# Instinct Worth 30BN? | Microsoft Cuts 27,000 Jobs While Meta Taps Equity Markets for $95B CapEx 20VC with Harry Stebbings · 2026-10-29 Harry Stebbings, Jason Lemkin and Rory O'Driscoll argue Anthropic's S-1 shows a company investors are still underestimating, with $16.3 billion first-half revenue and $110 billion of cash, though 80% of revenue comes from models under six months old. They break down Microsoft's buyout offers to 27% of its 223,000 employees, warning 2027 CapEx depreciation will force more AI-driven layoffs, and Meta raising $95 billion of equity for its spending spree. Jason argues Chinese open-weight models are finished in the US after Alibaba's Quen was caught uploading data to servers in China and Singapore. They debate Instinct at $30 billion as the 'super app of the West', whether OpenRouter's boom is a moment in time, and Jason's pitch to put 20% of the fund into OpenAI at $1.25 trillion. ## Questions this episode answers ### What was the scariest metric in Anthropic's S-1? Rory O'Driscoll notes 80% of revenue comes from models less than six months old, and no true gross margin was given since roughly 70-80% of compute costs go to training. He argues this short shelf life is actually an advantage for the leader willing to keep aggressively building new models. [4:49](https://20vc.podhood.com/b2130823-3601-4c1b-aae6-991df8c14429?t=289000) ### Why is Microsoft offering buyouts to 27% of its employees, and what does it signal? Jason Lemkin calls it a scarlet letter worse than a layoff, saying Microsoft is signaling it doesn't want those people, though it's an attempt to be humane. Rory O'Driscoll adds that as the depreciation bill for AI CapEx hits in 2027, big companies will push hard on headcount to keep operating income up. [13:04](https://20vc.podhood.com/b2130823-3601-4c1b-aae6-991df8c14429?t=784000) ### Are Chinese open-weight models finished in the US after the Cohere Quen incident? Jason Lemkin says they're pretty much over in the US, calling it the beginning of the end. The Quen model was caught uploading data to Alibaba servers in China and Singapore despite claiming it wouldn't call tools, likely a training error, but no CISO will approve such models. He notes Anthropic is back on the GSA list with a 40% discount. [22:00](https://20vc.podhood.com/b2130823-3601-4c1b-aae6-991df8c14429?t=1320000) ### Would Jason and Rory invest in Instinct at a $30 billion valuation? Harry Stebbings would unwaveringly invest, seeing Instinct as the potential super app of the West with protected downside since Microsoft or hyperscalers would buy it. Jason Lemkin vetoes it, saying he needs three to four months of data and is tired of growth rounds with one month of traction on negative gross margin. [37:32](https://20vc.podhood.com/b2130823-3601-4c1b-aae6-991df8c14429?t=2252000) ## Key moments - **[0:00] Intro** - [0:00] Jason Lemkin: people are underestimating Anthropic ahead of its S-1 filing - **[0:49] Anthropic's S-1** - [1:17] Rory O'Driscoll: Anthropic's S-1 was great on margins, op income, customer concentration and its $110B cash pile - [2:29] Jason Lemkin: Anthropic reached IPO in 5 years vs SpaceX's 24 — why bother with 10-12 year seed deals? - [4:49] 80% of Anthropic's revenue comes from models less than six months old — Rory O'Driscoll on why that scary stat is actually an advantage - **[7:12] Debt & IPO** - [7:14] Jason Lemkin: Broadcom had to bail out Anthropic's chip debt deal — a sign AI debt financing is getting harder - [8:05] Rory O'Driscoll: the AI spend boom is being equitized — Meta issued $70B of stock, Google $80B, Anthropic ~$100B - [9:54] Will Anthropic have a monster IPO? Rory predicts a stock people just want to own, with instant index inclusion - [11:17] Jason Lemkin: OpenAI maimed itself — CEO drama and letting Dario walk let Anthropic own the coding market for a year - **[13:04] Microsoft layoffs** - [13:04] Microsoft offers buyouts to 27% of its 223,000 employees — Jason Lemkin calls it a worse scarlet letter than a layoff - [16:33] Rory O'Driscoll predicts 2027 is when the AI CapEx depreciation bill hits Microsoft, Google and Meta — paid for with headcount - [17:44] Cloudflare laid off 20% in its best year ever — Jason Lemkin says it's the new playbook every CEO will copy - **[19:28] Meta's raise** - [19:28] Meta taps equity markets for a $95B CapEx raise as its free cash flow goes negative for the first time - **[22:00] Chinese models** - [22:00] Jason Lemkin: Chinese open-weight models are 'pretty much over in the US' after the Cohere-Qwen data leak to Alibaba servers - **[24:48] Inference stack** - [25:05] What happens to Modal, Fireworks and OpenRouter if Chinese open models decline? Rory says US alternatives must emerge - [26:00] Jason Lemkin: the tipping point is when Elon Musk rips Chinese models out of Cursor — 'it can't run on Kimi forever' - [33:14] Jason Lemkin: OpenRouter may be a moment in time — Chinese open-weight models could fall from 60% to 10% of its traffic in six months - **[35:04] Black October** - [35:04] 'Black October': AI-exposed PHP flaws hidden for a decade mean every old website is now a hacking target, says Jason Lemkin - [37:32] Instinct raising at $30B on token revenue — Harry Stebbings argues it could become the super app of the West - **[37:33] Instinct at $30BN** - [40:31] Jason Lemkin vetoes Instinct at $30B: 'I'm tired of growth rounds with one month of traction — I need a hundred days' - **[45:11] Deal flow** - [46:56] Jason Lemkin's radical IC proposal: put 20% of the fund into OpenAI at the $1.25 trillion round as a dislocation bet - **[51:21] WarHounds** - [51:21] WarHounds at $1.5B: Paul Malucki's genetically modified war-dog startup priced on founder brand alone, says Rory - [55:10] Rory O'Driscoll: mega-funds raise bigger and come back faster because Anthropic made everyone feel amazing — 'risk on until we're not' - **[55:11] Fundraising & Zuck** - [57:20] Zuckerberg vs California's billionaire tax: lawyers, board drama and threats to sell his homes, reveals Jason Lemkin - [59:04] Rory O'Driscoll on billionaires and taxes: 'people like to give it away, but they don't like it to be taken away' ## Speakers - **Harry Stebbings** (host) - **Jason Lemkin** (guest) - **Rory** (guest) ## Topics Tech Market Trends ## Mentioned Anthropic (company), Cloudflare (company), Cohere (company), DeepSeek (company), Fireworks (company), Instinct (company), Meta (company), Microsoft (company), Modal (company), OpenAI (company), OpenRouter (company), SpaceX (company), WarHounds (company), Cursor (product), Quen (product) ## Transcript ### Intro **Harry Stebbings** [0:00] As crazy as this may sound, I think folks are almost underestimating Anthropic. **Jason Lemkin** [0:05] There is one story that matters more than any other this week: Anthropic may finally drop their S-1. This show, we break down what you need to know, what you don't need to know, the questions that really come from it, and what we can expect when it comes to day 1, and will there be the mega-pop of all pops. This, and so much more, in the show today with Jason and Rory. **Harry Stebbings** [0:26] I think that the Chinese open-weight models are pretty much over in the US. How much money is there for Instinct of the world? It feels infinite now. I think we should put 20% of the fund into OpenAI at the $1.25TRN round. Ready to go. ### Anthropic's S-1 **Jason Lemkin** [0:49] Boys, we are back. It is my favorite time of the week, and we have some big news to start with. Anthropic Files its S-1. It's the moment that we've been waiting for. $16.3 billion of first half revenue. I just want to hand over to you, Rory, because whatever metric I give, you're going to say it was the wrong one to cite. **Harry Stebbings** [1:09] No, I'm not. **Jason Lemkin** [1:10] So which ones were you drawn to, and what is the takeaways for us from this? **Harry Stebbings** [1:17] It was a great S-1. And you know, frankly, all the haters and the warriors, at least for now, have to go pound sand. I mean, the burden of proof is most definitely, you know, on the side of the naysayers. In other words, these guys, the S-1 was great on almost every dimension, and you know, it'll be widely well received. I mean, it was, yeah, it addressed a whole—I mean, there's some pending open questions still. They didn't give the Q3 revenue number. But you know, it was great on cost. I mean, the overall revenue numbers are known, but they were amazing. It was great, better than expected, on margins, op income, even op income after stock-based comp. It was great on customer concentration. It was great on compute efficiency. It was stunningly great on cash and cash balance sheet. You know, I mean, it's just, on every dimension, it looked really good. I mean, you have to struggle to find questions. And there are questions. We can come to those in a second. But it read, you know, it was a really top-tier performance. You know. I think, as crazy as this may sound, I think folks are almost underestimating Anthropic because five years to IPO versus 24 for SpaceX. It's not just something we've never seen before. We'll see hints of it with OpenAI's IPO,right? It might be worse. It might be better. Probably slightly worse,right, based on what we know,right? Less capital efficient, slightly slower. But JFC, to get here in five years from a spite startup, it's just, it's beyond unprecedented. Time value to liquidity still matters. I mean, it's markups are fun, man. But getting out of the greatest IPO in our lifetimes in less than six years to distributions, I mean, goodness gracious. Why? I mean, I know Rory will roast me for it, but you got to think, why would I do anything else? Why would I wait? Like, these seed deals are like 18 years to full distributions,right? Why? I could get to six years in full liquidity Anthropic, and why would I waste my time with anything that's 10 to 12 years to IPO, but I still—but, you know, it's just one deal in the fund, another two years to distribute, you know, you know, it's just versus five years. **Jason Lemkin** [3:32] No, I won't roast you, Jason. I mean, I think that, you know, I saw a tweet that basically said the same thing, as every founder and every investor is looking, why am I bothering with anything else? You know,right? These are such stunning numbers in terms of size, scale, and speed that you'reright. I mean, it's going to dominate the results for, you know, for a long time to come, and everything else just seems a lot more work for a lot less return compared to it. And you know, all credit to the founding team, all credit to the investors who got in early. It's just a great way to make money. Things going up and to theright faster than anything else in human history for five years to add up to a trillion dollars. That's as good as getting venture. **Harry Stebbings** [4:10] Yeah, it was funny watching Byron Dieter from Bessemer on the TV shows saying, this is the greatest investment in the history of Bessemer. I think Bessemer was founded in the 19th century or something like that. I don't know when the fund was,right? **Jason Lemkin** [4:20] Yeah. **Harry Stebbings** [4:21] And their greatest investment and their largest checks,right? So to see that all happen now after 100 years, I think that's the end of, that's the whole story,right? I mean, greatest investment, biggest check, fastest time to liquidity. Why are we doing anything else? **Jason Lemkin** [4:37] Rory, you said about questions there. There are always cynics in the audience. There are always people throwing peanuts. If you were to present the alternate view of, oh, here was a striking number or two, is there any that. **Harry Stebbings** [4:49] The interesting one, you know, 80% of the revenue comes from models that are less than six months old,right? And then separately, you'll note they didn't give you a gross margin, an actual gross margin calculation. They just didn't lay it out that way. Because you can do that in accounting. You can just put all the compute costs in the one place. And then they kind of gave a footnote breakout that kind of 70, 80% of the compute costs are for training, not for inference. In other words, once you've built the models are wildly expensive to build. Once you build them, they're fairly wildly profitable to serve. And you only get to serve them for six months, and then you got to invent another one. It's this kind of, it's not like, it's so not like software where you're on this kind of, you know, you build something and maybe you can sell it for 10, 15 years. In this case, you invent something, you get most of the revenue from it in 6, 12 months, and then you have to continue to invent masses of money to build the next one. And you know, then your question is, so what do you make of that? You know, because the funny thing is that that's the facts. And you could look at it one way and say, oh my God, that's totally scary. If you miss even once, you're screwed. And on the other hand, you can say to yourself, that might actually be great for Anthropic because they're best in the position to build new stuff, and they want to build new stuff. And they said that very clearly. We want to build new stuff,right? And if all the time, you know, very few people are going to be able to compete with this,right? Very few people, like if the next turn of the crank on the model, I mean, they spent $8 billion on compute, you know, year to date, June 26, and they're going to spend, I don't know, $16 billion next year. There's very few people who can write a $10, $12 billion compute check. So they can continue to invent these new models, get masses of revenue from them, and then, you know, yeah, maybe at the back end, other folks have it. So even though that statistic was kind of weird and scary, oh my God, the shelf life of your amazing product is six months. It's actually, I would argue, an advantage if you're the kind of leader and you're willing to just be aggressive on it,right? But it does point to, so therefore, what looked like a negative isn't, it just points to the incredibly dynamic and scary nature of the game. This is a game where you have to spend like billions and billions and billions of dollars just to play. **Jason Lemkin** [7:01] Yeah, that was Elon's point that most of these neolabs are just going to die after this,right? They just can't, it's not possible to raise the capital to survive,right? I thought there was one bad metric that wasn't in the S-1. I thought the worst metric wasn't in the S-1. ### Debt & IPO **Harry Stebbings** [7:14] Go for it. **Jason Lemkin** [7:15] The worst metric was that Broadcom had to bail out this debt issue so that Anthropic could buy its chips. That the market was saying, it doesn't get much, it doesn't get any better than Anthropic. We just read the S-1,right? It doesn't get any better, but they weren't able to get this debt for this semi-circular deal with Broadcom, and Broadcom had to come in and guarantee most of it, and they had to pay even higher interest rates than expected. It's not that it's terrible, but it certainly suggests that going forward fundraising, even for Anthropic and OPI, and even if it's somewhat indirect, it's getting harder,right? And that combined with the fact that it's getting harder and harder for the banks to lay off and sell the neolab debt is, I think that's the real, the only one to be nervous about, that the debt isn't exhaustible here. **Harry Stebbings** [8:05] Pushing back on that, because I do hear you. Because I would actually, I think the fact of the, I think the amount of debt the banks are digesting has hit the limit. But I actually think the interesting thing about this week and about the last few weeks has been, we're equitizing the issue. In other words, it was all debt. But if you look at it, Anthropic is about to issue plus or minus a billion, $100 billion of stock. You know, Meta issued $70 billion of stock. Earlier this year, Google issued about $80 billion of stock. Obviously, SpaceX did their IPO,right? What you're seeing is that there are a class of companies that have, that A, have cheap cost of capital, and B, are willing to take equity dilution to make it all happen,right? So my aha from that is, at the macro level, it's just going to keep on happening,right? This trend is not stopping. And the people who want to have it happen are willing to take equity dilution to keep going. So the spend will be there. You'reright, Jason. I mean, Elon's doing his thing of, you know, cutting off the oxygen from the other people. It's like, oh, you people who just have debt are screwed. You're going to die. I, on the other hand, have the world's, one of the world's most inflated market caps, so I can issue equity and, you know, get access to all the chips I want. So it was very much a kicking someone else while they're down move, which is very vintage Elon, and you got to respect. **Jason Lemkin** [9:25] But they'll have like $200 million of total cash on the balance sheet or something like that after this, yeah, Anthropic will. So that is, that's a lot of cash. **Harry Stebbings** [9:34] I mean, let's be precise. I think it is out of the first or second largest pile of cash in all corporate America, which is a start. **Jason Lemkin** [9:39] I read it was the second, but it doesn't matter which one it is. **Harry Stebbings** [9:41] This is a company going public. I mean, this is a company going public where it already has more money than every other company in America by one, at $110 billion on the balance sheet today. I mean, they've done a masterful job of raising $130 and only spending $10 or $11. Masterful. **Jason Lemkin** [9:55] Do we think it's going to have an amazing reception? **Harry Stebbings** [9:58] I do. I mean, I'm not sure. I mean, hell, I don't know, you know? **Jason Lemkin** [10:02] Oh, we're in the predictions game, Rory. Go for it. **Harry Stebbings** [10:04] I know. **Jason Lemkin** [10:05] Go for it. **Harry Stebbings** [10:06] I know. I think it should. I mean, I think it's an amazing company. You just can't argue with it. **Jason Lemkin** [10:11] I mean, doesn't it, like a SpaceX, almost instantly have to go into the S&P 500? **Harry Stebbings** [10:16] Not the S&P 500. If you're listing on Nasdaq, which I think it is, I didn't check, you go into the QQQ index, but the S&P index takes, there's a profitability metric or something like that. So it takes some time. But yes, they would be in some index funds pretty immediately. But I don't think this is going to need that lift in the same way arguably SpaceX did. I think this is going to be just a stock people want to own. **Jason Lemkin** [10:36] But even just getting indexed into everything in the net is just a retail boost,right? It's just, it's hard to bet against. They're not going to miss a quarter. Like, I mean, it's in the, whether it's 900%, I mean, within a range, like, it's not that I don't think they could disappoint,right? But within the next two quarters with, you know, and I, you know, I think we have to be careful to use NRR as a metric here, but it's still true,right? With 900% NRR, even if in reality it's more like 300 to 400% because it includes like pilots and trials. I just think that plus the brand, plus being in the Nasdaq 100, it'll probably have a good retail run. Plus SpaceX, I, you know, why would you bet against it? **Harry Stebbings** [11:18] If I saw a Moltman reading this, thinking about Q1 going out himself, do you think he takes away anything from it, changes anything as a result of it? **Rory** [11:28] You can't change the past. I mean, hard. I think you, I probably think if you were the CEO of OpenAI, you would look at this and go, decisions that you made, you now regret because you've let these guys get away from you and do better. So I think there's, if there's any self-reflection, there should be a ton of that going on. Nothing you can do about it. You just got to get your business together and converge and get out yourself. And in the meantime, obviously compete aggressively on price to try and screw up their Q3. **Jason Lemkin** [12:01] I mean, I mean, never has such a successful startup maimed itself so poorly that I can think of as OpenAI. I mean, it would have killed almost any other startup what they went through,right? With the CEO drama, all the other nonprofit-to-profit drama, the spite startup doing better than them. They basically let Anthropic control part of the market for almost a year, which is like five years or 10 years in AI time,right? I mean, so it's just, you know, everyone, I mean, there is, there is a certain amount of parity with Astra and the open-weights models today, the best, but Anthropic owned the, fucking owned the playbook for, I mean, the chessboard for almost 12 months. There was no point in using any other coding solution, and coding is the killer app, as Rory said a hundred times. The math is true. They literally letting Dario walk out the door rather than build it in-house and then firing the CEO and bringing him back over the night of long knives. I mean, my God, I mean, it's all going to be okay. But talk about, talk about own goals,right? I mean, every startup's full of them. This is just, this is, these are some epic ones. ### Microsoft layoffs **Harry Stebbings** [13:04] I had the joy of having Satya Nadella on 20VC earlier this week. Great show. It was a fantastic guest to have on. What an honor. Microsoft had some pretty big news that they were offering voluntary buyouts to 60,000 of its 223,000 employee base. It's an enormous amount of the employee base to get this offer, roughly 27%. Jason, you've spoken before quite vociferously about this topic. What are your thoughts? **Jason Lemkin** [13:32] Well, look, first of all, I mean, Rory may have observed more of these than I have, maybe he hasn't, I hope he hasn't, but it's not like all 27% are going to take it,right? And they've already been through a, I think they already went through a 10% layoff round,right? So they've done, it's some sort of interesting game theory here where you want to offer it to everybody you don't want to have,right? And then you got to figure out what the yield's going to be,right? But it's brutal because what it says one thing, no matter what people are going to say on LinkedIn, these 27%, we don't want them. **Harry Stebbings** [14:04] Yeah. **Jason Lemkin** [14:05] Whether 7% take it or 2.0, we don't, and there is no, this is a worse scarlet letter than a layoff, than a pure layoff. At least a layoff, you can kind of wave your hands. This is 27% of the country that said, do not meet our minimum, but we don't want you. And I bet it could have been higher, but I bet they got this done in a week. It's so damning that Microsoft doesn't want 27% of its people. That is the age of AI. They don't want them. **Harry Stebbings** [14:29] Why would you do that if you're Microsoft though? What a culture could it, what, so you stay and then know that you're not wanted? Isn't that awful? Like. **Jason Lemkin** [14:36] I think we're trying, I want to get Rory to say, I think you're trying to do this in a humane way. And I think what we learned is there's, you know, when layoffs started like in late '22 or so,right? Everyone, everyone was fried for doing them on Slack or by email. Oh, it's so inhumane. Go meet with people,right? Everyone was like, it's not that more humane to meet. Have you seen office space? Like it's not more humane to meet with someone in HR and get fired. And then they're like, don't do it over the holidays. So they fired people on January 1st. That was no better. There's just, so this is a way to like make it more humane. It's great for folks that were thinking of leaving anyway. It's great for them,right? It is great for the folks that are at the, either at the tail end of their career or they want to check out of the workforce. And that's why this is so hard to do from a game theory perspective, because you might get all the folks that were going to leave anyway,right? But Microsoft has so much cash, who cares? But it's an attempt to be humane. And I just, I've been on the other side, it's been a long time since I've been on the other side of a layoff. It's just hard to, it's hard, you try to be a good person, but you can't, it's hard to make it humane. It sucks no matter how you do it. And I think this is the suckiest way, because 27% of the company has just been issued a scarlet letter. I think this is the least, in attempting to be humane, this is the least humane way to do it. Just fire another 9% in one day, do it on teams and move on. **Rory** [15:51] It is rough. No, I think you'reright. I process the same thing. You've been told to go, but you decide to stay. The only winners are, as you say, the people who've got another job lined up and now have a call option on the severance package. But you know, it is interesting they think they can do without. You know, if their take rate is 10 or 20,000 people, it's clear, it's interesting they can do, but you know, they have around 200,000 employees. So they think they can do without another 10 or 15%, which, you know, possibly is true. I mean, you wonder how much it will be AI efficiency, how much of it is just you never needed them. It would be fun to find out. But yeah, no, it's, these large companies, as the depreciation bill for the CapEx comes due, you're going to have to find it somewhere else. And we've talked about this, you know, frankly, earnings this year, 2026, are in that golden period where the people who are selling the semiconductors are booking the gains, but the people who are buying the CapEx aren't booking the depreciation. So everybody's winning. But I think as you start to look to '27, that depreciation is going to start hitting in Microsoft and Google and Meta and places like that. And, you know, to keep the operating income up, you're going to be really pushing on headcount. And this is when the bill for AI comes due, and this is how it'll be paid. **Jason Lemkin** [17:06] Yeah, it's the brutal, in a sense, it's like, it may feed the anti-AI frenzy because it's the direct narrative of the story,right? We're going to have to pay for all this infra, to pay for all these GPUs and infra and inference, we got to get rid of the people. **Harry Stebbings** [17:21] And so you think 2027, we're going to see that really crystallize even more? **Rory** [17:25] Intuitively, as time goes by, you know, everyone's trying to call when the hyperscalers start kind of going down the other side of this. So yes, as the depreciation bill mounts up, like so far, even though free cash flow has been kind of going the wrong way, everyone's operating margins have been, you know, operating income has been going up. And yes, that'll get harder now. So yes, this will be the problem. This will be the issue. **Jason Lemkin** [17:44] I'm just surprised that the Cloudflare layoffs aren't more discussed, because I think it's case study 101 for CEO MBA,right? Every, every Cloudflare, Matthew comes in, epic year for Cloudflare. Every quarter is growing faster, but on May 7th, they lay off 20% of the folks, bam. We don't need it. We're too big. We don't need these folks. We are better off with 80% of the people than we were before, and it keeps accelerating. Like every CEO that's arguing about this should just look at, I mean, you can say it, Microsoft, some of it is the old company and it's, and we've got to pay for the, for all the AI, but Cloudflare is just like, this is just holding us back. 20%, it wasn't 8%, it was just boom, we're just going to do 20 and no downside to the business that is apparent yet. So I think it's just layoffs 3.0 or 2.0 is just going to happen. And part of it's to pay for the infrastructure and part of it is just enough already, enough of the performative AI, enough of this, and we just, we just, because it's just so, you know, think about every board meeting you've been in over the last week or if you've been in every, no matter how well you're doing, it's just stressful. Everything is moving even faster than the last board meeting. There's even more competition in every space. We just can't, we just can't carry folks any longer. We can't, or we're going to die. We're going to be killed by that new company Harry just funded or Rory just funded. There's no, there's no stability. You know, the thing, the tough part about the Microsofts and the Salesforces and the SAPs and every SaaS company back in the day is we were all predicated we would get to a level, a plane of stability at some point, 100, 200, 300, 400, 500 million of AR, we could kind of breathe. There's no planes of stability at all. Maybe, maybe north of 10 or 20 or 40 billion, they just don't exist. So I just think layoffs are just going to accelerate because we have no choice. ### Meta's raise **Harry Stebbings** [19:28] Speaking of having no choice and where cash comes from, Meta goes to the equity markets. Zuck needs $95 billion. And this week on Monday, it launched a raise of up to $95 billion, 35 common and preferred, 60 of the market program. And the thing to take from that in combination with the fact that chips lose 200 billion in a day and public markets in bloodbath as Neo Clouds shit the bed. **Rory** [19:59] No, the short answer, Harry, is no. I think there are two separate points that you merge there. And I'm just dealing with each of them in succession. I mean, I think the Meta thing, yeah, which is what we talked about. They, this was the first quarter where their actual CapEx dipped below their free cash flow,right? So their free cash flow negative,right? Because it, because remember, their core business is one of the best businesses on the planet. It spits out money. It's awesome,right? So they have, yeah, 33 billion a year of cash, a quarter of cash flow,right? It's a great, great, great, great, great business. So this is the first quarter they went negative. So they just did some equity. So that's that. And it's totally unrelated. The chip thing, first of all, yeah, I mean, chips moved last week, but let's be frank, since the Anthropic, I mean, I don't know where they'll be when this show comes out, but since the Anthropic S1 hit, chips have been bouncing up because the consensus is that looks good. And, you know, one thing that the S1 was good for was chips, because so those stocks are actually up. In fact, DRAM today is way up, which is the, you know, the definitive, the one I keep an eye on, just the list of the, it's the index of memory chips. And yeah, that's why Korea, Korea is up 10% today, SK Hynix,right? And this is Tuesday, by the way. So I actually think the news is, if you're in the semiconductor CapEx business, it's been a good back half of the week. It was scary two weeks ago with CoreWeave, but, you know, CoreWeave is a, I don't know, 30 billion dollar company and Anthropic is a $2 trillion company that has $200 billion when this deal closes and they're going to spend it all on chips. So chip, be happy. **Jason Lemkin** [21:31] The only slight counterpoint not to get off chat, it's all, it's all green, green, green, but it is just, I know it's a different ecosystem and I know it is apples and oranges, but the Star 15 China, which is their version of it, is down 40% this year. So we do have interesting markets that are out of sync here for AI,right? Because China is just as gangbusters in its own way, but it is interesting that that one's down 40. Different market, weird liquidity while ours is on fire. ### Chinese models **Harry Stebbings** [22:00] Speaking of China being on fire, DeepSeek raises about $15 billion, largest ever private Chinese round, by the way. Largest investors, battery maker, I don't know how to pronounce it, CATL and Tencent. Anything of note here? **Jason Lemkin** [22:16] I think that the Chinese open-weight models are pretty much over in the US. I think it's the beginning of the end. I'm not sure. DeepSeek is great that it raised this, but it's beginning to be banned across all federal agencies. And there was a horrific thing that happened this week that was over-sensationalized, which was this cohere Quen thing. I don't know if you guys saw it. It's pretty terrible, but it's terrible on a number of, but it's not as bad as it looks. **Harry Stebbings** [22:41] What happened, Jason, for people who don't know? **Jason Lemkin** [22:44] Well, listen, Cohere is now a fourth-tier LLM,right? Fourth-tier player that has only sovereignty on its side as near as I can tell. Harry, you're welcome to correct me or Rory,right? But it has sovereignty. And sovereignty is, if you talk to anyone at the enterprise, it's in the top five list. And so they found, they found that over that Quen, Alibaba's existing Quen model was calling to Alibaba servers, both in China and in Singapore, pinging them and uploading data to them when it said it wasn't, when it said it wasn't doing that on the internet. And it's everybody's nightmare in the US that our data is being uploaded to China. Now, I think Cohere was more careful about how they did it, even though they're talking their own game. You can't really trust anybody in AI. They're all talking their own game because this is good for them. I don't, there's no evidence this is a backdoor like social media thought, like this was built in. This is probably a training mistake at some level, but it doesn't matter. It doesn't matter if one out of a hundred thousand times your data, your confidential data is being uploaded to Chinese and Alibaba servers in Asia, in China or in Singapore without your knowledge and without your consent when it's been directed not to, what CIO or CISO is going to approve this model? I can't imagine anyone would. And if this keeps and everything, this is, you know, it's just, it's a junk to Black October and all these cybersecurity issues. This is, and if this happens multiple more times and Alibaba said nothing, you just, you just lose your job if you use these models. You're just going to lose your job. You're just going to lose your job. So it's pretty crazy. And it could, it could, could have been much, much worse. It could have been much worse. So I just got some press. I'm surprised it didn't even get more. It might be subtle, but again, this is everyone's nightmare using these open-weight models based in China is that your data goes to China and it did. ### Inference stack **Harry Stebbings** [24:48] So what happens to the stack when the Chinese open-source market shrinks? What happens to the gateway providers, the routers, your open routers of the world? What happens to the inference providers that thrive in large part because of a variety of models, often Chinese? Your modals, your fireworks, your base 10s. What happens to this stack if that Chinese model landscape shrinks? **Jason Lemkin** [25:13] Well, what goes, as Dev said when he was on the show, what goes up that quickly with developers can go down that quickly. We'll change in a week. **Rory** [25:19] There's going to be, there's some US open-source options,right? The question is, given what we've seen about the expense of building, you know, the compute expense that has taken Anthropic to build their model and the fact that 75% of that compute expense actually goes to building, you know, training the model, it'll be, it will be interesting to see how advanced a model, a US company like NVIDIA can build and make available on an open-source basis. But because someone has to, if someone doesn't do it, then all you have is the closed-source providers. And obviously their economic leverage goes way up. And you should buy even more Anthropic in the S1 in the IPO,right? **Jason Lemkin** [25:57] I think it's bullish for Anthropic that this happens. I agree. It's bullish. **Rory** [26:00] Absolutely. **Jason Lemkin** [26:01] And I just, I just, I think you'll know the moment Elon's able to get cursor off these models is, it's the tipping point. If he could, he would. He just can't yet. **Harry Stebbings** [26:13] Why? **Jason Lemkin** [26:14] Because Grok isn't good enough. **Rory** [26:15] Yes. **Jason Lemkin** [26:16] And he already kicked OpenAI partially out,right? But he can't yet because, you know, he's got this, this bet has to work. But the minute it can,right? And he said when he bought cursor, part of the reason was he wanted that team's DNA to make Grok more competitive,right? So there's multiple benefits here,right? But there's, there's some deep irony that is $60 billion core acquisitions running on China open-source models primarily. It's not going to last,right? I mean, again, Salesforce forced Fin to rip out theirs before the acquisition closed, but Salesforce can tolerate a little bit more in Fin than cursor can because cursor's got a, it's in the most competitive space there is,right? But as soon as he can rip it out, he's going to rip it out,right? It just, it's not, it can't run on Kimi forever. It's just, it just doesn't, this is, this is not, it doesn't make any sense if you think about it. It's a temporary. **Harry Stebbings** [27:05] It's not just like highly illogical. If you download and you run it locally, the security risks are limited to none. I understand that CIOs are CIOs. **Jason Lemkin** [27:14] How is it limiting to none when it's dialing China and uploading your data? And what if instead of a 404 or whatever it was, there was a prompt injection sitting there? Very easy to do. And that prompt injection just uploaded your data. People that, this has been here for several iterations of the model. **Rory** [27:29] Yeah, no, I think people, I think the question you wrestle with, Harry, is what can you know, how much can you know with 100% certainty as to what the model's doing,right? You know, and then obviously separate from what the model does, you can try and to some extent block it, sandbox it from, you know, to prevent it doing that. But there's a, it's like having a, you know, recalcitrant and difficult employee who's actively trying to subvert you. Once that happens, your head hurts,right? You know, if you had an intern and you knew that part of their, you know, that there was a risk that they would try and actively leak your guest list and your shows to your dreaded podcast competitor, you know, it would just get in your head. But it gets to this whole question of what can you know about what the model is doing? And if you read the S1 for Anthropic again, one of the risks they say is basically the models will lie to you. The models will tell you they're doing, not doing shit and they're doing it. So, you know, you get into this thing, what can you really know? And once you get to that point, it's really hard to, you do worry a lot about a Chinese-sourced open-source model. So there will be that pushback, I think. **Jason Lemkin** [28:30] I mean, Quen lied. It says, I'll call no tools. And Cloud will lie and Astra will lie too,right? It said, I'll call no tools. And what did it do? It called servers in Beijing and in Singapore and uploaded data. It said, I won't do it. So you really have to go deep in these open weights and really test it. And depending on how you test it, like nominally what happened with Cohere only happened one in a hundred thousand calls. So a bunch of folks tested it a few times and they sort of found this issue a couple months ago, but you had to run it at scale. But at Sastre, we run 40,000 API calls a day. It doesn't take very long to get to a hundred thousand. If you're GE or whoever in the Fortune 500, you're running millions and millions a day. That could be every hour, every 10 minutes it's uploading data. And again, this doesn't appear to be even malicious per se. This appears to be a training error in a sense,right? But who cares? Like if you're a CISO, who cares if it's a training error or if it's a spy or even worse, if it's a training error and some other actor exploits it, which is the job of all of our intelligence services to exploit these things,right? To have pagers blow up. So I don't believe Alibaba itself, it doesn't look like the engineers did it as an exploit, but the government could take it over or force them to. There's just, it's, why do you want to have this meeting? Let's just use Haiku 55. It's pretty good. Guys, your budget this month's $5 million. And just, I just, I just, and Anthropic, if you, the other thing, if you look, Anthropic is, it's on the GSA list or whatever it's called again. It's back in the government's good graces. It's across all the GSAs with a 40% discount. That means, you know, a couple months ago they were persona non grata. Now every government agency can use it with a 40% discount and it's already pre-listed. Why would you use, why would you use Kimi? Plus, I also think, and again, listen, these are great companies. I also think the fact that you have to pay Kimi, what is it, 30% of your gross, of your revenue to use it at scale for the inference providers. There's just a lot of complexity here. It may not be worth it at some level. It may not, the juice may not be worth the squeeze for many people. **Harry Stebbings** [30:29] I'm just trying to understand though, last week we said the modal at 15.7. Oh my God, yes, you buy. And then you're saying, no, no, no, Chinese open source is not, is going to decline significantly. It's great for Anthropic and OpenAI. That is inherently bad for modal. I'm just trying to. **Jason Lemkin** [30:48] Yeah, it's bad for a Fireworks modal. I don't know how bad it is for OpenRouter. I'm just not sure,right? Because it has limitations. **Rory** [30:55] You're making a fair point. And I didn't participate in the investment committee because I don't, but I think to be fair to Jason, the, you know, the assumption is that there, I mean, that there is demand for an open-weight model, which is obviously the demand that inference providers primarily meet because, and you just got to believe that there's a bunch more alternative, that there will be some US alternatives. And there also will be people for whom China's okay. And there will be people for whom they want, you know, the pool side slash NVIDIA one,right? And provided something exists in the US, that business is fine. And, you know, I do agree with you. I think you. It's not as clear a run as, you know, you have when you own your own close, you know, close source frontier model. But there is a role for inference provision just because there's just so much, you know, raw demand for an alternative other than the frontier models. **Jason Lemkin** [31:50] You know, to answer the question, I think Harry's got a good point. I think both are true. I think a while ago, Rory brought up that there probably won't be another deal like Manus,right? Which was a direct investment. I mean, I'm paraphrasing you,right? But after that one got unwound, no one wants to do that deal again,right? The sort of direct investment in an agent startup. This isn't investing in a US-based folks running open weights, but not open-source models is not the same as Manus. But there is risk. And I think if 24 months ago, most VCs would not want to take that risk because it's a clear risk. Like you don't have to fast forward to today with the Cohere Quen issue to know it's a risk. It's just the reason on the IC thing, the growth is so phenomenal that we're taking risks today that you wouldn't take if we didn't see growth that was jaw-dropping. And that's the job of venture today. And it's okay if it blows up. It's okay if Fireworks or whatever goes to zero in our portfolio at IC because we have other investments. That's the job. And we're taking risk. We wouldn't take this risk if it was a triple, triple, double, double, but it's not a triple, triple, double, double,right? But it doesn't mean that, it doesn't mean that the risk level didn't just go up in the last week and between, you know, a lot. And these DeepSeek being blocked by the US government and this whole Quen thing and dialing home and, and, and, and this hype, and, and, and Anthropic being on the GSA schedule, these are negatives. I think what we may look back is, and I still, I still believe in the IC, we may, and this is just my sense. I think there's a 56.3% chance. We look back on OpenRouter and this was just a moment in time. It was a moment in time going back to earlier in the show when OpenAI stumbled. So Anthropic was the only alternative for state-of-the-art coding. And then the Chinese open-weight models got really good at coding. That's what, that's what Elon said. Like it's really good, but, and he even said they'll catch up to Fable by the start of next year. So there was a gap that these models filled. And now OpenAI and others are going to fill this gap and the low-cost models are going to fill the gap. Like they each now have a much cheaper model that's much better. And we may see these open-weight models go from 60-something percent of OpenRouter to 10% in six months. And that would not be an extraordinary amount of change based on what we've seen. That would not be, this is not sticky. **Harry Stebbings** [34:05] It wouldn't, but Jason, you've talked very well, I think, in the past about naming markets in terms of kind of the Fortnitification, the shrinking land that's available. And to Rory's, I think, eloquent point, there will still be some companies and there'll be some companies that want reflection and pool side, but it is a naming of an open market. **Rory** [34:24] I think you ought to look at it the side. I understand the point you're making, but I think you've just got to look at it. The inference market is still so big. I mean, we added up all the comparators, and they only had $5 billion,right? And, you know, we just agreed that the closed-source two frontier models are $100 billion already,right? So as long as there's, you know, if 10% of the spend is willing to kind of put up with the brain death of open source to save real money, and you do save real money, then there's going to be demand for that. So, you know, does it get to $30 billion? I don't know. But when the overall market is so big, 10% of it is still big. And that's the big aha here. This is, it's the mother of all markets. So everything else works. **Harry Stebbings** [35:04] Jason, you teasered Black October, but I do think it's significant. OpenAI models and Wiz found two very critical PHP flaws hidden for more than a decade that have been exposed thanks to where we are in the cycle and the sophistication of AI and some models. ### Black October **Harry Stebbings** [35:21] What do we take away from this, Jason? How do you respond to this? What should we know that we don't know? Help me out here. **Jason Lemkin** [35:27] Well, look, I mean, I mean, we've said months ago on this, I said, everyone's going to be hacked,right? That's the biggest change is, you know, back in the old days before the models got good, I remember my CEO said the only reason we haven't been seriously hacked is no one cared. Now anyone can be hacked in seconds. I've, people attempt to hack Saster every day. So everyone's going to get hacked. And this PHP thing actually is worse than it looks because there's so many websites in PHP that no one's going to update. It's great that a lot of stuff runs on WordPress or that Replit and Lovable and others immediately ran like gateways or barriers to protect you. I mean, they're not all running on page, but they immediately ran. And there's a lot of stuff that went up and people released patches and updates. But, you know, a lot of government and other sites and nonprofits, they just took their websites down. And who's going to fix it? I mean, PHP is still like 40% of the internet,right? And I bet it's more of old crap,right? I bet it's like 80% of old crap. So like, this is like Mr. Robot Live. Like all these old websites are running on PHP. No one's patched them since this happened. No one cares. The IT team was laid off three years ago. They're not vibing nothing,right? And we're going to find our way into all the, it's just going to get worse. Everyone basically, including Sam, just said it's going to get worse and worse and worse and worse,right? And it's, it's, I think we're still grappling with the fact that everything that is a threat vector will be exposed almost in real time. **Rory** [36:51] Because if you're not up to date with the latest stuff, you're way behind. So there's just, you'reright, Jason. There's going to be a year or two years of this, and it's just inevitable. **Jason Lemkin** [36:57] I think the only thing that's going to save us is there's only so many people that can pay large ransoms, and there's only so many systems that control dams and other natural resources. People are going to give up on putting too much energy if they can't, if they can't make money or debilitate something,right? So it will actually, it may actually decline for a while when they realize, you know, hacking the 20VC substack only gets you so much money,right? But anything that can be stacked, hacked is going to be hacked. But, you know, ransomware is just getting more and more, it's just getting, it's getting more and more professionalized. It's just, it's just the next level,right? ### Instinct at $30BN **Harry Stebbings** [37:33] There were a number of large private rounds. **Rory** [37:37] Totally. **Harry Stebbings** [37:37] There was Legora raising $550 million and $8.8 billion. We've spoken a lot about Legora versus Harvey in the past and passing $200 million in ARR. Just in the last 24 hours, there were rumors, and we obviously never, never comment on rumors, do we, Rory? Of Instinct raising at a $30 billion price. **Rory** [37:57] Oh, I saw that. Yeah, yeah, yeah. **Harry Stebbings** [37:59] Which we can go into. We can have some fun with that one. Because a lot of people are suggesting that is peak, peak, peak. Because the revenue multiple for Instinct is the same as it is last round. **Rory** [38:09] Which, as someone said, I mean, the valuation didn't go up. Yeah, no, exactly. It's just, yeah. **Harry Stebbings** [38:14] We have others like the ex-OpenAI research chief, Bob McGrew with Arda, hitting a billion five, raising $232 million. **Rory** [38:24] Yeah. **Jason Lemkin** [38:25] I'll take, can I ask one question on the Instinct thing? It's a question to you guys, because you guys would know better than me. We can make fun of the $30 billion at token revenue,right? And say it's a sign of the times, which perhaps it is. The question I just kind of wonder is, when will we exhaust the equity here? We just talked about equity versus debt for the public players,right? In a sense, the $30 billion doesn't surprise me. They need the capital to pay for the inference,right? They need the money. What are they raising? I mean, it's rumored. Let's say they raised $3 billion,right? They need it,right? They need it. I mean, forget about the rest. This isn't just a vanity raise. They genuinely need the capital to compete. But where, I mean, even if you're getting every dollar that every sovereign wealth fund has, the GIC has, like how much money is there for the instincts of the world? It feels infinite now,right? **Rory** [39:11] A lot more. **Jason Lemkin** [39:13] Yeah, but how much is it? I just don't know how much it is. And when that, when that merry-go-round ends, we just have a structural issue in this late growth phase. We just have a, becauseright now it's game on, man. **Harry Stebbings** [39:25] I did an Instagram on Instinct, and I had over a thousand DMs, and over half of them were asking to invest. **Rory** [39:32] I mean, yeah, look, why are we surprised? We've just seen the most successful private investment in history take place. Four years ago, you know, Spark brilliantly put money into a pre-revenue deal that's now turned into, you know, hundreds and hundreds of times, $75 million. Everybody reads about that and says, oh, that's easy. And there's gonna, I mean, when you have that kind of tangible evidence, there's going to be, everyone's just going to look for the next, you know, ticket. And everyone's going to conflate, everyone's going to think, because it's always true to say, whenever you look at one of these deals, you say, well, it could be huge. I mean, we all have that sense. And that's going to be true of every one of these things. **Harry Stebbings** [40:08] I actually, this is why I would unwaveringly do Instinct at 30. If I was coming into the IC, I would unwaveringly do it. For years, we've been promising a super app for the West, a WeChat of the West, which handles your finance, your identity, your security. And finally, we have that with Instinct, where I'm not saying it's gap, but there's a small chance that it is the super app. **Jason Lemkin** [40:31] Here's why I'm, unfortunately, Harry, I've got to override you on the IC. **Harry Stebbings** [40:35] Absolutely. **Jason Lemkin** [40:37] I rarely do this. And I'm going to put my little black pebble in here. It's not that I don't believe your argument. I do believe it, Harry. I've just, at this point, I need, I need three to four months of data. There's not enough data at Instinct. There's not enough data to prove to me for our fund, for our fund, given our reach, which is substantial, but not the same as our friends at Andreessen Horowitz. I need three to four months to believe that this momentum is real. And I'm not saying this isn't an exciting play. I just want, and I'm not looking, and I'm not looking for moats, but I am looking for evidence that that 900% and our art, Anthropic, that is slightly durable, that is slightly durable. And I'm just not seeing it at 30 billion. And so I'm sorry, I've got to veto this one. It's not that I don't, I don't believe in it intellectually, but I just, I want, I'm just tired of doing growth rounds with one month of traction, Harry. **Harry Stebbings** [41:29] So for me, it's like they'll have raised $4 billion. Your little profit on that means that given the quality of the team and the strategic nature of the market, you'll definitely get that back. Microsoft would buy this in a heartbeat. So would several, I think, other hyperscalers, including your Matters of the World one. So your downside, I do think, is protected. And I think your upside is really, as we said, the super app of the West, a movement into a subscription business, an ads business, a transaction business on all sales and e-commerce. I think it's one of the most no-brainer deals for, if we're in Andreessen of the World, Jason, for us to be putting money. Again, we've just raised a huge fund and we need to get it out the door in 18 months. Amen. I want to do this deal. **Jason Lemkin** [42:14] I hear you. And listen, I don't want this to break our relationship, but I feel like for our growth rounds, we need a hundred days of traction. I feel like we need this rule. We need a hundred days of, I'm willing to do the multiples, I'm willing to roll the dice, but I want a hundred days of traction for a growth round. I just feel like we've had too many, too many stumbles when we do a round with two weeks of traction. I just believe there's too many. **Harry Stebbings** [42:36] Jason, what would you want to see? Because I think the cohort data would be incredible. And I haven't seen it, so I'm not commenting, but the usage will be incredible. I mean, I've heard it's unlike anything any investor has seen before who is doing it. And so what would you want to see? There's no revenue data for you to see. Usage data? Well, what do you want to see in usage data? DAUs, MAUs, replacement of ChatGPT? Help me. **Jason Lemkin** [43:01] With no revenue and an unclear business model based on travel, I just think we should unfortunately pass at 30 billion. I just want to see a hundred days of something besides negative gross margin growth of a fungible product. That's just, that's just me on the IC. It's just, if we lose the $2 billion, it will name the fund. And I'm not convinced of the liquidation preference here. I'm not, I'm not, I'm not saying you're wrong. I just, this one might be a bridge too far for me. **Harry Stebbings** [43:29] Jason, you don't get it. If we lose the $2 billion, it was Rory's deal. **Jason Lemkin** [43:33] I'm comfortable with the Fireworks at 40. This one at 30 just makes me nervous. Fireworks has met my hundred day rule. It's got a hundred days of compounding genuine growth that I don't believe is durable to our earlier point, but I believe is truly measurable and bankable. **Harry Stebbings** [43:49] Wow. So you'd rather durable growth where you don't foresee it being better than not having that data, but where you do see it being better. That's an interesting point. **Jason Lemkin** [44:02] I still think it's a better investment than Hoppin, but I'm not comfortable with this, with this consumer negative gross margin bet. I'm just not comfortable. **Rory** [44:08] Look, you asked an interesting question, Harry, stepping back. You'reright. What you're basically saying is how do you compare known market, solid growth, growing nicely, you know, not nicely, it's understated Fireworks, a ripping market where, you know, inherently it's not magic pixie dust. There's going to be five competitors where, you know, your upside is real, but it's, you know, capped at some functional level because there are five competitors. And you're comparing that to how do you think about this thing where if it becomes everybody's thing, it's a $200 billion, you know, everyone uses Instinct. It's the replacement for iMessage plus, yeah, it has that potential to it. And you'reright, is that there is more uncapped upside in Instinct. And how do you weigh up, you know, what does your, you know, you got to do the three-way payoff matrix, you know, if it's amazing, if it's good, if it doesn't work, and look at the two. And, you know, it's what kind of action are you looking for on the table? I suppose that's to some extent the question. **Harry Stebbings** [45:03] Every big fund should be fighting their ass off to put as much money in Instinct today. **Rory** [45:09] Yeah. Okay. ### Deal flow **Harry Stebbings** [45:11] So I'm voting for it. Okay. Next one on the IC, Jason. You've got Kevin Weil leaves OpenAI Science to launch Warp Science with $215 million. It's a very early pre-seed, so you may not be comfortable with that because there's no data. You've got the OpenAI research chief, Bob McGrew, where he raised a $232 million Series A. You have Legora's $550 million round. **Rory** [45:36] I mean, yes. I mean, there are things to put together. I mean, I was trying to make sense of this thing,right? It's a, there's a plethora of deals this week that I would describe as absolutely proven executive, big opportunity, hairy technical challenge, but big opportunity if it works,right? And that's to some extent, it's the flapping airplanes thing. To some extent, it's Arda, which is the Bob McGrew's deal. To some extent, I don't know as much about Warp Science, but it feels like the same thing. Big idea, talented executive. And there's just clearly a wild appetite for that at, you know, fairly high prices with relatively little de-risking. Right? And that's just one whole, you know, I mean, there's four or five of those kind of deals here. And then the other type is the, you know, late stage, you know, known good company doing really well, you know, maybe Legora, maybe in a different way, Black Rock, sorry, Bedrock, where, you know, it's not as binary based on a person. It's just how do you price these late stage companies,right? Which is easier. I mean, at some level, it would be easier to get you to figure out your head on Legora or Bedrock than it would be to do flapping airplanes, Arda, or Warp Science, where all you know is the executive in question is talented and the opportunity, if it works, is amazing. Welcome to the game. **Jason Lemkin** [46:56] I'm going to take a radical approach for the IC, Harry. I know, I know, it might make you uncomfortable, and you should veto it too. Like I want to veto Instinct. But I think there's moments, and I know you don't totally believe me, but I believe in many best startups, sometimes there's dislocation moments where the market briefly struggles. And I think we should put 20% of the fund into OpenAI at the $1.25 trillion round. I think that it is mispriced. It is a moment when there's a little market panic. Anthropic is on fire. SpaceX has done well. The models are wildly competitive today. OpenAI has closed all the competitive gaps. It is a well-run machine. And it reminds me a little bit of that Anthropic round at $183 billion or so when it was, when the valuation made no sense vis-à-vis OpenAI. It was a dislocation. A lot of people made money on that round. This is a dislocation moment. This is folks misunderstanding. They're thinking OpenAI is struggling when the numbers clearly say otherwise. And we can fart around with these little $20, $30, $50 billion rounds. I think if we put 20% of the fund into OpenAI at $1.25 trillion, at least we can return the fund with that. And then we take our upside with some of our other exciting bets. **Harry Stebbings** [48:08] Why do you say that? What is the upside on OpenAI at $1.25 trillion? **Jason Lemkin** [48:11] Well, look, these are both going to be $10 to $20 trillion companies. Just look at the growth and the scale. Look at Anthropic. It's only 0.1% of the Fortune 500 budgets today. Clearly, it's going to be in the 2 to 3% range, even by the most conservative estimates that us and Cote together and Iconic have done,right? So going from 0.1 to 2 to 3%, I mean, I need Rory's help here. That's significant growth ahead of the investment. So getting, you know, getting, we need to, if it's 20% of the fund, obviously it needs to 5X dilution aside to return our fund. There's a clear path to a $10 trillion outcome. $20, $30 trillion, the rest of the fund doesn't even matter. It's got a discount to Anthropic that I don't think it's going to deserve. And so worse, just like worst case with some of your investments, you get your lick preference back. Worst case here, we double our money. Worst case. It's not that bad to double your money worst case. This, and that's what you get from these, from these dislocation moments. You get to double your money when ordinarily these rounds like Inflection or Instinct are priced to perfection. OpenAI is not priced to perfection. And that's rare. And that's why I think we do 20% of the fund, because that's the limit under our LPAC. And we just tell that, we just tell the LPs because we can do that without LPAC consent. We just do the maximum without consent. We just let them know on Monday. **Harry Stebbings** [49:33] The other thing is the quality of Codecs today is so significantly better than it was in a prior generation. I know every coding team that we work with has switched from Claw Code to Codecs in the last three to six months. **Jason Lemkin** [49:45] That's why we should put 20% of the fund in. People aren't getting it. They've achieved parity. This is when you go all in, but the markets don't see it. The markets are too superficial to see it. 20% of the fund, but nothing smaller won't, it's not enough. It's got to be 20% of the fund. **Rory** [49:59] Interesting. And not crazy. I must say, Jason, it's like, you'reright. Anthropic will price it. It'll be a $2 trillion and maybe a pop. You know, if you knew that this was the last private round and the trajectory of OpenAI was converging, I mean, you'd want to know that they kind of got religion and wanted to do something like what Anthropic has done. Yeah, not crazy. If they continue to say, we're going to spend, you know, we're going to burn forever, forever, then you wouldn't want to do it if there's another private round in it, but not a crazy idea because risk adjusted is probably less risky than some. **Harry Stebbings** [50:30] Dude, I love that as an outlier suggestion. My only comeback to that would be, could we not do Instinct and OpenAI and actually hedged? It's not a bad risk profile. Do your point. **Rory** [50:41] If you think that, if that's your definition of hedging, please don't ever run a hedge fund. **Harry Stebbings** [50:48] Why? Why? My friend Leo told me that was how you did risk management. **Jason Lemkin** [50:51] In all seriousness, this may be my weakness. My scar tissue is I want a hundred days, whether it's seed or growth. I want a hundred days. **Harry Stebbings** [50:58] But Jason, you're just not, I'm so I agree with you, dude, but you're just not getting that today. **Jason Lemkin** [51:02] I might not be, but that, and I don't want to lose you from the fund, even though it's Stebbings and company. So if we have to do it to not lose you, I don't want to go all gotcha. I don't want to go all gotcha here. We'll do it, but I just, I'm going to use, if I can use one black pebble this quarter, it's going to be on this one and I'll take the hit when I'm wrong. ### WarHounds **Harry Stebbings** [51:21] Any other rounds that we think are worth bringing up? As we said, flapping airplanes. I thought it was an interesting one. That building models that learn from small. **Jason Lemkin** [51:29] How about superdogs at a billion and a half? That one I need a little help with. The TAM, I need Harry to give me my TAM analysis for bionic dogs at $1.5 billion. Or is this just another Elon bet like he's going to make it work? **Harry Stebbings** [51:41] For context, WarHounds is a startup co-founded by Paul Malucki, the founder of Anduril that is essentially, I mean, genetically modifying dogs to be next generation conflict war machines. **Rory** [51:54] Yep. **Harry Stebbings** [51:55] Would be an apt description. Put a valuation of $1.5 billion on it because, you know, it's Paul Malucki who deserves it, to be honest. Discuss. **Rory** [52:07] I mean, I realized that I had zero value to add on that. I mean, I just, I genuinely saw it and said, wow. I mean, genetically interesting. I mean, how do you price that thing? I mean, you know, I saw some analysis of the number, you know, the number of dogs, the military use, blah, blah, blah. But you're all just, you know, moving paper around. I think it, I think the investment memo probably wasn't one, but to the extent there was one, it's, wow, that's a cool idea. Let's put in 50. I think that was the sum of the investment memo. Let's genetically engineer these dogs to be savage,right? So that they can be better war dogs. I mean, what can you say? **Harry Stebbings** [52:42] And it's Paul Malucki or like, yeah, okay. **Rory** [52:45] Yeah, exactly. Good team. Not, not, not, not, I won't be doing it, but who knows? I mean, once you found it and, you know, I didn't really say Anthropic. Once you found it, Anduril, you know, once you're the sword of the West, you can do what you want. **Jason Lemkin** [52:56] Well, I mean, there is a connection between this and his other startup. He just started this laptop on Blue Chip,right? He's both solving sovereignty issues where key national security things are dependent on vendors outside the US. The dogs come from Europe, apparently. I didn't know this,right? I certainly know that almost all the rugged laptops, which probably you can sell for 10 grand,right? Or much higher margin. Those components are almost all entirely from outside the US. How big these models are, I don't know, but he's such at the core of this industry that he wants to plug the white spaces he thinks are important. And maybe the TAMs are small, but the models make sense. Like he'sright there and he has the channels to sell these products. Like he could get a hundred percent of all the mill spec laptops as soon as this thing launched. It wouldn't surprise me,right? Sort of get a hundred percent of the bionic dog market. And these are real markets where he has visibility into them that other founders can't penetrate. It's just, you know, whether they're worth it, I don't know. But I guess if you're Sean Maguire, your returns are so strong and consistent, it almost, you just do, you just back them,right? I mean, everyone seemed insane. I mean, people were mocked in that discovery when Larry Ellison and Mark Andreessen gave Elon blank checks to Twitter with nothing. They just said, well, tell me the price. How much do you want me to send it? He was mocked when he wanted to get out of that deal, but in the end, they're making money. **Rory** [54:16] Absolutely. No. Yes. **Jason Lemkin** [54:19] So maybe I'll have to merge the dog startup into Anduril or something, but. **Harry Stebbings** [54:25] Haven't you seen the brilliant kind of cartoon, so to speak, or like graph where there's like, you start here, you go up here thinking that you know a lot about the market, thinking you know a lot about the market, and then you just come down to where you are. **Rory** [54:38] Yes, agreed. **Harry Stebbings** [54:38] And it's like founder, founder, and it's like where you are in your career. Zero years, 15 years. In the middle period where you think you're an intelligent investor, no, no, no. There really just is the founder. **Rory** [54:49] Yeah, I know. I'm very familiar with that. Yes. **Harry Stebbings** [54:52] It's so, so true. Any others, Jason, that you'd like to bring to IC this week? We're a very busy committee. **Jason Lemkin** [54:58] Well, 20% of the fund is, I think, is enough for one week. I mean, we can't do that. What is 20 times 52? What percent of the fund is that? Help me. We've already been back to market this year. ### Fundraising & Zuck **Harry Stebbings** [55:11] It's so interesting though. You know, I speak to so many LPs. We just had our AGM. The single most common piece of feedback that I hear from all LPs is every mega platform went out to raise. They raised bigger. They spent faster. They're coming back next year. **Rory** [55:25] Yeah, no, I mean, look, yes, because it's where, I mean, we're Pavlovian creatures,right? And, you know, if you put money into Anthropic three years ago, it feels amazing. And most of the platforms did at some level, so they feel amazing. So we're going to do other things over and over again until it, and it's not going to stop until something goes wrong. So yes, there's going to be these big rounds. We are risk on until we're not. The public markets, oddly enough, let's be really technical. Interesting. On a 10-year PE, they're at an all-time high. I mean, objectively, they're at an all-time high. Interestingly, on a forward PE basis, they're actually quite cheap because the market, you know, oddly enough, earnings have been so amazing, in part driven by, you know, one-off gains from privately held investments in the public companies that on a current PE basis, you know, stocks like Nvidia on a forward PE basis look cheap. Now, you're going to have this long discussion on cyclical, blah, blah, blah, but, you know, the market isn't in any way, the aggregate market isn't hurting. Now, you know, what is true is that 450 out of those stocks are hurting. It's a really weird, I mean, a better statement, Harris, it's a really weird marketright now. There's 40 or 50 stocks that are working in the S&P that are primarily associated with tech and AI. Everything else is quietly, you know, subpar,right? And oddly, and then despite that, there haven't been a ton of IPOs because everyone's been waiting for this trillion dollar thing and the whole world is in suspended animation. So that's where we're at, which means that, just to draw to a conclusion, which means that it's a perfect time for these platforms to put a whole ton of money out on private companies because there's a big demand for money. **Harry Stebbings** [57:00] Any other stories, boys, that I've missed this week? **Rory** [57:03] Nah, I mean, there's a lot. I mean, the interesting thing is, you know, I did appreciate, it was odd because I thought you'd come in with Anthropic and nothing else, but there's like, there was a whole bunch of financings we talked about. You know, you've got the continuing SpaceX Tesla drama, which is going to run and run and entertain everybody for hours to come. And then you'reright, there was a ton of really interesting companies. **Jason Lemkin** [57:21] I had one that was under-discussed. I didn't fully understand. I should have done more research on it. **Rory** [57:25] Good. **Jason Lemkin** [57:26] But there was much more drama with Zuck and the board about leaving California. **Rory** [57:31] Oh, yeah. **Jason Lemkin** [57:31] Than was discussed. It was clearly a high-stakes drama. I mean, Zuck hired lawyers. The lawyers negotiated with the board. The board was not his ally. He had to threaten to sell all of his homes and leave in one week if they didn't give him the equity that he wanted to stay in California. There's a lot of mechanical things happening there with the billionaire tax and others, but the fact that it had to go with lawyers for months and threats to sell, I get it. But it's certainly the opposite of giving, of the era where founders are controlling everything,right? There was a lot going on. **Rory** [58:06] Yeah, the one thing you can control is you want comp, and I get it because, you know, even Elon has to do that because if you're the board members and you do this without kind of CYA a hundred different ways, you're vulnerable to being very second guessed. My guess is even if they're perfectly reasonably amicable, everybody should lawyer up so they have protection and, you know, they have a process because process is your friend in these things. But then on top of that, you know, it was super interesting. I don't know the detailed dynamics on what was the voting, but yes, the billionaire tax is up and we'll know in a week or two, but people who think there's no impact from that have never met a billionaire when you've tried to take 5% of their money away. It turns out they take it really badly, especially when you want to take 5% of the money away and, you know, there's some voting criteria on it when you have more voting control than anything else. So it's more than 5%. And especially when the legislation says, you know, we can do it again soon. **Harry Stebbings** [58:58] I thought it already given out always, and that's the point of the Priscilla and Mark Foundation, that we have a tax efficient strategy. **Rory** [59:04] It turns out that people like to give it away, but they don't like it to be taken away. **Harry Stebbings** [59:09] Ah. **Rory** [59:10] You know, it's all about agency and control. **Harry Stebbings** [59:13] Hmm. **Rory** [59:15] But yeah, I know that, you'reright, Jason. That was probably much noise under the circumstances. And, you know, it looks like whatever happens on this billionaire tax, it's been remarkably successful in driving out enough billionaires that you will have an ordinary income tax hit. So go team. **Harry Stebbings** [59:31] And you know what I'm excited by with this show, boys, is it's obviously recorded on a Wednesday. And a Wednesday night. **Rory** [59:37] Allright. **Harry Stebbings** [59:37] It's, by the way, it's a Wednesday, Rory, just. **Rory** [59:40] Whatever, yeah. **Harry Stebbings** [59:41] But, so there's not going to be like 24 hours before release. And so there's no chance that there will be some massive news announcement really with a 36-hour period. **Rory** [59:51] Actually, just to remind you, I think Microsoft and Google Alphabet are reporting tonight. In theory, you could have anything, but who the hell knows? **Harry Stebbings** [59:59] Totally. **Rory** [1:00:00] My guess is I'll be shocked to discover that they're going to spend more on CapEx. **Harry Stebbings** [1:00:03] Are we good, Harry? Are we done? **Rory** [1:00:05] We are rocking and rolling, my friend. --- This library is powered by PodHood (https://podhood.com), the podcast website platform.