Capitulation? Have VCs given up hope?

Speaker

I've got sunshine on a cloudy day When it's cold outside, I've got the month of

Speaker

October the 27th, 2023, the last Friday in October, I've got sunshine on a cloudy day in October and if it's Friday it must be that was the week with my friend Keith Teer, Mr VC and this week the theme of his wonderful newsletter that was the week is capitulation. It comes with another terrible AI piece of imagery. Keith, when are you going to use some real art for this newsletter? Well you know these VCs will not pose for pictures like that so you've got to make them up. I think you put them in now to annoy me because that is a particularly bad one. All these guys with beards, why are they all got beards? Obviously Dali believes venture capitalists have beards and that's a white flag of surrender in the middle of the table. Yeah and Dali of course the one company that doesn't need VCs is OpenAI there. Now what are they buying back the shares from the employees? That's a good company to be working for. Yeah you know they have an internal market where they organize share buybacks from employees regular intervals to create liquidity, partial liquidity. I think there's a lot of rules around how much you can sell but the average senior engineer there is making between 900,000 and 1.2 million a year. So Keith you and I should probably go and work for AI, OpenAI. Exactly, I mean don't they realize? I don't think they'd give me a job, they might give you one. Yeah maybe sweeping up, definitely not doing the AI. Well you know sweepers get good money at least in the Premier League. Capitulation, another VC theme Keith. Is this good or bad news? Are VCs abandoning hope or is hope abandoning the VCs? What's going on? So this was triggered by two very good essays, one by Charles Hudson, the other by not an essay but a set of slides by Manu Kumar who is the, he runs a venture firm called K9 as in the letter K, the number nine. And they both talk about the end of an era for early stage investing and most of our conversations about venture up until now have been about the decline of late stage investing. These two are focused on seed and A-round investing and it kind of goes along with what Sam Lessing said a couple of weeks ago that the era where you could invest in something and someone else would come along and turn it into a unicorn is gone. That's roughly what they're saying and for venture capital that's a big deal. We've talked about the power law of venture capital. Yeah and you write about that in the editorial. What is the power law? The power law is a fact. It's that most venture funds over, a venture fund lasts about 10 years. It usually has finished investing after three years and then the manager raises another fund. So within a 10-year window they have three funds, one every three years and across three funds they might get one, two or three investments that are good enough that when they exit it returns the entire fund or maybe even more than one fund. That's the power law. It's the fact that most of your investments don't make enough to justify the effort but one every now and then. I don't understand what's changed. I mean given you link also to this OpenAI story. Isn't OpenAI a classic power law company? Is that whoever invested early are going to make ridiculous fortunes like Uber and Airbnb and Google and Facebook and every other investment is a disaster. But what's changed? Why is anything changing? Very good question. What's really changed is that for a small number of years roughly starting in 2019 and ending last year there were lots of power law companies, way more than is normal due to the fact that late stage investors were writing large checks at high valuations often quite early in the life of a company. So seed investors and ARAN investors were experiencing an unusually high ratio of power law companies and what Charles and Manu are saying is that that isn't going to be happening anymore. I think of it as organic unicorns versus artificial unicorns. Yeah I have to admit Keith that organic unicorn sounds like a square circle. I mean what does that mean and nothing's changed. Everyone's always pessimistic until the next big thing comes along and then everyone changes their mind. A good example of organic would be Google or Amazon or Tesla that literally built a company stage by stage. A big investor didn't swoop in and give them a high valuation early in their life. They had to build a real company. So an organic unicorn is a unicorn that achieves a valuation due to success in the market. So I would notice for me to rephrase your language but these are real unicorns. In other words they're not just fake unicorns of some absurd valuation which crash. Yes correct. That's exactly the meaning. Yeah you can you can haggle with my words. So OpenAI is a real unicorn. I mean this is not going to just crash. It's a real company. They're already on 1.3 billion revenue so by definition they're probably a decacorn and they may according to that story they're worth 80 billion. So that's roughly 80 times revenue a little bit. And a couple of weeks ago you suggested that in five years they'll be the most valuable tech company around. That they'll be worth 500 billion. I think as long as Mark Zuckerberg's evil plan to defeat them doesn't play out they will be. Well you link to that story as well. Do you really think Zuckerberg can take on AI? OpenAI? Can Zuckerberg steal any of the AI thunder from anyone? Microsoft? Google? Yeah well he's you know he's a very predictable guy Zuckerberg. He copies anything that works and tries to tell the story that what he's built is a better version. Didn't Microsoft used to do that as well? I mean don't all smart companies do that? Microsoft and Apple are cleverer. They look at what works and build their version of it inside their environment. Later on when we get to startup of the week we'll see Apple doing that right now. But what Facebook doesn't do that. It basically goes top down and copies an entire set of things like it's Reels copying TikTok for example. So Zuckerberg is a copycat whereas Microsoft and Apple learn from the market and iterate. So they're a little bit different. Well when we get to our, I don't want to give away our X of the week because that's always the end of the show, but one very prominent VC Vinod Khosla suggests that that's the nature of things. So what Zuckerberg is doing is no different from what Bob Dylan did or Steve Jobs or anyone else. Well that's definitely true. The funny thing with Zuckerberg is he almost always fails. You don't like him. Has he pissed you off this week Keith? You're usually quite sympathetic to these tech titans. What's so bad about Zuckerberg? Why is he worse than anyone else? Actually you know Facebook and WhatsApp and Instagram are awesome and I'm not against him at all. I just don't think he can succeed copying up OpenAI. You can't copy OpenAI. I mean we'll talk about OpenAI as a copycat too but how can Zuckerberg at Facebook, however many tens or hundreds of billions of dollars he pours into it, how can it replicate OpenAI? I mean Google's trying to do that too. So what Zuckerberg is doing is really an attempt to slow down the pace of OpenAI success by releasing open source versions of large language models. And now that can work because developers want access to models. So he will definitely have some impact but what he probably won't do is replicate chat GPT. There you're right.

Speaker

And there's lots of talk of AI as open source. How does Facebook hijack it for its own benefit and not just for the open source community? Why does that help Facebook? I think Facebook is almost 100% preoccupied with audience and engagement. And so the more time we spend on OpenAI the less time we would spend on Facebook. So by slowing down OpenAI it probably hopes to do a rearguard action a little bit like Google with Android really was trying to slow down the move away from the web. By building Chrome and putting Google search inside Chrome on an Android phone and taking some of the market share from mobile, it was really trying to make sure it didn't get too punished from the rise of mobile. Right, certainly. I don't know how he's done it. Zuckerberg, he works in unusual ways. He's stolen the thunder on this show and you don't even have any links to Facebook. So let's move beyond that. You have an interesting link on Essays of the Week, the Westminster Declaration, Keith. A lot of people who have signed that have been on the Keenon show. Is there an intellectual crisis of free speech? Do you believe it? I have to admit I don't. But what do you think? Well, I think there's a whole new lexicon around free speech. Words like misinformation or disinformation are basically used to render a point of view more or less a lie and often depicted as malicious and sometimes even illegal. So I think what the Westminster Declaration is trying to say is let's just take ideas at their face value and say what we think about them. Let's not label them because misinformation is another word for opinion. And, you know, it would be weird if we expected social media to only have facts. So who are these Westminster people? Who are they against? Matt Tiabi, I know you're a big fan of him. Julian Assange, I'm personally not a great fan of his. But who are they against? Who are they saying? Who's the Westminster Declaration against? It's not clear to me. It's really against everyone who's trying to reduce what is published down to things they agree with. And that's a wide spectrum. I agree on that. So what do you make of this week? I don't want to turn this into a political show, but you do the Gilmore Gang on that. What do you make of people trying to shut people up on the Middle East front of saying you think of embarrassing them and threatening them? Yeah. You know, it's kind of interesting because my empathy in what's happened in recent weeks, 100 percent lies with the Israelis needing to make sure Hamas can't ever do again what they did. So I have I have a bias in support of Israel. That said, an Israeli friend of mine published something this week talking about Palestinian terror. And I said, I corrected him and said, you should be talking about Hamas terror because otherwise you're equating Hamas with Palestinians and that is not appropriate. And what you will find is that almost any point of view coming from the pro-Palestinian side can easily be depicted as anti-Semitic and therefore rendered not allowable. And so I think inevitably when something as extreme as what it happens, people people choose sides and it seems somewhat illegitimate in the context to choose any side other than Israel's. On the other hand, there's more than two sides. There's a lot of hands here, Keith. You're running out of time. So you're being a bit mealy mouthed here. So what is going on? What are you saying? So what what I'm saying is you should tolerate all points of view and disagree with them where you disagree with them. And that's not exactly news. Who would disagree with that? Well, that's well, the Westminster declaration, funnily enough, is signed by people across the entire political spectrum. Yeah, which makes it sort of boring. And these people are so pleased with themselves. They're so annoying, all of them. Yeah, but you I'm going to guess you agree with them. But it's obvious who wouldn't agree with who would disagree with that? You know, even even the Jews that want to shut the Palestinians out and the Palestinians who want to shut the Jews up, they would probably agree just when it comes to certain issues, people get all peculiar, right? Isn't that the problem? Well, they get confused. I mean, if if the Middle East was as simple as Israel versus a mass, it would be easy. But in addition to Israel isn't a single thing. There's lots of different points of view within Israel. And, you know, on the issue of Palestine, Hamas is a tiny fragment of a range of opinions. So once you get into the details, it gets complicated. And that's why it requires thought and conversation. If it was just Israel versus Hamas, it would be so super similar like Man United versus Man City, Keith, much simpler, much simpler. Maybe that's how they should decide. They play a football match like Man City versus Man United. Whoever wins gets the land. How about that? Israel would win that match, actually. I don't know. I mean, if if if Salah was allowed to play for the Palestinians, I think it would be a different situation. He's Egyptian, so he wouldn't be allowed to. I thought the Egyptians were Palestinian. Some of them are actually. We can't get into this. We're going to trouble get thrown off the Internet. Let's get back to technology. Two interesting pieces you link to on techno-optimism. We talked about Marc Andreessen's rant on techno-optimism last week. One piece from Max Reid, who you say is always a good writer. He says techno-optimism is a sign of VC crisis. I'm not sure about VC crisis, but certainly VC nervousness. There's a certain weirdness around the VC community. You're not really a VC, but you have a front row in it. Are the VCs getting a bit weird, Keith? Well, they definitely have to justify their existence to the world. The predominant meme is that venture capital is a bit of a shell game. I invest in this company, you invest in it next, and then someone else invests in it after that, then it IPOs, then the public buy its shares, and then it crashes.

Speaker

Bit of a Ponzi scheme, a bit of an SPF. There is a meme that that is the case. The meme is wrong. It's because people don't understand the power law, but that is the meme. Andreessen clearly, when he did the techno-optimist manifesto, is trying to make a case for innovation requiring early stage investments in the best prospects for human beings. Cynics or smart people, depending on how you look at it, are pouring cold water on that and claiming that he's only self-interested and it's a sign of his even has to make the point, because if it was obvious, you wouldn't have to say it. I'm not a great fan of Marc Andreessen. To be fair to him, the one thing that I don't think he can be accused of is making all these techno-optimistic manifestos in defense of venture capital. He's so wealthy, it makes no difference to him one way or the other. He's not doing it from the venture point of view. He's doing it from the historical philosophical point of view, whether you agree with him or not. He really believes what he's saying. He doesn't care whether it makes VCs rich or not. He's trying to represent a point of view about history, meaning the future, not the past. It's a legitimate point of view and certainly progress has to be... I agree. It's not the point of view, it's the way he puts it. You refer in this week's newsletter to one of my favorite economists, Brad DeLong, who's been on Keynote a lot of times. He lives over in Berkeley. He's a lovely guy, slouching towards utopia. He's a highly erudite, one of the world's leading economists. In his new book, Slouching Towards Utopia, he basically makes the same point as Andreessen, except in very, very different language. Why do you link to DeLong's stuff this week? I link to it because Rohit Krishnan wrote a review, a very long review and glowing review, by the way. I, like you, saw the correlation between that and what Andreessen was saying. I do think it's a wonderful thing when a body of work comes along that is optimistic about the future. Almost any future view, and you can see it the best in science fiction movies and books, is kind of dystopian. The human race has never produced dystopian futures in its whole human race. It's gone through dystopian periods, but always came out the other end closer to utopia than dystopia. There's no big reason to believe why that would change. When you get bodies of work which have optimism at the core, it seems to me not only should you read them if they're well... It's a big book. I know you're not a big fan of reading books at the moment, but I wouldn't. I mean, I think that the DeLong thesis is complicated. Tell me what he says, because I haven't read the book and you probably have. Well, I've certainly... He's been on my show a couple of times and he argues that we are slouching towards utopia in the sense that, in spite of all the technological progress, we're miserable and he doesn't... And he tries to make sense of that. And he also suggests that DeLong's 20th century came to an end in 2008 and he's very troubled by the inequalities of the 21st century after 2008. So, to call DeLong a techno-utopian, I think, is a little bit of an exaggeration. I'm not sure what... It would be interesting, actually. I should get him back on keynote. It'd be interesting to see what he would think of Andreessen's techno-optimistic manifesto, but it's complicated. But it is true that he notes that over the long 20th century, I think he dates it between 1870 and 2008, we've made enormous economic progress. But after 2008, he notes things have changed. So, it's a complicated story. He's very good on globalization as well. Yeah, he's a great guy and he's a really nice guy. We should actually, if we ever got a third person on the show for a guest appearance, we should get him on. Speaking of Ponzi schemes, you link to an SBF piece this week on... A brilliant piece, actually, in The Verge about what a fool he's making of himself in the trial. Is he going to go to jail for a long time? You've always been... You've always said, well, let's wait for him to have his day in court. He's having his day in court. What's your take on SBF? There doesn't yet appear to be any evidence that he can get out of serving a long time in jail. His latest, and this is as of yesterday, his latest explanation is that everything he did was signed off by lawyers. Which is an absurd defense. I mean, it's ludicrous. Yeah. Since they're lawyers appointed by him. Well, the interesting thing about lawyers is they're paid to achieve what you ask them to achieve. So their job is to make what you want to do possible. Their job isn't to tell you what you kind of can't do, although that does enter into it. If you want to break the law, they'll tell you you can't do it. But if you want to stay within the law, but achieve something, even something, let's call it dodgy, they'll find a way, if it is indeed possible, for you to do it legally. And that sounds like that probably is what happened. Yeah. The piece is Sam Bankman-Fried is going to talk himself right into jail by Elizabeth Lopato in The Verge. It's a really good piece. Yeah. No, it is. Yeah. Well, what else is happening this week, Keith? You went on a bit of an anti-Mark Zuckerberg rant, but you also have a piece about Threads doing rather well, closing in on 100 million users, according to Zuckerberg. Is he lying? You suggest he usually lies. No, I think that's probably true. But we should contextualize that that's 100 million out of 2.7 billion Facebook users. And in the first week it came out, that was the number he said installed it. So I think he's spinning the story a bit in his favor. It's about three or four times smaller than X after, what is it, two months or three months? Right. X, you didn't link to any X pieces this week, but there's always anti-mask, anti-X stories every week. But there seems to be more and more of them in mainstream media. More and more people are arguing that the traffic is down, the revenue is down, the interest is down. There's always a core group of X users who aren't going away, maybe hardcore journalists, but most people just don't have the time or the interest in it anymore. Have you? You've always been a Musk fan, a supporter of X. Are you getting tired of it? Is it annoying? It isn't actually. I mean, it's a massive source of learning for me about people and opinions, and often quite deep. I mean, quite a few of the stories I read, I originally discover there. You've got to remember, it's still three or 400 million people use it weekly. So you're still a believer, Keith. What about all the stories about revenue being down, users being down? Those are real, aren't they? The best context that I've seen around that is that Musk isn't trying to save Twitter. He's trying to build X. For example, this week, they launched video and voice calling on X. He's trying to turn it into the global equivalent of what in China is called WeChat or WeChin, which is the app for everything. Yeah, and that's always been his goal. He's always made that clear. But can he actually do it? I wouldn't bet against him. I mean, the fact that he's retained that number of users this far in, in the face of what I think is essentially an old left woke kind of cancel culture attack on him. It says that he's winning against them. They've all gone to threads and they're talking to themselves. Well, Mike, I hate X, but I hate all of social media. And whenever I go on X, it seems to be the same people talking to themselves. But who knows, we will see at some point, this will become clearer. Another interesting story you link to is Apple TV plus cancelling the Jon Stewart show. What's happening here in terms of online media? I was very surprised with that. You think Jon Stewart is a winner one way or the other. He must have invested considerable amount of money in it. Yeah. I think we don't really know what's happening. We can read between the lines and guess. But the guess is that Jon Stewart was going to do a show that was very negative about AI and that Apple, as the owner of the network, took an editorial offence at his direction, led to an argument. And Stewart quite rightly said, well, you're taking away my right to my opinion, so I'm not going to do this. And they fell out. That sounds roughly like... Maybe he'll sign the Westminster Declaration. Maybe he'll put his show on X. By the way, that said, I'm a fan of Stewart's. I loved when he did the Daily Show. His Apple shows have been hard to watch. They're way too earnest. And it's a bit like he's Jesus speaking from the mountain every show. He definitely should be signing the Westminster Declaration. Yeah. But so Apple doesn't come out of that. I mean, maybe they come out of it looking okay. No, I think they look bad. I can't... You've also included Apple though. So Apple is good and Apple is bad. Good apples, bad apples. Apple is the startup of the week, a little company. What is it? $3 trillion? Why are they startup of the week, Keith? Well, they've produced a new app that you can only see right now if you get on the beta of iOS, which is free. You can sign up for it and your phone will update to iOS 17.2. There's an app called Journal. And it's a journaling app, which is one of the successful sub sectors of the app store is journaling. What does journaling mean? You mean writing about yourself? Writing about yourself every day. Actually, I drilled into it and it means a lot of things. So some people use it to talk to themselves. Some people use it to remind themselves of good advice. Other people use it as kind of a diary. Sounds like X. It could be. It's a subset of X, really. Now, what's interesting is ChatGPT is also being used in this way. So a lot of people are beginning to use ChatGPT as a kind of a therapist, a daily therapist, documenting their daily challenges and asking ChatGPT for advice, which is probably a way better version of Apple's Journal because it talks back to you. Whereas Apple's Journal is just one way. So I personally wouldn't journal, but I might talk to ChatGPT every day. Talk to me. Maybe I can be your therapist where you are. Unpaid, of course. Maybe I should charge you finally. And we talked about and this is a very interesting X of the week. We talked about intellectual property, ownership and copying. And you have something from Vinod Khosla, one of the smartest and most successful VCs around, also perhaps not the most likable fellow, but so what he talks about. I had someone on my show actually this week who's suing one of the AI platforms and artists for stealing their content. But he writes to restrict AI from training on copyright material would have no precedent in other forms of intelligence that came before AI train. There are no authors of copyright material that did not learn from copyright works, be it in manuscripts, art or music. You can't separate Gauguin's influence from Matisse, Velasquez from Dali, Picasso's from Pollock. I don't know if Picasso came before Pollock. Beyonce's from Taylor Swift, nor Charles Taylor's from Yuval Noah Harari. The list goes on. So what's his point? Well, he says it very well, doesn't he? He makes the point there's no such thing as original thought that isn't derived from prior copywriting. Yeah, but there are laws still around it. You can't just write a book that just copies somebody else. The music industry is very strict on just taking music or content from someone else. Yeah, but copyright covers copying. That's why it's called copyright. AI doesn't copy. Neither do great artists copy, but they do derive.