations over 2025 for the most part. What do you think? I don't think that that is a public perception.

I would say, like, the most popular topic. Well, beat the bubble will pop in 2025 allegations. Like, at the start of the year, people were predicting that, like, open AI would not be, would not do an upround, for example. And, like, that just didn't happen. It was, like, all up rounds for basically all the major players. There were a few that got kind of stuck in Quagmire's. Yeah, I would say that we've, in general, as a basketball. stabilized since the infamous podcast, right?

Yes, yeah, totally. Like, if you were long AI broadly in 2025, you did very well. And you did not see some sort of collapse. Yeah, Tyler. Yeah, there's actually a cool chart that Axios put out about, like, the 2025 news cycle, and it's like the distribution of, like, how often the search was. Yeah, yeah, I really like that chart. So you can look at the AI bubble, and yeah, it did peak, and we're like definitely not at the

peak right now. We're way down. Tyler, vindicated. Yeah. Vindicated. Never lost faith over there. Jordy was skeptical. I was maybe a little skeptical from time to time.

Tyler never lost faith, and that's why he's here. Sticking around, riding up. Max contract. Max contract. Let me talk about linear. Still with us. Meet the system for modern software development.

Linear is a purpose-built tool for planning and building products. So in the other section of the Dan Wong piece, he talks about competition, and I thought this was just an interesting element to read. He talks about Sputnik moments, and we got to read a little bit of this. So one might have expected the U.S. to have roused itself after this bout of the trade war. But there have been too many declarations of Sputnik moments without commensurate action.

I hadn't thought about that, but at one point we were meming like Sputnik moments for Sputnik moments. Remember? Do you remember this? Yeah, that was super. That must have been like a year ago. Yeah, it was like a year ago or something. But you know when it becomes like a meme that we're like, you know, everything is a sputnik moment. Like it really is overused.

Yeah. So Barack Obama declared a sputnik with China's high speed rail. Mark Warner repeated with Huawei's 5G. Mark Andresen called it with deep seek. The more that people use the term, the less likely that society spurs itself into taking it seriously. It's a good point, right? So he says, I think the U.S. continues to systematically underrates China's industrial progress for several reasons. First, too many Western elites retain hope that China's efforts will run out a fuel by its own accord.

Industrial progress will be weighed down by demographic drag, the growing debt load, or maybe even a political collapse. I won't rule these out, but I don't think they are likely to break China's humming tech engine. Dan has so many interesting insights. He had this one where he talks to people. He reflects a lot on talking to people in San Francisco because I think he just moved from Yale to Stanford and spending a lot of time in SF. And he was saying that everyone he talks to about Taiwan thinks that China only, wants to invade Taiwan because of TSM and semiconductors. And he has to be like, no, no, this goes back like 60 years.

Multiple, like, yeah. Yeah, like back to the civil war that led to the exile and whatnot. Like this is not some like new trendy thing where like, oh, they need this particular resource this day. This is not about chat cheap E, T, little bro. Yeah, truly. Deeper than that. Yeah, I mean, it's this, I mean, yeah, it was interesting because you could say this was a

2025 letter, but this felt like the best possible summary of the geopolitical economic dynamic between China and the U.S. There was a paragraph here that stood out. He said, third, Western elites keep holding on to a distinction between innovation, which is mostly the remit of the West and scaling, which they accept that China can do. Jan says, I want to dissolve that distinction. Chinese workers innovate every day on the factory floor by being the site of production. They have a keen sense of how to make technical improvements all the time. American scientists may be world leaders in dreaming up new ideas, but American manufacturers have been port building industries around these ideas.

And so, again, there's some interesting commentary in here as well. I forget the exact section. Everybody should go read it. But he basically says, like, in SF or within our world, it's become popular to just say, like, oh, yeah, we need to figure out how to build things again, but it's okay because AI can do it. And it's like the thing that I would say you can be most sure about right now

is that AI is not operating with like, you know, you have all this intelligence, all this intellect, this horsepower, but it doesn't have any agency. Like it's not just sitting there in the background like fixing American manufacturing, right? And while we have this idea that AI will automate manufacturing

and create these sort of like feedback loops that allow us to start making stuff again in the U.S., currently, like, manufacturing jobs in the U.S. are actually declining. Yeah. Month every month. Yeah. I mean, at the same time, there are a bunch of people in SF who make a somewhat convincing argument about, like, the software-only singularity and bootstrapping by kind of everything

off of it. I agree, it's a little bit more of a sci-fi scenario, but there are somewhat, there are some good arguments on both sides. But he does make this point about manufacturing that even when you, you, how do you do an apples to apples comparison? Well, you look at Tesla and how they operate the gigafactory in America, and then how do they operate the same gigafactory, building the same vehicles in China. And he says, I sometimes hear that the U.S. will save manufacturers through automation. The truth is that Chinese factories tend to be ahead on automation. That's a big part of the reason

why Chinese Tesla workers are more productive than California Tesla workers. And he says, what does he say? Gigafactory, is that in here? He says, according to to Tesla's corporate disclosures, a worker at a gigafactory in China produces an average of 47 vehicles a year. A worker at a gigafactory in California produces an average of 20. That's like a huge gap more than twice because they have more automation over there. And you think of it as like a copy-paste thing, like the giga-factor here should be the same as there. But there really are like more machines, more automation just walk across the street in China and there's some sort of, you know, CNC factory that has 10,000 CNC machines or something like that. So, um,

There's just a lot more to build off of there. So China's automotive success is biting into Germany more than anywhere else. I keep a scrapbook filled with mournful remarks that German executives offer to newspapers. Quote, most of what German Mittlesland's firms do these days, Chinese companies can do just as well, said a consultant to the Financial Times. Quote, in my sector, they look at the price point of the market leader and sell for roughly half that, the boss of a medical device maker told the economist. It's never hard to find parades of gloomy Germans.

Now more than ever, it looks like their core competencies are threatened by Chinese firms. Yeah. Anyways, go read it, but he called Europe cooked and chopped. Well, you know what's not cooked and chopped? Finn.AI, the number one AI agent for customer service. If you want AI to handle your customer support, go to fin.a.i. The last section here. So Beijing has been working relentlessly to build up its resilience.

While the U.S. talks itself out of Sputnik moments, Beijing has dedicated immense resources to patching up its own deficiencies. It's not a theoretical fear that Chinese companies might lose access to American technologies. So the state is pouring more money than ever before into semiconductor makers and research universities. It is investing in clean technologies, not so much because it cares about the climate, but because it wants to be self-sufficient in energy. And that's a huge, huge, just economic incentive. And it's rewriting the war. the rules of the global order with caution because it's been a giant, giant benefit.

We got to try to get Dan on again. I'm curious to ask him get his read on everything that happened in Venezuela over the weekend, just given that the Chinese delegation was there on the ground meeting with him, effectively the night before he was taking him. What a crazy story. Wild weekend. Let me finish reading this. So Beijing has been preparing for a Cold War without even.

for waging it while the US wants to wage a cold war without preparing for it. So here's a potential way that China succeeds. Beijing's goal is to make nearly every important product in the world while everyone else supplies its commodities and services by making the country mostly self-sufficient and by vigorously policing the outputs of LLMs and social media. Xi Jinping might hope to make China resilient. He is building fortress China stone by stone in order to outlast the adversary. Beijing doesn't have to replicate American diplomatic, cultural, and financial super powerdom. It might hope that its prowess in advanced manufacturing might deter the U.S.

and its success in manufacturing might directly destabilize the U.S. by delivering the coup to grace over the rust belt. The U.S. might shed a few million more manufacturing jobs over the next decade. The job losses combined with AI psychosis, social media, and all the problems with phones could make national politics meaningfully worse. Black pill, but then he says, I don't think this is going to happen, which is good. Yeah. Anyway, turbopuffer, serverless vector and full tech search,

built from first principles on object storage, fast, 10x cheaper, and extremely scalable. Well, should we talk about manis? Let's talk about manis. Let's talk about manis. This all started because Bill Gurley was an investor in Manus. And whenever he's on. Bill Gurley has the best voice in venture.

I agree. The best voice in venture, hands down. Who's got a better voice than Bill Gurley? It's amazing. He's got pipes. He's got pipes. He does have a good voice for podcasting, too. It's great to hear.

Well, he's going on a podcast tour. Yeah. And we're, we're, I think we'll be a part of it. Fantastic. I can expect to hear him on. So this all started with Alex Wang over at Meta, formerly scale AI. He says,

excited to announce that Manus AI has joined Meta to help us build. Jack Randall in the X-Chat. Did anyone check on Delian? Oh, yeah. That's funny because Jack worked with Delian. Yeah, I mean, it really is. Hey, Deleon might be taken a victory lap

because if Miami comes back because of this Welpax thing, Delian totally vindicated, right? Also that, and Patrick Hollison was bulbous. Oh, yeah. That's crazy. So did Deleon capitulate?

Did they capitulate too early? No, no, he gets to play musical chairs. He's like, what? I never said anything about Manus. I don't know about that. It's fine. It's fine, investment. It makes sense.

Did I mention that Facebook was a Founders Fund back company? Yeah, I worked in Founders Fund. Yeah, I mean, there was so, I mean, the Manus, it's great to skip portfolio company, acquire Manus. The moment in time that benchmark invested in Manus, it was a really, it was a crazy move. I was like, what's going on here? It was just, it was just very strange. It was, it was truly like, it was truly contrarian.

It was very contrarian. It was very contrarian. Yeah, I mean, the vibes were like, it was the most, it was the most intense, like, part of the trade war. Trump was really putting the screws. He's back in, so he's going to go extra hard. And it's not like Biden took the, the, the, the, his foot off the gas of, you know, the trade war with, with China. Like, the initial volley of chip restrictions happened under. Biden. Like Chris Miller writes chip war during the Biden admin and everyone really wakes up all the Anderol folks are talking about it. Yeah. So here's the other thing. So AI 2027 was released in April of

2025. Benchmarked did Manus in April of 2025, same month. And so obviously a lot of people process AI 2027 as like, cool, this is like sci-fi. Kind of like, you know, let's just extrapolate a bunch. It was kind of a thought exercise, but at the same time, this was kind of the peak in some ways of like the AI war narrative. So to have a famed American venture capital firm backing a Chinese AI firm. Sorry, we got to go back to Dan Wong. There's so many good things. He talked about AI 227. I want to read it.

But first, let me tell you about Figma. Figma make isn't your average vibe coding tool. It lives in Figma. So outputs look good, feel real, and stay connected to how teams build. Create code-backed prototypes and apps fast with Figma. So Dan Wong says, the most read essay from Silicon Valley this year was AI 2027, the five authors who come from the AI safety world, outline a scenario in which superintelligence wakes up in 2027.

A decade later, it decides to annihilate humanity with biological weapons. My favorite detail in the report is that humanity would persist in a genetically modified form after the AI reconstructs creatures that are, quote, humans, what corgis are to wolves. I somehow missed this. I listen to the Dorcasch podcast. I read most of AI 2020. Did you pick up with a corgi line, Tyler? You read this, right? I don't remember that. Somehow the corgi line just got completely missed by me, but I love it. It's so funny. It's such a funny detail. I'm going to control that for corgi. It's not pulling anything up. No? I don't know. Okay, maybe Dan's taking some creative... Oh, you have to like choose.

an ending. Okay. Oh, you have to find the Corgi ending. Yeah. That's key. We'll take the golden retriever ending. It's hard to know. Yes, turn us into golden retrievers.

We're ready. We're ready for it. Already did. I'm not. Who's trying to be a wolf? I'm not trying to be a wolf, the lone wolf sigma grinds. I'm trying to be a golden retriever on performance enhancing drugs. Yeah.

It's a golden retriever with 40 delts. Yeah. I mean, we've been growing out our hair, like the golden retrievers, working on looking good. Everything's, everything's lined up for a golden retriever 2026. It's a good time. It's hard to know what to make of this document. He's talking about AI 2027. He says, because the authors keep tucking important context into footnotes repeatedly, saying that they do not endorse a prediction. Six months after publication, they stated that their timelines were lengthening, but even at the start, their median forecast for the arrival

of superintelligence was later than 2027. Why they put that year in their title remains beyond. me. Sort of an interesting, I know why they put it in there because it's engaging, it's entertaining, and it makes for good banter. Anyway, he goes on to talk a lot about the, the dynamic of how Silicon Valley works, how New York finance works, how it's different, it's very interesting. He, he, we should read a little bit more of this since we have time today. You want to go back to Dan? I want to go back to Dan. We got a Dan addict over here. Yeah. What do you say about finance? He said like they're wrong before breakfast or something. Finance. Let's see. One thing, okay, one of the things I like about the finance industry is that it might be better at encouraging diverse opinions.

Portfolio managers want to be right on average, but everyone is wrong three times a day before breakfast. So they relentlessly seek new information sources. Consensus is rare, since there are always contrarians. betting against the rest of the market. Tech cares less for dissent. Its movements are more herd-like in which companies and startups chase one big technology at a time. Startups don't need dissent. They want workers who can grind until the network effects kick in.

VCs don't like dissent, showing again and again that many have thin skins. Taking shots. That contributes to a culture, I think, of as Silicon Valley's soft Leninism. When political winds shift, most people fall in line. Most prominently this year, as many tech voices embrace the right. So he calls San Francisco very insular city.

He still has a lot of nice things to say about it. Anyway, let's go back. Since you're back here, I thought it was, he talks about humor within tech and the CCP. Yeah, yeah. He gives a line from Sam Altman that is, Sam says, I think that AI will probably most likely sort of lead to the end of the world. meantime, there will be great companies created with serious machine learning. I mean, that is so good.

It is funny. That is really funny. That is really funny. It's great. It's also like maybe true. I don't know. It's like a whole. No, and I think this is the key problem for the tech industry in 2026.

Yeah. Is like how you need to paint a more, like the founders need to paint a more optimistic vision. Yes. For AI. Yes. Because it's not working right now. Yeah. Because people see their energy bill going up or they even hear about the idea of their energy

bill going up and then they see some of the slop in their feed and the the the slop is getting better right you look at some of the videos coming out of the sloppel venezuela debacle right uh i mean there i saw i saw maduro like face swapping yeah yeah yeah yeah i guess that would so so the the the video is higher definition etc but people are people in all types of roles are worried about job loss, and they see the investment going into AI, and they're just scared, right? And so when you do quotes like this,

whether they're serious or not, saying, I think that AI will probably lead to the end of the world, people are starting to ask, like, wait, why are we automating all of this in the first place? Can we just stop? Right? Like, I don't, actually, I don't want my job automated, but I'll keep doing it, right?

So the industry needs to figure out how to paint a more optimistic, you know, because there's a lot of ways, There's a lot of very optimistic, you know, I think Dan talks in here about what is the line he says around young people, you know, people, you know, maybe in 10 years reminiscing about a time when young people didn't just have shelter and an income provided for them. They had to fend for themselves, right? And so, so anyways, I think that's real. he also talks about CCP humor, which was he saying the PM2.5 back then was even worse than it is now. I used to joke that it was PM 250. What is that? He's just saying, like PM25 is, 2.5 is a measurement for air, like air quality.

Oh, oh, okay. Particles per million. Yeah, yeah, yeah. It's like PM 250. Yeah, yeah, yeah. Let me, I have a few rebuttals, but first, let me tell you about. Restream. One live stream 30 plus destinations. If you want to multi-stream go to restream.com. So first rebuttal on humor. Some of the tech people are funny and I was noticing this. I was remembering this as we watched our year-in-review video that Jackson on our team put together. And in there one of the clips is is us talking to Sam Altman who Dan highlights as like famously humorless because he's like trying to speak in many levels of and it's It's this thing. Best Sam joke ever, him coming on our show. Yeah. I mean, he was making the joke on the timeline. Yeah. But nobody got the joke. Yeah. He was saying, I'd never buy a car for $250,000. Yeah, it was hilarious. So someone, someone posted this, this, like, what was Gt3RS or something. And they were like, would you buy this. And Paul Graham says, like, oh, what a waste of money. Like, I would never buy this. And Sam just said, I would never buy it either. And everyone was like, oh, like, he's so out of touch. He has a Konig's a F1.

And he's ridiculous, of course you'd buy it. And then he came on our show and explained, but no, the joke was that you would never buy a car for $2.50 because he wants a multi-million dollar car, which is hilarious. And a lot of people that I've talked to who've interacted with Sam a lot behind the scenes, like in the right context, he is truly very funny,

but it's just hard because he's trying to speak seven different languages anytime he's on screen. Because he has to speak to politicians who want to regulate him, to people that might be losing their jobs, but to people also who want to come work for him, and also to investors who want to work investors who want to return. I think it was so much easier. In the 20th century, it was like, you know, it was easy mode being a politician or a CEO. Because you could go into a room of

business executives and said, okay, you currently spend, like if you were running like an AI company before the internet, before everything was recorded, before everything hit, you know, every single distribution point immediately, you could go and say like, hey, you currently spend, you know, a billion dollars a year on payroll. I'm going to be able to reduce that by 80%. You're going to be able to conduct lots of layoffs. And then you could go on TV and say, like, AI is going to create an abundant, you know, future.

And the clips wouldn't kind of circuit everywhere. Or politicians could go talk to a labor union here and then go talk to business leaders here. Yeah. And you just didn't have to do. I mean, we see this on our show where people come on and we're all having a conversation. And we're all like, oh, yeah, this makes sense. we all have the context. And then a clip goes out and people are quote treating out of context and being like,

oh, I hate this. I hate what they're saying. And we're like, well, we weren't trying to get them to say something controversial, but they did outside the context of our show. Go back and watch every episode of TBPN of 2025. Before you criticize us. Before you criticize us. Just to get some context.

Yes. There was, oh, there was one more thing. I want to go to this Dan Wong thing. He has a great line in here about feedback he got on his book. Obviously, there's critical. criticism and stuff, but he has this great quip. He says, I learned of Leo Rostin's quip that it is the weak who are cruel and gentleness to be expected only from the strong. I thought that was a very good quote. Just very interesting, just saying that like, like the, if you're, you know, like,

like the negative comments usually come from a, from a place of weakness. But that doesn't mean that there should be empathy. There's one more thing about the AI future and the positive future. It reminds me of my one and a half-year-old who will bite my three-year-old. There you go. Because it's weak. And so he has to take the cruel action of trying to chomp the three-year-old, three-year-olds. Like, hey, look, I'm a little mass monster over here. That's hilarious.

Let me tell you about the two AI futures that are sort of dueling right now that I think are interesting. And then we've got to go back to Manus. You still have to give me your take on Manus. We're going to do Mass. But let me tell you about Plaid first. Plaid powers the apps you use to spend to save, borrow and invest, securely connecting bank accounts, move money, fight fraud, and improve lending. Now with AI. So, um, to be partnered. So we, we will have to dive into these pieces in more depth at some point. But basically, I do think that,

even though Dan says that, you know, tech is very insular, it's less contrarian, there's only, you know, everyone aligns themselves around one vision of the future and just like runs And I think he is correct in some ways. There is some dissension and at least conversation back and forth, mainly between Dwar Cash right now and Ben Thompson. Ben Thompson wrote a piece on AI and the human condition where he sort of reacts to Dworkesh's piece with Philip Trammel, Trammel, called Capital in the 22nd century. And Dwar Keshe is talking about capital accumulation, inequality. how things might need to be redistributed in the future. It's unclear exactly the timeline.

And if you go out really, really far, it becomes much easier to wrap your head around all of this. Like at one point they start talking about buying galaxies, which is sort of like a funny thought experiment. And at that point, like if you were to just say, hey, we're going full communists with the galaxies, we're distributing them equally, it's like, well, there's trillions of galaxies.

Like, you personally would get a thousand galaxies. And it's like, yeah, that's probably like enough. Like it takes 25,000 years to get to the nearest galaxy. Like, imagine the life, imagine the life you're living where you're immortal and you're like, just to go between my two, my two galaxies that I own, it takes 25,000 years. Like, I don't know that I'd be so pissed off that like, oh, I wasn't able to get more than one, one six billionth of the galaxies out there.

I don't know. Has anybody tried to kind of stake a claim? I know America, the moon's ours. It's not legally yet, but it's ours. It's somewhat inevitable. I mean, it's crazy no one's been to the moon. Like, we went there, like, 50 years ago, more, and everyone was just like, we're good.

Lost art. Lost art. You'll send some robots. But literally, this is a crazy thing. No other country has landed a man on the moon. I don't know why. What are you waiting for? Get somebody up there.

Anyway, what are you saying about acquiring galaxies? You want one? You want more than one? You want more than a thousand? I want more than a thousand. You want more than a thousand? Say you're not a... A communist.

I want more. You want 2,000 galaxies? So I think that quote was actually referencing... Leopold. Yeah, it was Leopold. And people were arguing whether Leopold was being serious, and he was. It's sick. It's a great quote.

I love it. It is sufficiently sci-fi for me. I'm very, very excited about that quote. And it's thought-provoking. It's truly thought-provoking to think about what property rights look like in a world where it takes you 25,000 years to get to the place that you own. Like, your vacation home.

Unless, I don't know, I don't know if the AGI-pilled people are so AGI-I-pilled that they think we're going to defeat the speed of light. Every, like, physicist I've ever talked to has always said, like, no, the, like, that one holds forever. That's not like, oh, quantum computing, it can work, but it's a couple decades away, or fusion or fusion,

like, all these other things are, like, somewhat engineering problems. I think the speed of light is, like, going to be around forever. I don't know. Just need a faster horse. It's the year of the firehorse, by the way. What does that mean? It's 2026, Chinese Year of the Fire Horse.

Are they adding superlatives to the animals? I thought it was just horse, snake, dragon, and then they rotated through those. It's the year of the fire. We upgraded animals. It's amazing. It's amazing. I'm a fan. Manus is acquired.

I think... Okay, back to Manus. Great Chinese century coming up. I think we're going to be putting on those hats soon. So Ramp Labs modeled it out. Of course, Ramp Labs is AI agent for a spreadsheet software, so you can give it a prompt, and it will build the spreadsheet for you. Very, very cool.

And they did a little analysis on the meta acquisition of Manus AI. They said the estimated price is $4 to $6 billion based on AI M&A comps. It's the fastest to 100 million ARR in history, just eight months. Wow. and benchmark likely 8 to 12xed in under a year. So again, they're sort of estimating here. I think people were kind of putting the deal between 2 and 3 billion. Yeah.

Is what I had heard. Okay. So again, so yeah, I mean, I guess let's get in, let's get it. What was, I guess, what was your immediate reaction? The thing that I was excited about is this is Zuck buying a product that people love. Well, that's a debate. But I like your take. But, but they're,

Mark Zuckerberg has done a number of talent acquisitions. Yeah, that's what I'm saying. So, so, so he just went, he just went through a talent acquisition era. Yes. Yes. He, he, they, they're the, you know, biggest investor in scale, right? But, but they didn't acquire the product. Yeah, yeah. Still.

And I think they're actually acquiring data from other companies as well. And so historically, Zuck has been saying, you know, drum, you know, beating the personal super intelligence. Yeah. It's been very unclear what that means. Manus has been building high-quality agents. They're sort of time and time again, I talk to people that are just very excited about Manus. They're like, just use Manus. Like, it's the year of the agent. Yeah. Like, use it. You'll see the future. Sure. And so just given Zuck's history from a product acquisition standpoint, I get excited because this can potentially give us some clarity around what they're, like, one version of personal

super intelligence is you have AI agents that can go out on the internet and do things for you. They can go, I think in the context of meta, I think about like shopping, right? You see something on Instagram that you like and you can, you could trigger an agent effectively, like, go find this product. Yep. Like you see a car that you like, go find this product. The elusive GT3RS. GT3 bench. Yeah.

Yeah, basically like you see a, like, you see a, like, you see a. make any image or video shoppable. And in fact, I would probably argue that LLM's in the chat form where you have a bunch of information with a knowledge cutoff baked into some big model, a llama for, and it's vended through the Instagram search box that you chat with, that just, it can never fully satisfy the vision of personal superintelligence, a personal assistant,

something that actually can take actions for you. It feels like it's incredible. It's amazing. Obviously for knowledge retrieval, it's great. But it was never going to fulfill the real vision there. So this does feel like, I don't know, I would hope that they bake this in in a really interesting way. I wonder if we are talking about going out and shopping for you off platform

or if it's more doing things within the meta ecosystem. I don't even know what that would look like. But a lot of the image models that they're training, you can imagine that they exist almost entirely within the family of apps ecosystem. So if MetaTrain's an amazing image editing model, a nanobanana, right? Direct nanobanator competitor. What's that going to look like? It's probably going to look like you take a photo, you put it in Instagram, and then you say, hey, change the background. Make the sky less gray, right?

And it'll just do it really, really well. And that would exist purely in there. And maybe they open source it, maybe they put out. But that's not the product. I was also thinking of, I guess, one way that I am thinking about Manus in the context of what Zuck and the AI team at Meta are saying, which is personal super intelligence, right? They're talking about, they've talked about integrating AI into Rayban Metas, right? So when I think, when, and if you're using Rayban Metas, you can say, hey, meta, and then you can give it a task or ask it a question or things like that. And so I expect that Manus-type workflows will be heavily integrated into a, hey, when you say, hey, meta and you're wearing glasses, like, hey, build me a slide deck to help me prep for my final later.

And I can imagine, like, effectively the same product experience that Manus has today will be integrated into, like, meta-AI. Yeah. Let me tell you about Vanta, automate compliance and security. It's the leading AI trust management platform. I agree with you. At the same time, there's a lot of areas where it feels very natural for Google to expand into. It feels very natural for Microsoft to expand into, Apple to expand into. Apple has a spreadsheet app, right? They have a word processor. They, like, meta doesn't really have those. They did meta for, meta at work for a while, but it doesn't, this whole idea of, like,

go to open Instagram to do your homework feels crazy to me. And the same thing with open Instagram to build a slide deck. That feels crazy to me. Yeah, but everything that they're doing in AI, they're piping through meta AI, the new standalone app. So that's a bunch of, like to me, I look at that as like experimental area that you can just put anything into it. I was told by someone close, close to the action that apparently like the,

the meta-AI trough, the, the, the, uh, vibes. The vibe, yeah, meta-vives. Actually, the usage is actually insane. Really? Yeah. No way. So, so. But I mean, it's, it's terrible in the app store, right? It cannot be ranked. Even, even SORA is not ranking well. Like, can you just look this up? I don't know. Yeah. Well, we're looking that up. Let me tell you about phantom cash. Fund your wallet without exchanges or middlemen and spend with a phantom card. You see that ticker down in the bottom telling you the crypto prices, that's thanks to Phantom. Legends. Let me try and find, I don't think MetaVibs is going to rank, but at the same time,

they might have gotten a bump over Christmas because when you get the Raybans, you unbox those, you have to download the Meta AI app to integrate with the Raybans. So maybe they got a pump from that? I don't know. But Sora, let me see what... It's not in the top 25 right now. Ben Thompson recently wrote. It's also worth noting the relative popularity of human-generated content versus AI-generated content.

SORA is down to 59 in the app store. And I count double-digit human-denominated social apps that rank above it. Yet I get the argument that this is the worst AI will ever be, but it will also never be human, which is what humans want most of all. There's a bunch of great posts about content where all this goes. And I think Ben Thompson offered a very, very compelling white pill. Dwar Keshe is a little bit more like, this is going to be extremely rough,

so it requires, you know, aggressive, aggressive maneuvers and rethinking the entire social contract and potentially the entire economy. Ben Thompson is a little bit more like things can continue as they have been, for the most part. At least that was my name. One thing with Manus,

this is maybe the first acquisition of, in the billions for a company with an Isle of Man domain. They're Manus.com. I am. You don't see a lot of multi-billion dollar dot I am acquisitions. I love man. Tyler, is that where you're from? I love Chad?

Did you get into looks maxing over the break? I've been bone smashing. It looks like you got into looks maxing. Doesn't it? Doesn't his midface ratio look different? I got to fix my recessed maxilla. I feel like his maxilla is looking different. It's looking better.

He doesn't have the filter on. Wait, do you guys have the filter on? I think, yeah, I think they do. they do have the filter on. They do. Wait, Tyler, go like this, do the. The double josser. Oh, yeah, definitely, they definitely have the filter.

Definitely, definitely. Well, speaking of Dwar Cash, he's sponsored by Label Box. We're sponsored by Label Box. Label Box. We're delivering you the highest quality data for Frontier AI. We love Label Box. We're considering making a physical

label box. And putting Tyler in. We will see. We don't want to leak it. We will figure out if we can figure out how to physically... He's wanting to put Tyler in a box. I just like boxes. So Signal asks, can someone explain why Meta bought Manus? What's the one pager for Zuck? Other than we have an F ton of money, so why not?

And Lucas Beyer, Metis List researcher, says they know how to build... They know really, really well how to build good agents, and maybe that's enough. Nick Dobo says, best tool set on the market. They bought a Swiss Army knife to hand to any A. model, wide research, virtual browsers, VMs, code interpreter, PowerPoint slides, app builder connectors. So yeah, they're just good at product. Yeah. Yeah.

Eric Mehir says, head in the box. His handle has box in it. He says, I left meta because I made a bet that models were going to be commoditized and the value would be in the products on top of the models. But Metamate and Gen. I were highly politicized, sucking up all the oxygen in the room. As always, I was right. hilarious line. As always, I was right. Ex-metta people are getting spicy. Do you see the Jan Lacoon Financial Times deep dive?

We should dig into that, but he took a bunch of shots of a lot of different people, basically just saying, like, you can't tell a researcher what to do, especially not me. He was very salty over Alex Wang coming in and becoming his boss, effectively. At the same time, it feels like... You got level. Well, yeah, it isn't awesome. I mean, it's a bold choice to put Jan Lacoon on the bench. He's a, I think, Touring Award winner.

He's, like, one of the greatest AI researchers in history. Where did he land on the Mettus list, Tyler? Jan Lacoon, how did he do? But he's back with a $3 billion fundraise valuation. But the interesting thing is, like, I think there's two ways that you can read into the Ilius Sutskiver. We are in an age of research story. Tyler, do you have yawn on the mess? Yeah, he actually wasn't on there.

He wasn't on there? I think we were trying to rage bait him or something. No way. Yeah, he wasn't on there. Wow, we are so bad. But anyway, you can read Ilias... To be clear, you were trying to... We don't participate in rage baits.

We don't have anything to do with the rankings. You can interpret the Ilya Sutskiver age of research concept as, hey, we... if you're a hyperscaler, you need a really, really solid research team to go extra hard into research because you want to be the first one to have that research breakthrough. So you need a lot of, you need a lot of true researchers. Or you can read into it as like, hey, this is kind of an AI winter. And the research is sort of stabilized. So we can pull in all the best research. Yeah, we need to get to the frontier, but we can be a month behind. If we productize really well, we can win.

And we can have a really great outcome here. And then, you know, when Ilya comes up with something great, It's going to get copy-pasted all over the industry because of the, you know, SF house parties and, you know, papers that get written and, you know, open-source implementations that happen pretty quickly. Maybe, maybe not. Maybe this is the one time that technology just stays completely siloed, but typically it doesn't. Yeah.

Anyway. Yeah. Ollie from Databricks was, at the end of the year, was talking about how he's great. He's fantastic. I've got to have him back on. But he was talking about putting people in different buckets. So, like, there's a super intelligence, right? There's a super intelligence bucket.

I think that when you look at, like, Zuck can say we are trying to build super intelligence. That is a much more compelling vision than we're trying to make great consumer products that we can monetize via advertising and commerce. That is not a compelling vision that you rally a team around, right? In some ways, in some ways Open AI is doing the exact same thing, right? there are different groups that are taking, you know, maybe the Anthropics, taking a different approach in terms of saying, like, yeah, we're actually trying to build super intelligence and we're doing it in a very opinionated way.

But anyway, so I just like to see, I will be very excited to see, hopefully see them try to take what Manus is built and dramatically scale it. Yeah, yeah, for sure. Quick shout out to 2.6 in the chat. He says he's down 99 pounds. That's fantastic. He's up 20K. And he says life is good.

So what an amazing start to the new year. I also saw Sam Schaeffer in the chat earlier. He launched a new product over the break called Timelines. It's like a daily series of games that you can play to better understand the news. So I'm excited to check this out. And congrats to. to Sam. Well, let me tell you about the New York Stock Exchange. Want to change the world? Raise capital at the New York Stock Exchange. We like NICI. Potentially the best tagline of all time,

the best marketing copy. Want to change the world? Raise capital at the NICCII. That's how it works. Yeah. Anyway, GROC. So there were two major blockbuster deals before the end of the year. Unclear. I mean, we're both. of these zombie acquisitions or were any of them just clean normal deals? I guess we don't fully know. We know that the GROC and VidiA deal was very much a sort of ghost ship type deal where you take the team, take some of the IP license, pay out the investors. $20 billion deal. What should we do with that, John? Should we hit that app-loving gong? We hit that app-love and gong. We have a new gong, everyone. Look at this gong. It's bigger than ever. We

actually, it's so big, it's so loud. We have. have to be kind of careful with it. I mean, it's a John-sized gong. So John is 6-8, if you're just tuning in for the first time. It's a massive gong. It's a massive gong for a massive company, fantastic company. Yes. Applovin, which we're incredibly excited to partner with for 2020. You're not going to believe this, but Apple oven's next in the stack. Profitable advertising, made easy with axon.a.i. Get access to over one billion daily active users and grow your business to be. So there's been a bunch of chatter, says Dan Premack,

about how GROC employees made out in the NVIDIA deal. He made some calls to find out. And in short, they did very, very well, even if not fully vested. We did hear that if you were there a very short amount of time and you were not senior at all. You might have had a bad run, but we'll have to dig into that a little bit more.

But it seems like in general they did, everyone sort of got paid. It sounds like Chimoth did very well on this. He was a very early... Oh, I mean, Chimoth haters were in shambles. We need to check on them. We need to check on them. If one of your friends doesn't like Chimoth,

call them. If you haven't heard from them, I would be worried. Are you doing okay? Let's just go hang out. Call your friends. Call your friends. Yeah, take him out.

Touch grass. Touch grass. It's going to be rough for a while. Chamath is on a tear. And didn't he like, Didn't he basically give back the LP capital right before? So it's like 100% of his own money or something. I saw some take about that.

I'm not sure if that's when that happened. But I know social capital kind of converted to a family office. Yeah, but that was way after. That was way after. Okay. At least the first. I mean, the haters will hate to see that he invested molt back to back to back. I mean, with conviction, with authority, really.

And got to give him credit. So Alex Heath over at sources. dot news has new details on the Nvidia GROC deal. The whole process took less than two weeks and was personally driven by Jensen Wong. No other bidders. Wong wired money early and wanted the deal to close before the new year.

What a chat. What a chat. Send the $20 billion wire. Sir, we haven't fully executed the docs. I've good for my $20 billion. That is actually crazy. Yeah. I mean, very interesting.

GROC is, we were talking about this. Like, why is NVIDIA buying GROC? is a response to the TPU. We'll have to get some of the semi-analysis folks on the show to dig into it. But I think that there is, you know, GROC has been, they've been on this like super crazy rollercoaster. I think not a lot of people expected them to wind up at $20 billion.

A lot of people assume that the... Yeah. Because, I mean, there's two sides of it. One is like, when, if you read the actual social capital, the actual social capital, like, memo that Chimoth put together. It makes so much sense to invest at the valuation he did because Jonathan, the CEO and the founder,

was early on the TPU team, the father of the TPU. The guy understands how to make a chip that does one thing really well. This was, what, seven years before people in tech broadly knew what the TPU was? Totally. Very, very early. Yeah, crazy. But the, so there was always like that sort of bull case that it was like, it was a real team,

real expertise, building something real. But with a lot of the custom silicon, the custom chips, you sort of like point your bow and then you release the arrow and then you just like wait like five or ten years because like tape out, just tape out just like going from the design of the chip to actually getting the first one made. It takes like 18 months or something like that. It takes years and then you're sort of locked into the specific spec and you're building an ecosystem of developers that can run on your chips. And so like there's this you make a bet on certain architecture, and there's other firms right now that are betting on like, we're going super long with transformer architecture,

or we're going super long cerebrus, like wafer scale. We want to put the whole model on one chip instead of a rack, instead of a rack of a whole bunch of chips. Or we want to be really memory constrained or have a ton of memory or something like that. So you make your decisions about what tradeoffs you're making, and then you kind of just wait. And you hope that people come up with a use for what you made. And with Grock, people were like, oh, okay, it's fast.

When Lama came out, people were like, oh, he can stream tokens really, really fast. It's very cool. Some of the demos are really cool. But there were also a flip side where people were saying, well, it was very expensive to do those demos, and maybe it wouldn't scale properly. But I think there was a little bit of like the team is great, obviously. And I think GROC was probably getting comped. Like you were effectively looking at GROC v2 versus like TPUV7.

and you weren't necessarily accounting for like, okay, if you continue to invest in this, like Jensen will, obviously, and the team's great, and they have somewhat of a direction, and there's also some sort of bifurcation of the models where sometimes you need a really big model, sometimes you need a really small model.

Sometimes you need a model that's in your phone. Sometimes you need a model that's in a data center. Sometimes you need a model in space. And sometimes you need a chip company that sponsors the winner of the Constructors Championship. Grock sponsored McLaren. Really? No way. They won.

It was a good omen. Jensen saw that. He was like, I got a lot of cash. Why not buy a winning? I mean, the chip company that's sponsoring the winning team. Pretty compelling. Meanwhile, Red Bull, loser.

No chips. Works with Oracle. Oracle stock. Yeah. Been down in the dump. Ridiculous. Just doing entirely F1-F-1-based analysis. Sponsorship-based analysis.

of markets. Anyways, continue. So, yeah, I mean, I think as this, I heard an interesting take that one of the wafer scale bull cases that if compute does go to space, you don't necessarily want a rack in space because not only does that take up more space, but there's so many different points of failure of like, is this connected to this? Is this connected to that? Whereas if you just have like one big, you have the entire model just on one chip, you could potentially put that in space much easier. But Groch made a space more. But Groch made a space a small. specific bet, and it's still, it's not like it totally panned out, and it's like they're winning right now. Like, TPU is the hot thing, but then there's, you know, Nvidia, you know, GB200s are

still, like, flying off the shelves, and there's all these debates about Blackwell and stuff, and, and VINIA is doing great, but I think if you look to the future where there's different models for different things, sometimes you need something really fast, sometimes you need something really smart, then you want to be able to control, you know, all of the different chip architecture's all the different chips. And I remember, didn't Jonathan tell us he doesn't fab at TSM? I believe he told us that. And I'm pretty sure that's true. And so there is also, where would he, uh, maybe either Samsung or global foundries, I believe. And then I think by the way, we got to give a shout out to Ryan, Ryan Siebert in LinkedIn chat. He says,

W stream best content on LinkedIn. Let's go, Ryan. Thank you for watching. One of our two. We typically will have like three viewers on LinkedIn. Thanks to restream. But, yeah, thanks to restream. We are there. So thank you. Thank you for tuning in. So closing out, GROC, there's a few different takes.

Tyler Hodge says, the peers GROC will return over $4 billion to social capital and Chamath. I don't know how big the fund was, but it's safe to say this is one of the best venture outcomes of all time. Unbelievable. Power laws rule everything around. Okay. Yeah, just getting, so did you read this as,

like defensive at all. This? Tyler? No, no, no, no. I'm saying just buying grind. Like, is this offense or defense for Jensen? Clearly he's got the, he's got the, he's just got a lot of cash. He's got to do something with it. He's not going to just dividend out cash. He's willing to take bets with it.

He's investing in all the labs, all the neoclouds. Yeah. Right. I mean, I think he's like, almost, like, even though he's not literally the richest person in the world, he's, run the biggest company, so he has the most capital, like fire around at places. And I think to put this into perspective, this is like us saying investing in a nice camera, right?

Yeah, yeah, yeah, yeah. Like, it's not the kind of, like, it's, we take like, Jensen took, Jensen personally let this process, it took him two weeks. He wired early. Yeah. You would do the same thing for a nice camera. But I don't, I guess the bigger question is like, is Nvidia spread too thin? And they don't even know it in a social network.

Every other MAG7 company owns a social network, right? Facebook. Amazon has Twitch. Maybe they should. InVitya chat just for AI researchers on Rumble. Something like that. I don't know. But it feels like they did DJAX leptone, the cloud that was competing with some of their customers.

It feels like they sort of pulled back on that. It feels like they've remained as much of a pure play as you can at that scale. And so this feels like, I don't know, maybe it's a little defensive, but it feels just like a very natural extension of what they already do, which is selling chips. It would be very different if we were seeing them like, oh, they're buying power assets and they're going to build their own data centers and they want to get into, they want to step on Oracle's toes, and they want to step on Crusoe's toes or anyone else's toes. So, I don't know. But let's go into Gavin Baker and what he says about Nvidia buying Grog. He says, NVIDIA is buying GROC for two reasons. Inference is disaggregating into pre-fill and decode.

S-RAM architectures have unique advantages in decode for workloads, where performance is primarily a function of memory bandwidth. Ruben CPX, Ruben, and the putative Ruben-S-Ram variant derived from GROC should give NVIDIA the ability to mix and match chips to create the optimal balance of performance versus cost for each workload. Ruben CPX is optimized for massive context windows during pre-fill as a result of super high memory capacity with its relatively low bandwidth, GDDR, DRR. Rubin is the workhorse for training and high-density batched inference workloads with its HBM DRAM striking a balance between memory bandwidth and capacity. The grok-derived Ruben SRAM is optimized for ultra-low latency, agentic reasoning inference workloads as a result of SRAM's extremely high memory bandwidth.

and at the cost of lower memory capacity. In the latter case, either CPX or the normal Rubin will likely be used for pre-fill. Two, it has been clear for a long time that SRAM architectures can hit token per second metrics much higher than GPUs, TPUs, or any ASIC that we have seen yet. Extremely low latency per individual user at the expense of throughput per dollar.

It was less clear 18 months ago, whether end users were willing to pay for the speed, SRAM more expensive per token due to much smaller batch sizes. It is now abundantly clear from Cerebrus, and GROC's recent results that users are willing to pay for speed. And so we've seen this with a lot of the, you know, the meme of like, you're sitting there waiting for a deep research report in chat GPD,

or you're firing off clod code, and then you're just waiting, and you're scrolling or you're doing something else. And there's just, there's a lot of, like, you know, sitting around, waiting around, and now the really good developers are able to orchestrate six different cloud code instances all at once. but everyone would be happy with faster responses. And most people are in this weird in between

where they're paying $200 a month for something that they're using constantly and is doing all of their job. And so would they pay $2,000 a month for something that's 10 times faster or even two times faster? In many cases, yes, if they're doing extremely economically valuable work.

And so anything that can offer a different dimension of optimization on that Pareo Frontier is going to be interesting. And that's why you're seeing folks try and become not just GPU rich, but like ASIC rich on a particular, on a particular scalar vector of the performance tradeoff curve. We should get Andrew from Cerebrus back on the show because he came on for five minutes.

It was one of those days when we had like 20 gas. And we were like, this guy's like a legend. $8 billion company. I'm glad we could ring the gong for you, but this feels disrespectful at this point. Anyway, let me tell you about vibe.com, where D2C brands, B2B startups, and AI companies advertise on streaming TV, pick channels, target audiences, and measure sales just like meta. I'm super pumped to partner with vibe. We love advertising. We love advertising.

I have some insight into their metrics, and this is a company that you should potentially join if you love advertising. I would be excited to go work there. So I'm pumped to partner with them. Should we talk about DoorDash? DoorDash. Well, no, we're not talking about, we're not talking about DoorDash. What are we talking about? We're talking about a post that people assumed was about DoorDash.

Yes, yes, yes. Okay, so the post is titled, Holy Fing S. I'm not swearing, really hurts the quality. Or you can read it on the screen. But he's very sensational. So it's a screenshot from Reddit. It has 36 million views and 207,000 likes. It's a massive post.

It got 11,000 upboats on Reddit. So it says, I am a developer for a major food delivery app. The priority fee and driver benefit fee go 100% to the company. The driver sees zero percent of it, zero dollars of it. So here's where it gets conspiratorial. It says, I am posting this from a library Wi-Fi on a burner laptop, because I am technically under a massive NDA. I don't care anymore.

I put in my two weeks yesterday, and honestly, I hope they sue me. I've been sitting on this for about eight months, just watching the code get pushed. I hope they sue me, but I'm on a library Wi-Fi with a burner. We're going to dig into all the different aspects of this, but this is crazy.

Some people in the chat, I read this, but we're going to go through it. So just watching the code getting pushed to production, and I can't sleep at night knowing I helped to build this machine. You guys always suspect the algorithms are rigged against you, but the reality is actually so much more depressing than the conspiracy theories. I'm a back-end engineer. I sit in the weekly sprint planning meetings where product managers, PMs, thanks, discuss how to squeeze another 0.4% margin out of, quote, human assets. That's literally what they call drivers in the database schema. They talk about these people that, like they are resource nodes in a video game, not fathers and mothers trying to pay rent. First off, the priority delivery is a total. scam. It was pitched to us as a psychological value add. Like I said in the title, when you pay that extra $2.99, it changes a Boolean flag in the order JSON, but the dispatch logic literally ignores it. It does nothing to speed you up. We actually ran an AB test last year where we didn't speed up the priority

orders. We just purposefully delayed non-priority orders by five to ten minutes to make the priority ones feel faster by comparison. Management loved the results. We generated millions in pure profit just by making the standard service worse, not by making the premium service better. Wait, so it does, this is, this is incorrect because he says it is, it does nothing to speed you up, but like, you are getting a faster delivery there. I mean, it's like, it's, even in this scenario, it's like maybe unethical, but like, this is actually a bull case for paying $2.99. Because, like, if the default service is actually slower and $2.99 gets me five to 10 minutes faster. This made me think they should add a feature that's, uh, whenever you get around,

I'll take it whenever you get around to it. And so it's just like, it's just like, surprise me. Really, really. I'm feeling lucky. I mean, no, it's just like the lowest cost version. Oh yeah, yeah. You might not get the pizza for three hours. Do you remember that or Uber pool those? Amazon does this for packages too. Oh yeah. You get a credit if you like ticket. Yeah, yeah. You get like a Kindle store credit or something at Prime. It's kind of interesting. Um, but yeah, uh, so, so I use DoorDash pretty regularly. I never pay for priority delivery because the time always seems about the same. And so I just don't care and I never push that button.

But this makes me feel like the button works, even though he's saying it does nothing to speed you up. And so that's kind of odd. But anyway, it goes on that to much darker places, he says, but the thing that makes me sick, the main reason I'm quitting is the desperation score. We have a hidden metric for drivers that tracks how desperate they are for cash, based on their acceptance behavior.

If a driver usually logs on at 10 p.m. and accepts every garbage $3 order instantly without hesitation, the Algo tags them as high desperation. Once they are tagged, the system deliberately stops showing them high-paying orders. The logic is, why pay this guy $15 for a run when we know he's desperate enough to do it for $6? We save the good tips for the casual drivers to hook them in and gamify their experience while the full-timers get grinded into dust.

that feels like that might not work. Like, it might just wind up putting like a strain on the entire system and just slowing. Like, you feel like you would just want to, like, optimize for the fastest possible delivery at all times. But I don't know. Anyway, theoretically, this could be happening. Then there is the benefit fee. You've probably seen that $1.50 regulatory response fee or the driver benefits fee that appeared on your bill after the recent labor laws passed. The wording is designed to make you feel like you're helping the worker.

In reality, that money goes straight to a... corporate slush fund used to lobby against driver unions. We have a specific internal cost center for policy defense, and that fee feeds directly into it. That's so, that's so interesting. I feel like, that's not really how money works, right? It's not like, it's like, route, you know, this is the earmark for this. It's not a guy in linear being like, tag the, you know, revenue from this product. Yeah. Send it to that. No, no, you know what this is. This is, this is, this is this is like dude logic for being like oh oh i got i got a i got a i got a five thousand dollar bonus like this is watch money i should buy a watch or something like that like this this money is for

this thing right it's like oh i got a bonus i'm buying a car with this with this money instead of just being like no you have a pool of money that is that is completely fungible and liquid and yeah i would say so so in restaurants around l.a you'll see like an employer well-being charged, which is like one, you know, one to three, four, sometimes five percent that is going. And I would assume that is going entirely to the employees, right? So there's two different things. The place we go for breakfast has a charge that they advertise as going to the health care

for the employees. And so if it turns out that restaurant owner was taking that money and kind of putting it towards rent, I would be very upset. Yes. So I can understand the general framing here. Yeah. All of this is, I mean... Well, so the key difference is like, is the money actually fungible or is it a function?

Because in theory, like, tips go to the service workers. And if the restaurant has a great year and lots of people come in and they all tip a lot, then that money goes directly to the workers as a bonus, right? And so they get that delivered to them. If you say, hey, the tips go to the workers and they don't. Just say you haven't worked in a restaurant. What do you mean? Oh, because it's pooled.

Well, I mean, when I worked at a restaurant, we'd get the tips. I mean, most of the tips were daily. You get paid out. Yeah, yeah, yeah, yeah, yeah, exactly, exactly. But it is a function of the business. It's directly aligned with the business progress. And so you would expect that if this delivery service, which everyone is alleging is DoorDash,

So if there are a lot of deliveries, there are a lot of people paying $1.50 for driver benefits, and so that creates a larger pool for driver benefits. If the amount of drivers stays flat, then the amount of driver benefits should increase. If that's not happening, then that is a violation of this promise, right? Basically. Yeah. In reality, the money goes straight to this corporate slush fund, they say. So, and regarding tips, we're essentially doing tip theft 2.0.

we don't steal them legally anymore because we got sued for that. Instead, we use predictive modeling to dynamically lower the base pay. If the algo predicts you are a high tipper and you'll likely drop $10. It offers the driver a measly $2 base pay. If you tip zero, it offers them $8 base pay just to get the food moved. The result is that your generosity isn't rewarding the driver. It's subsidizing us. You're paying their wage so we don't have to. I'm drunk and angry. Ask me anything before this gets taken down. I mean, really, I mean, first of all, got to feel bad for this person who, if it's real, they just quit working a delivery app. Yeah.

It's a software, and back-in software engineer. Now they're drunk in a public library on a burner laptop. Are you allowed to be drunk in a library? It feels like you shouldn't be drunk and angry in a public library on a burner. Imagine, when did they buy the burner laptop? Did they get drunk before they bought? Did they get drunk and then go by the burner laptop and then hit the library? I don't know.

Very premeditated. I don't know. It's a, yeah. No, I do. So, so I get extremely, I'm extremely frustrated. I will say this. Like this feels like, this feels like fan. Yes.

Really? What restaurant? In high school. I worked at, I actually, I first job, my first job outside of refing soccer was picking up cigarette butts with my hands for two hours in the morning. Yeah. And then I'd go at a restaurant and then I'd go work at a surf shop. Then I'd actually worked my way up the restaurant from picking up the cigarette butts off the

cigars. It was just pure cigarette butts. I was like, I think I was. Wait, people could smoke indoors or something? It was like a, it was a brewery. Okay. So people would sit outside. They'd sit outside late at night.

And so they would go until like, just a child in this smoke-filled environment. Yeah, and I'd go in and pick up the cigarette butts with my hands. I didn't even wear gloves. Very, very, very gross. Yeah. Disgusting. And so I, so I, so, this feels like it's fan fiction.

Yeah. There's, there's certainly some truth to it, which is the truth is that like, if you're using a delivery app, you should assume the delivery app is trying to get as much money as possible from you. Yes. And going to psychologically manipulate you. Totally. Into feeling better about the order, feeling better about the value, feeling like you're

You're supporting this person. It's on its way. They're going to do everything they can, UX-wise, to make it a delightful experience. And as somebody who... I appreciate the art of tipping. There's an art to it. It's something I've always enjoyed doing

because I'm always getting a service. And I at multiple times in my life worked for tips. So having this sort of digital, intermediary that is messing with that relationship sucks, right? Like if I order, if I order groceries and somebody's driving all around, I want to give them a meaningful amount of money for doing this service. Totally. And so if you have this intermediary that's kind of messing with that, and that's super frustrating. Again, I don't, I don't know exactly where this goes, right? This is clearly a product that participants on both sides are happily, maybe not happily, but they are still

like, you know, you can say you hate these platforms, but you're still using them. Yeah, of course. And people are still accepting these jobs. There are people that as much as they don't like the experience, maybe, of being the labor side of this equation, are ultimately still like the flexibility of being able to earn $10 here, $20 there, et cetera. Yeah. The thing that always bugged me was when there's not enough financial education among the workers

that you saw like Uber drivers who would be like renting a car or they'd buy a car with high depreciation and they wouldn't factor the depreciation into their wages. And so they were actually upside down on the deal and losing money every month. That felt like, whoa, okay. Like there needs to be something that helps that. I know that there, at least with the Uber days, there were some forums where people would go and calculate the highest ROI vehicle you could possibly buy. And sometimes it was a Tesla. Sometimes it was, you know, oh, you get this version of the escalade that's four years old.

So it's already depreciated. And then you can do Uber black and it's the highest ROI possible. But if someone's not doing that, then they could actually not be making money, which is very rough. But anyway, Tony rebutted this and said, this is not DoorDash. And I would fire anyone who promoted or tolerated this kind of culture, the kind of culture described in this Reddit post. There's so much wrong with this post. Dasher's are not human assets. Having a metric like a desperation score is an abomination. We've never had a driver benefit fee. Why would you charge for faster delivery, but not make it faster? We're not perfect by any stretch of the imagination, but we work every day to make our platform better for everyone who comes to it. What's described here is appalling.

And if true, whoever's operating in this manner should be ashamed. And DoorDash, the official account quoted as well and said the awful claims in this post clearly hit a nerve. To be clear, this Reddit post is not about DoorDash. Given how differently we operate, we thought we'd take a moment to share what our approach is actually like. And there was a whole bunch of back and forth. Some people were screenshoting it saying that it might have been AI generated. And I think that's not a good rebuttal because, first off, just, you know, I read the whole thing. It doesn't feel like it's written by AI. But even so, if you're a whistleblower and you have a bunch of information to share. I don't see a problem with passing that through an LLM to punch it up and get it spell

checked and grammar check. It might not do you any good if people feel like it's written by AI and they feel like, oh, this is just someone trolling on Reddit anonymously. What I would like to better understand is how many people, what the actual churn is like on the driver's side of these businesses because, you know, people talk about, you know, this acceptance rate and who gets what jobs and things like that. In a traditional environment, like in a restaurant, if you are like a server, let's say, and your manager is like, hey, like, you're in this kind of zone. Like, you're in the zone of the restaurant. Any table that comes in here, you're handling today. And the server's like, well, I don't want to do that table.

because I don't think they're going to tip very well. Or that person always comes in and they just sort, they, they're needy, but they don't order that much, and it's not going to be a big bill. Like that manager probably end up, like, firing the person, right? You have to, like, kind of accept the good and the bad, right? Whereas for this, it's different because DoorDash doesn't actually employ these people. The drivers, I think, like the flexibility of the work.

I think that's, like, a big value prop. And so, again, like, it creates, this scenario where you have to use a bunch of weird incentives to make the to ultimately make the entire like if half the time you went on DoorDash like your food never got

delivered the service would cease to exist because people would just be like well I'm not going to I'm just going to drive and go pick it up myself and I'm sure some people would argue that that that's actually good but anyway. The big question is did he tip his librarian

He was at the library drinking. I think if you're getting drinks at the library, I didn't even know you could get drinks served at the library. That's awesome. I want to go to this library. So at least hit the librarian with a buzz ball. For sure. This is a life hack.

If you have a 30 rack, if you bring a 30 rack into the, throw it down on the desk. Throw a buzz ball to the librarian as a sign of respect. Yeah. librarian, beer me, beer me, brother, if you're going to be in here posting on Reddit and chugging beers all day long, tip me, tip me. Oh, well, I think, I think, I hope that 2026 is a year of the buzz ball. You're into buzz balls. I was, buzz balls were post my time. I never had one. They were big

when I was in college. And what I appreciate now, in college, people were ready for buzz balls. Like, you were, no, you were, you were trained in the sense of, like, if somebody threw a buzz ball at you, your arms are staying at the side. Oh, Wait, wait, wait, oh, this is a, this is like a prank. I'm completely unfamiliar with it. Wow, you don't know. Tell me the culture of the buzzball. The culture of the buzzball.

So you get a buzzball. Yes. Obviously, like, near the checkout at any type of, like, yes, I've seen them. I've seen them. And so the idea is, like, you're going to buy the thing that you're actually going to drink. Okay. Just pick up one of these.

They have a few, it's the equivalent of a few shots of just terrible alcohol. That's pretty strong for single. Yeah, it's very strong. It's more than a single serving about. Yeah, it's very strong. It tastes terrible. So it's like a punishment. You would never drink one of these for fun.

Sure. And you go back to whatever, your house or apartment, and you throw the ball at somebody. If they catch it, they have to drink it. Normally, if you throw something at somebody and they're not expecting it, they're going to catch it. They're like, what's going on? Of course. But in kind of at least the heyday of buzzball, for me, you knew this was something that could, a threat that could come out of nowhere at any time.

It could come out at 9 a.m., right? 10 a. And so you're going to keep your hands down and you're not going to catch it. you'll side step it. But now... What if it, like, smashes in the TV? Are there, like, buzzball disasters that happen? I'm sure there are.

Results in a lot of destruction. But now people are not... I hit my dear friend, Ben Taft. He's a venture capitalist. He's my neighbor. I hit him with the buzz ball. When? A couple days ago?

Like, three weeks ago. He was fully unprepared. Well, I would be unprepared. I was like, I feel bad that you caught them. Scoot in the chat. says, I already got the CEO of Substack to drink a buzz ball. This is good lore. This is good, Scoot.

This is great. This is good, Scoot. See, is he, buzzball. Is the, is the, is the culture of the buzzball alive and well on college campuses? Or do we send you back? I mean, I'm not 21. I never had out of the ball. Oh, okay.

Oh, yeah, yeah, that makes sense. Uh, well, maybe we have to send you back for, for one, one non-gap semester day on your 21st birthday to go experience, uh, college life and, and the life of the buzzball. Anyway, let me tell you about Gemini 3 Pro. It's Google's most intelligent model yet. State of the art reasoning, next level vibe coding, and deep multimodal, understanding. Okay, let's go to one of the more viral posts from the last few weeks from the holiday. Kick us off, read this.

Oh, you want me to read it? Yeah. You want me to do it? No, no, I'm a three or four posts. No, no, no, okay, I'm kidding. We can trade off. So Justin says, am I just a monster? It's been four years since I became a father, and I'm beginning to fear for my soul.

The truth is, I just don't like being around kids for very long. Historically, this is not uncommon among fathers, but today it feels almost illegal. It's causing me a lot of confusion and anguish. The ideal amount of time I would like to spend playing with my kids is probably about 70 to 140 minutes a week. Roughly 10 minutes each day, maybe two times a day, taking breaks from work. My feelings of love towards them are perfectly strong, but I have to watch them or entertain them for more than about 10 minutes. My blood starts to boil. I just want to be working or accomplishing something. I try to be grateful, but it doesn't work. It's 9 a.m. this morning, Saturday, January 3rd.

It's a sunny warm day here in Austin. My four-year-old son is begging me to play catch in the street. I was drinking coffee, still waking up, so I didn't really feel like it. But at this age, his desire to play is insatiable. He begged and begged so I conceded and with a smile. I have no problem being a kind and loving father. The problem is I only, is only that I do not enjoy it. brutal very brave to to post this it's not that I'm trying

to maximize but we haven't seen the other side of this what if his son is like yeah bro I don't enjoy it either yeah what if this is going to be like a generational

you know father son rivalry where some some patricide going on yeah we don't know anyways Justin continues it's not that I'm trying

to maximize my personal pleasure it just seems wrong that I experienced so little delight when my dad friends all claim to experience so much. It was beautiful. We live on a picturesque tree-line block. I am even relatively relaxed from the holiday...

Give it up for tree-line blocks. Give it up for tree-line blocks. Great places to raise family. I don't know how to make him anymore. Playing catch with your son is supposed to be an iconic peak experience, yet for every single minute. On the inside, I just don't want to be there. I want to be drinking my coffee in peace.

Then I feel guilty and absurdly ungrateful and ashamed when we're done. I know that when he is a teenager, I'll long to have these days back. I have all of this perspective rationally, and I've been very patient and steadfast trying to digest it, but nothing fixes me emotionally. Am I a terrible person, or is my feeling within a certain range of historically normal? And it's modern parenting norms that are off. Whether it's my fault or not, I don't even care.

I just want to figure this out. Something is wrong, and I no longer have the excuse of being new to this. Okay. Well, a lot of people weighed in. 14 million views. This is, this is, this is, this is, this is, this is, this is, this is, five thousand like, I mean, I would, I would ultimately say this is the worst like-to-view ratio. It's rough.

It means that a lot of... And seven, more replies than likes. Crazy, crazy stuff. Anyway, I like Justin. He's been on the show. He came and talked to us about Nick Land. He's a very interesting thinker. Very interesting writer.

And one thing that part of what he talked about was like how you can make it as an independent writer or creative. Yeah, yeah. And so I think you have to view this post from the lens. of somebody who is, like, their primary focus in life right now is escaping the permanent underclass. But also outside of institutions. Like, he's not in an academic organization. He could be doing, like, writing or in-house somewhere

at some tech company or something. I think he was talking to us a little bit about that. But he's been very, very independent, and that obviously creates the type of work that consumes all, it seeps into all sorts of cracks because there's no moment when you're like, okay, I'm logging off and my boss doesn't expect me to do anything until Monday. It's like, oh, you could always be doing a little bit more when you're effectively an entrepreneur. So there's a little bit of that in here. I think, in general, skill issue. Skill issue. I love Justin, but skill issue. You have to use what I call the Dan Belzerian method. This is a parenting. The Dan Bilzerian method, especially, this works especially well for four-year-old boys. So you just assume your, you're, you have to use what I call the Dan Bilzerian method. This is a parenting. So you just assume your, Dan Bilzerian, but instead of entertaining, like, an Instagram girl, you're entertaining a four-year-old. So you're like, hey, want to get in a really fast car and drive around fast? They're like, absolutely. That sounds amazing. Want to go look at fine watches? Want to go,

you know, want to go look at guns or whatever else? Like, whatever Dan Bilzerian does, want to learn poker, buddy? They'll be like, absolutely. I want to play cards. Four-year-old boys have the mind of Dan Balsarian, effectively. And so before you say, oh, yeah, I can't find, I can't afford a fancy car like Dan Balsarian. A four-year-old cannot tell the difference between an SF-90 and a Dodge Viper.

Just get the Dodge Vip. Get a red car. Get a red car. They will be so stoked. My son wants to have a black wing themed birthday party. And I'm like, we're not doing a Cadillac CT5V blackwing themed birthday party. It is a very weird thing.

I got this car half jokingly. I know you love it because it goes room and it goes very fast and he's very excited about it. But truly, like, you have to return to like four-year-old boy mentality. And that is the mentality of Dan Bilsarian. And so if you adopt the mind of Dan Bilsarian, you will have a very enjoyable time with your four-year-old son. But I think the bigger picture is that like Justin probably just doesn't like playing catch. And that's fine. Like, my son is into soccer.

I think it's slightly deeper than that. I think if you're not satisfied with where you are in life as a man whose job is to provide for your family, if you're not satisfied with your life, you will not be satisfied by parenting. Unless you're behind the wheel of Dodge Viper. Yes, yes. But what I understand here is my reading of it is like he doesn't feel sad. Like, he doesn't, he, he, like, it's maybe a sense of, uh, stress. Yeah.

It's hard to enjoy just playing catch or going for a walk or, uh, yeah, because he can't you know, over the last two weeks. I was spending all day with my wife and kids and it's raining, right? So you're kind of like stuck indoors and you're, and our options are like going outside. Like, I would go outside and we would walk around and like pick up slugs and like fucks, right? because Dan Belzerian. Dan Bilzerian, four-year-old mode. And I felt at peace.

I was enjoying that, in part because we worked so hard last year. I was like, you know, historically on Christmas breaks, I would be annoyed that it was the holidays because I'd be like, it's inappropriate for me to email somebody and say, like, hey, let's talk right. It felt like hitting, you know, needing to slow down for like this couple speed bumps when you're trying to go 100 miles an hour.

Yeah, yes. And so, yeah, I think that's, I ultimately think that's, that's just a huge part of it. It's like the more satisfied you are by your life as a father, the more satisfied you'll be just doing the simple things of life with your kids. The other thing, the other thing that's, that I've noticed is the more time you spend with your kids, the more enjoyable it is. It's actually like, if you don't spend a lot of time with your kids, you'll spend, you know, 10 minutes, you'll get that sort of endorphin rush, and that's enjoyable, and then it kind of drops off from there. Yeah. But the more, like, there's increasing returns to scale.

Totally, totally. Yeah. Yeah, I mean, I've been playing Mario Kart with my son, and he beat me for the first time ever just yesterday. And it was like this, like, very interesting moment, very cool. It felt very fun. But also it's like, I just enjoy playing Mario Card. So, yeah. And same with like when we do Legos now. And so now you guys can start.

I feel like it gets better and better. You can start gambling his allowance on Mario Car. Introduce him to, introduce him to putting some money on the line. Maybe you should get an allowance. Every once in a while, he like acquires coins randomly somehow. And like, he's very proud of that. But yeah, I do think that there's something to be said for trying to find activities that align. Because a four-year-old, specifically a four-year-old, specifically a four-year-old,

boy, I think. I mean, that's not my only experience, but a four-year-old boy with a father is, like, the most memetic creature possible. Like, he's literally just looking to you for everything. So you can say, I like dinosaurs, and he'll be like, I'm so into dinosaurs. But if you're not into dinosaurs, you can be like, actually, I want to go look at horses today or something. Whatever you're into, like, they will be alongside, like, there's times when I can, like, whenever we're in the car, like, he doesn't request. Get ready to read the Wall Street Journal. Get ready to read the Wall Street Journal, buddy. I actually am excited.

I actually do that all the time. I will read all sorts of stories to him. And yeah, I have to stop and like explain them in like I'm five, literally explain them like I'm four. But it's still fun. And same thing with the music. Like he wants to listen to my playlist. Like he wants to listen to rock music and like the songs that I like.

And yeah, I do a light filter so that there's no like. Yeah, part of that is almost leadership. I think, uh, my, son loves dinosaur is probably, you know, if there was a Spotify rap... And you took a class on dinosaurs. Yes, yes, yes. No, but what I was going to say, if there was like a kid's book, Spotify wrapped, he'd be in the top point 0.001% of dinosaur enjoyers.

The problem is like, I read to him before bed every night, right? Seven days a week we're reading. If he was just like deciding what we were going to read, it'd be dinosaurs every night. Yeah. And it's so hard to pronounce. Dude, you've got to give me some dinosaur book racks? because we... No, but let me finish.

So where I got to is I'm like, okay, like, we got to introduce more books because like if we're, there are like kids oriented books even for like the four-year-old range that are more, that are like interesting parables, stories, et cetera, that are just like timeless. Timeless and worth reading. And so I've like enjoyed massively over the break,

like switching it up a lot. We got like a comic book, comic book style, like action comic of the Bible, right? So we're reading that, right? And that's like he, enjoys it, I enjoy it. It's just like, you've got to be like switching it up and not at the same time like you want to be leaning into their interests, but not not overly just doing whatever. Yeah. Generate your own ideas. Yeah, yeah. Turn this semi-analysis deep dive into a children's

book so that I can read it to myself. That's the TPUV-7. There's something there. Something there. You've got to give me some recommendations for dinosaur books because my son found a dinosaur toy, took a picture of it, IDed it with one of the LLMs, and then he was like, I want to know more, I want to read books, and I realized I'm actually like, woefully under-resourced when it comes to dinosaur books. I have like poop, and none of them are particularly good. So anyway, there's a lesson. Breaking, breaking news. We got some breaking What's breaking news? What's breaking news?

Department of Energy just announced that it awarded general matter 900 million for uranium. Wow. Hit that app-loving gong, John. It shakes the camera. It shakes the camera. That's amazing. I think we got to figure out how to hang the gong from the rafters. We do.

Yeah, yeah. We're refining. We're refining. Let's pull this. Well, while we do that, let me tell you about Lambda. Lambda is the superintelligence cloud, building AI supercomputers for training and inference that scale from one GPU to hundreds of thousands. Let's move on to our breaking news. Let's see.

So the U.S. awards $2.7 billion worth of orders to boost uranium enrichment. So it sounds like there's a few players here. Department of Energy announced Monday. it was wording orders 2.7 to three companies to boost domestic uranium enrichment over the next 10 years in a broader effort to reduce U.S. dependence on Russian supply. Yeah. No, no, no. Like a lot of the, a lot of the uranium comes from decommissioned missiles and stuff. So it's already been processed, and then they break it apart and we, you know, just buy it from them because they're like basically denuclearizing, I believe. Pretty, pretty well. Yeah. So it's American centrifuge.

operating General Matter and Orano Federal Services secured the order. Nice. The department said, if you want to sell to the government, maybe put federal services in your name. Well, I think that might be not a startup. I think that might be owned by the government or something like that. Is it or is it a private? I don't know. It says it's a private company.

But the contracts would require the companies to meet specific milestones to provide enrichment services for low-enrich uranium and high. assay low and rich uranium for existing nuclear power plants and new smaller modular reactors today's awards show that this administration is committed to restoring a secure domestic nuclear fuel supply chain capable of producing the nuclear fuels needed to power the reactors of today and the advanced reactors of tomorrow said Secretary of Energy Chris Wright Russia is currently the only country that makes HAL EU you rich uranium enriched to between

5% of 20%, which is said to make new high-tech reactors more efficient in commercial volumes, funds to make the fuel domestically in the United States were included in a law to ban uranium shipments from Russia fully by 2028. Big news. Very cool. Well, let me tell you about CrowdStrike. Your business is AI. Their businesses securing it.

CrowdStrike secures AI and stops breaches in their tracks. Check out CrowdStrike. So the Dwarkech piece, it's sort of big. We might want to go through it. Let's save it for tomorrow. Let's save it for tomorrow. Let's skip through this. The big shift to Miami, Miami is back.

This was from Teddy Schleffer in the New York Times. He writes, News, Larry Page and Peter Thiel are making moves to leave California by the end of the year to avoid a possible billionaires tax. that could hit them where it hurts. Patrick Carlson said yesterday, Miami posting might be a phenomenon whose time has passed, but having just spent a few days visiting, it really does feel like a boomtown in a way that no other American city that I've spent time in over the past few years does. In some way, it reminds me of Chinese cities.

Why was he in Miami? Was he just there for the Soho Beach House? Was he there to... F1, maybe? F1. F1. F1. It takes place in Miami.

Miami every year. Oh, you think he was in there like this week? He said having just spent a few visiting. Okay. I don't know. Because I mean, to be clear, like Patrick Collison would be subject to totally the billionaire tax that Roe Kana, Roe Kana. Former guest of the show. Former guest of the show. It was funny. It was funny. So Roe came on the show and I asked him about something that he had said like a week prior. And he was like, I don't agree with that. It's like, it seems, He seems to be a guy that does, it's weak beliefs weekly held. I don't know. So, anyways.

But yeah, it was, yeah, David Sacks announced the, David Sacks announced that he was going to Austin. Yeah. PT, great press release. PT clearly press release for Specter. December 31st. Rokana said Peter Thiel is leaving California if we pass a 1% tax on billionaires for five years to pay for health care for the working class facing steep Medicaid cuts. I echo what FDR said with sarcasm of economic royalists.

When they threaten to leave, I will miss them very much. So Rokana will not be missing, everyone that leaves. There is, the debate sort of shifted back to like, okay, if billionaires do leave successfully, what will the impact be on the technology industry? And I'm wondering if the question is not like Silicon Valley exists in San Francisco and Silicon Valley or like, you know, the delian idea of like let's move Silicon Valley to Miami.

Like that didn't really happen. But is there is there a world? I had this theory that Miami could become, you're not moving Silicon Valley to Miami. You're moving Sand Hill Road to Miami. And like no startups move to the bay and are like, I need an office on Sand Hill Road. They're like, I'll get an office at San Francisco when I want to raise money. I'll drive down to Sand Hill Road. Yeah, but almost every venture capitalist also just got a presence in San Francisco.

No, but what I mean is that, like, there is a fundraising hub that founders go and visit, and whether it's in Sand Hill Road or the Presidio or Miami, it is accessible if you're doing a roadshow. Like, if you're taking a Series B seriously, you're probably. going to spend a week in New York. You're probably going to spend a week in Miami at this point. And maybe Austin, too. And going around to different locales for fundraising doesn't seem that hard. Yeah. In 2021, people were doing Miami road shows. And they were pretty effective. You could go to Miami and get around down. Miami Tech Week was in many ways that. Yeah. And you were sometimes just calling back to the West Coast doing Zoom calls, but you could meet a bunch of ECs.

Yeah. You know, tight. And then also, like, most of these funds have, like, one or two billionaires at them. And then the vast majority of employees of the fund would not be employed. So you could have the founder or the core GP somewhere else. And then the entire team there on the ground taking a ton of meetings. And if it's time to meet the big guy, you fly out or you do a Zoom call. And so I don't know that this whole idea of like transplant everyone.

I guess it is a question, it is a different question for the founders. Because like Sam, Dario, these guys, like they cannot be in Miami and actually run an organization in San Francisco. I would imagine they just have to pay this. And it'll be very odd. Well, I guess not Sam because he's like, who knows where he is in terms of equity? I mean, he's probably above it broadly. Every definition that it's like. He's like, would you like some nonprofit credits?

No, every, every, I'm sure Sam is close to a billionaire just off of his striped. Yeah, just off of other stuff, for sure, for sure. Okay. The car collection. Here's a theory. Okay, Sam is actually behind this, right? Because it's going to force Dario either to pay a massive fine or they're going to have to move, right? That's better for opening eye because he doesn't have the, yeah, get the tinfoil hat on.

So Sam is actually behind this. Maybe. He's getting everyone out of Silicon Valley. so we can stay there. Maybe. I somehow don't, I somehow don't think that's going to, that's what's going on.

Anyway, Roe Connagher's. Yeah, I mean, the funny thing that this whole wealth tax debate was happening almost at the same time as the Somali Daycare thing. Somali Daycare thing. Yeah, which was very odd because it's,

there's a question of like, how effective is government, like once you get the money into the government, it sounds very nice that it's like, oh, it could be redistributed, but if it all gets like stolen. Which was funny, which was funny,

because we've had a recurring bit on the show where we said, like, is PT really a billionaire? Like, I've never seen him in Chrome Harbour. Oh, yeah. I've never seen him wearing Rick Owen. Yeah, and then the guys were wearing Chrome hearts. I've never seen him with an RM, a racing machine on the wrist, right?

And the Somali Daycare founder was actually rocking Chrome for a press conference. It's crazy. You know, on the question of like, is the government effective at just redistribution broadly? I was reflecting on this fact that you can kind of put the billionaires in two camps. There's like, let's call them like the bleeding hearts that like they sign the giving pledge and they say, I'm going to give away half my wealth or all my wealth. And then there's like the ruthless capitalists say, I'm not giving away anything. I'm just going to pass it on to my kids, right?

And so there's two camps. And let's call the bleeding hearts. Like they really do want to give away. They want to leave their capital behind with other people in better ways. they wanted to be given away, right? No one that I'm aware of of the giving pledge folks has ever just said, I'm going to give my money to the government. Like they've never just been like, oh, actually, I should just give it to the federal

government because the federal government is an efficient way to pay down the debt or just like distributed to people and a tax rebate or something. Like it's always been, okay, I got to go and do like this health care and this cancer research and this doctor, it's always something more specific. Yeah, Bill Gates's new pledge is specific. specifically going to deploy a lot of capital into Africa. Yeah, but it's never just, okay, yeah, like I built my wealth in California. I'm just going to give my money to the state of California, which is odd.

But Rokana puts a lot of context here. He says, my district is $18 trillion, nearly one third of the U.S. stock market market market market. In a 50-mile radius, we have five companies with a market cap over a trillion dollar. And we have five companies with a market cap over a trillion dollar companies. This is not AI written, but it maybe should be. If I can stand up for a billionaire tax, this is not a hard position for 430,

for other members or 100 senators. Those saying we wouldn't have a future Nvidia in the bay if this tax goes into effect are glossing over Silicon Valley's history. Jensen was at LSI Logic and his co-founders at Sun. He started Nvidia in my district because of the semiconductor talent. I do sort of agree with this.

I think the effects would take a very, very long. time. No, no, the, the, the issue I have is the, is, uh, it's just a slippery slope down, uh, the, the, the asset seizure. Yeah, yeah. They really, they really got to build the high speed rail. Like, like, all of this is just so, like the high speed rail is just such an easy dunk on all of this. Like, high speed rail, you know, like, oh, you want more right? High speed rail. Like, high speed rail didn't work. So, like, you need nothing you do works. If I was Gavin Newsom, I feel like I'd just be like, okay, I'm actually going to go finish the high-speed rail thing, and then you can't say that anymore because I did it.

But it's still, and it's like we were even talking about the high-speed rail thing was a meme like three years ago. It's like you've had three years to go build more high-speed rail. And it just hasn't happened. And I think that's like it's, I mean, it was the cornerstone anecdote of Ezra Klein's book, Abundance with Derek Thompson. It's like the main anecdote for like a new order among like liberal Democrats. And I thought it, I thought it did.

It broke through really well because it met people where they were, which is dissatisfaction with the lack of progress, lack of building, lack of abundance on the left. And Ezra sort of like reset the discussion a little bit. So I don't know. I'd be interested to see where he weighs in on this particular. Is he a fan or does he think that this is a good one? Friedberg had some of the best commentary on this. He absolutely mauled, Roe, Kana.

It was a brutal mocking. There's a quote from the All-N-Podd X account. They say, Freedberg, California's billionaire taxes a Trojan horse to go after the middle class's private assets. Freeburg, the reason they're calling it a billionaire tax is to make it easier for people to vote for it and sign up to this entirely new tax system that they're proposing to put on all Americans at some point and for the first time ever degrading our private property rights. forget how much wealth you have, forget about how rich you are, forget about the term billionaire, millionaire, whatever it is. We're creating or proposing the creation of a new tax system that allows the government for the first time ever to come in and audit everything you own. All the jewelry, your grandma gave you, the value of all the couches in your house, the value of your car, the value of all

your stocks and bonds, and the government can come in and for the first time look through the veil into your personal property and say, here's how much all this stuff is worth. I'm charging you a percentage of that. That's what I need to get paid. And it doesn't matter that it starts with billionaires. What matters is that we're giving the government the right to look into our private property and take a percentage of it every year. The total net worth of billionaires in the U.S. is $8 trillion. The net worth of the U.S., the middle class, and everyone else is $170 trillion compared to the $8 trillion of the billionaires. Chumas says they need a way to open the doors to think after the real honeypot. The real honeypot is not 200 people. And Freedberg continues

just so everyone understands what the real goal of this is not to tax billionaires because there are other ways to tax billionaires. You could charge them a capital gains tax if they borrow against their assets that they haven't paid capital gains tax on. Very simple. That can resolve this. Another thing you can do, you can raise the capital gains tax rate. Sounds unpopular. I don't agree with that, but that's another way to deal with this, which is to take the capital gains tax rate from 20% to 30%. You could do that. The real goal of this is to create for the first time in American history a private property asset seizure tax because they're going after the 170 trillion, not the $8 trillion that the billionaires have. So, yeah, again, the founding fathers would be,

absolutely disgusted with the state of our current tax structure. They would be throwing up. Cover the New York Post would be crazy. Yeah, I just think people need to realize like how far this can go. And I think that the political story of the next, of the next seemingly decade, two decades, potentially as our lifetimes is politicians running on, I will raise taxes, I will take assets from people, I will take the assets from private citizens, and I will redistribute them to you if you vote for me. And that feels like an incredibly vicious cycle

that I don't see how it ever stopped. So I'm so blackpilled on it. I don't even want to talk about it anymore. Well, unfortunately. The number of billionaires has grown globally from 140 in 1987 to over 3,000 in 2025. With enough inflation, everyone will be a billionaire eventually. If we go through a period of hyperinflation, you could see everyone making billions of dollars in a year. Introduce billionaire tax.

Print. Yep. Hyperinflate. Hyperinflate. I mean, you know there's like countries where like there were so much inflation that people would be making billions of dollars for loaf of bread, basically. A quadrillion dollar bills. I'm reading a book on the banking system in the 14th and 15th century right now. And they introduced various governments who would introduce wealth. They were always like going to war with other city-states.

And so various governments would like introduce a wealth tax. And then so people were just constantly trying to argue like, well, this isn't worth that. Or they'd be putting, you know, they'd be putting assets into a bank under a bunch of different names. Like there was tons of tons of tons of tons. of fraud and, but ultimately people do not enjoy earning, earning dollars, buying assets and then having those assets confiscated. It does feel like it will create an immense amount of jobs, like assessing all these assets. I mean, like the family offices of these billionaires themselves employs like dozens of

people to track everything. And then you have the flip side of that, which is the private, or the, the, you know, the state effort to monitor all those assets. It's so, it just becomes, yeah, it's, it's, it's, it's going to be truly, it's truly an impossible, like, you know, we, we spend all of our, not all of our time, we spend a lot of our time covering the private markets, right? And so if you're a business owner, uh, and you have a, uh, some, uh, somebody in the government coming and saying like, hey, you made X amount, you made, you made, you made a million

dollars. We're actually looking at public comps right now. And, uh, companies in your, uh, category are trading at between 20 and 30 times earnings. So actually you're worth, you know, you're worth 50 million dollars. And we have this tax where if you're worth 50 million or more, which, which Elizabeth Warren has proposed a wealth tax, I think like eight, she's talked about doing like 8% above 50 million, something like that. And so yeah, you're actually, you, you call, congratulations, you qualified for the wealth tax. And the person's like, well, we only have this much net income last year because of this one-time thing. And suddenly you're having to argue against the value of your own. Yeah. It'd be very weird because a lot of startups want to get the

valuation as high as possible for recruiting purposes. And then you have like for tax purposes that they want to minimize the valuation. There's already a little bit of a time. It's never, it's never going to work. I was talking, I was catching up with a portfolio company founder over the break. And technically, I have shares in the company that are marked at close to seven figures. And he's considering a pivot right now, which means that, like, the value of the equity is, is, is worth, like, a tiny fraction. Sure, sure. The last round, because it's, like, completely changes.

Yeah, it's starting over, start basically, like, starting over from zero. Also, there's scenario that he just shuts it down. Right? And so it's like, I, I kind of, I kind of, of put the even though even though like investors in the company have that equity marked yeah at tens of millions of dollars yeah I value it at effectively zero yeah but the government will not the government will be like oh look at this reputable VC that backed that that that backed the company and they just did it they did it in the last year of course how can you say this is not worth that this is worthless stock this looks like it's worth quite a lot and and anyway so anyway let me tell you about shop

and then let's move on to some outlooks for the future. Shopify is the commerce platform that grows with your business and lets you sell in seconds online in store. The backbone. On mobile, on social, on marketplaces, and now with AI agents. The backbone of the internet economy. Backbone of my career, I've been using Shopify since 2012, I believe, 2013. So, yeah, over a decade on Shopify.

Wow. All right, let's talk predictions. What did Scott Belski predict for the future slash 2026? Number one, he said, massive amounts of talent arbitrage. There is much fear and discussion about job losses as a result of AI, but the often young and fearless AI native talent that is most aggressively engaging with new tech has a fascinating advantage in the current workforce.

People with knowledge of the better way will run circles around their colleagues and bosses. For example, companies want to hire answer engine optimization efforts or experts over search engine optimization practitioners. Most marketing leaders are still not conducting old-fashioned focus groups rather than using AI breakthroughs and market research. 2026 provides a precious window for young or overlooked talent to gain an advantage as early adopters. Two, the buzzy concerns around AI in Hollywood will be grounded by the reality of what audiences increasingly crave craft, meaning, and shared experiences. The industry, this is particularly interesting because, of course, he works at A24. The industry will start to realize that there's a stark difference between content creators,

those willing to trade control for speed using AI to make ads and social media content, and artists those not willing to trade control for speed, who only favor emerging technology that preserved control and precision to help new and better stories be told. The technology and models that ultimately elevate the craft will get lasting traction in Hollywood, while the prompt-based slop tools will focus more on social content creator use cases. the idea of personalized films with audience cameos will be humbled by the realization that people favor shared experiences. Ben Thompson's talked about this a lot, the shelling point. Like Taylor Swift is a specific event, and when you go to that Taylor Swift era's tour

and you run into someone else who also went to the same show, you can talk about that, and it's the same thing. And people, even though they like personalized content, they also like being able to say, oh, do you agree with Dwar Keshe's take? Not just, do you agree with what AI told you specifically today, and you saw something completely different? People do like that shared experience. I think that's something that's a little more tractable than people give credit, too. Finally, the idea of, so people want to be inspired by craft, and they want a common experience

to discuss or even share in theaters with friends. Did you see any movies over the break? Yeah, this is why people enjoy watching TVPN, is they can debate, is Tyler using the Gigacad filter right now. It provides a shared experience that you can have a conversation around. What did you say? Did you see any movies over the break? Sicario, one. You did. Let's go. I recommended that. Give me your review. I mean, masterpiece. Masterpiece. Goes in my top five films ever. It's fantastic.

That's because you've seen four. That's right. By default. Top five by default. Top five by people. Did you like, you know the difference between one and two? Like a completely different team. So one is regarded. Taylor Sheridan didn't do two?

I forget exactly the difference. I need to look it up. But Sicario two, I believe it had a different, I believe it had a different team. The director, he wrote Sicario two but didn't direct it. Exactly.

Yeah. And I think it got worse reviews than the first one. But did you like one? better than what? Zootopia 2. He's no, as I'd recommend Zootopia. Just skip Zootopia 1.

Zootopia 1 is great. Go straight to 2. Great movie. Yeah, anyways, what movie do you recommend I watch next? Ooh, that's a tough one. There was a list of movies that was going viral from, was it Bill Simmons movie list? Movie list.

This one... Did you watch the ping pong, Marty Supreme? I didn't. I watched it. You watched it? What did you think? What's your review? It was pretty good. I enjoyed it. I like Safty Brothers, their movies. Yeah, me too. Okay, here's the thing. The marketing was so good and so intense that it feels very likely without having seen the film that the marketing overshadows the film itself. Do you think that tracks? Yeah, maybe. What will we remember more in five years? The marketing or the film? Yeah, I think that's definitely an argument you could have, which is like not, you don't usually, like, say that about movies. So, yeah, just by saying that, it's like, yeah, probably the marketing.

Yeah, maybe it's a lesson. It's like, don't make your marketing too good. Otherwise, people... They feel like they got the experience. This is something Derek Thompson talked to us about, and Dan Wong called out again to go back to his piece. He said that, like, a really, really short TV hit would sell more books than going on a... really big podcast for an hour because if somebody listens to someone lay out their thesis for their book over an hour. Tell the most interesting stories. Exactly. They're like, well, it was, I got one hour of this. Do I really want to have six hours of this on the audiobook? I kind of got the audiobook version, the summary. Whereas if somebody sees like on, you know, some daytime TV show, somebody shows up for two minutes and it just gives like one take, you're like, oh, like I could go more.

And then also demographically, the TV watchers are more likely to buy books. But we're not talking about books, we're talking about movies. And there are some recommendations here from Bill Simmons. He calls these the most, the most re-watched movies of the century. Devil wears Prada, social network, anchorman, the town, the departed, Miami Vice, Hangover, Super Bad. Have you seen Superbad? I think that would be a good one for you. It's very funny. It holds up. I think you'd like that. Stepbrothers? You've never seen stepbrothers? Stepbrothers. I feel like we did a meme around stepbrothers. The movie that I think you might like, if you like, Sicario might be taken. Have you seen Taken? Liam Neeson? Great movie. Great movie. Team loves it. Studio goes wild. Studio goes wild.

John Wick. If you haven't seen the original John Wick, that's a great movie. John, have you seen Prisoners? Yeah, Prisoners is great. For sure. Prisoners. Any De Niveville Nove movie is good. Jordy, maybe you should go like Art House. Have you tried any of those? Art House. How about Capitol? House. You have any recommendations? Margin' call? I saw, I saw Sicario as a double feature. I went to the

movie theater, bought two tickets, saw Sicario, walked straight into the Martian, saw that. It was great. Great day. Well, this is the year of the social network, too. Yeah. I somehow think it's not going to be as good as the first one. It just feels like it's going to be myers.

I think it depends who plays Dwar Cash. Yes, yes. I hope that they really go into that. It's about, it's a It's a free AI. It's entirely on Jan Lacoon's hot takes around the Lama 4 overfitting of the benchmarks. That should be where it's... Anyways, back to Scott.

He says, behind the scenes, proof of craft content will enter the mainstream of advertising and entertainment. As more AI generated content fills our feeds, we will develop a membrane of doubt. That's fake. We'll become a default reaction as we become. That was kind of the slower reaction to this company, pickle.com,

that came out. People at first were like, wow, this is super cool. And then people started being like, wait, is this real? Is this fake? We can get to that later. Scott says that's fake will become a default reaction as we become increasingly unimpressed by attention-grabbing content.

This is how I've been with AI-generated kind of launch videos and vibe reels. I just, I haven't watched a single one. I assume you'll tell me if one of them is good. AI-generated launch vibe reels. Maybe you're not watching them either. Tyler probably watches them still. pure AI purist over here. Andy two cents. That was pretty good, I thought.

That was AI generated. Yeah, but that had, that had like great ideas in it. It was, it was a lot more, it wasn't just, here's a bunch of cool images that are trying to make you feel something. I'm skipping over those. We got a couple recommendations from the chat. Glenn Gary,

Glenn Ross, highly recommend. That is a fantastic movie. Also famous watch movie, if you're into Rolexes. Alec Baldwin's character, says, look at this watch. This watch costs more than your car. It's a Rolex president. Margin call. I feel like

I saw that a long time ago. It's been a wild lioness. I haven't seen that. That looks pretty good. I'll have to check that out. There's a couple others here. Anyway, cognition. They should make a movie about cognition. Let me tell you about the software engineer. Devon, the AI

software engineer. Crush your backlog with your personal AI engineering team. We've got to get Scott back on the show. Yeah. Update on how things look in 2026. He had some crazy predictions that came 100% true this year. He came on the show very early and predicted the IMO gold medal from AI exactly as it played out. That was a great call. And he's always been incredibly tapped in there.

So Scott continues, says multiple industries from insurance to health care will be impacted by the implications of materially better lifespan, health span, and joy span. That's interesting. Well, we have Justin Mayors from Truman Ed coming on the show in just 25 minutes. And he's obviously deep in all over this and can tell us more about this particular trend. Scott says, as health wearables, routine blood testing, increasingly accessible preventative body scans and AI health coaches proliferate the population. Humans will have materially more insight and impact over their own health. As we begin to live longer, multiple industries will change U.S. life insurers with de-risk. denuity exposure that are most impacted by mortality, like global life, among others, are poised to benefit from the insured policy living longer. We will also see growth among the underlying plumbing companies that manage blood testing, like Quest Diagnostics, biomarkery detection, and preventative scans of the major consumer apps and whoop-like devices proliferating our lives.

We will also see shifts in travel, entertainment, dating apps, and beyond, as the societal implications of longevity become a mainstream conversation. You got to get on TrueMed. We're having Justin on the show in just a little bit. We've got to get on TrueMed and figure out the ultimate looks maxing stacks and make sure you're paying for your looks maxing regimen. They offer the peptides? Yeah, with pre-tax dollars. You've got to be looks maxing on pre-tax dollars. That's key.

That's key. And thanks to Trumet, it is now possible to looks max with pre-tax dollars. Hardware becomes a more popular moat amid a surge of hardware startups. Interesting reflecting on this year, there was a surge of hardware startups. We demoed a lot of hardware in 2025. I was asking you about the X-Reel, those AR-V-R glasses. I was super bullish on those. But you churned, I churned.

Haven't really been using it. Wait, were you actually bullish on them? Yeah, because when I used the Apple Vision Pro, the one experience that I really liked was watching movies in it. So I watched Citizen Kane, another great movie you should watch, from start to finish. But Citizen Kane is an extremely difficult to film to watch by modern standards. Like, Sakari was like, boom, boom, boom, there's always something happening. It's really exciting. You know, you're watching the tactical stuff, the camera's moving, those sweet shots of like the SUVs moving from the aerial.

It's amazing. Like it's so visually stimulating. The color grade's amazing. The sound mix is amazing. Citizen Kane is so slow. It opens with these like really slow title cards and like it's just the easiest movie. ever if you're not locked in and really into it to like pull out your phone and just get um and just get uh uh you know just like tuned out is tyler watching a movie now i mean citizen can there's

much more boring movies i will say that i've seen no i agree i agree it's a great it's a great movie but but it's just like any movie that's even more even before the 80s i feel like has a slow start there's lots of exposition it's just less it's just less ticotified versus like a modern Fast and the Furious movie, even if it's not a good movie, it's going to hold your attention because there's always something happening. And so... Yeah, that's probably true. But, I mean, Citizen King is known for having, like, the...

I forget what the term is, but it has, like, the thing that starts the whole story, right? The, like, Rosebud, like, is supposed to, like, get you... Yeah, yeah, it's a good hook. It's a good hook. You know what it is? The initial Rosebud, where he drops the... It's the Snow Globe, right? Yeah.

The Snow Globe. Have you ever seen those brain-rot TikToks where it's throwing the bottle of, like, marbles down the stairs, and it smashes, and they bounce. and they bounce everywhere. No. You've never seen those? It's like the most brain-rondy content possible. And it's kind of like that.

Somebody should make a version of that where it starts out. It just smashes and just completely goes everywhere. Anyway, but I was bullish on this because my experience in the Applevision Pro was extremely positive when it came to watching films. Like, I enjoyed watching movies in it. You could put on Avatar, and it's in three-in-in-an-screen, which you just can't get at home otherwise. So it was basically like a home theater just for yourself.

Now, I never had any time to use it, and so I returned it. It was not something that was going to go into my daily driver. But we tested a lot of these on Guinea Pig Tyler over there. We gave them a Quest 3S Ultra Xbox Edition. Didn't play that much. You didn't prestige, right, in Cod? No, I didn't have prestige. Skill issue.

And the X-Rails, you wore them on one flight. You enjoyed it for that. Yeah, on the flight, it was great. But you just took a flight, and you didn't bring them. You didn't think, ah, I got to put those in my backpack. Yeah, I mean, that was... Not even aspiration. That was in the summer, I think.

Yeah. I last used them. RIP. And the rabbit and the humane. Well, the only, the hardware that I care most about this year is getting a couple great racing sims here in this, here in the studio. We got to see if we can get one on TrueMed. I wonder.

It is a, it is a sort of a lifestyle upgrade. I wonder what is the current meta? I've been watching a lot of Instagram reels about sweet racing sims set up. and I was wondering what the, is the meta, is the best, is the best racing sim VR or is it physical screens these days? I wonder which one is like the true. Still physical screens. It's still physical screens. Like the F1 drivers use physical screens. Yeah. They don't do VR. Because I've done some racing in VR and it's really cool. It's, it's amazing, but it is a heavy thing in your, on your face. Anyway, there will be more hard. Scott continues. Ordinary data becomes a less valuable

mode. These days it feels like every company is trying to sync everyone's data and doing so is becoming easier than ever before. Soon enough, all your products will have all of your data, whether it is via connectors now or computer vision plugins later. As connectors, data between apps become ubiquitous, you can expect, one, the growing importance of proprietary graphs, who knows who, who works with who, who has access to what, et cetera. Two, more projects trying to make personalization portable for consumers in three real-time data sources, whether X or the weather or ocean tide patterns becoming more valuable and increasingly captured by robots. graphs, portable memory, and real-time data are the new proprietary data modes.

He says seven ambient listening and summarization will go mainstream. Well, many teams use services like granola or Zoom to record and annotate their meetings. Not me. I talked to some people that loved the granola wrapped because it had all the context on every, like they... You said nothing on my end, thanks, 400 times. But, I mean, there were some really interesting insights because it had... if you run a remote startup and the vast majority of your day is spent in recorded Zoom meetings, you kind of have a record of every word, and you can compress that down, summarize it.

And it can actually give you some, at least entertaining insights. I don't know that it's going to completely reinvent your business. But people, it definitely delighted a lot of people. So that was cool. What if Granola this year just has a pop up and it's like, do you know you can automate this entire role? for $1,000 a month, you can replace this. Brutal.

I have to imagine that's part of their pitch to investors, right? It's like, we're going to understand, like, what work is actually getting done at these companies and who does what. And how things get made and how things get done. Kind of far-fetched. But, yeah. I mean, Will Mnitis was posting about the, like, it was, like, corporate, it wasn't, like, corporate espionage, but it was some post about, like, corporate, about, like, corporate, like backstabbing and like Machiavellian behavior within corporations. And I think his point was just like there are there are so many secrets within businesses

about like how things actually get done that don't typically get documented. And so you can get closer to that. Just because like if you're running some business and the secret to your brand is that you do things a specific way and that's not documented. I mean, we notice this with, we'll, we'll go on podcasts and get interviewed and we'll explain exactly the thesis of TVPN. like how it works. And even then, some, there's still some like lost in translation. When we lay it out,

it doesn't just come out of just watching it. And I'm sure that that's, and I think that, like, that's clearly true for a lot of businesses regardless, like there's just secrets that are kept within internally. And if Grinnell is like a way to access that, like it could be valuable. I don't know.

What's the next one? The power and consumer AI will shift to tightly coupled hardware and operating system providers, the desire for local AI, like ambient listening and summarization will coincide with a shift to AI privately running on your device

as one opensource models that you can download change and run locally on your device become more capable. Two, consumer hardware, Apple and Android phones, whatever Open AI is brewing and our computers become capable

of running powerful LLMs locally while the world is going to change yet again. The implications for the AI stack are tremendous from the chips that become valuable, the evolution of operating systems, the devices we use in the role of open source AI. I suspect these changes will be major area of focus over the next year. I'm very excited to see what Apple releases this year with

Gemini. I expect it to be if it's better than the Photos app. If it can be... I have noticed that the text-to-speech feature has gotten better. I feel like they updated the model at some point, and if I'm on a blog post and I click, like, read this to me in the car or something, it will do a good job and it sounds very human and they clearly updated that model. I haven't tested a lot of other stuff. But the thing that strikes me as crazy is just that Apple really seems still stuck on this annual release cadence. And in the boom cycle of AI, it feels insane not to just be like, yeah, Gemini deals going

live tomorrow. Like, it's like, okay, you did the deal. Like, just put Gemini in Siri today. Yeah, I mean, people have been talking about there were some some, some research paper about whether like AI tools are actually benefiting companies. And then you, you look and like it was the models that were out in October. It's like that's already like too late. Like yeah, yeah, yeah. Opus is is much better than the previous plot model. Yeah. And you think just getting like actual customer feedback,

starting to troubleshoot these things. There's also like just a benefit to, I feel like Apple is, in many ways, very cautious with their brand. They're very thoughtful. And they don't want, like, you know, when Google launched their first image model, now they're loved for nanobanana. It's remarkable. But remember the first launch, people were doing, like, the accurate 1940s soldier, and it was very controversial because it got the race wrong.

And that was sort of like a backlash that Google was, like, moving fast enough that they got through that. think of a deal. And obviously they had no negative intentions with that. It was just a mistake. But Apple's not really in that in that mode of oh yeah, like let's release things and then deal with like the backlash of things that aren't fully polished. But it feels like there's a little bit of an advantage to actually releasing so fast because like open AI over 2025 probably had like seven different crises of like AI psychosis and adult mode and all these bad. There were so many bad headlines, but then there were like 22 really great days where people were like,

this is amazing, thanks so much, this is cool new model. And so like, it's sort of netted out, whereas Apple was sort of like at this low din of like, oh, what are they doing forever? But I'm very optimistic about the next version. I'm going to skip over a few. One more from Scott. He says, we'll see the rise of internal development teams as companies replace single purpose bloated SaaS products with their own apps that collapse functions in magical ways. New internal application development teams will be spawned in large, companies and they'll buy code tailor-made software and agent-driven solutions for functions around the company. At first, these efforts will replace SaaS products, especially those that are expensive,

private equity-owned, clunky solutions. We support private equity here. We do. And they're clunky solutions, but I can see many of them getting replaced. Over time, these homegrown tools and workflows will start to collapse functions into one another. For example, why must legal and finance tools be so different in an era where the distinct functions of social marketing sales and customer support are blurring together shouldn't new solutions collapse these functions into one. As companies build their own solution, they'll be optimized for a higher level of cross-functional work. Let's give it up for cross-functional work. Yeah.

I mean, the Mark Leonard, the founder of Consolation Software, was sort of talking about this, right? It's great. He just came out, sat on an investor call. It's over. It's a good run. But I mean, like, so I was, I was wondering this because on, it was January 2nd or something, the market opened and all the SaaS companies were selling off. Did you see this?

And people were sort of like, oh, everyone's waking up to the Claude Code moment. You'll be able to, you know, vibe code any app and replace all your SaaS. And so it's bad news for all the SaaS companies. And I was trying to think about that. So I pulled some stats on, um, on, on, you know, the actual revenue of the large SaaS players. And I have some here. I mean, they're all, they're all growing. So Salesforce, Adobe, Service Now, Palo Alto, Shopify, Workday, Atlassian, Zoom, CrowdStrike, Snowflake, Datadog, HubSpot. These are, these are basically the biggest

sort of pure play software companies. And they're all, they're all growing revenue somewhere between, you know, five and 30 percent. And for this year, I mean, they're all still up. I guess the question is, like, if one of them has a down year, that would be a real indictment. But it would be, it would be, my initial thesis was like, it'd be hard to justify this idea of, like, vibe-coded software, truly replacing SaaS. If all of the big SaaS companies are still growing, then it's like clearly a Jevins paradox thing. We're just getting more software. but maybe it's the PE-backed small point solution thing that's actually going to get ripped out. And that might be more of like a collapse that we see.

And sure, the big software companies, they're so big and they do so much that, yeah, you're not going to one-shot it with Claude. So like you're not going to do that. But you do need to be worried about this like long-tail software that just sticks around and just raises prices and doesn't add any features. Yeah, the threat to some of these, I guess more, if you think, about like the Joe Lamont style companies where built to die built to die right you have the software that you know an investor allocator like Joe would be comfortable buying knowing that the revenue is just going to decline and eventually cease to exist and you can imagine this being like somebody that makes software for large vending machine companies to manage inventory you know something like that right

in a kind of overlooked category, not big enough for venture. The question is, like, will those companies, like, be super quick to adopt? Will they be super quick to adopt AI products themselves, or will they be using, will they be, like, vibe coding? I don't know. The question is you potentially get this sort of, like, like, explosion of, like, web developers that just were, like, thousands, millions of people globally that are going to, I'm going to go to business owners and I'm going to make them a website, right?

and you could see something like that happening with like vibe coders. Yeah. That's literally his first point is massive amounts of talent arbitrage. People with knowledge of the better way will run circles around their colleagues and bosses. So there'll probably be a lot of people that understand these tools and then can go into a mid-sized enterprise and say, hey, we're paying so much for this old piece of software. Or they'll go into that company and say, hey, we can improve it and keep our customers. We can go back into growth mode economically. and maybe this thing doesn't need to die.

Maybe with just a little bit of extra work, we can make it great again or make it bigger. I want Tyler's response, but first I want to tell you about MongoDB. Choose a database built for flexibility and scale with best-in-class embedding models and re-rankers. MongoDB has what you need to build. What's next? Tyler.

Yeah, I think there's some distinction. Like you mentioned of the software companies, it's like Snowflake or MongoDB. Like people aren't vibe-coding their own vector data. databases. It's like a tool. It's not like the full solution. I would be very surprised if those like are the chief, you know, if those are heard a lot by vibe coding. Yeah. No, I know. I agree. Also, I mean, anything with regulatory, anything with proprietary data involved, there's just a lot of different modes that companies build up. And I think that even companies that seem simple where they, it feels like something you could vibe code, but, you know, they've been aware of moat building for 20 years and they've been doing it. And you might not know that, oh, they actually have some really solid relationship with this legal decision and this thing. And they can navigate this for you. And it feels just like a SaaS app. But there's something

else going on behind the scenes. I think that that creates more durability. Anjane has 2026 group chat consensus. One, extraordinary capabilities, progress from Anthropic and Google. Leaving out. Leaving out opening. What's going on? there. Two, sovereign. I mean, he's, he was an angel and anthropic. He's, he's got his team. Sovereign nations as largest customers of open source models. Three, compute scarcity at unimaginable levels. Four, violent public backlash against perceived AI job losses. Hopefully no violence. But the perceived AI job losses is notable. I saw there, there was a pretty viral video that popped up on my ex this morning, and it was like a montage of like big tech layoff, and they were positioning

it, like it was all AI job loss. And so I think that is going to, that's the story right now. On one hand, it's CEOs and management team saying, hey, maybe I should do more with less. In some cases, it's, I'm doing more with less because of AI. But in a lot of cases, there's just like, you can just exercise willpower, right? This was the Twitter story. which was like, hey, let's do the same, let's provide the same platform with 75% less people. Yeah, yeah, I agree. Railway. Railway simplifies software deployment, web apps, servers, and databases run in one place with scaling, monitoring, and security built in. Go check out Railway.

Tomas says 11 predictions for 2026. We got a bunch of prediction posts here. Every year I make a list of predictions and score last year's predictions. 2025 was a good year. He scored 7.85 out of 10. Here are his predictions for 2026. Businesses pay more for AI agents than people for the first time. Whoa, that's a bold. I don't know that he's saying all businesses. I think he's saying some business. Waymo rides cost 31% more than Uber on average, yet demand keeps growing. We saw blacked out Uber, which is, or Waymo, which was iconic. I want to get to that one. Murdered out Waymo? It was the real murdered out Waymo?

It might have been AI. I don't know. We got to dig into it. Vector databases resurge as essential infrastructure in the AI stack. 2026 becomes a record year for liquidity. I saw something interesting where, so he's predicting SpaceX, Open AI, Anthropics, Stripe, Databricks, IPOs, with SpaceX and Open AI ranking among the 10 largest offerings ever, the pent-up demand. And so a lot of people were saying, like, this will drive a lot of liquidity into new companies. I mean, it doesn't seem like new companies are having trouble raising right now, but in theory, you know, if you're a venture capital firm, you return, you know, 50,

billion dollars to your LPs. They probably want to sign up for the next fund, and the next fund probably gets bigger. And we've seen a couple charts where it looks like venture dollars are declining, and the number of funds are declining, but you could see sort of LPs become a little bit more cash rich, where they, over the last five years, they've been really, really allocating towards venture. They're finally going to get their payoff if they're in the right funds. And now that they have liquidity, maybe they go and do something else with it. Maybe they go fund new managers. Maybe they go double down on the growth funds that have been providing them return to this far.

Maybe they buy land. Maybe they buy land. There is a question. They're into land maxing. AI models execute tasks autonomously for longer than a work day. According to meter, AI task duration doubles every seven months to two current frontier models reliably create complete tasks, taking people about an hour extrapolating this trend by late 2026. AI agents will autonomously execute eight-hour work streams. That will be crazy.

You will be able to just say, hey, go spend a full day on this problem. come back to me and it will come back to you when it's done. Now it will come back to you before eight hours, right? It's the equivalent work time. Yeah, I think there was this, there's good blog post that came out, I think, a couple days after the break started. It was like critiquing the meteor evils. And the main thing was like, if you actually look at what the eval is doing,

for the, I think it was for Opus, the headline was like, it can do four and a half hours or four or something. or something. And so the way you do that is like you give it tasks that take people on average four hours. But it's pretty hard to find tasks that do that, right? So if you look at it, they only had like 14 actual tasks that gave that benchmark. So it's a very small sample size. So it's pretty hard to, you know, reliably see it. Is that actually true? At the same, are you laughing at the chat? Bobby says that you guys hear about opening eyes, quote, pen. And Goldrock says, is it a vape pen you

can gamble on from. What's that I mean? That's the most 2026 pen. I wouldn't be a surprise. There are a lot of AP pens that have full on LED screens and microcontrollers

so people will run Doom on them or run Tetris on them and vibe code on them and stuff. There's a lot of fun stuff there. Let me tell you about graphite code review for the age of AI. Graphite helps teams on GitHub ship higher quality software.

Faster. Now part of cursor. It's wild. Well, if you want a vape pen, maybe you should try and pay with it with pre-tax dollars using true med. We have Justin Mayors. He's going to tell us that you can't do that. I'm going to tell him. I absolutely can. And I will. And I will. I will be buying a vape pen on this. We were, we, Justin, welcome back to the show. Great to see you. Happy New Year. John wants to vape Chinese peptide. No, no. So I don't. I actually don't. But I do want to, I want you to tell me. I want you to tell me. how I can use TrueMed to build the ultimate looks maxing stack. I want to pay for it with my looks maxing stack. Yeah, what are you used? What's your stack? You've clearly been, since you've been on the show, you've gone.

Your cheeks are red. I can see you've been bones smacking before this. You're just talking about the fit, right? That's incredible. Your midface ratio is completely different than the last time we saw you. Something's clearly going on. TrueMed clearly helping.

Anyway, how are you doing? What's new? I'm doing phenomenally. Doing so well. And, you know, one of the things that I've been using from a health standpoint is obviously bone broth and all of the health benefits that come one hand washes the other wait so so you literally smash the bones to make the bone broth correct so you've been in bone smashing for a decade now his family actually made their money in bone smashing

it is a bone smashing company might be the best pure play bone smashing company there is we were decades ahead of clavicular i think you were yeah we need a looks max index for sure for sure Exactly. Kettle and Fire can be, when you guys go public, we'll put it in there. Yeah. Anyway, just reset the story for us. What's the news? You picked the worst possible day to announce it, but give it to us.

Maybe the best. You guys broke through the, there wasn't a lot of noise. It was sort of a quiet day. But yeah, yeah, take us through it. Very quiet day. Take it. Yeah, I mean, so we finally announced that we raised $34 million for TrueMed as part of the series day. And so.

The first hit of the new call. Hit it. I love that. I love that shakes the whole camera. It's so good. It's a bigger dog. The first... 2026 is going to be a big year for you guys. The first 2026 gong hit. Anyway, so $34 million. Wait, wait, wait. So when did the deal happen? What was the thesis around launching it, the announcement when you did?

Yeah, so it happened like earlier last year in the first half of last year. Sure. But we basically had a bunch of different things happening at the company level. We wanted to tie it up with a bunch of different announcements. we grew three times, which was fully baked by the end, which is great. Yeah. And so kind of like packaged everything up and decided to announce and kind of mention that, you know, this thing that we're doing and this movement that we're building

around incentivizing people to invest in prevention and invest in their health, well, they still have it, is a big thing. And it's a thing that tech and other people are paying attention to and waking up to. So I want to do announce it later than the year. And has the elevator pitch changed to consumers, or the companies that you're working with, like how are you positioning the actual value prop of TrueMed, like getting people on board?

Yeah, I mean, so to brands like Peloton, AteSleep, Lifetime, those are our partners. You know, it's quite easy. For many of these brands, yeah, like for many of these brands, exactly, Aid Sleep, there are many of these brands like, they realize that they are selling a product that oftentimes is expensive.

And so if someone can use pre-tax funds and save 25 to 50%, you know, if you're like in the max tax bracket and living California, you can save 50%. Buying, getting an 8 sleeper, a peloton at 50% off is like quite a material thing. And so we have merchant partners where we're doing 15, 20% of their total sales are going through true med. And for the individual, you know, if you're someone with an HSA or FSA account, historically these accounts have been used under this idea of like, wait until you get sick,

get cancer, need surgeries, need like pharmaceuticals late in life. And then you have a tax advantage account that can pay for everything once you're sick. And, you know, the IRS has been clear, like, the average American is sick today. We should be spending these dollars on things that studies show are effective at treating or preventing different diseases. And so that's what Trumanet enables. And what's the status of the, of, like, HSA? It's usually, like, an option and you have to pick a specific health care plan to be eligible. Is that right? Like, like, are HSA's booming? Are people adopting them? Like, what's the actual flow to get one in I maxed out my HSA this year because of TrueMed on our TBPN plan.

Yes. Yeah, I mean, HSAs are growing quite a lot. They grew around 11% last year. I think of the last three years that that's been about right, like 10 to 15% growth. Our thesis, though, is that, you know, you ask the average person, like you, Jordy, before TrueMed came along, like, what is an HSA? Do you have one? Why should you care about it?

The average person just has no idea. Like, they just do not care. It's considered almost like a boomer. savings account. And so one of our hypotheses is that these accounts historically have been high friction and you can't use them on anything that you actually want to prevent disease. Like, who cares about buying Band-Aids tax-free? And so our hypothesis is like if you can buy something, pay for lifetime membership, pay for AIDS lead, pay for these different things that, you know, assuming that you qualify, that many, many more people will actually sign up and fund these accounts.

And like, I think that over the next decade, there could be a trillion dollars in HSA and FSA accounts, even without any regulatory changes or anything like that. Is it possible to pull a PT and put startup equity in an HSA? Could I throw a few shares? Do you have any availability? Like trade in the HSA? Is that a thing people can do? That's definitely a thing people can do.

It's not something that TrueMed necessarily supports yet. But it's a thing people can do. I mean, I don't know if there's an HSA PT out there yet. Why do you think we can move? So, so TrueMed is a company that, to me, is shockingly political. Like some people hear about what TrueMed does and they get angry. They're like, no, you should only be able to spend your HSA funds once you're deathly ill. You shouldn't be able to use an exercise.

You shouldn't be able to use it to buy something that helps you exercise that, even though we know exercise extends your lifespan. So it's just like crazy, crazy that we're in a place that, you know, spending, spending HSA funds is like political and people are angry about it. Do you have any, do you have optimism around that changing over time? Is this just kind of the nature? So much of health is just deeply political. Yeah. I mean, honestly, one of the biggest disappointments that I've had in the last couple years is seeing health become political.

Like somehow you have a former, you know, Democratic candidate. for president that is now part of the Trump administration who's talking about health and it like means it's a right-wing issue somehow. I think that there's always going to be people that look at this. Like within the industry, you know, I came as an outsider from the industry. I started kettle and fire before this. And within the industry, it's still like controversial somewhat, whether like bug spray should be HSA or FSA eligible. Like the industry is just very, very behind and very stuck in the ways of and under the assumptions that the average American is healthy and HSA and EFSA accounts are there and exist to help people.

pay for health care when they're sick, on the rare occasion that people are sick. That was true, like 22 years ago when these accounts were started when Congress basically created these accounts, I just think those default assumptions are no longer true. Like, you look around and the average American today is sick. Like, you know, 70% of Americans are overweight or obese, 93% have at least a metabolic marker of dysfunction. Like, we are an incredibly sick country among the sickest countries in the world. And in my view, HSAs and FSAs are one of the best tools that we have in our admittedly very broken healthcare system to try and direct funds towards prevention and towards root cause interventions that can actually make people healthier and treat and prevent

meaning of the chronic conditions people struggle with. Is there a lot of volume happening on TrueMed that's going towards like Wagovi and people paying out of pocket for weight loss treatment products, or is that separate? No, so as of right now, we don't support GLP-1s. Like, what true, you know, I would say that the healthcare industry as a whole, the payment rails are relatively good at allowing people to pay for pharmaceuticals. Like, if you want to pay for pharmaceuticals, you can use an HSA today. But aren't a lot of people paying for GLP-1s out of pocket?

Yeah, they can, but HSA for many people is considered out-of-pocket as well. And because it's like they can direct where and how they spend these funds. they're their own funds, just they're tax-free. And so what we are basically doing is we are trying to make it legible and bring online a bunch of effective lifestyle interventions, exercise, sleep, supplements, things like that, that the traditional healthcare system, like right now doesn't really know how to think about and people don't really know how to pay for it.

So is there like a big roadmap for you on new verticals that you want to bring online? Is there a process for like actually getting something approved to work with TrueMed? Like did you start with eight sleeps and then you added Peloton? and it was like a different process? Or is there just one bucket and you can throw whatever you need into it and you just sign the companies as they come up? Yeah. So we basically, we have a medical advisory board and a medical team.

And we basically look at when an intervention comes to us, when a brand or a merchant comes to us and they say, hey, we want to work with TrueMed. We basically look at a couple things. Like what do the studies say about how effective a certain intervention is? Based on that, like what conditions will that intervention be effective for? You know, like you can't get exercise to treat like a toe fungus or something like that. Like that would be a bad intervention. Like a doctor would never prescribe that.

And so what we are basically looking for is something like exercise. What are the conditions that exercise as an intervention can treat reverse or alleviate? That's the IRS standard. And assuming that there's a good amount of data research supports that intervention, and then we talk for a clinical team and we decide can we support or not a specific intervention. And so we have a bunch of things, exercise, sleep supplements, and the life that we support today. A big thing that I want to unlock this year is food. Like if you look at the data, like medically tailored meals have incredible efficacy, incredible

ROI, but they're very complicated to, you know, it's very complicated. And the IRS, frankly, has like a heart attack when they hear people are using HSA or FSA to spend on food. Because like everyone needs food. On raising canes. Chicken fingers. Chicken finger dream. My chicken finger dream is chicken fingers via my Truman account. No.

No, you're talking about like an actual, like, you know, properly portioned meals. There's been like a whole history of companies that have tried to do sort of like portion control meal prep. There's been a number of different kids. Obviously, there's varying levels of efficacy. But in theory, that's a really great intervention. I mean, it just seems so obvious. Like, you go to any health and wellness influencer. there. And before they tell you about the six different types of magnesium or something,

like they're going to lay down the law, which is just sleep diet and exercise. Sleep diet and exercise. And so it feels like those three should be on TrueMed first before, you know, even the creatine gets on there or something and the supplements. But obviously, like food, it feels like it's a challenge. So you're working on that. So is that a goal for 2026? 100%. I would say that is one of our big goals for this year, especially as like tens of millions of Americans are going to be on GLP-1s. Things like eating more nutritionally dense food, getting more protein. Like, these things really, really matter.

Or even things like GLP-1s to work over any long period of time. Yeah. And that's something you want to watch. Right. 100%. Is that a place where people miss out if they're on GLP-1s? Yeah. How big is a team?

We're 51 people now. 51 people. And like what's your kind of like personal ethos as a CEO around building a company like this as, you know, every single day we have more and more AI progress. Are you pushing the team to, are you pushing everybody on the team to like, you know, use the tools as effectively as they can? Like, where are you getting the most leverage, all that kind of stuff? Yeah, I mean, we're certainly pushing the engineering team to use things like clock code and whatnot all the time. I think that where we are getting the most leverage and where we have

started to focus is leveraging a lot of these tools, specifically on the prototyping side of things. We recently stood up a team within TrueMed that it's basically me and two other people, where we're calling it the Venture Betts team. It's basically the team that just moves fast, tries things, prototype stuff, and sees if it works or not. And the goal there is basically to unlock a new line of revenue over the next 12 to 18 months.

And for that, we are basically using almost entirely AI tooling to spin up, let us. landing pages, many products, like all these sorts of things that are not exactly built to scale, they're not built for six-nines, uptime or anything like this, but just with the pure goal of like get as many shots on goal and as many iterations as possible. And so that's how personally we're starting to use it a lot. I feel like do you identify with the label like FinTech? Is that a fair category? Or do you think of yourself as like a different category? He'll say he's finance tech.

finance tech. Yeah, I would say definitely we're a fintech company right now. Yeah, that's fair, right? Yeah, for sure. The question I ask is because do you have any, like, are you at war with people that are trying to do fraud on the platform? I feel like that's like a permanent background noise if you're a fintech company, but I don't know if you operate at like an abstraction layer where that's kind of like,

oh, it's not a headache for you. Is it a headache? Or most payment like a company that does payments online typically is incentivized. Yeah. It's set up in a way. It's like business joins platform moves money. Yeah. You want more businesses joining the platform moving money.

I think that you guys are, I'm sure people have tried stuff, but feel pretty insulated from that kind of like large. Yeah. I mean, it's not a payroll company. You can't just immediately do money laundering. But is there like a like a fraud fighting team or how have you addressed that? Is it even a problem? Yeah.

Yeah, so I mean, it has been a small problem, but it's been a thing that, like, honestly, Stripes tools are pretty darn good. And, like, we build on top of Stripe. They're quite good. Sure. And so it hasn't been the thing where we're, like, throwing personnel at it or anything like that. But, yeah, I mean, certainly the most possible boring explanation of TrueMed is, like, we're a B-to-B-B-SAS company that does compliance for HSA and FSA.

You know, and, like, given that, like, we're definitely a fintech company. News Resolutions. have any, do you think they're effective for living a healthier life? What do you think of New Year's resolutions? I personally, I find them great. I actually just had my first kid though, so I'm sort of like resolutions out the window. Sure, sure, sure. Just keep on, keep it on. A hundred percent. But we were talking, we were talking a few days ago and you said, it's amazing and super underrated, which is a, which is wildly different than another person that we were talking about earlier on the show that said, are my missing something?

Yeah, yeah, similar name, similar name. What about, what about, like, big trends in 2026? I feel like you have seen basically every health trend between two and six years early. What do you think we're going to be talking about this year? Well, let's get into, I think, the thing that's been going the most talked about. I mean, that is, in my world, that's the most. Tide to that, but peptides. Oh, peptides.

Sure, okay. Let's start with peptides. So I did my first peptide cycle probably five years ago at this point. I'm now at the point where so many people are trying so many different things. And I did this at a time where I was like, there's really no large-scale studies on this. But anecdotally, I knew people that had done it. Sure. And they'd tried them for, you know, you do it as a cycle.

Nobody even back then was saying, take this and never stop taking it, right? There were also some really big high profile, like the Hugh Jackman Wolverine, like everyone saw that and we're like, oh, whatever that he did. Is that peptides or just anabolic? I think that was BPC 157, correct? I don't know. Anyone know? I don't think so. I don't know.

No, that's the Wolverine peptide. Oh, okay. I just mixed them up. It's Wolverine peptide because it helps like. Oh, recovery. Recovery. Okay, anyway. Maybe it was just gear.

But my concern with peptides is like you basically, I just look at them as like slightly toned down like anabolic steroids that are more targeted in terms of what they do. And so when you have like millions of people that are just trying to buying them from random sites online, they can be contaminated. I have a lot of like a lot of red flags around that. I'm not kind of diving in any deeper.

So peptides. Faustian bargain or not? I think that these are going to be incredible tools. I think that the FDA is probably going to release much needed guidance around them this year. Like, there's just too many Americans that are buying cheap for research purposes only peptides right now. You know, it's just gotten too big that the FDA has to do something. Like, I personally know a bunch of people that swear by these things, the efficacy is there.

And, you know, they're certainly, like, they're indulgence, like your body makes them in many cases. I'm very bullish on them from an efficacy standpoint. I have tons of questions about quality. And I think for me, the thing that you just have to wonder on all these things is, like, one thing that our current FDA does a very good job. job of, but pharmaceuticals is figuring out, are there any sort of like side effects? So like, does pancreatitis risk go up 12% when you take this thing over a year period or whatever? And like, no one is looking at those sort of long-tail effects of peptides. Certainly not when you're

looking at staying on these things, combining them with other peptides over, you know, some long period of time. So I would say that I'm like bullish but cautious, whereas I think most of the people in peptides right now are sort of like rithing it. Purely bull. Yeah, exactly. It does feel like we're in a new era of just experimentation and just like, I don't know, like a couple decades ago, like a TRT or something was something that was like buried in a body building forum and yet had to know a guy. And now it's like, here's the website that was perfectly coded and will just deliver it to

your door. And it's like a completely different experience. It's way, yeah, yeah, yeah. It's way more like user experience focused and like just, way less risk. I mean, we've seen this with all sorts of substances and just activities where, you know, we've productized them in the way America does best. It becomes a button. And I mean, that's totally true. I also think to be fair, this is like the health stuff is the biggest problem in the country in my mind. Yeah. The average American is sick. The average American has bad

sleep. They have bad energy. Like the demand for something that can promise you feel better, your joints feel better, you're sleeping, you're losing weight is like off the charts has never been higher. And so I think given that, like, there's just infinite demand for products that can make someone feel better, give them more energy, lose weight, sleep better, whatever, which is why I'm just, I think peptides are going to be one of the biggest trends of the next five years. Certainly they're going to come with side effects and downsides, but they're going to be here. Yeah, yeah. Underrated story about American competition, the whole story of how Novo, like, came up with all this stuff and then just got, like, kind of beat in the public markets by Americans just being like, where, like, the American

biotech industry, like really ripped it. Anyway, other trends? Nova Habinan had been in series. I mean, basically. I mean, they had that problem with like the, they didn't file the IP rule or something. What's the, what's the state of Austin? Oh, yeah. Sacks is opening an office. Lots of people are getting out of California. Make the pitch for Austin. Yeah, give us, like it. Give us an update on Austin. Yeah, look, Austin's great. I think that Austin will do well to the extent that California decides to self-immolate. And it seems like right now, California is like just pouring gas on the fire. So, you know, we'll see if the wealth tax thing actually goes through or even gets a medium amount

of popular support, then, yeah, like there is going to be a huge exodus from Silicon Valley, from L.A., to other places all over the country. Is it Austin or Bust? Is it Austin or Bust or are there tech people that are going to Dallas or anywhere else? What about the, I could see some allocators ending up in Highland Park. I don't know. Highland Park is in Los Angeles.

No, no. Highland Park is a neighborhood in Dallas. Oh, it is? Okay. I've only been to Dallas once. I don't think anyone's going to Dallas. No one's going to Dallas. So if a tech person's leaving San Francisco, they're going to Austin.

I think it's basically Austin, maybe Boulder or Denver, maybe Miami. Although I think the Miami thing was kind of not a real thing. I personally, personally, I'm more of an Abilene guy. Abilene. Straight to the data center. Energy. Straight to the data center. I love it. I love it. Yeah, I'm Alaska guy. I say get up there. It's nice and cold. There's bears. Yeah. Yeah. Yeah. You're locked in. But anyway, any other, are there any other predictions for just health trends or foods or anything that's like rising and falling? Are we finally reached peak slop bowl? Are we going in a different direction?

No, I'm going to go get a protein cup from from Chipotle later. Is that real? Are the restaurants real? Are the restaurants real? like downsizing in reaction to GLP-1s, or is that just like, you know, some random organic thing that's happening? You are seeing certain grocery chains, especially ones that sell a lot of junk food, are starting to downsize or see, like, junk food sales go down. Okay. Mainly because people are less hungry and most of junk food is predicated on, like, people eating them infinitely while snacking all there.

Sure, sure. And so I do think there's going to be a renewed focus on nutrient density, on protein, on things like this. I think you have to imagine, like a big trend that I'm, I'm, I'm, bullish on is these GOP-1s are going to be everywhere. We have the oral version of Ozempic coming this year. There's going to be 50 million Americans on these things in the next five years. I think there is going to be a backlash where people start to see what are the downsides. They start to get caught up in like micronutrient deficiencies in other sorts of things that just come from,

even if you're eating a bad diet, but you're eating less of it, you're still going to have health problems even if you may be, you know, less overweight or less obese than you were previously. I think we'll start to underwrite that more over the next couple years. Makes sense. Well, it'll be interesting to track. Well, this is your second time on the show? Yeah. Let's make it a, let's make it a monthly thing.

Yeah, next time you're in our life. There's a lot to talk about, and we love you. This is fantastic. We love that. I love you guys. Thanks for the merch. We'll talk to you soon. We'll talk soon.

Goodbye. Bye. Let me tell you about 11 labs. Build intelligent, real-time conversational agents. Reimagined human technology interaction with 11 labs. Or in our case. theme songs. Our theme song is generated by 11 labs.

Do you want to do some timeline? How long do you want to go? Because we could close with this post. This is very optimistic post from Elon Musk, replying to Mark Andreessen. Mark Andreessen says, it's time to grow. This is in response to 4.3% year-over-year GDP growth in Q3. Kind of crazy.

Didn't see that coming, says Calibhammer. And Elon says, double-digit growth is common. within 12 to 18 months. If applied intelligence is in proxy for economic growth, which it should be, triple digit is possible in five years. Let's pray for triple digit GDP growth. It would truly be a golden era. A lot of bottlenecks to that, but let's hope it happens.

I like the sound of that, John. That's a good place to end. I think that's a great place to end. We'll cover Lulu's post tomorrow. Lulu's actually coming on the show this week. And also, yeah, tomorrow. So Gabe, you got exactly what you're looking for. And also we'll have to discuss the prediction about Open AI buying Pinterest.

That's an interesting one to kick around. It was very funny because it was just a prediction post from the information, but people were reporting on it like it's a rumor or like it's like a fact or something. But there's a lot to dig into there. But anyway, thank you so much for watching. It is so good to be back. So good to be back. So good.

of you. Thank you for tuning in. We seriously missed doing this. As a couple of Irishmen, we yearn for labor, we learn to be on the mics, tilling the RSS feed. And thank you, thank you for being a part of this. We're so excited for this year. Yeah, it's going to be a lot of fun. And we hope you have a great Monday. Talk to you soon. Goodbye.