Isaacman Back in the Cockpit at NASA, Google Takes AI to Space, 𝕏 Timeline Reactions | Cliff Obrecht, Jerry Murdock, SHL0MS, Shehzan Maredia, Mina Fahmi, Alessandro Chesser, John Maslin, Eugenia Kuyda, Anish Acharya, David Risher & Erin Brewer

29:10Maybe he was actually wrong about the NFT question. But of course, the truth is that he was correct about NFTs being a bubble and just happened to be also wrong about us all dying to AI in a matter of days. Both can be true. You can swing and miss on certain things and you can hit it out of the park on others. Oh, we mentioned it yesterday. Polymarket was tracking the Momdani election in New York City. I think we were tracking it at like 95 % and then of course it went to 100 % because Momdani did get elected. And Rex here has a reaction of Michael Burry or Christian Bale in the big short, biting his fingernails, saying, a socialist just got elected mayor in the heart of the financial world at the top of the greatest bubble of all time.

30:03How are those related? I don't know. But yeah, people are not happy. They're moving to Florida, I guess. I don't think, I mean, I think a lot of the people that said they were going to move are waking up this morning and it seems, you know, we'll see. The big issue for New York State from a tax revenue standpoint is they don't actually need a million people to leave for it to have a material impact on budgets. Like if like a certain. Yeah. But the question, I think people overestimate how communist New York can become in a year and underestimate how communist New York can become in a decade. And so I generally think that people are freaking out, thinking that there's going to be a 45 % wealth tax next week.

31:02We'll see. We'll see what actually gets put in place. But good luck to all the folks over in New York City. The market seems to be liking it, according to High Yield Harry. How are the markets doing today? Oh, we're ripping. The NASDAQ is up 1.15%. We should be in white suits. Bitcoin's up 3%. After the last week, we should be at White State. The Dow Jones is up 0.6%, and high-yield Harry said, Zoron's America. Wow. Bitcoin touched 99. 99? Oof. At 1 a.m. That's stressful. Thankfully, while we were all sleeping. This post from this Aaron Slodov resurfaces email that I think is - It's a classic.

31:45Instant classic. Peter Thiel wrote it to Mark Zuckerberg, Sheryl Sandberg, Mark Andreessen. January 5th of 2020 said there are many themes that could be developed more here but let me let me make a few quick points for now. Nick I certainly would not suggest that our policy should be embrace millennial attitudes unreflectively. I would be the last person to advocate for socialism but when 70 percent of millennials say they are pro-socialists we need to do better than simply dismiss them by saying that they are stupid or entitled or brainwashed we should try and understand why. And from the perspective of a broken generational compact, there seems to be a pretty straightforward answer to me, namely that when one has too much student debt or if housing is too unaffordable, then one will have negative capital for a long time and or find it very hard to start accumulating capital in the form of real estate.

32:35And if one has no stake in the capitalist system, then one may well turn against it. So feels like basically called called to a T, uh, you know, the last, uh, or at least like could have easily predicted this type of, uh, election outcome, uh, five years ago. Somebody asked me, do I have it here? Where is it? Somebody asked me, um, about, oh, I need to find it in my DMS. Um, like, you know, should you go to college, something like that. Oh, I really wish I could find it. And I was kind of noodling on it. It was after we talked about the Palantir, don't even go to college, just go work for Palantir straight out of high school.

33:26And someone was saying that I think that they had the opportunity to do it. They wound up going with college or something, and they wanted to know my thoughts. And there is this weird question of student debt that I think gets completely left out of the equation when, and it's so, so important. It feels like if you, if you want to, if your life's work is like, you're wanting to be a doctor, like you just have to go to college. Like you, you're just not going to get on that track. Just being like, yeah, I've been, uh, I've been cutting up people in my, uh, in my basement for the last month. So I'm ready for surgery.

34:01I vibe coded the scalpel. No, it's not going to happen. You got to go to college. Uh, But if you do want to be something that's a little bit more flexible, a little more creative, a little more entrepreneurial, you can probably drop out of college. But it depends on like a whole bunch of other factors. Like if you're rich and you're going to graduate without any debt, there's pretty limited downside to going to college because you get to just hang out and vibe code apps and do whatever and clout farm and aura farm and do whatever you want. Like, it's like, it's a pretty flexible environment. It just sucks to go through that and then come out with$200 ,000 in debt.

34:43And so, so that's the thing. Then, then there was like the question of like, if, if the, if the Department of Defense is going to pay your, for your full ride, like, I don't know what's going on, but if you can get a full ride and you can get no debt, well then, uh, then it's probably worth going. But also if you're so elite that people are just throwing money at you to go to college, you can probably take that money and not, and just teach yourself and not go to college. What do you think about that framework? Like this idea that, that, that like that the, that the actual debt is a very important factor into like, is college valuable?

35:22I feel like so many people boil it down to like college good or college bad, as opposed to like college, a good bargain or college, a bad bargain. Yeah. I don't think it's that under like underrated debt question. Yeah. It seems like, yeah, obviously, like if you're not going to take out any debt to go to college, then it's like basically just like have a good time for four years. Yeah, yeah, go have fun. Yeah, there's like the cost of like your time, but it's like, you know, you're young. It's not that big of a deal. Yeah. It does feel like we're not making a lot of progress on it. Yeah, I think that the people that go to college take on debt and then get a job that doesn't really require a college degree, just acquires some amount of agency.

36:06Yeah. Like really kind of shoot themselves in the foot. Yeah. I wonder what would happen if we deregulated some of the tracks that require college. Like it is extremely hard to become a lawyer without going to college. because in order to take the bar, you have to have a degree from a law school. And in order to go to law school, you need to go to the undergrad. There is a loophole where I believe if you are studying under, if you're like mentoring under a lawyer, they can advocate for you, something like that. But in general, do you know? Yeah, it depends on the state. So what you're saying is true for New York, Maine, and Wyoming, where you can have an apprenticeship.

36:53Some states, you don't need that or a degree. You can just go take the bar? I could just walk in and take it and become a lawyer? Basically. No, you need to do an apprenticeship. No, yeah, wait. I'm pretty sure in many states. Apprentices under a lawyer or judge. Some of them, it's just a lawyer. And then some of them, you need to go to law school. Yeah, so in general, it is a regulated industry. Let's read from Brandon Jacoby. Brandon Jacoby. Jacoby from deep, he says. Before we tell you about what Brandon Jacoby said, we've got to tell you about his favorite tool, Figma. Think bigger, build faster.

37:33Figma helps design and development teams build great products together. Let's hear from Brandon Jacoby. What did he say, Jordy? Jacoby says, as someone who had the decision to start a job at Cash App and get a big tech salary or go further into debt to do my fourth year of college financials is a big factor. And, uh, I think it, I think it is, it just feels like it's, it, it does get reduced a lot. And I think there's this interesting thing where like, there's almost like a, there's like a inverted U sort of bell curve thing going on because, uh, if you can get the full ride, then you might want to drop out and just go straight into entrepreneurship because you're cracked.

38:18But if you're not so good that you can go to the Ivy league for free, but you can go to the Ivy league, but you go, but you don't have to pay that it's worth it. But if you do have to pay that it's not worth it. And it's more of like this big matrix than just like, is it good or bad? And how much does it cost? Yeah. The other thing, school is a way to buy time to figure out how you want to spend your time like with with brandon jacoby yeah he was design doing graphic design and like a teenager and so the opportunity to go work at a company like cash app instead of finishing college made a lot of sense yeah but if you don't know how you want to spend your time and how you want to spend your career how you want to start your career then school i think a lot of people are just doing it to kill time so they're not just sitting yeah do you know that there is a job in tech, in venture capital, where the number of, it's extremely prestigious job.

39:15And this year, the number of job openings for that role has doubled. Do you know what I'm talking about? The job I'm thinking of is steward of Sequoia Capital. It's doubling in size. And if It keeps doubling. There could be thousands of positions. As a steward of famed venture capital. Exactly. So it doubled from one steward, Roloff Bota, to two stewards, Alfred Lynn and Pat Grady. Next year, if it doubles again, we could see four stewards, then eight stewards, then 16, then 32, then 64, 128. Eventually, we will need the entire human race. We'll be stewarding Sequoia Capital. Yes, that's exactly what I'm thinking.

40:10Roloff certainly understood compounding. He went on Jack Altman's podcast, Uncapped, and said that there is a lot more talent than really interesting companies to be built, and I think we're spreading a lot of that talent thin right now. Venture is a return-free risk, and Shield Monat says. And with that comment, he's out. This reminded me of Mark Leonard's from Constellation, some of the comments that he made on that investor call a while back, shortly before he took a step back from Constellation. But very different situations. Did you see Rolf Winkler from the Wall Street Journal taking shots at Roloff Bota?

40:59on the timeline. Sarah Guo, friend of the show, is very happy about this. She says, congratulations to my better half, Pat Grady, the new senior steward of Sequoia, along with the wonderful Alfred Lynn. Investing firms must be led by great investors. Sequoia's motto at one time was, we are only as good as our next investment. If that's true, the firm is in good hands. So she is talking about Pat, who has invested in ServiceNow, Zoom, HubSpot, Okta, and Snowflake, and then Harvey Open Evidence and OpenAI in the AI era. Pretty solid portfolio. That's some good stuff. But Rolf Winkler comes in from the top rope and says, Roloff, he led a$300 million round into Bird and a$250 million round into 23andMe.

41:52Both are zeros at this point. Are they not? Are they not? And Sean McGuire fires back. Scott Kapoor fires back as well. Scott Kapoor says, is it too much to ask that WSJ reporters who cover finance actually understand the asset classes they report to report on? Roloff, Bota, and Sequoia are world-class investors who have generated amazing returns for the LPs. VC is not a downside minimization asset class. The other thing is 23andMe did go public. Byrd did go public. Yeah, you don't know how much secondary they solve. not even saying they could have just been selling well bird went public too right yeah yeah it's like figure out it might not have been that much capital incineration but yeah i mean i highly doubt the whole job is incinerating capital baby you got to be burning you got to be making you got to be ripping wild checks every once in a while um it's uh it's totally totally acceptable um sean says sequoia distributed over 50 billion while roloff was running the I was running Sequoia US since 2017.

42:59One of the first things Roloff coached me on when I joined Sequoia was to look at each fund's write-off rate. Any fund with a write-off rate below 40 % wasn't taking enough risk. That's interesting. I feel like Sequoia also has a stat on their page that's like 97 % of the companies are still in business. Don't they have some stat like that? Yeah. So there's something a little bit in the math that I'm struggling to square. Sequoia led one round in Bird at a billion-dollar valuation. They then invested another time at$2 billion. But the company IPO'd at$2.3 billion, which means their effective valuation was somewhere in the middle of$1 and$2.

43:44So at least at the IPO, which of course had traded down, they would have been up on that investment. Are they in Vanta?

43:55sequoia yeah uh yeah of course they are you're so lacking the automate compliance manage risk and accelerate trust with ai vanta helps you get compliant fast and we don't stop there our ai and automation power everything from evidence collection and continuous monitoring to security reviews and vendor risk whether you're starting up or scaling uh I was still running the numbers because they also did a series B at two and a half billion. So they did a few rounds back to back. They were obviously were, were betting big. Yeah. Um, but, uh, I think they, they, they sure they ultimately did. Um, I don't know.

44:33I'll defend, I'll defend Rolf Winkler. I think it's funny that he's taking shots. He's just out there talking trash. What's wrong with that? Uh, Ryan Pearson gets to talk trash. Ryan Pearson, since this is Sequoia passed on Flexport so many times that I don't really care who runs the place, to be honest. And Sean McGuire says, we love you, Ryan. And we're dumbasses all the time, especially me, which is funny. But, you know, people get to have fun either way. Scott Kapoor was firing shots. Yeah. So we read this one. He's firing back and forth. He's defending. So, yeah. I mean, Scott's question is like, is like, should, is like, uh, is like, should the Wall Street Journal reporters understand that, uh, that they take roll-offs, uh, or that they take write-offs, um, and that's maybe not what Rolf is doing.

45:28Like he might just be talking trash. Like he might actually understand it. He might be like, very possible. He might be like, yeah, like, I don't salt in that wound. I'm still going to rub it in. There's nothing about that post that says that Winkler does not understand the game. You could reply to the post and say, to be clear, I understand finance. I'm just throwing salt in there. I'm just talking trash. But there is a risk of legacy media not understanding the asset class. It does happen every once in a while. We run into this a lot. The Financial Times had this sort of breakdown on - It happens all over the place.

46:12And certainly this is not like a great look for the Wall Street Journal because it doesn't read as - He should have just said, that guy sucks. We should - Yeah, he sucks actually. We can do anything. We love the Wall Street Journal. We love Sequoia Capital. If we can do anything to kind of bridge the divide. Bridge the divide. We're here. We're here. Maybe we can get both The Wall Street Journal and Sequoia Capital On graphite.dev Code review for the age of AI Graphite helps teams on GitHub ship higher quality software faster That might be Find some middle ground with graphite Ryan Peterson also followed up and said Clearly should have pivoted to an AI lie detector bot Write it to your video calls And it adds a bigger and bigger Pinocchio To the other participants' faces I mean Don't give Roy Lee any more Stunt ideas I mean this is a banger He's such a good, stunning marketer.

47:07Roy had a great response to some commentary that we had yesterday. He said, explaining the Clue Lee pivot and brief history, nine months ago I built InterviewCoder an undetectable translucent desktop layer that lets you cheat on lead code interviews. That's crazy. That was only nine months ago? Is that real? Yeah. It feels like a lifetime ago. Oh, yeah. I knew it had viral potential, and I posted about it constantly until I did go viral. Then I did it again and again and again, and then it made me$1 million in profit. Holy F. Apparently you can build a GPT wrapper, go very viral and become a millionaire.

47:41Could I do it again, but bigger? So I started clearly with two big questions. Where else could this overlay be useful outside of LeetCode interviews? Can I keep getting this much attention? The second answer came immediately, yes. I'm very good at getting attention. And thanks to the attention, we got hundreds of thousands of users ridiculously quick and were able to use the data to figure out that our stickiest power users were using it in meetings. So we just built for that. The deeper learning, though, is that we seem to have cracked the formula for organic virality. Yeah, one thing I would say is the website copy is framed as an AI note taker, but it's still leveraging that functionality that Tyler is so dependent on or historically has used to, you know, if we put him on the spot, like, Tyler, what's the capital of Nigeria?

48:36Let me think about that for a second. Abuja. There you go. See, that's clearly in action. But he says, no other tech startup has ever been able to do this reliably or at this scale. It's really not luck, and most importantly, if done right, it works to grow companies. interview coder proved it and can you clearly continues to prove it growing faster than ever in a saturated space the world is vast there are tens of millions of people in meetings every day and 90 of them don't know what an ai note taker is the market is big enough that you can be the most controversial person in the world and still win and as long as you deliver the magic moment to the user so um yeah i would not be surprised if they are out marketing all of the uh other companies in the category, even if it is competitive.

49:20But thoughtful response. I have more to talk about here. I wonder this idea of, can I, what do you say? The deeper learning though, is that we seem to have cracked the formula for organic virality. Like, what does that actually mean? Does that mean like his claim here is almost like organic virality is effectively high ROI marketing, low dollars in, high dollars out. It's just great ROI. And to do that

50:04organically, consistently for a long, long time is very, very hard. It's very hard. And the folks that wind up doing it, like when you think about Mr. Beast, it's like, that's his full-time thing. That's all he does. And I wonder, I don't think it's necessary that that in 10 years we could be looking at clearly the um the ai note taker for meetings and yeah they still don't run google ads they just do stunts and people are still can't get enough of these stunts i i just i i have to believe that this is a market entry tool and not a permanent piece of the infrastructure versus like you talk to the ridge wallet guys they're like we are good at buying performance ads, and we have been for a decade, and we will be for the next decade.

50:55I don't know that that's something that you can just be like, yes, let's scale our organic viral content forever. I feel like it caps out, but maybe it doesn't. Yeah, I mean, I think they're going to run that test. Yeah. It does feel like as we tracked the stunts throughout the year, the quality and the effort going into them, and even some of the creativity was not deteriorating. Like I would say the team was executing at the same level or better throughout the year, and yet the impact got less and less and less because - You're saying for us? For Cluely. Oh. For Cluely. What I'm saying is like, Yeah.

51:37Cluely, like the first stunt - Yeah, that's the issue of like things that go viral organically. over time, the effectiveness kind of deteriorates. People do get, like, if you scale organic, like eventually you will saturate that market. Also with rage bait marketing, it's like, once I've been enraged, it's harder to get me enraged the third or the fourth time. Whereas if you show me an ad for a value prop, you can like hit me with that ad again and again and again, and it's probably similarly effective. But if you've gotten, if you were the type of, like, remember there was a moment when everyone was mad about clearly, like truly like it was, the timeline was in turmoil.

52:23Lots of people were chiming in. It was this big wedge issue. You were either for it or against it, but a lot of people were against it. Like the people that were against it, they were rage baited successfully in March and then a little bit more in April and then a little bit less in May. And then more recently, like they just, the rage bait doesn't work on them anymore. Because they're like, oh yeah, that thing that I don't like. Because I made my call on it. But to Roy's point, there's so many people that have never been enraged. They have never been enraged. They have never had the opportunity to be enraged by, clearly.

52:54I think a lot of their marketing is less rage bait. Like on other platforms, when he shares stuff, it seems much less rage bait. Oh, I completely agree with this. Yes. Much less rage bait focus. Yeah. And so, yeah. I don't know. I think a lot of people, if they see funny marketing and like X is just a really small place. Yeah. Like even back in 2021 building Party Round, like the difference in reaction when I would meet somebody that was like extremely on X versus somebody that didn't use X at all. If somebody was on X, they were like, you started Party Round? Like, no way, blah, blah, blah. And they'd like wrap off a bunch of stories of ways that like our marketing had entertained them.

53:34and then if he met somebody that was like an Instagram LinkedIn power user, they'd look at you with a blank space and they'd be like, they'd look at me and be like, party round? Like you have a fintech company called party round? It's just like very different reactions. Yeah, no, totally. It's fascinating. Well, you know who else did some stunt marketing? Julius, the AI data analyst. Connect your data, ask questions in plain English, and get insights in seconds. No coding required. fool me I'm getting like deeply can't get fooled again iconic iconic quote by our former president I'm getting like FOMO over this Tyler Cowen interview because we keep seeing keep seeing transcription segments hit the timeline but I can't listen to it because we're live and this seems good I will listen to this later today I think the way to Sam says I think the way to monetize the world's smartest model is certainly not hotel booking, but you want to do it nonetheless.

54:40We're getting agents that can book you travel, folks. I think the way to monetize the world's smartest model is certainly not hotel working, but you want to do it nonetheless. That's very funny. Anyway, oh, the Substack X relationship is going all over the place. Have you seen this? Yeah. I can't tell what's real It's hard to So Maybe we should just have The most recent post is from Chris He says even correcting for fake views Traffic to Substack links from X is up substantially Full post read, signups, etc. Also track, we're so back So Anyways, I think this is good I thought it was very unfortunate That X and Substack Got in such a fight in such a fight.

55:32It was bad for all the writers on the platform who are some of the best posters on X. Their businesses were impacted by it. I don't really think the two platforms are that competitive. Obviously, Substack does want to be more of a social platform, have a social layer, but I just think X is shockingly durable. I don't know how much of a threat the social product of Subsec is. So yeah, what's interesting is like the, uh, uh, I don't know the, like X had review, I believe was the, it was the email newsletter product and it never, they never really invested in it. They, they ultimately shut it down.

56:19I, and it's interesting that no other, there's something about like the email and, and, and taking the audience with you. That's just so revolting to a true social media company. Like if you're a social media company, it'd be so easy for Instagram to have an email newsletter product or LinkedIn to have an email newsletter product built in, right? And yet they don't because they want control and they see it as counter-positioned. And so for some reason, the folks at Twitter bought Review and were doing email newsletters and then fully pulled back from it because it was just, it just did not make sense for them to keep going.

57:03But I don't know. We'll see.

57:08SoftBank is down 10 % in the last five days. Is that good, John? That was not good. They just reported strong first half net income of 348 billion Japanese yen. The company's revenue reached 3.4 trillion Japanese yen, underscoring its robust market position. Despite impressive earnings, SoftBank faces challenges. What is this writing? SoftBank faces challenges with financial strength indicators such as a low Altman Z score. I've never heard of the Altman Z score. Is that just relationship? It's a financial model that uses five key ratios to predict a company's probability of bankruptcy. Wait, what?

58:01It does have anything to do with Sam Altman? The Altman Z-score, according to Investopedia, is a financial metric used to evaluate the likelihood that a publicly traded company may face bankruptcy. How did we just discover this? This is the fagest thing I've ever heard. This is insane. Wait, Edward Altman? Is he related to Sam Altman? He was born in 1941, professor of finance at NYU. No, the nominative determinism here is so crazy. Send this to the group chat immediately. This is hilarious. He's the brother of Stuart Altman, a noted healthcare economist. Ellen, this is so wild. I love this. The Altman.

58:43Okay, we need to run the Altman Z score on all the MAG-7 to understand. I bet they have very, very low scores. uh is that something chat gpt can do can you try and and have uh you know uh like a gpt5 pro run the altman z score for all of the mag 7 i really want the altman score and then we need to come out with the the altman z score and then the altman x score or something where it's like how how close are they overlapped to with uh with sam specifically or like how big is their contract with Sam right now. Maybe this is the true X and Y axis we can work on. What is your Altman Z score if you're in the MAG-7 versus how closely is your business tied to Sam Altman?

59:33And is there a correlation there? That's funny. Is Sam Altman related to Edward Altman? Okay. We have to do more research on this. This is fascinating. Absolutely, absolutely insane. Um, but sorry, back on the, on the X link thing, uh, Aaron back on the link thing. Huh? Yeah. Aaron P six 13 says, here's a list of domains that X has excluded from using the new in-app link viewer on iOS for, and it's apple.com wayfair.com grok.com, instagram.com, fb.me, tick tock v.com and open.substack.com. And so what does that mean? It's like if you click on those, you won't go to the new in-app link viewer on iOS, and so you need a different website for those or something like that.

1:00:25I'm still confused on what that list, what this actually implies, but Chris Best shares some sort of emoticon that feels like he's not super happy about this. I don't know. Is that why we skipped it? Yeah. Oh, well. somebody was thinking that the sell-off from SoftBank was tied to the interview over the weekend. The pod? Who knows if they're tied at all. SoftBank is still up massively year to date. But apparently Korean equities have been selling off like crazy. There was a limit down. It's terrible. which we hate for the Korean retail army. Don't like that at all. We might have to check in with the Korean TVPN on how they're doing.

1:01:15Have they been going live? We got to check in with them because if they're not going live, they maybe were a bit too long for the chaos. Captain Nemos is absolutely insane to me that one of my friends hasn't bought the Getty House in the Berkeley Hills yet. Listing price is only$5 million. the perfect techno monastery or at least a rationalist AI dev polycule house would be a tragedy if it got in the hands of someone boring I wish we could we need more photos on this house this is a crazy house do you know about this house Tyler? have you been here? no I've heard of is this where your polycule meets up?

1:01:54what? it says can't believe you fools with liquid cash from AI lab tenders let this slip through your fingers sad So I guess someone bought it. Huh. Wait, what news do you have for us, Ty? NFM, they are still, so they did a stream five days ago. And before that, the last stream was a month ago. So I think they're slowing down a little bit. They're also not wearing suits anymore. They're not wearing suits. Oh no. Well, actually, no, no, no, no, no. Actually, I like that. I like that because when we talked to them, we said like, okay, they're gonna start with something that's like clearly an homage to TBPN.

1:02:30They were drinking the tap water. like they got all the inside jokes they were like very much like doing our bit uh and then what we talked to them about was like clearly if they wind up taking it seriously they're going to shift and change and find their own voice and it's like when you learn guitar you know you might play some covers and then eventually you write a real song and so um i was like super thumbs up on you know like put on the tvpn costume but then over time find your own style and so i don't know what their ultimate style and their ultimate product will be. But it was just fun to watch them come in to the live streaming game like a bull in the china shop.

1:03:13If you put The Technological Republic, which is Alex Karp's book, and Abundance, which is Ezra Klein and Derek Thompson's book, Side by Side, says Young Macro, you'll note that the defining bipartisan tendency of the late 2020s is, quote, we need to become a lot more like China. And with this, we vindicate Nick Land as the only serious thinker to have correctly identified Nietzsche's prescience as that of a supply-side reform pundit. In opposition to the normie right who read him as a conservative, the continental canon who read him as a bad word that I'm not going to say. And as the Anglophone left liberal offshoots of the Continental Canon who haven't read him.

1:04:05And so what Young Macro is saying is that both the abundance, who you could think about Ezra Klein, this is the left, this is the New York Times, and Alex Karp and Peter Thiel and the palantir, this is the conservative side, the left and the right are actually unified in saying we need to solve the supply side. We need to reform the supply side. We need to build more. We need to create abundance through capitalism. Young Macro, undoubtedly one of the greatest thinkers and posters of our time. He made sure this post was fact-checked by real landian neo deep status yes thank you for for fact checking good thank you those i love those posts so much the fact check ones but yeah i mean it's like makes total sense this is also a dan wang's take right breakneck it's just like we basically are getting smoked by china get ready to learn chinese more like them get ready to learn chinese buddy get learn get ready to learn state capitalism buddy that's basically what he's saying uh get ready to learn fall the generative media platform for developers.

1:05:11The world's best generative image, video, and audio models all in one place. Develop and fine-tune models with serverless GPUs and on-demand clusters. Interesting post from Sam Altman on Bose BarakTookCS. That's a long name. Thoughts by a non-economist on AI and economics. Tyler, you've been working a little bit on this. Has this entered your research process towards understanding our like the scale of the build out? Yeah I mean a lot of that post I quickly read through it and fully read through the whole thing but a lot of it is just about like okay we have you know straight lines on logarithmic like graphs.

1:05:56If you want to read the whole thing we can get subway surfers for you set up. I need a Minecraft parkour and subway surfers together. You know what I want? I want to see the straight lines, previously straight lines on log graphs. So show me, there was a period where iPhone adoption or smartphone adoption looked like a straight line on a log graph, correct? Sure. There was a time when miles of railroad construction probably looked like straight lines on a log graph. And then what happened? I want to know what happened after it was a straight line on a log graph. But I mean, there's never, I can never have too much intelligence.

1:06:37I want, I always want, I'm always, I always need more. You're insatiable for slop. Yeah. Like, like, like, you'll never, there'll never be enough slop. Well, I guess energy is a bad example. Yeah. But that's, I think. Because what happened with that? What happened with energy? But that's regulation. It's not that demand really was so bad. I think it's most of that is just regulation. Yeah. And we won't regulate this time, right? No chance. No chance. People are going to be like, I'm getting to the voting booth and saying like more AI. for sure that's what's gonna happen yeah they're not gonna they're definitely not gonna do what they did with nuclear this time it's gonna be different it'll be different we won't have we won't have state-by-state law yeah yeah we're not we're definitely not gonna have state-by-state laws why would we do that that would make sense they're gonna they're gonna they're gonna make it illegal for claude to make mistakes no i so the marcia blackburn thing is amazing nominee determinism that your name is marcia and you go to the swamp and you just dominate uh i love that but also uh hilarious because it's the ultimate AGI test.

1:07:34Because if you're open AI and you have all these draconian state-by-state legislations, like on the Brad Gerstner BG2 pod, Sam says, I don't even know how I can comply with this. It's like, get ready to use some super intelligence to figure out how to comply, right? If you're truly AGI-pilled, you should just be able to say, Hey, Codex, fork my entire organization 50 times, rewrite the code, retrain the models for each state, and create 50 new LLCs, 50 new C-Corps, 50 new nonprofits that are all 150th the size so that we can actually perfectly, perfectly comply with state-by-state regulation. Problem solved.

1:08:18It's one prompt. Yeah, we just need to get, we need the next model to do that. Yes, the next model. Yeah. Yeah, the next one's going to do it. So once we get that one, then it'll solve everything. Then we're good. We need 50 runes. We need 49 more runes because we already have one. But he can only be the California rune. We got a shout out on the Tyler Cowen podcast. I did. I love that. Both guests. Actually, all three of them were guests. Okay, we have some breaking news. Breaking news? Joe Weisenthal. And I just actually want to go through Joe's profile after this if you want to pull it up. Because there's so many good insights.

1:08:50One of the top posters of our time, of course. But the CFO of OpenAI says people are not exuberant enough about AI. Bloomberg has an article. OpenAI Chief Financial Officer Sarah Fryer suggested the market is overly focused on anxiety about a possible bubble in the artificial intelligence sector and should muster more exuberance about the technology's potential. I don't think there's enough exuberance about AI when I think about the actual practical implications and what it can do for individuals, Friar said in an onstage interview at the Wall Street Journal's tech live conference in California on Wednesday, we should keep running at it.

1:09:28I got some exuberance for you right here. AI! It's coming. Artificial intelligence. It's coming. Yeah. It's coming. Palantir at 600 times earnings at a 600XPE is not exuberant enough for me. I want to see four figures. Four figures. But anyways, this is trying to think if there's any other.

1:10:04What else on Joe Weisenthal's profile do you want to talk about? Oh, there's a couple others. I'll just scan through. Joe Weisenthal says, not surprised that Bitcoin has fallen so much lately. I've been talking about a Bitcoin bubble for over 10 years. Good point. Good point. I mean, that's the bull case for Palantir. There's another one here. Traders think Trump will lose the tariff case at the Supreme Court. So there's been some developments on that front. And today, there was a New York City office landlord, Vornado Realty Trust, is down 3 % today in reaction to Mom Donnie. And there was that other post that I wanted to cover from Joe.

1:10:58He was frustrated because he was asking ChatGPT to do something. and it was just basically completely rebelling. Oh, yeah, it was being lazy. That was funny. It's not feasible. It's not feasible. Chat Chippity is basically saying like, sorry, I understand your request, but it's not feasible. And he's like, why is it not feasible? Good question. And you're absolutely right to press on that. It is feasible, just tedious to do manually. It's just like... I love lazy AI. It's just like, I don't want to do that. And so here's the thing. Here's the thing. When I get these instances of laziness and I am giving them the maximum amount of money that they allow me to give them on a monthly basis, I want to churn.

1:11:48Sure. Because I'm like, what's the point of paying you$2 ,400 a year for you to not do things that you're capable of doing? Obviously, there's a good reason for that. Rocco's Basilisk. You've got to be nice to the AI because eventually it's going to be in charge. And if you're bossing it around now. I've tried threatening to churn. That's not going to work out for you well. It has not worked so far. It's not working now, and it's not going to work in the future. I thought it might work. I literally said recently in a chat, if you do not do this. Don't do this. You need to apologize. I will. If you do not do this, I will churn immediately.

1:12:23You need to apologize. It completely called my bluff. That was rude. It called your bluff. Joe Lonsdale replied to Joe Weisenthal and said, LOL, OpenAI working on its margins. Okay, nerds, do you realize how much all this energy costs? We need the chatbots to be lazier. That's funny. Anyway. In other news, Bitcoin has now performed worse than U.S. Treasuries in 2025. Oh, yeah. Yeah, this Joe Weisenthal post about the laziest OpenAI thing. uh joe lonsdale actually quoted it and said lol open ai working on its margins okay nerds you do you realize how much all this energy costs so funny um yeah so you said uh you wanted to move on to bitcoin is that right no i just i just i just covered that bitcoin just kind of like like round tripped right uh suspended cap went pretty hard on pretty much every company in the mag seven he said meta s-h-i-t-t-y frontier model no cloud share culture is f'd microsoft sensible approach to azure build out what do you wait so on meta what what do you think the probability is that they launch a gpu cloud in the next five years or two years even or like a cloud like like aws so we have aws gcp and azure but then anthropic and op open ai are also in the in the like the cloud business in terms of like selling inference, selling their models.

1:13:55Like Meta has all this infrastructure. They've never done cloud hosting, but it just seems so logical that they could get into that market. I think. Even NVIDIA has been talking about like, oh, we're going to build like an inference like cloud, like our DGX cloud. Seems greater than 50 % chance. Yeah. I put it at, I personally put it at like. Do you mean like selling infants or like renting like the actual like. Yes, yes. So, so they have a ton of, they have a ton of GPUs, right? They might have slack capacity in the future. They might get caught in a glut. They also have a good Lama, a good model, Lama 4.

1:14:39They might have a better model soon, Lama 5 or whatever they wind up building. And they just, they have the ability to serve all sorts of stuff. because they have a lot of infrastructure. Yeah, I think that's very high, right? Because it seems like they're not gonna do open source. So it's like, why are they making the model? How high is it? Because they've literally never done enterprise, never done B2B. It's like a completely different motion for the company. It's highly competitive with Google, Amazon, and Microsoft. It would be a huge departure from their core business. I don't know, I can see it not happening.

1:15:14Anyway, do you wanna run through the rest of this? uh yeah pretty pretty pretty aggressive takes here but entertaining uh but they do think apple is going to look look smart for sitting this one out um google is going to be the biggest company in the world by the end of 2026 in suspended capitals uh opinion so um good luck to them in other news Pinterest is down 22 % 22 % is where it landed they had an earnings miss and a weak forecast market did not like it I wonder kind of tough right now you can beat you can beat and your stock will go down and you miss and your stock goes down Pinterest is a 17 billion lose dollar company I wonder how it fares next to something like Reddit.

1:16:08So Reddit's twice the size market cap wise. And I have to wonder, but the PE ratio, Pinterest is at a 10 and Reddit is at 111. How does this make any sense? What's going on? Well, Reddit is like the AI data broker company and Pinterest is like kind of just getting slopped up. Like if you've actually gone on Pinterest, it's a great place for inspiration. Reddit is headed for the same future. I don't know. For some reason, Reddit's been able to hold on. I don't know. But very different narratives. Would be interesting to dig in. Would be interesting to think about where Pinterest goes in a post-AI future.

1:16:47The product, I would use it to go and find inspiration for brand design and colorways and different stuff like that. And the AI slop was infiltrating pretty quickly. And it was very frustrating. because what percentage of new reddit like written content is ai i have no idea because it's possible it's extremely high but it's siloed like i want to believe that it's just like the same question as like x like on x it should be the easiest thing to ai right 140 characters 280 characters like it's super easy and yet what percentage of the posts that you actually read and interact with do you think are AI?

1:17:30Like 1%, right? I don't know. On X now? Yeah, on X. It feels like every third post that I read is written by... I don't feel that way at all. I feel like maybe it's because I feel like it's more like my following tab, but I feel like I'm seeing like a Joe Weisenthal post. I know he's not using AI. Then I see a Joe Lonsdale post. I know he's not using AI. Then I see Brad Gerstner post. He's not using AI. Then I see Tyler. I watch Tyler post. He doesn't use AI for his posts. I see a you post. Like when I'm scrolling through the timeline, I'm seeing people that are just not using AI for whatever reason.

1:18:08It's not, it's like, it's just not, that's not the point. And that's not like the whole structure. And so he, I don't know. I would imagine that certain subreddits are tight enough where it's very easy to clock if they've been like taken over by AI. There would be value to that, but I don't know. Anyway, without further ado, our first guest of the show today.

1:18:36Cliff, what's happening? Good to see you. Welcome to the show. What is happening? Are you in Australia? I'm doing good. I'm doing good. Can you hear me all right? Yeah, yeah, we can hear you now. John asked if you were in Australia. It might be a silly question. I am in Australia, yes. Got it. How you do? Yeah, give us a super excited to have you on the show. I know you have a bunch of product updates to talk about today, but catch us up to speed on the latest at Canva, the shape of the business today, all that good stuff. Because I feel like you guys are maybe a bit quieter on our corner of the internet, but every month or so people realize how much revenue you guys are doing, and it's always pretty shocking numbers.

1:19:28So we're super excited to chat. Yeah, we're in Sydney, Australia. We like to be quiet achievers over here. Business has been going well, really well, actually. We are about 260 million monthly active users. We'll close the year close to 4 billion in revenue. 4 billion in revenue. Well, close to, close to. Close to, close to. It's a profitable company for eight years now, growing high 30s. Business keeps trucking along. We've got an amazing team building an amazing product for our amazing customers. And we just focus on delivering user value and really feel by building the best product possible and delivering as much customer value as possible.

1:20:13The good things like the users and the revenue comes quite naturally. So we've been investing hard, obviously, last couple of years. I think the world shifted under every established technology company with AI. We've had to kind of figure out what does that mean in this new world. And where we've landed is it's really just reinforced our mission to empower the world to design, to create anything, design anything and publish anywhere. And that's evolved. And now it's easier and better than ever. And we're owning more and more of the workflows, helping organizations create on-brand content on-brand content at scale, helping them deploy, understand how that content's performing once it's deployed, and really facilitating that feedback.

1:20:56So yeah, it's been fun times. Internally, walk us through your thought process around the opportunities or even competitive threats around Gen AI over the last few years. Because from my point of view, it feels like gen ai is is uh uh it's just an accelerant is is like could be an extremely meaningful like um accelerant to your guys's business it just makes it easier to create cool things uh on your computer feels very aligned to the core business but how have you guys thought about it over time did you feel like you were caught off guard or were you experimenting early like i'm curious to know what what the development cycle has been like?

1:21:37We weren't caught off guard. We kind of knew this was coming and we were working on AI design in a more structured programmatic ML way for like a long time before the image models came out. The speed at which those image models came out and got really, really good, that's what kind of shocked us and with new methodologies. That led us to acquiring a company called Leonardo. We really needed to infuse the AI DNA into the company and we really needed to accelerate our sort of roadmap in that area. And so we acquired an AI company. We actually acquired an AI company for background removal about five years ago.

1:22:16So we very much knew it was coming. It was just the speed at which it hit and scaled was a bit of a surprise to us. We think we caught the wave well. It really, as you said, it's an accelerant for what we can do. So we create videos, we create social graphics. We recently just launched the world's first design model that actually generates layered designs that you can edit. And then when things like you're creating a video, we set out the timeline and the storyboards and you can generate AI clips within that. So you can essentially prompt, create, upload images. And actually, all the great video models and whatnot that are coming out now really help our customers achieve their goals better than ever before.

1:22:55So we have kind of two-prong approach. We work with the world's best models and world's best AR companies. And then where we've got a strategic data advantage and we've got strategic insights, like when it comes to design, that's where we build our own models and build our own foundational models. What about on the product side? There's this interesting dynamic in, I feel like, most design pieces of software where they start as discrete applications and they all kind of bleed together. Like eventually, I think you can technically edit video in Photoshop and you can design a thumbnail for YouTube video in After Effects.

1:23:30And like the entire Adobe suite's kind of bled together where they can all do the same. There's like overlapping features. And I'm wondering about how you think about creating content with AI. Is that just something you need to stuff into all the existing applications? Or do you need a new application that can be like AI first? And then maybe you add the features to the other applications. but do we need like a specific new application for generative AI? So I feel like this is almost a planted question because this is something we think about a lot. And we've actually... It wasn't planted for the record.

1:24:07We never prepared for interviews. Our audience knows. I don't even get any. Normally I get prep questions or something for these things. We do it live. We do it live. Freestyling. um but yeah no so the platform we've built over the last 10 years is actually perfect for ai so when you think about it we've built a unified document model so whether you're creating a presentation a social media graphic a long form text document a website a video all of these documents have been built off the same engine right with simultaneous collaboration and everything across all of them then at the found so that's the dock type that's what we call our visual suite up the top.

1:24:48At the bottom, we have our infrastructure layer. So that is things like our digital asset management system. It's your brand kit. It's all the core engine that allows you to share and publish to any platform. Okay, so that's what we had. In the middle, we've put this AI layer. So when you want to create a design, let's say I want to create a QBR report, like a big quarterly business review for a customer using this data and their brand. It calls on your data from Salesforce or whatever. It calls on that company's brand kit. It calls on your company's brand kit. It then understands what you're trying to achieve.

1:25:28It then generates it as a presentation and then allows you to deploy that and understand, is it getting clicked through, blah, blah, blah, blah, blah. So it's actually connected this entire ecosystem that we call the creative operating system, which has sort of all come to fruition and it's really leveraging that prior 10 years of investment to make Canva an incredibly powerful design platform. A lot of the other companies, they've built these bespoke products so they've built a Gen.AI image generator, they've built a Gen.AI this or that, they've built a presentation tool or a tool to create vector graphics.

1:26:03At Canva, it's all an integrated creative OS which is really, I don't know if we planned it that well on purpose but it's just, it's kind of, we can't dig. Yeah, it's also, it's nice if you get a little lucky sometimes. Where do you see the end of the map or the edge of the map of like what you want to do creatively in Canva? I'm just thinking about like, if you get really, really crazy into, you know, media creation, you can wind up in a 3D modeling software. And like Adobe has substance and they have that app dimension, but you can go into Houdini or, you know, Maxon products like Cinema 4D or Blender.

1:26:47And it feels like with the AI, with the generative AI boom, you might not ever need to actually go that deep down the stack or go into like an After Effects competitor, motion graphics, video tracking. There's so many things that might never be exposed to the user because you're building at a time when the abstraction layer is now rising thanks to generative AI. But how do you think about the long-term functionality, like the stuff that's the furthest out on the prosumer, professional, enterprise-level creator frontier? Yeah, another really good question. Another, I feel like, planted question, but not planted.

1:27:30um so you want to abstract regular users knowledge workers etc etc from all of that complexity as much as possible if you want to create a 3d graphic we just launched a 3d product just last week um so you can now generate with ai 3d products in canva edit them rotate them do all those things so um you can do that as a regular knowledge worker with no design experience we also 18 months ago acquired a company called Affinity. They are a professional design suite that has a vector graphic tool, a layout tool, and a photo editing tool. We just relaunched that last week as well. That has a million downloads within the first week.

1:28:10So, and it's a hundred percent free for everyone. Yeah. That's awesome. And so that's really the craft side. So if you want to get super detailed as a professional creative and create a logo or a vector graphic, or get really down to the pixel manipulation side of things. That's what Affinity is absolutely incredible for. And then it seamlessly integrates to Canva. So you can push that vector graphic into Canva for the rest of your knowledge workers to use within the organization. So you can create templates, you can create them in Canva, or you can create them in Affinity, push them into Canva, and then scale them throughout your organization.

1:28:42So these two things work hand in hand. And you really want the power of those tools if you want them. But most people that aren't professional designers don't want that power and they don't want the complexity that comes with that foul power. What is sentiment like within the community at Canva around AI broadly? I'm curious if certain types of users are much more optimistic and excited about new AI features versus maybe like traditional graphic designers, but what's the general feeling around the technology today? I think people, when they wake up in the morning, they don't wake up and say, I want to create a social media ad.

1:29:28Or they don't wake up and say, I want to create a presentation. Sometimes I wake up and I think that, just to be clear. Not an ad. But people create an ad because they want to grow their business. People create a presentation because they're a startup wanting to land that pitch deck and get investment funding. So they wake up in the morning with a goal. And what we're evolving our business to be is a company that helps people achieve their goals, not just create a design. And AI has accelerated our ability to do that like a million times. Yeah, that makes sense. You're the COO. Give me some best practices for acquiring companies.

1:30:11post-merger integration, what makes it successful? What are you watching out for? What are the pitfalls? What are the nightmares that you've heard anonymously from your mentors about what can go wrong when a large scaled software company like yours acquires another software company and the integration doesn't go quite right? What are you looking for? so we've acquired 10 companies now about 10 companies um of various scale i think it comes down to the founders and i've done my last three acquisitions on the balcony of uh my favorite pub and so i honestly like on the balcony so it's like if this doesn't go down you're going over you're one of those guys you're kind of like old patty down at the pub is going to throw you off break your legs if you don't do the deal is that what's going on?

1:31:11I mean that's yeah kind of we're not threatening anyone but like you've got to get the cut of the jib of the founder you're acquiring what are their motivations, why are they in it does 1 plus 1 equal 10 so the companies that we find best to acquire. They've got an amazing piece of technology, they've got an amazing founder, and that founder wants to get the power of what they've built in way more hands than their current distribution channel. So if we can plug the power of their technology and team into our distribution channel, one plus one equals 10, and you can see if they get a glint in their eyes when you talk about that.

1:31:47And it's that glint in their eyes and just deeply understanding their motivations that is going to make it successful or not successful. If they're just after a quick exit, But that could be fine if you're just after technology or just after their customer base. But if you don't deeply understand the founder, and that's why I don't have an M &A team. Like we do have an M &A team, but we don't have a head of M &A because I feel when you're acquiring companies, it is so deeply personal. And the person that's going to sponsor the success of that company once it joins your company needs to be involved in that process.

1:32:20And they need to be the ones making the decision. And when that's delegated down, I feel that leads to a lot of carnage. And most of the founders that we've acquired and some we've acquired like eight years ago are still with us today and are in leadership positions in the company. So I think that's sort of testament to the success of a lot of the acquisitions we've done. It's very cool. What's the chat wants to know the biggest fish you've ever caught? Oh, yeah, I love fishing shit. Some big ones. Oh, you should see this thing. I've got this whale above this as well. I've got all this stuff in this one shop.

1:32:58Look at that. Wow. Do you have any violet crumble on you right now? No, that's not the best Australian chocolate. It's not? Oh, I always think of it as the pinnacle. It's my favorite chocolate, and it just happens to come from Australia. Amazing. Well, chat loves you. It's great to finally meet, and congratulations on all the launches. The scale at which you guys are operating is really incredible, and it's super fun to chat. Yeah. We'll talk to you soon. Appreciate it, guys. Have a good rest of your day. Cheers, Cliff. Before we bring in our next guest, let me tell you about Turbo Puffer, Search Every Byte, Serverless Vector, and Full Text Search, built from first principles on object storage, fast, 10x cheaper, and extremely scalable.

1:33:41Our next guest is Jerry Murdoch, the co-founder of Insight Partners. He served as managing director until April of 2011, led investment strategy and developed many of the firm's portfolio investment. Jerry, are you in the restroom? Wait a room. Are you in the TV and Ultron? Welcome to the show. How are you doing, Jerry? What's happening? I'm really good. Thank you so much for joining. We're super, super excited to meet and chat. Why don't you give a quick introduction? What is venture capital? What is growth equity? What is growth equity? What is investing? Yeah, I actually love to know a little bit about your journey, how you wound up co-founding Insight Partners.

1:34:20It's obviously a firm everyone knows, but I'm not sure everyone knows exactly the earlier story here. Early story? Well, because I'm old, that's why. But where do you want to start? Maybe the moment you... When did you know you wanted to get into asset management? Yeah, exactly. I'd love to know just the minutes before you sign the incorporation documents. That's always interesting. Okay. So Jeff Horian and I were, Jeff was an associate at a big venture capital firm called Warburg Fincus. Oh, yeah. He didn't even have an office. He had a desk under the stairs. I was a consultant, and we had this crazy idea to start a company called Open Vision, and we wrote a business plan.

1:35:08And the partner, Bill Janeway, said, well, if you can recruit the president of Oracle, I'll fund it. and at the time the president of oracle was a guy named mike fields that worked for larry and uh and mike said well i think larry's gonna fire me so i'm open to it and then a month later larry fired him and he became our ceo and open vision merged with veritas and a 60 40 merger and was worth 30 billion in 1999 and wow that funded jeff and i into insight that's got you into business. Why did he get fired? Yeah. What's the story there? Is he coming for Larry's next? You're good at Larry. He'll tell you why he fired him, but Larry has his own reasons, you know?

1:35:51Yeah. And then take us through a little bit of the journey on the early Insight Partners days. Sure. Yeah, I just want to hear your craziest stories from the 90s. You have some crazy stories yourself. I mean, what a wild decade. What was the timing overlap between the IPO you mentioned and founding Insight Partners? Yeah, I think so. Insight, if you look at the website, we say January 1, 1995. In truth, we shook hands in August of 94. We funded the very first company December 30th, 94. So we said, okay, January 195. But we didn't close the first fund until July of 1996. We were investing friends and family money in what became the first part of the portfolio.

1:36:55And we had an advisory board because we didn't have any credibility, right? Jeff and I both had acne. He was 30. I was 35. We didn't really know much about what we were doing other than we got lucky with open vision. But I lived in Aspen and still have. I still live in Aspen. And I was involved in the Aspen Institute. And I invited a bunch of people to join me in a little roundtable. And they became my advisory board. So the advisory board for insight was eric schmidt who was a vp at sun a guy named scott cook who was the founder of intuit a guy named ray lane who replaced mike fields at oracle as the president um and uh those and a few other people but those guys really gave us their money they gave us their time gave us credibility and they helped us succeed what did they they you said you were didn't have experience or credibility back then, but clearly to get those kind of people to spend time with you, trust you with their capital, what made you guys special?

1:38:01What did they see in you? I think, like, look at you two guys. You get everybody on your TV show, and why are you getting them, right? You guys aren't on 60 Minutes or somewhere else. You guys got personality. You got brains. You got aggressiveness. You got a gong. That attracts people. Yeah. Okay? Come for the gong. What's current AUM? Can you share any AUM figures? I'd love to ring the gong. Somewhere around 100. So right now it's 110, I think. There we go. Congratulations. Not every day we get to hit the gong for 110 billion of AUM. I mean, we're super in the weeds tracking how venture funds today are positioning themselves in the AI boom.

1:38:46folks will say, I'm only betting on one foundation model company, but I love the infrastructure layer. I'm doing some energy investments. And then I think there's a lot of opportunity in the application layer, for example. How were people in venture talking about the dot-com boom or like the internet boom? It was obviously now we collapse it just down to internet or dot-com. But back then, there had to be dividing lines between themes. And how was it being discussed amongst venture capitalists at the time? Okay, so let's take Insight. Insight, we thought we were really genius and said, we're not going to invest in dot-com companies.

1:39:28We're going to do infrastructure and applications, but we're going to avoid the dot-com bubble because it's crazy, right? I mean, I sit there and said, look, there's not enough. Everybody's on dial-up. You're not going to do commerce on dial-up. We don't have enough broadband for this to work out. And so we avoided it thinking we were going to be safe. And we did all this infrastructure deals. Well, guess what? March 2000, the stock market dropped 40 % for tech. And we all got killed. Everybody got killed. And so our safe strategy was a total failure. Everybody went down. And then in 2001 with 9-11, we got trashed.

1:40:09And so we had the worst year in venture capital history. I think our 99 fund returned a whopping 1%, maybe. I think we were top quartile or basically a break-even fund in 99. So break-even was top quartile. Yeah, that's right. That's right. That's insane. So were you guys feeling, it being top quartile, were you like, okay, that was rough, but we made it through. we still look better than most of the industry, so we're going to be able to continue to build momentum? Or was there ever a point where you thought it was over? No, we just were cheap, basically. We just wanted to get every dime we could possibly get back.

1:40:52So we had to work hard on three or four investments from 2001 to 2005 just to get back to break-even. That was like rolling up your sleeves in and like getting operational with the companies to make sure that yeah yeah we're yeah helping helping build a team and and change the strategy for a new world the one thing we learned from that lesson is that when the boom ends and the crash begins is that all those companies on the old platforms the old technology are not going to be very attractive with the next boom and the next thing happens right so think about web companies in 2008 they were before that crash you know they were looking great and then mobile companies became what was important after 2010 yeah right so if you weren't mobile you weren't very attractive and i think we're going to have the same thing here when this next cycle goes down there's three major trends right we only talk about ai today but crypto i think is becoming more and more important and eventually not sure if it's five years or 10, but quantum computing will become a major platform.

1:42:03Absolutely. And so it's just, so the question becomes is when does the crash happen? And then how long does it last? And then when you come out of it, what are the new technologies that all the companies have to be built on? What was your, what was your state of mind in let's say 1998, 1999, and then those first few months of 2000? Like, did you, what kind of conversations were you having as the deals and the IPOs got crazier and crazier? Was there a sense at any point in your, even, even, you know, friend group and network and advisors that there was another, was it, was, was it late 1999 and people were like, we got at least three to five more years of this?

1:42:48Or, or was there, was there any type of sense or real concern about how? I think it was early 99 or late 98, Jeff Horne and I were walking down Fifth Avenue with two advisors, Bob Rubin, who was former Secretary of the Treasury, and Steve Friedman, who was former CEO with Bob at Goldman Sachs, going to Jack Walsh's office to sit down and listen to him, you know? And both Bob and Steve said, you know, the banks are not accounting for risk. that was pretty big number they knew they felt that it was not they didn't know where the risk was coming from we didn't know exactly which area was it subprime was it real estate was it corporate bonds was it swaps you know there wasn't obvious which one of those were the greatest risk because all of them were risky and so we were walking to listen to him and and uh on the way down there, a guy called me and said, yeah, it was, I think it was an analyst of Morgan Stanley.

1:43:54I forget his name, but I said, look, I said, but we're in the bubble. And he says, yeah, but it's an iron bubble. So I thought, okay, an iron bubble. I got to process this. What does iron bubble mean? Right. And, uh, and bottom line, what he meant was it's going to be a iron bubble until it's not. And, and if you're a believer in making money in booms, you have to stay in the game to the very end. You can be smart and hedge your bets. But if you bail out too early, you're going to miss a lot of great returns. And so that's what that guy meant by that iron bubble that it was, you know, it was just going to be there for a while.

1:44:32And sure enough, it was an iron bubble for over a year. In March of 2000, inexplicably, the stock market shifted, people moved out of tech and 40 % would drop everywhere across the board. And that was like, okay, now we've got to work hard. How are you thinking about recycling or managing LPs in that year where you are top quartile, but you only return 1 %? Are you set up as a fund to kind of, I mean, it seemed like every company, even if it sold off a ton, they were IPO-ing, they were getting out. So were you able to get some liquidity and then reinvest that? Or did Do you have to go back to LPs and raise new funds and kind of restart the whole process?

1:45:17Or were they already kind of bought into giving you another run in 2003, for example, 2002? Yeah. I mean, no. Nobody was bought into anything. It was a pretty nasty thing that was going down. And so we just had to slug it out with the funds we had, which got us through until we were able to raise the next fund, which, of course, was a smaller fund. We had raised fund four just about the time of the crash. So we actually had capital, but it was still ugly because we had a bunch of investments that we knew were not going to be great going down the road. and so we didn't really raise fund five until four years later so it was the four toughest year in venture capital and a lot of our people who were started in 95 didn't make it right so i mean if you look at who started 95 benchmark did and they survived and they did great but a lot of other funds didn't yeah remarkable how would you know if you were still running a platform venture firm today, like how would you be currently assessing the market?

1:46:33Because there's a hot debate right now of like where we're really seeing a bubble, right? If you look at, obviously there's plenty of companies in the public markets that don't make any sense. There's some, you know, there's hyperscalers that don't, that actually seem priced pretty reasonably uh and then in the private markets you have you know if you're an ai company you know you're you're gonna you know probably with a solid team pretty easily get a 100x 200x revenue even if you have little to no margin on that revenue yeah i mean look we don't know what's gonna what's gonna be the straw on top of the camel's back that ends it we don't know when and there's all kinds of political things that you don't want to bet against, right?

1:47:23I mean, we could end up with zero interest rates in March, you know, and you don't want to bet against that. So, I mean, a lot of people who bet against the Fed, if you remember that five years ago, or during the COVID times, lost out huge. So, look, anything rational, like I could say, oh, the laws of physics still apply. You know, I'm on the board of the Santa Fe Institute, where we've got some of the smartest minds in physics and math in the world, irrational things like bubbles can last a long, long time. And you can say, oh, it doesn't look like a bubble. Well, I hate to say it, but I think a lot of these hyperscalers are going to regret the CapEx commitments unless they can get out of it.

1:48:04We'll see. I think what's interesting to me is I think they may be underestimating the sort of drop in token price. I mean, I really think that the amazing power of the chips and things that Jensen has done is going to outstrip our ability to get electricity to really power this stuff. Yeah, that's what Satya has been saying this for over a year now. I'm more power constrained. I think if you look at the amount of chips he sold and look at it out there, I don't think there's a power out available today. They have to find a new source in order to be able to sort of benefit from all this build out.

1:48:49And if there is an economic recession, then the enterprise, the expected enterprise explosion of AI usage will be delayed. right and if there's any kind of a sort of major security problem around ai that'll scare off the enterprises that'll also delay so i don't think you can i don't think you can continue the ai boom without fortune 5000 getting on board it's not going to happen without that and so if there is an economic interruption a serious one then the demand is going to is going to change that has to happen. How do you think about different opportunities to buy assets by whole companies throughout a cycle?

1:49:37There's the companies that are, or the firms that are kind of waiting for a crash and maybe going to buy up stuff as after post crash. There's also firms that are going and buying Fifth Avenue and they want to get the best asset right now. There's other folks who are purely founder bet, pay any price, others much more quantitative. How have you thought about your overall investing thesis throughout tumultuous times? Well, the difference here is AI is affecting everything. Bezos is right. It's an industrial bubble, not necessarily a financial bubble, because the value of AI is not going to go away.

1:50:19That is absolutely the future, and it's here now. It just isn't here at scale yet, in my opinion, at least not for enterprise. So I do think that the problem with buying up companies is, what are you buying? Is it going to be valuable down the road or not valuable? And it depends a lot on the quality of the management. It depends a lot on the technology. And so buying SaaS companies, hoping they're going to be more valuable in five years, I think that's a questionable strategy. I think you have to think about, like I said, from each – if you assume that there must be some disruption, like if it was a COVID-type disruption where it's literally a few months, and the Fed bails you out, then people just keep going.

1:51:08The bubble bounces back and it just keeps rocking. If, on the other hand, it's a serious economic issue. Right now, M2 money supply is all-time highs. So you've got massive run of capital out there. It's not a liquidity crunch. It's just going to be a challenge with the overall economy. And then there will be a liquidity crunch because the debt that's out there is massive everywhere. So I think here in this particular case, if you're going to strategize to buy up a bunch of things after a bus, that's a bad strategy. Most of the companies are going to be first generation AI and they're not going to be the right technology platform.

1:51:52Right. I mean, first of all, most people are betting on these large language models. and what we learned at Santa Fe, we had in middle of March, we had the top research scientists from all the hyperscalers, the 40-person conference that I hosted with the scientists. And the thing that came out of it is everyone's too focused on LLMs, not focused enough on the complexity between the humans training them and the humans consuming them. And I think the secondary part of that is that LLMs are, we're going to find a new technology and cheaper forms of models that are going to come out. I'm certain that what China did with DeepSeek, a release in open source, there'll be many, many more open source models in a year.

1:52:37And you won't need to spend all this money on an LLM. I have a feeling that it'll be a multi-model world with lots of free open source models going down the road. So if you buy a company that's based on foundational models, technology, it may not be so valuable in the next wave of AI. Can you bridge this to trade for me? I heard an interesting stat last night that America has more data centers than the rest of the world combined. And it feels like America might actually be winning in terms of industrial power on the data center side, but lacking in many other industrial capacities. How do you think the way that AI and the AI infrastructure build out is changing America?

1:53:23How do you think that flows to trade policy? I'm not sure we understand or anybody understands American trade policy right now. It's like a kangaroo. It's jumping up and down constantly. Keeps you on your toes. Yeah, I don't know anyone really truly understands it, and I'm not sure policy is the right word. I think it's an innovative game that's being played with trade policy. So I don't know what that's going to look like in four years. There's a lot of uncertainty, and that's part of the problem, making these long-term investments. I think that the data centers themselves, the actual physical properties, are important and will be important long term.

1:54:14The question is, if Jensen keeps his word and the chips are going up by some multiplier every year, how valuable is the chips in the data centers being built? The depreciation is just going to be crazy. Yeah, yeah. And so, look, I think the one thing that's for sure about this bubble that's different than anything else is that the innovation is real and the innovation is profound. And I think that innovation and what it's going to unlock is challenging. And you have geniuses like Elon Musk and others out there that are, how can I say, they're inventing so much so fast. And they have so many resources.

1:54:59If you think about it, the CEOs of the Magnificent Seven and Sam have a lot of influence on how the world is going to play out over the next three or four years. You could add a couple of chip guys. You could have Hawk and Hawkton and CC Wei. And you could add maybe there's three women that are up and comers, maybe Fei Fei and Mira and Lynn at Fireworks that could become more influential. but right now the assets and the power is in the magnificent seven and sam altman and those guys are going to have a an amazing impact on how the world rolls out in the next five years three to five years so i mean when you think about that those guys are going to shape the politics of what's actually happening in fact i kind of think that we're in for a different world where right now we're worried about immigration and we're worried about all kinds of you know social issues i think in the future when robots are walking around we're going to be more interested in like wow who has who are the ai haves and who are the ai have nots and i think that better or worse the magnificent seven and eight are going to change the political dialogue in this country pretty dramatically within three years and uh and i think they know it which is why there's some pretty good ideas that I've heard about giving every child, you know,$1 ,000 in the stock market, something like that, that could be a great idea.

1:56:27That could be a great idea. I heard Jensen and Brad talking about it. That could be very helpful because there will be a have and have not world in America. And that'll be the primary dialogue, I think, of the future, not the stuff we're going through right now. How have you processed the boom in private credit? it's going on a long time and i don't think there's enough uh it's so big it's so profound i think the banking system in america you know because of the rules they're not really accounting for that risk right i mean in the normal way you would think about it so i mean it depends on the disruption right what happened in 2008 or what happened in in in 2001 that would be challenging for private credit.

1:57:16Big time.

1:57:22Where do you put that on

1:57:28in terms of how the probability that private credit plays a significant role in the next financial crisis? Well, put it this way. The banks have made these investments, and right now, private credit looks reasonably safe, right? I mean, people seem to have enough capability to overcome it. The question becomes is how deep and how bad is the future liquidity crisis and how challenging is it? How much? And we don't know. So I'm not betting that there's going to be another banking crisis. I'm just betting that there's going to be an end to the bubble and there's going to be some kind of recession, whether it's short term or long.

1:58:15Whether it's a crisis or not depends on the nature and the severity of it. Right. Right now, it's always a break of trust. Right. I mean, what happened in 2008, people had to take the tarp loans because trust was essential. When Hank Paulson told him the story about what he was dealing with, it was basically forcing all the major U.S. banks to take TARP money. He said, look, when you're boiling in oil, you don't ask how hot it is. You just act and you move. And I think he saved the country. I mean, I'm not necessarily a fan of anybody during that era, but I do think he saved the country by restoring trust in the banking system.

1:59:00I don't see private credit being that big of a threat, but it absolutely could play a role. That makes sense. What single deal are you most proud of? As an investor? As an investor, it doesn't have to be like it generated the greatest return, but something that you look back on, you know, 1995 to 2011 index. Look, I mean, the current portfolio of AI investments I've done, like Avin, I mean, Sadi Khan is one of the truly great CEOs of the generation. I recommend you put him on your show. But Vinod Koshal and I talked about it because Vinod's one of the large investments there. but fireworks AI E2B Lotus AI Dynasty is one that I'm really interested in because I just think trust is something that people like me have used successfully for a long time and I think entrepreneurs should be using them more often give the pitch for Dynasty you know you can put it simply but for those that are hearing about it for the first time yeah okay so look If you start a company out there and you basically don't have a lot of money, but you've got your shares, you should go put those shares in, whether it's a dynasty or somebody else.

2:00:24You should put those shares in trust for yourself and your family. And depending on whether it's your mom or your dad or your wife or potentially kids or whatever charity you want to put it in. And you should, it's a way of protecting it. And it's definitely a way of taking advantage of QSPS exemption, which came in during Clinton, which Bush, you know, embellished more and which Obama did more, which Trump did more. Every president since has pushed it. And what's happened as a result, I think, is you've got a lot more startups. You went from about a half a million startups a year in the 90s to close to 5 million startups a year now.

2:01:10And so for inside and I think for most other funds, right, I think about 80 % of the investments return less than 1.3x. Okay? But that means over 20 % return multiples. And those 20 % return multiples, those entrepreneurs should be thinking about putting their assets in trust. and not just stock, but also their crypto, you know, and protect it. And there's lots of reasons why. It doesn't cost a lot, thanks to the dynasty. They're democratizing it. So it doesn't cost a lot. You're crazy not to take advantage of the tax exemption that's associated with it. Yeah. Well said. Well, thank you so much for coming on.

2:01:52I'm sure we could keep talking for hours more, but I really just appreciated the stories and insights. and we'll have to have you back on again soon. Yeah. Well, thanks, guys. And look, you guys are the most fun entertainment show out there and that's why I'm here. So keep it going. It's fun. Thank you so much. That means a lot to us. We'll talk to you soon. Thanks so much, Jerry. Great hanging. Have a great rest of your day. You're the man. We'll talk to you soon. All right. See you. Bye. Before we bring in our next guest, let me tell you about Google AI Studio. Create an AI-powered app faster than ever.

2:02:26Gemini understands the capabilities you need and automatically wires up the right models and APIs for you. You can get started at ai.studio.com. Our next guest is Shlomes, the anonymous poster on X. That's right. I've been a fan of Shlomes' work for a long time. Always found the, I don't even know, art, stunts, marketing. It's everything in between. It covers so much. It's always cooking. Always enjoyed it. And I'm very excited that we get to have him on the show today. I believe he's in the Restream waiting room. But we are going to work. Because he's anonymous, that means voice changer. That means facial recognition.

2:03:10You've got to confuse the AI algorithm of the future so he's undetectable. So I will tell you about Profound. Get your brand mentioned in ChatGPT. Reach millions of consumers who are using AI to discover new products and brands. And I will also read this post, which made it in the timeline, from Sisyphus Bar and Grill, where it says, oh damn, the highly produced launch startup video starts with an outtake of the founder sitting down from an offset angle. Now I know it's going to be unique and awesome. It really is an overused trope, but let me just steel man it for two seconds. So I was in the business of making videos about startups, basically.

2:03:49I made a variety of videos. And when me and Ben figured out how to do the extra behind the scenes camera that wanders around while you actually do the interview and then splice that in, oh, it just takes you from like, you're doing a normal podcast to like, it feels like it belongs on HBO and it's so satisfying and I wouldn't get rid of it for the world. And I will defend it to the end of days. although yes it is truly played out and you should probably at least hit the drawing board live video tracking is the new meta yes yes it's just for sure i can just pace around the room i believe you have a lavalier mic on too so can you talk and can we hear you let's see are we are we getting audio from jordy on the walk he's not talking so i can't tell you got to keep talking if we're going to keep talking oh we hear it we're hearing him it's working i'm live you're live this is a new feature we're testing out everyone yes uh this is live camera tracking with lav so it's such a funny start to just pace around the room why don't you go take a crack at the gong i'd love to it's a gong for all the progress that the production i'm hitting the gong for jerry oh okay turn down the volume the the gongs getting hit hit the gong oh those worked Good, okay, we're solving it.

2:05:12There we go. That's great. There we go. See, this is good. This is good. The menacing walk with spinning the thing. I'm waiting for schlums. What's on that whiteboard? What's on that whiteboard, Jordy? Why are you walking in front of that whiteboard? This one? Yeah, that one. Oh, we were just doing some analysis. Yeah, what did you notice over there? No, this didn't actually go anywhere. You need to walk right in front of it. You still can't see it. Yeah, you need to walk right in front of it. This didn't go anywhere. This didn't go anywhere. we were it's interesting uh because you know people have been calling open ai maybe it's the next enron and if you if you trace the letters the e and enron there's an e in open ai that is n in enron there's an n in open ai and the r in enron the o in enron there's an o in open ai and at the end of enron it's an n there's an n in open ai and the p and the a and the p and the a as well yes and so uh those are the letters in enron and there's also letters in open ai and so you know i i just feel like the connections are a little bit it's a little bit it's a little bit obvious to be clear we don't believe that we are no we are as exuberant as ever i i am when sarah when sarah fryer said we should be more exuberant we didn't just listen we studied i like someone in the chat was like, he needs to be more exuberant.

2:06:35I'm shocked that those are real quotes. Yeah, they're wild. Well, let me tell you about Linear. Linear is a purpose-built tool for planning and building products. Meet the system for modern software development, streamline issues, projects, and product roadmaps. We might be coming back to Shlom's next time, maybe next week. Maybe we'll reschedule. That's the nature of live TV, baby. Let's go back to the timeline. Autism Capital has a post here. They're claiming that October 6th was literally the top. This was posted in December of 2023. We must find this man and make him our God. Nostradamus lives.

2:07:11All-time high. This was calling the Bitcoin. Okay, this is Bitcoin all-time highs. Basically, it was 1 ,064 days from 2015 to 2017. All-time highs. then 364 days from 2017 to 2018, then 1 ,064 days 2018 to 2021, and then 364 days from 2021 to 2022. The pattern would print this cycle's all-time high as the 6th of October, 2025. That was the actual top for Bigfoot. Okay, so I have no idea if this is Photoshop, but this is crazy. This really was the top, absolute insanity. Of course, it could just go back up next week and then this isn't true. Who knows if this is Photoshop? But it is fascinating, like, chart analysis.

2:08:05We need to get – wait, did we ever figure out – what is that called, Tyler, where people analyze the candlesticks? Oh, technical analysis. Are you familiar with technical analysis? With the concept of it, yeah. Technical analysis is like it's the head and shoulders pattern. So it has to go up. You just follow the lines, bro. Yeah, but it's not just linear regression. It's like if it goes up and down and then up again, that's the camel pattern, and that means it's going to break out. And people have a lot of fun with this. They draw all sorts of stuff all over it. It's been largely debunked, and if you're playing that game, you're probably going to get cooked by a real team of machine learning experts who are doing basically technical analysis at a much deeper level.

2:08:48But I think we need to bring technical analysis to private markets. I want to see people looking at, oh, the Stripe secondaries, this week they sold down 2%. They're up 4%. That means it's a bullish catalyst. It's going to go through the roof. I want to see a technical analyst bring that skill set. Okay, you see there's a down round here, but it's actually bullish. It's actually bullish. Exactly. It's prime for a breakout. Technical analysis, just looking at the charts, not even understanding what the company is, just looking at the charts, that's the key. Ben Sands says stockstrology. Stockstrology would be good.

2:09:26China has overtaken the U.S. in cumulative open source AI model downloads, says A16Z. Oh, we also have some updates on the solar panel company that we were digging into yesterday. Turns out it was built by the Chinese. It was built by the Chinese. So a Chinese company came over here. They built a Trina solar, built this facility, and then due to some regulatory pressure, they were a forced seller. And then it was taken over by what is now T1 Energy. And so everyone was like, wait, we know how to build facilities like this? And everyone got really excited, including us. And of course, it turns out that it was actually built by the Chinese.

2:10:14But I don't know that there's that big of a problem with that, right? Because it's built in America. That's the important thing. So let's study it. Let's study it. But also, I mean, I regard TSMC Arizona as a win for American dynamism, right? Yeah. This was that rune take where he says we should basically just bring over the Chinese. Yeah, or bring over whoever's building the best stuff and build it here. And there's a whole bunch of debates, like levels to the debate. Like, do you want, like, and it's sort of a yes and situation. Like, I want American workers to build American things in American factories, but also like, I'll just take a factory in America that's owned and operated by another country.

2:10:58Because if, you know, like there's some geopolitical crisis, like at least it's within our borders. Like that's better than it being halfway across the world, right? because you just walk over to it and be like, give me the solar panels, which is great. But back on the A16Z, they said, A16Z said, China has overtaken the U.S. in cumulative open source AI model downloads. And Lewis says, I'm going to make a series of bets on the little guy to start. We are going to be granting out compute up to 100K per project to support new experiments on GCP. If you have an idea for an open source model that you'd like to explore, I'd like to hear from you.

2:11:41So go hit up. I DM'd with Louis a little bit yesterday about this project. He's working on pulling together the compute resources, but very excited about this one, and it'll be fun to follow. Well, we have our next guest, Shloms, in the Restream waiting room. We will bring him in to the TVP. How are you doing, Shloms? There we go. Yeah, we can here you oh yeah we did it this is a crazy crazy uh a non-call we've had a few no one's ever sent is it safe to look at yes this is just my webcam i have a very old laptop sure yeah got it i i apparently the technical difficulty was that my display name was breaking zoom which i suppose is on brand whoa that's super interesting what is your what is your display breaking stuff uh so uh the backstory is that i was looking for the longest unicode character to break twitter in maybe 2017 um which is a very long arabic uh um sign um but then people told me that they didn't like that i was using that so i looked for the second longest unicode character which is a cuneiform symbol yeah and so you stack all this up yeah you've been you've been fantastic at kind of like hacking these systems and and breaking them in a way that you know really gets outside the box.

2:13:02Like, yeah, you have seen like the text that goes completely out of your, your UI designers, worst nightmare, basically. Uh, so many places I want to go with this conversation. We're super excited to chat. What, uh, what are you, who would you say you do here? Yeah. What exactly do you do here? That's a tough one. I mean, you know, I think a lot of artists are driven by wanting to be an artist, right? And same with founders, to connect it back to your wheelhouse. And they want to be the thing, right? And then they try to figure out whatever their angle is. Then there's people who are kind of like driven by whatever instinct or whatever is inside of them.

2:13:44And yeah, this is just me, right? I'm anonymous, so this is like a very specific part of me, right? I'm filtering. But I wouldn't say it's a character per se. I've always been driven to break things. Yeah. Do you like the term artist, hacker? Does it matter? Is any of this relevant? That's a good question. No, I don't think it matters. I think artist is a boring and kind of, um, yeah, it's kind of a self-serious term, right? And I don't want to take myself too seriously. Let's skip the titles then and jump to the projects. Uh, I always like this question. We asked Gabe Whaley from Mischief, uh, what is the most underrated project that you've worked on?

2:14:25something that you are proud of but maybe people don't already know you for oh that's a good question i mean i think most people know me from blowing up a lambo right yeah and can you for those who don't know can you tell the lambo story briefly sure one of my one of my favorite and most memorable moments from from that whole cycle it was so cool in so many different ways but i'll let shalom tell the full story yeah i think i think that um that project you know like a lot of things you know trying to ride the the line between like a satire and a serious like earnest engagement with something so that was my sort of take on what was happening in crypto art at the time um but yeah we physically blew up a lambo well before that was possible to sort of fake with ai or anything um took all the pieces to a very involved uh videography studio and took these like beautiful videos of all the fragments and sold those as NFTs.

2:15:25Wow. And like a post-mortem on that project, is everyone generally like happy with the way it went? I feel like there's so many NFT projects where like the expectations of the community get away and people are like, no, like it's going to be a whole metaverse. And it feels like, I don't remember you promising that. Is everyone pretty happy with like they were bought in on an art project and that's what they got? Yeah. I try to make things that stand on their own, right? I think that stood on its own and didn't need a roadmap. It was a thing that happened and people wanted to own it. I think that was always what NFT should have been and sadly didn't quite turn out that way.

2:16:01Yeah, no, no. That makes total sense. Should we go back to any underrated projects that you think didn't get enough love? Yeah, underrated projects would be cool. And I just want to, I mean, I think maybe we will get to some of that. The thing I'm interested is like, what's interesting to you about the current state of the internet? Interesting, yes. And specifically various platforms. I know you from Axe, but I'm sure you're active elsewhere. But how do you rate the current internet? So first of all, prediction markets, you're seeing sort of turning it to hyperstition markets. I think there's a lot of artistic potential there that hasn't been engaged with.

2:16:45So pay attention to that space. and you know I think you asked earlier if I view myself as an artist I think that I try to just do things right like I'm an agentic person or I try to consider myself one I think AI unlocks the ability for people to be agentic right so I don't view when people say AI art they think of entering a prompt and you get an image output I don't make images right like that's not what I consider my art but I do things and so i can use ai to more effectively do those things with fewer resources in a more asymmetric way so it's completely changed my life yeah i hope i don't sound too one-shotted so yeah no no i completely i completely agree i guess help me uh resolve this we were talking to gabe whaley at mischief not to comp you guys too much but uh i i do think of him as like another person who creates in the original original thinker original thinker that's not just not entire you know yeah commercial element but is not entirely just you know not not yeah the goal is not wake up every day create shareholder value it's like wake up every day and do novel which we would have no problem with we would actually prefer you know if you could stop fooling around my real passion is improving the monthly recurring revenue of b2b sass this is my fallback yes yes is that is that actually true do you actually no okay but so so so so here's my question uh it's It feels like there are all these really heady questions about AI and slop and what is good and will it one-shot you and paper clipping.

2:18:27And it's the best time to be a podcaster, a newscaster. Like the Dwarakash Patel podcast is fascinating. There's so many different ways to engage with it. There's books. do you feel like the art community is engaging with it in the proper way? Are there more projects coming? Is it harder to engage in a critique of AI art because it proclaims to be art itself? And so it's maybe a little bit more complex to tussle with? Or do you think that basically, like are artists driving the discussion uh around the all the trade-offs in ai effectively in your mind right yeah i think there are a lot of interesting parallels with like crypto art right because in early crypto art it genuinely did unlock some interesting things but then you kind of had an 80 20 where 80 of it was just like kind of boring meta stuff or people who probably shouldn't have been anywhere near that stuff pushing kind of garbage right um i think I think there are definitely creators who are doing some insane, just like on a technical level, some things that they couldn't have done before.

2:19:38Then there are a lot of people kind of pushing out kind of slop images that I would say is like a second bucket. And then you have kind of like meta AI art. So like you, you see like in like digital art spaces and like Instagram, if you go on your Explorer page and you're interested in art, you probably see like a lot of things of like bounding boxes or just like these kind of overplayed motifs about like surveillance or whatever, which, you know, there's a time and place for. But yeah, I think any new technology that people are interested in is bound to result in just really boring pontificating about it.

2:20:10And then like really boring engagement with it. And then like a very small subsection that maybe doesn't get enough attention of people doing genuinely interesting, innovative things with. And I'm not the person to do those genuinely innovative things because I'm someone who works within their own constraints of being a fucking idiot. So my game is to do something interesting conceptually about it with my own limited tool set and intellect. Yeah. Fascinating. You made a viral post on Sora of Sam grilling and eating a Pikachu. Oh, yeah. Yes. How did that? How many shots on gold did it take to come up with something that you felt like was worthy of sharing or something that kind of broke through the noise?

2:21:02Well, it's cool that you talk about shots on goal because I think a lot of this stuff, I wouldn't necessarily view that as an art piece. I think that maybe even more falls into my kind of like misinformation practice, right? Like trying to spread some sort of vile rumor or something inherently is going to be like a shots on goal game. You're trying to find some asymmetric way of hitting on like just the right court at the right moments. It's like my big example is like the Gmail sunsetting hoax. So I created this hoax that Gmail was shutting down. And, you know, for that one, there were a hundred other things that I tried to do that didn't quite hit.

2:21:33But like this was at a time that Google had just sunset something. So it was like kind of in the late insight guys. But I think the real thing there was like it was it was sort of this moment of people being like viscerally feeling their reliance on Gmail. Right. Which like if I just wrote an essay about how much we rely on Gmail, that doesn't really go anywhere. but if you like i think that's why misinformation or some of this stuff is like a useful tool but to bring it back to to um to the to the sam altman thing yeah i that was a bit of a lowbrow moment for me um but you know that that that day kind of was you know the thing was like there it was a lowbrow day on the internet yeah right yeah it was it was slop about pokemon all over sora and and your timeline right and yeah and this was just kind of a depiction of that but uh what uh Not my proudest moment, but it was fun.

2:22:20Yeah. How did the Gmail sunsetting hoax really catch fire? Like what was the inciting post? How did you seed it? How did it actually go viral? I'm always interested. John, you're asking me to give away the sauce, man. Oh, wait. You're going to have to buy my e-book. It's not something that I can just trace through? You're going to have to buy the course on misinformation. I didn't realize that that was not just – I would have assumed there was public. Sorry. No, no, no. I'm joking. Yeah. If you were to look back at that, there were a lot of tweets about that, and mine did okay, but there were other people who sort of picked it up.

2:22:57Okay. And it's actually not even necessarily clear to me which of those people were engaging with it earnestly or kind of realized what was going on and were trying to single boost it. Yeah. But yeah, I mean, you guys are part of the cool kids club on Twitter, I would say. And it helps to be at sort of like the, there's like a cantillon effect for attention too, right? Totally. If you're upstream of that, then you can really spread things. Yeah, I don't actually remember, but I could totally have seen myself amplifying that just as a troll on Google, even though I would have recognized that it's not real.

2:23:31But I'm still upset about Google Reader. And so I'm going to amplify the fake news, knowing it's fake, but just to cause chaos because I want revenge for my RSS reader that was killed in the cradle. Are you disappointed? I'll do you. We should spread some more. Oh, I'm ready. I'm ready. We are the fake news here. Are you disappointed with sort of a lack of creative uses for artificial intelligence that it's all kind of funny? it's like hey I made uh this musical artist do country music or I made my dog look like a doctor or I mean I you know as you guys kind of noted I'm guilty of the same thing sometimes I would say broadly you know I I want I want more generally that's why I do this right because there's like this kind of art that exists in my head that uh that I want to exist in the world and I'm grateful that it resonates with other people.

2:24:33There's definitely things that I think that people could be doing with it. But I don't want to be too misanthropic. I think it's a net good long-term because I think the art world is very stagnant. Not a fan of the fine art world, as they would call themselves. And I think any technology comes out. I'm sure they're not a fan of you either. No, but it's surface area, right? It's surface area for people who are willing to engage with something that the orthodoxy doesn't approve of to say something new and interesting and not have to play the same game of getting into a gallery or doing an oil painting or a sculpture or whatever.

2:25:08We need new games to play and technology increases that sort of thing. Do you think any of your work will appear in Sotheby's at some point? Maybe because of Sotheby's, I think I'm quite black-fledged from because I've pointed out some of their own practices. eventually they'll you'll still be going their team will turn over and they'll they'll yeah maybe maybe if they forget about some of the dubious possibly illegal behavior that I've I've I wouldn't bet on it you'd be surprised major things have happened this was super fun thanks for so much next time you have a particularly devious stunt feel free to jump on the show and break it down for us I'd love that next time we can go deep on prediction markets so yeah Yeah, I'm super, I mean, watching this category explode and then thinking about all the ways that you could hack, I think Brian Armstrong's kind of hacking it on the earnings call was fascinating.

2:26:10I think there's so many underexplored areas and surface area there and also some potentially dark scenarios as well. But the sci-fi future is certainly here. the last thing I would leave with is I really think that open source AI is important I don't know how much you guys have gone into that but we can talk about that one next time yeah that'd be great slums model coming soon I like it yeah thank you great to meet you yeah great to meet you too we'll see you out on the internet cheers before we bring in our next guest let me tell you about numeralhq.com sales tax and autopilot spend less than 5 minutes per month on sales tax compliance we have a return guest in the Restream waiting room.

2:26:57Welcome to the show, Shazan. How are you doing? Congratulations. We got the gong already. Sorry, we're running late. Give us the news. What's the latest in lava world? Can you guys hear me? Yeah, we can hear you loud and clear. So we basically announced Monday that we raised$200 million venture and debt. we brought on new investors like anthony pompliano eric jackson from opendoor you guys on before as well and we also announced what i think is the most important product i think in bitcoin's history so far other than bitcoin which is the point it's the world's first bitcoin line of credit so lava has been doing loans for years now but all our loan previously was structured like a Bitcoin mortgage.

2:27:49So you have to make fixed payments. There was a fixed term to it. But now, and that was very similar to all the other loan products in the market too. But with the Bitcoin line of credit, we are offering the most flexible loan product in the entire market. So basically you can borrow, you can keep it open for as long as you want, two years, three years, really forever. You don't have any fixed payment schedules. There's no monthly payments involved. You can pay down your loan whenever you want your line of credit and you can borrow back against it at any time. And we're offering the lowest interest rates, fixed interest rates at 5 % across the board.

2:28:26So really what we realized was that Bitcoiners are a very unique audience. They all have very distinct cashflow needs, income. A lot of Bitcoiners are retired on their Bitcoin. So having a product, like building a product with fixed terms, fixed payments and applying it to all the Bitcoiners was just not something that was working and it's not something that they wanted. Whereas this product, it offers the most amount of flexibility. So no matter what your cashflow needs are, no matter what your income is, you can use this Bitcoin line of credit, use it to use your Bitcoin and manage your Bitcoin wealth with the flexibility you need and at the lowest rates across the world.

2:29:06um yeah uh as i we we've been on the show multiple times we've talked about the business um i'm interested in to understand like why is fundraising happening in a traditional sense instead of sort of all happening on chain one bitcoin at a time uh like why why is there like a package tranche of of investment happening because we actually have our own uh yield product We announced that to you guys a couple of months ago, weeks ago. So we do have depositors that are coming on chain with stablecoins around the world that are depositing their cash, their stablecoins, earning yield, funding Bitcoin-backed loans.

2:29:51But those depositors can deposit and withdraw at any time, right? There's some permanent investment. Yeah, we're offering fixed. So to make sure that we're never in an asset liability mismatch, we need to make sure that we are raising the credit lines available to refinance loans. And these credit lines are at fixed rates, fixed terms. So we always are kind of matching liquidity in a way. We're just providing on-chain banking in a way. So we need to have depositors. And I think that's the real value of stablecoins is that they're global. so you can kind of give functionality and savings to people with stable coins.

2:30:32Obviously, Bitcoin is a savings asset, more of a long-term savings asset for some of this audience that is using this product. But then we need the credit lines as well to make sure there's never an asset liability mismatch. Does that make sense? Yeah, totally. What's the strategy around navigating lava through such a rocky environment? Bitcoin's down as of this moment, 16 % in the past month. There's been a lot of volatility and other, even more volatility in other tokens. But it feels like that, you know, you guys were started like a year and a half ago, two years ago. What was the timing? Yeah, around two or a little more than two years ago.

2:31:17So we've been building this for a long time. But yeah, so building it on an upswing and it managing through, you know, corrections is going to be what makes Lava, you know, super, super, obviously the solution is meant to be trustless, but brands still matter. Well, there's definitely trust involved. We totally like want to lean into that, right? So one thing is on the Bitcoin, yeah, Bitcoin price hasn't been doing well, but also Lava is Bitcoin only. So the only collateral asset we accept is Bitcoin, Even recently, there's been some protocols that were accepting other crypto assets, other digital assets.

2:31:59There's one, I don't know if it's fully public yet, that's actually recently blown up because the altcoin debt, the altcoins have been dropping pretty significantly in price. If you're lending against it and it's dropped way too quickly, you basically have taken on bad debt and then the depositors are no longer whole. actually a few weeks ago i'm sure you guys discussed this but there was a big auto kind of de-leverage a lot of coins actually instantly dropped 80 percent bitcoin was still fine in that world so on app one thing that helps us to manage liquidity and safety for our users is we're bitcoin only so as a bitcoin they know that they're not taking on risk of other altcoins when they're borrowing with lava the other thing is we always encourage people to add more collateral to their loans.

2:32:47The average LTV on a lava loan is 30%. So it is pretty safe. So Bitcoin would have to drop effectively like 70 % for the majority of users to get liquidated. And almost all of our users actually have extra Bitcoin that is available to them to also as collateral to kind of make sure that they are safe. And there have been some like 50 % drawdowns, but a 70 % drawdown in Bitcoin, I feel like that hasn't happened in long, like years and years and years. yes so it makes sense it is definitely a risk um when you're borrowing against your bitcoin but the other kind of counter to that is anyone holding bitcoin right now is also predicting that the price is going to go up right because otherwise if you do believe bitcoin is also going to drop 70 it's actually probably not the right idea for you to even hold bitcoin don't hold it it's actually probably the right idea for you to hold cash and then kind of rebound into bitcoin so you know we are not like providing this financial advice to people but most of our users are so believing that Bitcoin is going to do well.

2:33:45Otherwise, they would just hold cash. They're diamond-hensing. Well, thank you so much for coming on the show. Sorry for keeping you waiting, but congratulations on all the progress. Honored to hit the gong for you. We will talk to you soon. Have a good rest of your day. Good to catch up. Before our next guest hops on, let me tell you about Fin.ai, the number one AI agent for customer service, number one in performance benchmarks, number one in competitive bake-offs, number one in G2. I actually just heard about a bake-off just yesterday and someone said finn won the bake-off literally i'm not even making that up love it uh our next guest is mina from sandbar coming in the studio from the how are you doing welcome how are you thanks for having me it's great to have you thanks so much for hopping on the show uh give us an introduction uh explain the product some of the trade-offs and then we'll go into go to market and how everything's playing together and first i have to tell you, I saw the domain sandbar.com available at some point back in the day.

2:34:44And I was just thinking to myself, that is such a good domain. So as a domain enthusiast, I'm happy that you picked it up and are doing it justice. Appreciate that. It's got sand, it's got a bar, it's got everything you could want. It's got a nice feeling. I have positive feelings about sandbars growing up in California. But yeah, would love an introduction on yourself and the company and how you got to building Sandbar. Totally. So I'm Mina. I'm one of the co-founders at Sandbar. I originally met my co-founder, Kirak, at a startup called Control Labs that was developing these wrist-worn neural interfaces.

2:35:22And we got acquired into Meta and we're working on a lot of devices there. And in parallel, I started exploring a ton with LLMs and started trying to think about what it would take to have a more effortless interaction with these new speaking, thinking machines. So we started to explore a lot of different forms, and that eventually led us to two products that I guess we're announcing together that we call Stream and the Stream Ring. And Stream is, as we say, a self-extension for talking through ideas and capturing them into notes. We have this very sandy app that has a log of interactions you've discussed, notes that you've asked Stream to save, as well as the ability to zoom out of interactions and see what you've been discussing over time.

2:36:21and especially if you're on the go, your phone is in your pocket and you want to perhaps talk about an idea or save something into a note, that's where you can use the stream ring. So you just hold this touchpad, whisper into the top, and then release. And if I do something like that, I can say, hey, can you give me your hottest take on AI wearables. Sure. The hottest take? Most AI wearables chase novelty instead of intimacy. They brag about features, but the real frontier is emotional calibration. You know, why don't you start a list with some of those hot takes?

2:37:09And it'll scurry away and save those away. There you go. Sure. For when you need a list of hot takes. That's right. I loved it. I love the live demo. It's always risky, but yeah. So the key, the value that you're delivering to users is not, you're not trying to say, I'm going to be this personal assistant that's going to do everything in your life. It's simply like you're just trying to build that interface with effectively just building this new surface area for a mobile device that people are already using in a new way to just input ideas and thoughts into a computer and have them be organized.

2:37:52Is that generally right? Exactly. We're really focused on letting you talk through ideas and save thoughts into notes. And normally I wouldn't have my phone out. I would have it in my pocket and maybe I would have earbuds in. And already so much of our thinking either happens in the go or lives in Apple Notes. And it's this beautiful mess of grocery lists and letters to a parent and career plans. And we see Stream as the most effortless way to either build those ideas through conversation or save them away. Very cool. What's the timeline until you guys are shipping the product? I know you're doing pre-orders now.

2:38:35Totally. So we've been building now for around two years, and it's good to be coming out of the cave. We are actually right now kicking off our next build with our manufacturing partner in Taiwan, and then we'll be fully ramped to mass production in summer of 26. So that's when we'll start shipping. Very cool. Very cool. What are your hottest takes on wearables? obviously it sounds like you maybe trained your your uh your at your your internal model to here's here's one do you do you think like how confident can you be in predicting success of a wearable because it feels like it's something where you need to just run a lot of tests and the speed of each test is a little bit slow because you have to iterate on the actual hardware and you might one-shot it, but that might be a lot of luck.

2:39:32How much luck is it? Or do you think that there is a way to empirically know that you have a hit on your hand before you go to market? Yeah, I mean, we definitely didn't start with this. We've worked on a lot of different devices, a lot of different use cases. And now we've been living with some form of stream for the last, call it a year and a half. And Karak and I wanted to get to a point where we felt like it was useful in our daily life. And then the people who are having test the device, our friends and our family who we've given units to are themselves finding it valuable. And we have a marketing professor who will be driving to work and she'll talk about her lesson plan.

2:40:13And then she'll watch her kids in the playground playing and she'll be able to keep her eyes up as she takes notes on what she wants to improve for the next day. We have a body work coach who will walk her dog and be talking about clients and also riffing on Eastern versus Western medicine. And after enough time living with it, we got to this just really high degree of confidence that it was super useful in our lives. And we're exploring other forms and other use cases, but they always have to hit that bar. And I would say having built always in New York where we get to wear devices easily outside and not be noticed and also be in touch with people outside of tech has helped really ground us in something that we think will be useful for everybody.

2:40:59Amazing. Thank you so much for coming on the show. I was laughing thinking about a future where the average human has exactly 10 rings. I was thinking the same thing because the aura ring is so popular. It's like, this is my thinking ring. Well, then you get into the gauntlets and the, yeah, yeah. And the chains bracelets and the You pierce both of your ears for the wearables and eyeglasses and headphones, all these things. Yeah, well, I'm super excited about this. John knows that I have a hilarious way of keeping track of the thoughts in my mind, which is texting myself. I'm constantly texting myself stream of consciousness things that I need to remember and do or whatever.

2:41:37It's unhinged. I'm excited to try this out. Yeah. Have a good day. Yeah, appreciate it. Thanks a ton. Thanks so much. Yeah, great to meet you. Talk to you. Before we bring in our next guest, let me tell you about Adio, customer relationship magic. Adio is the AI native CRM that builds scales and grows your company to the next level. Our next guest is Alessandro Chesser from Dynasty. Let's bring him in. Alessandro, welcome. Thanks for having me. It was so fun having, I think, your lead investor, Jerry, on earlier. He's got so many amazing, amazing stories. But before we get into Dynasty, why don't you give a quick intro on yourself, background, all that good stuff?

2:42:21Yeah, absolutely. So, you know, born and raised in the Bay Area, worked in financial services for about a decade, wound up at Carta as the first sales hire in 2014, ended up becoming VP of sales and owned a lot of revenue for about eight years. uh so help take the company from zero to 300 million in ARR you know hired and managed hundreds of people um and then left to start Dynasty uh what was what was the critical inspiration for for Dynasty and and how walk us through kind of the idea maze around it so in the beginning of you know in the beginning uh when I joined Carta me and my co-founders who are also early Carta We were helping onboard thousands of cap tables, and it became very clear that the most successful founders created lots of trusts.

2:43:17They didn't just hold the shares in their own personal name. They would sometimes have like 10 different trusts that they would split their shares into. And it was really strange. I didn't understand why. At first, I thought they were just being super generous. but we started doing some research on the strategy and we learned that the reason why they did it is because they got a separate QSBS exemption, which basically means each trust that they created was eligible for$10 million in tax-free capital gains, like literally zero taxes on 10 million for each trust that they create. So you create 10 trusts, you get$100 million in tax-free capital gains.

2:43:56And, uh, yeah. So, so from that point, like, I guess, uh, what I think the, yeah, the question, the question that, you know, we, we talked off air a while back and the question I had is like, uh, obviously, you know, great businesses often start with like a simple, you know, vision for like a kind of narrow customer base, but what's kind of the focus today? How big can Dynasty get just serving the kind of like early stage founder venture community? And then where do you want to go long-term? Yeah, absolutely. So QSPS stacking is very narrow. You know, it's basically have to be a large shareholder of a C-corp.

2:44:43And it's not just venture-funded technology companies. Like we've been signing on porta potty founders. We've been, you know, people that make pipes for plumbing. Like there's all types of QSPS eligible entities. They just have to be C Corp and they have to not be an excluded industry like any regulated financial services or anything like that. So there are a ton of companies like maybe we don't know the exact number, but maybe somewhere between 250 ,000 and 500 ,000 companies in the U.S. could qualify for QSPS. And so we got a lot of greenfield. Like each company has sometimes two or three co-founders.

2:45:19And so that's a, you know, it's a very narrow market, but we think we're going to be working on it for quite some time. But the broader vision for this company is like, so we are a licensed Nevada trust company. We are the only venture funded licensed Nevada trust company. The only ones that are trying to tackle this space from one of one from a technology standpoint. And so we think that – so the richest people, Jerry, for example, the richest people in the U.S., they don't create – they live in California. They live in New York. They don't create California and New York trusts. They use Nevada trusts because Nevada has the best laws in the country, maybe sometimes even the world when it comes to taxes, asset protection, privacy, control.

2:46:02Like you can have more control over these Nevada trusts than you can typically over California and New York trusts. And so we think that goes much further beyond QSBS stacking. Like, you know, why shouldn't the farmer in Oklahoma be able to benefit from a Nevada trust? Why shouldn't the young professional with$50 ,000 in crypto be able to benefit from a Nevada trust? And so we want to open it up for everybody. That's our vision, to build a mass-market, AI-powered Nevada trust company. But to your point, we have to start narrow. We have to solve one problem and solve it really well. And so that's why we're tackling founders.

2:46:37uh amazing what uh what is what is the most common setup for call it like a yc founder for trust fifteen hundred dollars a year gives you up to 40 million dollars in qsbs eligibility if they have kids they they make their kids a beneficiary maybe their wife if they don't then they're creating trust for their parents and their siblings Very cool. Well, yeah, I think every founder should be at least taking a look at this. For sure. And like Jerry said, there's other ways to do it, but I highly doubt that the other methods are nearly as streamlined or founder-friendly as what you guys are building.

2:47:19So very excited for more founders to be able to optimize for this, not just the second or third time founder that has experienced the gain or the pain of realizing that if they have a$50 million liquidity event and only getting that tax-free gain on the first 10 million would be quite painful. So excited that you're building this. That's exactly the problem. The problem is people usually do this later because it costs six figures to set up. And by the time they do it later, they're limited by their lifetime gift exemption. And so they can't stack that many trusts. And so that's why we exist. So founders should be doing this from day one.

2:47:56That's when your gift value is zero. You create your corporation. That's the best time to do it because there's no gift tax. The longer you wait, the more gift tax implications you're going to have. Yeah. Amazing. Well, thank you so much for joining, breaking it down. And I'm jealous that you got Jerry on your cap table. That guy is a, that guy's a legend. So anyways, great hanging and happy building out there. Thanks for having me. Thanks so much. Talk to you soon. Our next guest is in the Restream waiting room. But first, let me tell you about public.com investing for those that take it seriously.

2:48:30They have multi-asset investing, industry-leading yields, and they're trusted by millions. John Maslin, the CEO and co-founder of Vulcan Elements. Got to get this gong ready. Just a tiny little deal. Just a tiny little deal. Welcome back. A wee little deal. A wee deal. With the U.S. government. Break it down for us. $1.4 billion. dollars. Absolutely massive deal, something that people were just waking up to. Obviously, you've been thinking about this for a long time. How did you realize that this was something you wanted to work on and then take us through the anatomy of the deal? Yeah. Hey, guys.

2:49:10Thanks for having me on again. It's really good to see you. So this is a problem that we've been focused on for several years. My background is former Navy. I was a supply chain officer. And when I was in business school was thinking a lot about the critical components that will define the 21st century technologies whether it's the data centers that enable ai drones robotics and when you really boil it down it's only three components semiconductors batteries and rare Can you check your Wi-Fi? We are getting some technical difficulties. We lost you. I'm still there. Your audio is here. Your face is in the past.

2:49:54In the past. It's a time machine taking us back to a mere five minutes ago or one minute ago. Out in the field somewhere. Yeah. Probably doing important work. Yeah. What a cute freeze, too. Yes. I mean, it's kind of working if you just want to keep talking. I'm going to hit the gong for the$1.4 billion. talking. Okay. Yeah. Yeah. Jordy's going to hit the gong, hit that gong, Jordy. And then we will go back. Thank you. And I'd love for you to break down the structure of the deal, what it allows you to do. I know that there's a loan piece, there's an equity investment piece, there's investors in how are you describing the deal and what it allows you to unlock?

2:50:37Yeah. So what this deal allows us to unlock is a 10 ,000 metric ton rare earth magnet facility. And just to frame the problem a little bit, by 2030, the U.S. will need over 70 ,000 metric tons of these magnets today. We can't even produce 500. So what this is doing is this is going and ensuring that we can move at Operation Warp Speed in partnership with the United States government, the Department of Commerce, the Department of War, to ensure that we can go and meet the demand for defense, aerospace, and critical economic industries. And the U.S. government, Secretary Hegseth, Secretary Lutnik, they're serious about solving this problem.

2:51:18And we're very serious about ensuring that we go and meet the moment and the mission. What is the timeline for something like this? I feel like we're hearing about billion-dollar build-outs, data centers. This deal must have been very much in the works even before China pulled the rare earths card. Sure. yeah so we plan to have initial capacity online by 2027 and if we're able to go faster we will and we've been having conversations uh you know with everyone across industry and across government uh for several years and you know we've realized you know we need a whole of industry and government approach to go and fix this issue and our full intent is to go and build at a speed and Operation Warp Speed that hasn't been seen in this industry in the 21st century.

2:52:13How much, like, what else is going on in the supply chain? Is America mining enough rare earth elements? Or should we be, you know, is the job, like, are we on the right track now in terms of actually taking rare earths and producing magnets with what you're going to be doing? Or are there other pieces of the supply chain that you want to see other founders or yourself go after in the near term? Yeah. So it's a really good question. And the way that this works is you mine the material or you take recycled end-of-life magnets. You separate those chemically into a rare earth oxide. You turn that into a metal and then ultimately a bespoke magnet to meet a customer's spec.

2:52:59What we're doing with this deal is we're partnering with ReElement Technologies. They recycle end-of-life magnets and separate that material. And we're expanding that partnership. One thing that I think is really important to note is this is less of a raw material availability problem than a manufacturing problem. Today, China only mines 55 % of the rare earth material. They make 94 % to 98 % of the magnets. So what I would say is with this deal and other folks in the industry, We want as many smart people focusing on this industry as possible and scaling up capacity where I think that there are opportunities are downstream.

2:53:36As we think about drone motors, there's a gap. Servo actuators. As we think about orthogonal and complementary critical minerals, we need to do more around antimony and germanium and gallium and lithium and tungsten and titanium. There's a lot of wood to chop in the critical minerals industry to make sure that we have safe and secure supply chains here at home. Well, very, very exciting. The progress is just tremendous, and I'm sure you'll be back on soon. So thank you for doing this work. We've been following the story, and we're glad that people like yourself and your team are tackling the issue.

2:54:16Next time with a better Wi-Fi signal on my end. Or when you're in L.A., come out in the studio. Yeah, or in L.A., or just do it out in the field somewhere, off your phone. We'll make it work. Thanks so much for the update and great to see you. We'll talk to you soon. Have a good one. Thanks, guys. How did you sleep last night? We had to sleep in the hotel room. We were traveling. So no scores. But tonight we will be returning to the eight sleeps. You can go to eightsleep.com, get a pod five, five-year warranty, 30-night risk-free trial, free returns, free shipping, of course. Code DBPN. We have Eugenia Kuda coming back in the studio.

2:54:54Should we hit some timelines? Let's hit some timelines. So the real story that was shaking up the timeline is that Scotty Pippin is saying it's not over. He said, I think in regards to the crypto market, he says it's still the shakeout before the breakout. So Scotty Pippin remains bullish on the crypto markets, and we'll have to check back. How do we know this is about crypto? This could be about AI or something. Could be about anything. Look at everybody replying or crypto. I think we got to ask Dylan. I think he's been... He's a crypto influencer of some sort. Yeah. If you want to talk to Antonio Linares one-on-one, you can now do so for the low price of$15 ,000 an hour.

2:55:46$15 ,000. But it includes taxes. He pays the taxes. That's... He pays$2 ,603.13 in taxes for a one-on-one session. Where are those? I'm not sure who Antonio Linares is, but he took the advice. Naval Ravikant said, set an unreasonable opportunity cost so that you only do things that are of the highest value. And that's what Antonio is doing. What's the value to him of yapping on the phone with some random person? You know, he's giving up a lot of opportunity. We've seen a few of these screenshots from people sharing these unreasonably high one-on-one session times. I mean, there are plenty of people out there that would have higher numbers for sure.

2:56:3814K, 15K. I would like to see somebody throw out the$100 million number, right? You'd think that for$100 million, people would talk with... I don't think Elon would do it. I don't think Elon would talk to a stranger, a random person for an hour for$100 million. What about 30 minutes? We'll have to ask him. $200 million. You have to start a show. Hourly rate. And then you can get him. You have to get on adquick.com. Out-of-home advertising made easy and measurable. Say goodbye to the headaches of out-of-home advertising. Only Adquick combines technology. out of home expertise and data to enable efficient, seamless ad buying across the globe.

2:57:19I believe we have Eugenia in the Restream waiting room. Announcing a new company. Welcome to the stream. Whoa. Look at this. Call it in from the headquarters. Do they poach you? Are you a new GP? What's going on? No, just a new Portco founder. You wish? No, you're supposed to say. Still a struggling founder. Yeah, yeah, still a struggling founder. Tell us about the new company. We just announced today the round we did with the pre-seed round we did for Wabi, our new company. So time to announce the new company as well. So we're building Wabi, which is a personal software platform. Think of it as an app where you can discover, quickly remix, or even create personal mini-apps for your daily life.

2:58:07Okay. So, yeah, walk me through. I mean, I've created personal miniature apps with, you know, vibe coding platforms or, you know, coding agents. Kind of run them on my computer. Usually try and have them all self-contained in HTML or something. How are you thinking about the actual interaction? Is this like prompt to product, self-hosted, hosted within Wabi? Like, what are all the different tradeoffs? Because there's so many different ways that you could instantiate that idea, I imagine. So think of it as a YouTube for mini apps. So all Wabi mini apps compared to most app builders, they all only live on Wabi.

2:58:45You cannot download them. You cannot put them on the app store. You have to use them on Wabi. But that comes with a lot of benefits because first of all, the social graph, discovery, you can like, comment, see who your friends are using, is using what app, see remixes, follow good creators or creators you like, and so on. You can use apps with your friends. And then on top of that, of course, all integrations are included. You can immediately connect your Gmail account, your calendar, location. Our plan is to build 50 integrations before the end of the year of all sorts. And of course, persistent backend, data security and privacy, everything is sort of on Wabi's side.

2:59:31Because if you think about it, even if you build a great app with one of my VipeCoding tools and send me a link. If it's like an AI therapist app, I don't know if I'm going to fully trust to use this because I don't know where the logs are going. Even if you're a good friend and you don't mean anything bad, you're not a professional developer. What if you forget to pay for the hosting of this app? Or what if something leaks? We've already had some of these VipeCoding apps reach the top of the app store, like that Teabag app. I don't know if you saw that. The T-dating? Is that the one? I think so, yeah.

3:00:06Was that the T-dating? Yeah, yeah, yeah. That one's been very popular. So talk to me about the technical hurdle here. It feels like the app store doesn't necessarily love apps in app stores. Is this a beneficiary of the new regulation around allowing app stores in the iOS app store? Is that a unique unlock? So we're not building apps. We We don't call them apps even. These are mini apps. So think of it as like this small, very lightweight workflow. Some of them just like, it's almost like a lightweight UI on a prompt. For example, a simple mini app that turns your photos to like Polaroids from different decades.

3:00:46And so the App Store allows mini apps. And we're not planning to, or we're never going to be about like building full-fledged apps or downloading them or anything like that. Yeah, it is interesting. It feels like Apple sort of have to grapple with the idea of instantiating UI or mini apps on the fly. Because for a long time, the App Store review process has been very much like we want to review all the code. And if you are changing the code within the app or the functionality of the app, once it's already shipped and approved, Apple has kind of dealt with that in a wide variety of ways. but I love the idea.

3:01:28So let's just stick to the actual technical side of things. Are these apps built in basically like HTML windows that are then served within, or can you actually instantiate like React Native or like Objective-C on the fly? These are React. So all of our mini apps are React Native. And then Wabi, the platform, of course, is the iOS native Swift code. So I think sort of similarly to Replica, what was the native app and a Unity container inside of it. Here are the same. I feel like we're always going to be about like blending the two technologies together and struggling constantly with how do you merge these two things together.

3:02:19What do you think, what do you expect to drive the most usage? obviously it's early it's hard to predict with consumer but do you expect it to be people coming on board just playing around creating things and people are coming to the app to have this sort of creative experience like they might go on Suno and just start making music they might go on Wabi just start making apps and sharing them with their friends etc or do you believe it'll be like more of a hits driven business where there's a one or two mini apps that you know develop some organic virality themselves that really become the driver of real downloads.

3:02:57I think really we are going to be looking at two big categories, more retentive utilities, Think trackers, things that you might use with your family, your friends, yourself, learning, all sorts of like to-do lists, things, and everything that kind of like little workflows for daily life. And then more viral personalized mirrors or like viral prompts for image and not only. for example something that I really liked on Wabi was this mini app that takes in your Gmail and builds an action figure based on the information from your Gmail or like the same but the same but based on your bank activity through looking at your bank account so just stuff like that where people always want to see some cool stuff like that brag about themselves share and it's almost like a one-time use thing.

3:03:57You're not going to be coming back for that. But at the same time, if that was an app on the App Store, that's kind of too much. You'd need to download it. It's like it's too much. But here, it's just like this little mini-app thing that lives inside Wabi. You can try it. There's no cost to it. What about apps as memes or just purely for entertainment? Do you expect that to be kind of a category of itself, like this kind of idea of we can now make a mini-app that's really just meant to be kind of funny and make people laugh, but then it doesn't have to necessarily be durable or it doesn't have to be a standalone business itself?

3:04:35A hundred percent. I think we're entering this era where software is not going to be set. It's not going to be just the static. It's the app I created and you use it and you can't do anything like that. Instead, it's going to be this ultra-personalized, very malleable, very lightweight software that can change that is truly built on the platform of you. I would think that kind of the operating system of the future is the one where you'll have a few apps, big apps, like TikToks, Instagrams, of course, that will always exist. And then there may be a few apps built by the community that you discovered.

3:05:11Maybe some of them you remixed a little bit. Maybe some you actually built yourself. And then there are going to be some mini apps built for you by AI. Like, for example, if I'm going to New York next week, it should, Wabi should go through email and suggest an app, pre-build an app for you. Hey, I know you're into art. You're going to New York. Here's an app that basically recommends art shows around your Airbnb for the time. That's really cool. I feel like, I feel like this is a product that, uh, once people have that, that like people will have a moment where they're like, wait, why did I, I mean, the, the idea that you could go on a trip and generate an app that was almost like some quasi uh even just tracker for your trip of like oh these are all the things i want to do and check them off and then you can make so many other layers that you can you can build on on on top of it uh how much are you what what do wabi like how much do you feel like you'll need to incentivize creators on the platform early on like would you ever create like some type of like creator fund to incentivize people to come on and experiment?

3:06:22I mean, a lot of big social media platforms have done this over the years. You guys have raised quite a bit of cash, so I'm sure you could soft circle some of it, but maybe you don't need to at all. Curious how you're thinking about it. I think you're going in the right direction. That's what we're also thinking about. How do we incentivize the best creators to create the best apps out there. And what really struck me the other day was that actually it's easier to create a cool mini app than it is to create a YouTube video, a great YouTube video. It doesn't require that much production in a sense.

3:06:59Like for example, even yesterday, I was putting my daughter down and it was bedtime. She likes to play these puzzles where I give her a puzzle and she tries to guess what it is. So we wanted to just find something online for it, but it took me, it was a lot faster. It only took me a minute on Wabi to build this simple mini app where it basically is a puzzle and then just four pictures that she can click on. And because she's into Princess Elsa, we made it Princess Elsa themed. And because she goes to an Italian preschool, we made it Italian, in Italian. And so it literally took us like a couple minutes And it was like co-creation.

3:07:38We're creating that app together with her. So think of it as like Roblox meets personal software in a sense. So I do think, to a degree, right now it's almost becoming easier to create something than even to search for it online. Yeah, we got to hang out with the CEO of Roblox yesterday, and I think he's coming on the show tomorrow. How much do you think it can be dangerous to be too inspired by, in this case, something like a Roblox? Because in so many ways, I can imagine that an economy forms around Wabi in the same way that it has around Roblox. There's so many different ways you can kind of draw comps, but sometimes it could be a trap to be too inspired by something because ultimately it's a totally different kind of use case and value.

3:08:30than maybe Roblox has. I mean, our vision is different. Our mission is to set software free. And I think kind of the main, main inspiration for Wabi was that idea that we live in the Microsoft DOS era of AI interfaces and the GUI moment for AI is around the corner. And we do believe that with this almost like godlike technology that we created, it will require a different type of an interface. There will be the Windows, the macOS for AI. We're not going to be using AI through a command line forever. And so I think this is kind of where our North Star is. It's like how can we create a different interface for everyone to tap into all the capabilities of these AI models?

3:09:21Yeah, it's going to be really fun. I feel like we can sit here and predict how things will play out, but we should just have you back on once you have real data on how people are using it, how much is entertainment, how much is productivity, what the top apps are, and kind of how the community takes shape. Because I imagine there's only so much that you can predict once you build a tool. Yeah, that's the thing with Roblox. It's like, how would Roblox predict that Grow a Garden would be the most popular game in the world with teenagers? Yeah, or like a full reenactment of Call of Duty in Roblox.

3:10:00Like, that was probably not on the roadmap from day one. That's the beautiful thing about starting these platforms. But congratulations on the round. What's the waitlist timing? I'm on the waitlist now. What's the strategy around opening it up? Soon. We're building a few really exciting things. Right now, working on multiplayer as we speak. I think one of the really cool things about Wabi is that it's going to be that all apps come with multiplayer included. And even just simple mini apps. Like, for example, I built a Wordle game to test it with my mom so that we can play together. Obviously, we played on the New York Times, but it's pretty much like our way to connect with each other.

3:10:43And with Wabi, we do have a leaderboard, and we can see who's beating each other, who's better today, who's doing better. We can set up these notifications and all that. So a few things that we want to build before we roll it out to general audience, but we're getting fantastic feedback from our first users and it's honestly really exciting to, again, to be in that zero to one stage. I'm so excited about this. Amazing. Well, as soon as we have access, we're going to have Tyler make something like six or seven mini apps. We'll send them to me. Thank you so much. We will. Great catching up. I'm excited.

3:11:28We're excited you're going zero to one again and super excited about Wabi. Yeah, this is really cool. We'll talk soon. Thanks so much. Thank you so much for having me. Have a great rest of your day. Cheers. Bye. Before we bring in our next guest, let me tell you about Basil. Your bezel concierge is available now to source you any watch on the planet. Seriously. This morning, we're not going to dox him, but we ran into a founder we like who had quite a nice IPO a few years ago. And we did do a wrist check. Oh, we did. And he performed. It was great to see. Yes. Well. At the cover of the Wall Street Journal, there's a bidding war for obesity startups.

3:12:10Did you hear about this? Pfizer and Novo Nordisk boost their bids in an unusual fight for MetSera. I feel like when you say obesity startups, it's going to be the Wigoovy white labelers. And I was imagining this would go like HIMSS Roman, like that crew. But apparently, they are all fighting over the developer of the obesity drug. It's called MetSera. The latest twist in their unusual fight over the startup. MetSera said Tuesday that Novo Nordisk's proposal values the biotech company at up to$86.20 a share, or approximately$10 billion, up from its previous bid of$9 billion. So they're going back and forth.

3:12:53The valuation represents an approximately 159 % premium to MetSera's closing price on September 19th, the last trading day before the Pfizer deal was disclosed. So Pfizer's new offer value shares at up to$70 a share was$8.1 billion. Pfizer had previously struck a deal at$7.3 billion. They're going back and forth. The duel between two of the world's biggest drug makers over a three-year-old company, three-year-old company with just more than 100 employees and no approved drugs, shows the importance of weight loss drugs to the pharmaceutical industry. Weight loss is one of the hottest categories in prescription drugs, and companies big and small have been lining up to be part of the market with leaders Eli Lilly and Novo Nordisk.

3:13:49The global market for the drugs is now worth$72 billion, according to TD Cowan, which projects it will reach$139 billion in 2030. it's always so funny because i'm like oh 72 billion tyler 72 billion dollars a lot no wait like you mean what google's spending on capex just this year like that's just what just what one hyperscale is spending on capex is the entire weight loss market big tech is pretty big well we have our next guest the there he is what's up guys how you doing great welcome to the show Great transition. How are you guys? You just slid in there. Slid in fantastic. We're doing well.

3:14:29It's great to see you. We're doing well. Congratulations on the new deal. How did you meet Eugenia originally? Well, I mean, it's hard to not know her. If you guys do know her, what you do, she's a total badass. She's been in and around the ecosystem for years. She was kind of the first AI native consumer founder. Replica launched prior to ChatGPT. So perhaps ChatGPT was even inspired by Replica. companionship is now one of their biggest use cases. So she's been in and around and well known to the entire firm for many years. So when I heard that she was up to something new, I knew that we had to be a part of it.

3:15:02Very cool. This feels like a, what I like about some of my favorite ideas are not ideas that no one has had or talked about before, but they're ideas that have been kind of floating around, but nobody just decided to like tackle them in a really intentional way. This feels like one of those ideas that is kind of perfectly pulls together kind of the potential of this new technology that we have. The fact that you can generate software on the fly. You can make software so quickly now that you can make things that don't need to have a long, useful life. They don't need to be commercial at all. They can be software that would have taken six engineers a year to build.

3:15:46You can now build it in a few minutes at times if you're really one-shotting, and just two people can enjoy it. So this felt like, yeah, it just felt like Eugenia is the right person to take on this idea that has so much potential. Of course it's consumer. Of course it's going to be really hard. But this is like an idea that I want to exist. What was your kind of framing around this concept? had you been thinking about it for a while, and what kind of made you do the deal? Well, I'll give you two framings, Jordy. So the first is, when did software get so serious? It's like all the software we use is so clinical.

3:16:26Even when you see content inside Insta or TikTok or YouTube, it's the same frame, and it's sort of got all the same aesthetics. So if you think about the early 90s internet, it was really eclectic and weird, and I, for one, think a weird internet is a more interesting internet. So I think, one, there's a sort of software as a form of pushing the cultural edge. We've lost that a little bit over the last 20 years. I think the other thing is that the mistake we keep making in AI is that we think these things are markets when they're industries. AI code is not a market. There's not one winner. It's an entire change to the software supply chain.

3:16:59It's an industry, which is why Codex is working, Cursor is working, Replit is working, and Wabi is going to work as well. I think it's both an insanely delightful and approachable idea with really serious implications at large scale.

3:17:15What are you, like, it's hard to predict these kind of things. Like, I feel like with consumer, I was trying to get at earlier, is this going to be something like people are just playing around making things or is it going to be a mini app that kind of breaks out and goes viral over on TikTok and suddenly Wabi has, you know, an insane amount of downloads just because of a single video? So I can see it being this kind of, like, virtuous flywheel where people just enjoy creating and then they create something cool and then a bunch more people come in. But what's your kind of, how have you tried to predict the future around what will allow it to break out?

3:17:55Yeah, it's hard to predict. If you think back to YouTube, that's why I love the YouTube metaphor in 2005. You look at the shaky home video sort of, videos that were put on there, home camera videos and you pirated content and all kinds of random things. You could sort of easily look at that and be dismissive of YouTube and say, hey, this is non-serious. Hey, what are the YouTube native shows going to look like? Is that even a real thing to think about? And you now look at, it was hard to look at that and see Mr. Beast being implied in it. And the same thing is true today. I made a spooky meditation tracker for Halloween last week to track my meditations.

3:18:28There's a lot of fun AI creative mini apps. A great example of that is last week for Halloween, a big trend on TikTok was this sort of ghost face killer AI videos where you take a photo of yourself, you give it this very long and elaborate prompt, and it generates a sort of Scream-style, 2000 eras, you on a bed, the ghost-faced killer, the guy from Scream behind you in the doorway. And to actually replicate it, it looks amazing, but to replicate it, you've got to share this thousand-word prompt. That makes no sense. So now there's a mini-app for that on Wabi. It was super fun for Halloween. It's probably not to be relevant for another 51 weeks and then it'll be hot again.

3:19:05So by lowering the barrier to making software and lowering the sort of seriousness, you get this sort of explosion of really interesting native disposable perhaps apps that wouldn't have existed otherwise. Makes a ton of sense. I can't wait to get access. I wanted to ask you about, not to switch gears too much, but I wanted to get your take on the AI, kind of like gen AI and music space. You're on Spotify. I have some of your songs saved. Am I saved? I'll be on a drive and you pop up. But how are you thinking about that category and where you kind of see it going? Is it we were having a debate?

3:19:48I think we're having Mikey on from Suno later this week, but we've been having a debate. Is it just people that are playing around with these tools and enjoying the process of music making? which has always been a big part of the traditional music industry. Is it people listening to the music that they're making? Are they listening to other music? But what's your read as a musician? It's actually very, music and software are no different. People want to participate. And here's what's misunderstood about music. If you look at why was peak sort of record sales and music revenue in the 1990s, it's because that's when we had mixtape culture and mix CD culture, right?

3:20:21People were able to participate and make their own music and share it with their friends. And you sort of had this whole long tail of music being created. and then music somehow with streaming went back to being broadcast the way it was in the 60s and 70s before the cassette was invented. Software is actually very similar. Where we have software today, 20 million programmers in the world decide what 6 billion of us use. So I think the thing that drives love of the media, whether it's software or music, is the ability to participate and that's why people love AI music. So that's just like a general steepening of the power law between creation and usage.

3:20:57because with the Spotify AI music, yes, there's tons of people who are using AI to create songs, so there's more creators than ever. But then there are also AI music artists that are publishing and climbing the charts, and I wouldn't be surprised if we wind up having some crazy power law outcome, much like Harry Potter Balenciaga took over the internet for a day. That's right. And, you know, people are so, it's always the same criticisms, which is, oh, it's not real. It's not real. Like TikTok's not real, right? We're interested in entertainment and like whether it's real or not is completely secondary.

3:21:37I think the important thing is people are participating and that's, it's not just going to be a music thing. It's going to be every form of media. It's going to be software. I mean, that's the whole thing more people can now create. Yeah. Speaking of TikTok, do you, is it a prerequisite for growing a new social network to piggyback on another social network? Feels like TikTok did that with the watermark and just buying a lot of ads on Snap and Facebook, maybe. And then Threads, it seems like it's actually sort of working, but it's heavily piggybacked on Instagram. Ken, is it possible to just grow a new social network without having some unfair advantage?

3:22:13I mean, have you tried posting anything to Threads? Working might be generous. Yeah, I have. It's certainly not where the tech community is right now. It's where Connor Hayes is, my boy, running it over there. I would agree with you how much it's working, but I think the point still holds, which is like, I don't know. How could you even try? I mean, have you bet on a lot of it? Have you made a ton of, like, I'm going to start this new social network investments? I feel like A16Z is the place that takes that risk, and this is sort of a new entrant into that. Yes. Yeah, look, when networks work, what emerges are network economies, and that's the gold standard for everything, for modes, for business model quality, for increasing returns to scale.

3:23:02What's so interesting about AI has mostly been tools. There aren't really that many AI-native networks, and that's what Wabi is. That's one of the reasons we were so excited. But I think this general trend from tools to networks, it's a really important one to watch, and it's been this sort of like increasing topic within the investor community. So yeah, no, look, as for whether they need to be bootstrapped off of others, I think that the existing networks have become a lot more sensitive to that. And they're very cautious around emerging networks bootstrapping off of their networks. So I think it's a tactic, but it's a hard one to execute.

3:23:36You've got to kind of think about owning networks in new categories. Yeah, I mean, we see this even with just like the policing of links across networks. Like you can imagine it's so easy to like, okay, yeah, it exists within the app, but that you can share with a link. But all the, I mean, the walled gardens, people are, there's a bull market in building walls around gardens right now. Like I have friends that use TikTok and they send me TikToks and I don't have the app installed. And so I'll click it and it'll open in a web browser, but the web browser just will not play it for me. There's no button that I can click to be like play in the web browser.

3:24:10The only, it's just like download the app or get lost. And I'm like - You have to send it to our one friend who's on TikTok and say, can you send us a voice note describing this video? Just describe it. It's terrible. It's a mess. It's a mess. Mark said that, you know, voice notes are a form of violence. So that's a whole other topic. Yeah, he did. Look, guys, I do think the networks are really sensitive to this. If you look at it, you know, one possible explanation is that it's hard to actually build a network today. So the new network effect is the externalization about it. So if you look at MidJourney, MidJourney is not a network per se, But what happens is when MidJourney is consistently a part of the conversation on X, people are making tutorials on YouTube.

3:24:52If you dominate networks organically, that is the network effect today. It's sort of what Cluely does. So I think that's one approach to it. But no, man, it's like the last 10 years of books and blog posts have overtrained every executive in PM. And they're all like, hold on a second. No way we're going to let this get bootstrapped off of our audience. Yeah, the thing that I'm interested to follow is Sora created this new magical experience for creating AI videos, specifically the Cameo feature was like the innovation in my view that like was not, it wasn't just like model quality that made it broke out.

3:25:28it was like, it was good at making a certain type of like internet style, internet native video that, uh, that, and then it also innovated on the, on the feature level. And that's what caused this kind of breakout. The challenge is like, they made a great tool, which is what Wabi is doing and plans to do, but they couldn't retain the experience on the platform, right? Content wants to be free. Content wants to go everywhere. And so they created this amazing tool. Will they create a network? That's like TBD, right? But I think with Wabi, it's like you can create the software, but then it's self-contained.

3:26:07So if other people want to consume that software, they'll actually have to stay on Wabi, which should give it a real shot at building that kind of network and consumption. That's exactly the bull case. Actually, the interesting thing about Sora is that you've got Cameo as a new content primitive, but then you've got a question of what is the status game. You guys, of course, have read the famous essay We're All Status Seeking Monkeys. You have to have a status game. It might have been comedy for a minute on Sora. Without a status game, I think people quickly lose an incentive to continue to participate.

3:26:40I think the nice thing about Wabi is it's not just that creation consumption happens in a single platform, but the status game is creating the most opinionated, strange interesting app and because social is built in you can get all the kind of consumption and network effects on the platform yeah yeah that makes a lot of sense tell eugenia that we want to be at the top i want to be at the top guys i'm actually gonna i'm gonna make a tvpn app that creates the trading cards after this i'm gonna send it to you because i've always had trading card fomo man i want to be on one of those things oh yeah yeah make make one make one I'm planning to make it.

3:27:17We got to get Eugenia on there first though, man. She's a genuine badass. There's two ways to get on a trading card. You can do like a specific deal or something like that. Or you can get married. Or you can do something just normal in the real world, you know? I mean, I was tempted to actually make the round much, much bigger just to hear the gong. Okay. Oh, we're all hitting the gong for you guys. We love hitting gongs. This is not, we hit the gong for all major. It's fantastic. Love that gong energy. Thank you. Yes, super exciting one. Great stuff. Good to catch up. Thank you so much for hopping on the show.

3:27:53We'll talk to you soon. All right, lovely. Let me tell you about Wander. Find your happy place. Book of Wander with inspiring views, hotel great amenities, dreamy beds, top-tier cleaning, and 24-7 concierge service. It's a vacation home, but better, folks. Our next guest, we have David Risher. We're deep in the fourth hour. We are in the fourth hour, but we have David Risher. They said we couldn't do it. Aaron Brewer from Lyft in the Reacher meeting room. Welcome to the show. How are you doing? Congratulations. Massive earnings call. Looking good. You guys look more like you're kind of calling in from like a radio show or something like that than us.

3:28:28I love this setup. This is great. That's how I feel. I feel like we're like the morning DJ crew. Yeah. It's like day four mornings. Hot hits all day long. I love it. There you go. There you go. Take us through what happened today. What's the news? Okay, so just off our earnings call. So for those of you who don't know, a public company, I'm David, by the way. This is Aaron. Hi. And CEO, CFO. And what we do every quarter is we spend about an hour or so talking to analysts about what just happened the prior three months. So here's what just happened. Best earnings ever. Highest sales ever. You know, more active riders than ever.

3:29:10more driver hours than Heather. And here's a crazy fun fact. I love the air horn. So, but here's a crazy fun fact. So when Aaron and I started about two and a half years ago, we were sort of losing, you know, maybe$300 million in cash every year. Now we just had our first billion dollar over the trailing 12 months cashflow generation. So just a huge, huge... John, hit the gong for that. All right. That is amazing. We don't get to hit the gong for free cash flow as much as we'd like, honestly, on this show. So what's the point with the cash flow? Are you... We're zooming all over the place. We've talked to some folks who are in similar situations.

3:29:55They flip their business around. They start generating cash. Then it's time to buy back stock. It's time to reinvest in the future. It's time to pay dividends. What have you done in the past, and how are you thinking? How do you even make that type of decision? Yeah. Yeah. Happy to, you know, first of all, as David mentioned, like what a privilege to be here now, two and a half years later, be in this position where we're generating cash. We also guided out a few years in advance and said, we're going to generate even more cash than we thought over the next couple of years. So, you know, CFO's dream here.

3:30:29Over 2025, you know, if you think about what we've done, we announced our first share buyback program in the company's history. We said we would, it's$750 million in size. We said we would use about$500 million of that by May of 2026. In fact, we're going to, you know, things are going great. We're going to actually use that$500 million by the end of 2025. So, and then we've acquired two companies. We bought FreeNow in Europe, leading taxi provider in Europe. And then we most recently announced the acquisition of TBR Global Chauffeuring. We announced some investments we have coming up at Waymo.

3:31:04So if you think about that overall, pretty balanced, but very, very focused on, you know, we are a growth business. We are in growth markets. And so having the opportunity to invest against that, to do it in a way that's very conscious of driving value for our shareholders, not in terms of the growth, not only in terms of growth of the business, but in deploying some of that back in share buyback. So we like the balance. The question on everyone's head, on the top of everyone's head, I'm sure, is autonomous vehicles. You're probably getting AV questions constantly. Maybe set the table for us. How about another?

3:31:39Yeah, just walk through the high-level thesis, and then maybe we can dig into some of the particulars of the strategy. Yeah, 100%. It is the number one question everyone's asking. And you don't. Look, what a privilege, right? Because how often do you get to be in an industry that is clearly going to be transformed? Okay, so what's going to happen? We think the market size is going to grow dramatically. Like we think AVs for the sector, I'm not just talking about for lip, but for the sector, because it's a good product. If you've been in San Francisco or some of the other places where you can see AVs, it's safe.

3:32:12You can kind of zone out in the back seat. You can, you know, whatever you want to do. So, anyway, so it's a cool product. And so what that tends to do is it tends to expand the market. Okay, so what are we doing? You know, we're not in the business of making cars or even making AV tech. So we've got to partner. Who have we partnered with? Well, we've partnered with two of the biggest in the world. Waymo in the United States, who's clearly the market leader in the US, and Baidu in China, who's clearly the market leader there, that'll help us get cars and tech into Europe. And what you'll see is we'll integrate these onto our platform and create what we think of as a hybrid network.

3:32:44A lot of drivers on our platform, 1.5 million drivers, 1.6 million drivers on the platform, they're going to continue to drive because there aren't enough AVs in the world just to satisfy the demand with AVs. But then in certain markets like Nashville and others, we'll have more and more AVs in the platform and create something where holds greater than the sum of the parts. When are we bringing track cars to the platform? You want to track a car, you want a Ferrari Challenge or some sort of BMW that's been modified for the track, and you want to rent that on... I love the way you think. John, we did a track day on Sunday, and now it's the only thing I think about.

3:33:21But I do think that the actual... The car ownership pattern might actually change dramatically in the future. And this is my thesis. It's half a joke. I own a track car. I maybe take an autonomous vehicle on a commute or specific things. For purely functional, but then you drive for enjoyment. human drivers in the car for specific rides where there's either higher demand or specific weather conditions or specific road conditions. And so I feel like a lot of people want to hit, you know, oh, the new technology is going to come in. It's going to be 100 % of the market on day one. And that's just never the way it plays out.

3:34:04So how are you thinking about the way it plays out? And how are you preparing for that transition? I mean, you're putting your finger on something that I think as human beings, we tend to do to a fault, which is we sort of think binary, right? It's this or that. You know, it's X or it's Y. And it rarely is one or the other. It's almost always both. And this is the reason why, you know, frankly, being alive today is so interesting is you have these new modalities. So, like, okay, look, over the last – I'm an old guy. So, over the last many years when I learned to drive, I would drive on a stick.

3:34:35I still have a stick. It's great, but I also have a Tesla, which is not a stick. In fact, it's the opposite of a stick. Did you ever drive stick in San Francisco when you were first learning? Because John and I were in San Francisco last week, and I was showing him the different hills when I was 16 and a half, 17. Driving stick and driving a manual in San Francisco really is like a coming-of-age moment. Yeah, and then reversing into a parking space. Yeah. I will tell you a story. I grew up on the East Coast, but our daughters grew up in San Francisco, and I still smell in my head the grinding of our older daughter on Divisadero.

3:35:18You know what I mean? It's just burned in my head. But that's what it's going to be like, right? So it's going to be, yeah, I want to go – I want to take a road trip. Of course I want to drive myself. That's super fun. I want to commute tomorrow. I'm just going to let something else do the thing, and I'm going to zone out. And that's the future we're building for. I'll tell you one more funny thing, which is another weird human thing. If you ask people, do you want a driver in the car? A lot of people say, nah, I'd rather just be a lawyer. If you ask people, what's your favorite Lyft ride ever?

3:35:45They'll say, oh, my God, I had this amazing conversation with this driver. It was all about his life. He asked me about my day or whatever. That's the future. The future is not going to be just mechanical and virtual. And it's not just going to be physical. It's going to be this blend. And that's what we're building to. Talk about FlexDrive. Talk about how complicated the value chain is. Even if we see amazing performance in AVs and we get tons of autonomous vehicles on the road, it feels like there are tons of things that Tesla's not going to want to do. There's tons of things that Waymo's not going to want to do.

3:36:19There are going to be other competitors and other differentiators. Walk through the whole supply chain, how it fits in, how you're planning to fit in. so this touches on something again so important and i'll use a funny analogy for a second uh i don't know why but for some reason like my tiktok stuff or my youtube stuff like often they'll have like here's what disney world looks like to you and then they'll like zoom they'll like they'll do drone shots and be like that haunted mattress looks amazing like this huge building behind it looks nothing like a haunted mattress just a warehouse with this same thing with avs like it looks like a sort of a seamless digital thing but behind it as you say well humans driving remotely.

3:36:57Well, there is some of that. Sometimes. Okay, that's interesting. Sometimes. Maybe one person per five cars, but, you know, and we're getting there better, but the teleoperation will be occasionally. The teleoperation is the thing. It's valuable. Yeah, it makes sense. Well, but look, there's also someone's got to clean the car. Someone's got to charge the car. Someone's got to maintain the car. Someone's got to reboot the car. You know, so there's all that. That's called fleet management. We do it through a subsidiary called FlexDrive. Lyft has about 15 ,000 cars today. Today, these are not EVs.

3:37:25Today, these are EVs and normal, you know, ICE cars that are being managed by FlexDrive and kept available for drivers to use when they don't want to use their primary car. So that's a piece of, you know, it's a skill we have. It's great. And then you've got to keep those cars, if they're available, you've got to keep them utilized. You've got to keep them, you know, the demand coming in and matching that supply 24-7, even if it's raining, even if, you know, concert just got out, you know, even during the Super Bowl, you know, in the last five minutes of the Super Bowl where demand goes through the roof.

3:37:54So anyway, what you're referring to or what you're kind of alluding to is that rideshare is complex and it's very big scale and you add EVs and it's only going to be more complex. But this is what we do. This is sort of our jam. And we're super excited about this, the future for it. Yeah. Yeah. No, it seems like you're set up very well with the cash flow to actually play in other industries. It's going to be a very exciting time. Jordy, do you have anything else? Yeah, just how have you guys, like, what's been the focus internally with the team? It feels like the best way to build confidence as a team is to just, like, steadily execute against your guys' plan and roadmap, and you certainly have done that over the last couple years.

3:38:36I imagine when both of you guys took this job, there were some people that were saying, like, why would you sign yourself up for this? It's much harder to look like an incredible CEO or a CFO with a company that needs some kind of new momentum. And it's easier to look like a genius if you're joining a business that's growing at, let's say, 300 % a year or something crazy like we're seeing in some of these AI companies. But yeah, what has been the kind of unlock from a team or culture standpoint that has allowed you guys to deliver like this? I'll jump in here from my perspective, you know, thinking about joining Lyft.

3:39:20You know, number one, just incredible brand, right? This company just has an incredible, durable brand. And it operates in a market where scale matters, right? And so you continue to grow scale. You can drive very attractive economics. and then certainly as I joined the company, you know, this company has an incredible culture, right? We have a tremendous teams here, incredibly smart, really, really focused on our customers and inclined to always do the right thing, right? To think big, to do the next right thing. So, you know, honestly, the first six to nine months, I think for both of us was just, you know, if you think about it, the company was recovering from the pandemic, founders stepped aside.

3:40:03It was a big time of transition. So focusing on execution, focusing on just serving the customer, delivering a great service, and then taking it really step by step because that earns you the right, right? You build that discipline, that success. That earns you the right to do the next thing. Now you start doubling down on product innovation. For drivers, for riders, we start launching a slew of new products. That's super exciting for team members, for our riders, for the community out there. then we start generating a lot of cash. Great. How do we build from there? So, I often reflect on this a little bit with David.

3:40:41There was sort of an inflection point of the company, I think, when we both came in. And I'll speak for both of us. I think we both feel it now. It's like this next big leaping off point. And yeah, it's exciting. It's a fun place to work. I'll just say a tiny, tiny bit more. I mean, when you come in, look, you got to cast a big vision so that people get kind of excited about it. And we had a big vision around customer obsession and around serving and connecting. You got to make sure you got the right people. And I got super, super lucky. Aaron was one of the first hires I made, one of the best hires of my life, right?

3:41:13So you got to have that right team around the table. And our management team is incredible. And then, as Aaron says, you just got to execute, execute, execute. And that success begets success, begets success, unlocks opportunity. Now we can think even bigger. So we're just kind of getting started two and a half years in. I love it. I love it. I love it. I love it. success. I love seeing you guys win. Congratulations. David, we really enjoyed our last conversation. Aaron, it's great to meet you. Looking forward to next quarter. We'll talk to you soon. Well, I know that David was mentioning he wants to get the earnings call on your show.

3:41:45This is the closest we could come. Thank you for having us. That was awesome. Super fun, guys. Congratulations on an awesome quarter. Hey, good to see you too, by the way. Did you guys write that? We did not. all Mike Isaac it was a lot of fun we survived we did great to see you guys thank you take care let's check in with the timeline Stefan was saying please ask them about the biggest fish they've caught that should be an ongoing question what's the biggest fish you've caught what is the biggest fish you've caught Jordy have you ever caught a big fish? I've only gone fishing on the ocean you know that's where they have the big fish they do have the big fish there but I went out on I tend to like doing weekend activities that sound strange but are like constrained to like a few hours and going out we were trying to go for bluefin tuna this was sometime last year and we spent like 12 hours out there did not catch wait before you Tyler do you ever do a bet going do you think Jordy's biggest fish is over under 30 pounds what do you think uh I'm going under you're going under 30 pounds okay Jordy give us the final answer how big was the bluefin tuna that you didn't catch we did not catch a bluefin so have you ever caught a fish I don't think you must have like gone in a pond one day and got a minnow right I don't think I don't think I've ever you've never caught anything i don't think i've ever caught you've never caught a fish in your entire life but my my dad was wow into fishing wow we went we would go camping we need a trading card up never caught a fish i mean we should we have to we have to change this this is important yeah i it's it's funny growing up around the ocean surfing my whole life yeah never never thought to just throw in the old wine with the hook.

3:43:54It's so easy. You live by the ocean. You live walking distance to the ocean. This is a challenge for you. Today, maybe this weekend, walk to the ocean, catch a fish. Catch a fish. Catch a fish. You love Nobu. You go to Nobu all the time. You never thought, what if I did it myself? What if I DIY Nobu? Safe as Shinke catches my fish. That's true. That's true. Yeah. John, what's the biggest fish you've caught? 150 pounds, baby. Halibut. I was 10 years old. It was fantastic. The crowd goes wild it weighed more than me at the time which was electric because it was truly like conquering a beast that was bigger than me which is like so satisfying uh i so i need to look into this tyler maybe this is a good question for you hit the deep research buddy get ready get ready to research buddy but uh uh at my uh at my local beach place that i normally surf in malibu there are tons of leopard sharks and they hang out where the water is like three to four feet deep so after you catch a wave if there's good visibility you almost like every other wave i'm seeing a leopard shark and usually i just kind of just jump back on my board and paddle out because they're like i don't know they're like three or four feet they're not they're not massive or anything but since you've challenged me to catch a fish maybe the next time i jump down and shark is technically a fish yeah yeah that would count yeah they are yeah they're yeah so So I should just dive down and catch it with my bare hands.

3:45:20I like this. Maybe we could. If that doesn't work, I'd recommend trying to shoot fish in a barrel. I've heard that that's pretty easy. We'll get you a barrel. We'll get you a barrel of fish. And you can. Oh, I did. You can pull an AR-15 and just start shooting the fish in the barrel. You know, shooting a fish in a barrel, it must be. It's actually pretty hard to do, I imagine. Depends on the barrel. If you use a big gun, you're going to break the barrel, and then the water's going to go everywhere. My beautiful puffer fish. You caught that? My pride and joy. That doesn't really count. I want to see you catching an Atlantic bluefin tuna.

3:45:57They can exceed 1 ,500 pounds, with some estimates reaching up to 2 ,000 pounds. Wow. That's what we need to do. Anyway, thank you so much for listening to the show, for watching the show. Leave us five stars on Apple Podcasts and Spotify. Subscribe to our newsletter at tbpn.com. To be honest. Kind of want to keep going. You want to go to the fifth hour? I kind of want to break the fifth hour. We have guests here. We do have guests here. We have to hang out with our guests. Thank you so much for watching. I can't wait to be back tomorrow. John's cut me off. John's cut me off. We did get into a good riff right at the end.

3:46:35It was fantastic. Amid 10 guests today. New question. tomorrow we'll ask every guest we'll ask Dave from Roblox we'll be joining tomorrow we got Vlad from Robin Hood what's the biggest fish you've ever caught this is important we're doing a super cut how big is the fish you've ever caught the biggest fish and then maybe we'll create a boom in fishing maybe people will go fishing that's right thank you so much for watching we'll see you tomorrow goodbye