WEBVTT - Jerry Neumann on the Problem With Investing in AI Right Now 0:00:02.720 --> 0:00:16.480 Bloomberg Audio Studios, Podcasts, Radio News. 0:00:18.680 --> 0:00:22.200 Hello and welcome to another episode of the Outlass podcast. 0:00:22.280 --> 0:00:24.439 I'm Joe Wisenthal and I'm Tracy Allaway. 0:00:24.560 --> 0:00:27.160 Tracy our colleague here at Bloomberg at Harrison, had interesting 0:00:27.160 --> 0:00:30.120 newsletter today and it's actually something I've been thinking about 0:00:30.280 --> 0:00:32.159 a little bit lately, which is that for all the 0:00:32.240 --> 0:00:35.320 talk of the AI boom or the A bubble driving 0:00:35.320 --> 0:00:38.279 the stock market, there's no AI pure plays really that 0:00:38.320 --> 0:00:41.400 are publicly traded. Like in video is probably the closest, 0:00:41.680 --> 0:00:43.680 but three years ago people were excited because they were 0:00:43.760 --> 0:00:47.839 mining ethereum. Before that, it was like video games. You know, 0:00:47.920 --> 0:00:50.159 this was only an AI company in people's mind. For 0:00:50.760 --> 0:00:54.720 since late twenty twenty two, Google still is you know, 0:00:55.160 --> 0:00:57.520 we know they're all investing in a ton. There's actually 0:00:57.600 --> 0:01:00.520 no like no AI company that people are excited about 0:01:00.520 --> 0:01:01.280 in the public markets. 0:01:01.320 --> 0:01:03.000 I mean, I think it's true. Here you have this 0:01:03.080 --> 0:01:06.679 thing that a lot of people would say is revolutionary technology, right, 0:01:06.720 --> 0:01:08.920 but you kind of have to decide if you're going 0:01:09.000 --> 0:01:10.920 to invest in it, is it going to be like 0:01:11.400 --> 0:01:15.319 upstream or downstream? And there doesn't really seem to be 0:01:15.560 --> 0:01:17.960 that much pure play, as you say. 0:01:18.280 --> 0:01:21.240 As our colleagues said, Verma might like to say, people 0:01:21.240 --> 0:01:23.240 are investing in the picks and the shovels, you know, 0:01:23.360 --> 0:01:25.640 this gold rush. I don't think he actually said that. 0:01:25.640 --> 0:01:27.240 It's just a million other people's but it is. 0:01:27.240 --> 0:01:31.319 It shovels, it's fun, a completely reasonable commentary, That's what 0:01:31.360 --> 0:01:31.800 I say. 0:01:32.560 --> 0:01:34.720 Is there? You know, it turns out instantly picks and 0:01:34.720 --> 0:01:37.839 shovels have been great. You know, you could have bought Caterpillar, 0:01:38.000 --> 0:01:40.800 or you could have bought some old school HVAC company 0:01:40.959 --> 0:01:43.959 that's providing cooling or heating or whatever and made a 0:01:44.000 --> 0:01:46.280 ton of money. So actually, it turns out, at least 0:01:46.319 --> 0:01:49.280 for the last few years, all those awful cliches have 0:01:49.320 --> 0:01:51.240 actually been big money makers and I should not make 0:01:51.280 --> 0:01:51.600 fun of them. 0:01:51.680 --> 0:01:53.440 Well, here's the other thing I would say. It does 0:01:53.520 --> 0:01:56.960 feel like everyone kind of agrees at this moment in 0:01:57.080 --> 0:02:00.000 time that there is froth in the market. Maybe it's 0:02:00.120 --> 0:02:02.320 on a massive bubble, right, but there's some froth, and 0:02:02.400 --> 0:02:06.120 everyone is kind of admitting or saying that you're going 0:02:06.160 --> 0:02:09.520 to have some companies that emerge as big winners, much 0:02:09.680 --> 0:02:11.960 like the dot com era, and then a bunch of 0:02:12.000 --> 0:02:16.120 companies that like actually end up being losers, and I 0:02:16.120 --> 0:02:19.280 think again, like that is consensus at this point, but 0:02:19.720 --> 0:02:22.959 it doesn't really necessarily translate into actual investment, because of 0:02:23.000 --> 0:02:25.320 course the trick is actually picking the winners and losers 0:02:25.320 --> 0:02:26.240 in the market. 0:02:26.440 --> 0:02:26.959 Yeah. I don't know. 0:02:27.000 --> 0:02:28.760 It just seems like a weird point in time where 0:02:28.760 --> 0:02:31.640 people are like, oh, yeah, AI is great, but we 0:02:31.720 --> 0:02:33.480 all know that some of these companies are going to 0:02:33.480 --> 0:02:34.560 be massive losers, right. 0:02:34.760 --> 0:02:36.800 I think now they're all kind of big treated as winners, 0:02:36.800 --> 0:02:38.720 which is the other thee Yeah, anyway, it's a very 0:02:38.760 --> 0:02:40.920 weird time. You got to do more episodes on this 0:02:41.000 --> 0:02:43.399 because it is sort of the central question for on 0:02:43.440 --> 0:02:45.720 whether we're just talking about the market or talk about 0:02:45.720 --> 0:02:48.600 the economy, et cetera. Someone I've wanted to talk to 0:02:48.880 --> 0:02:51.640 for a long time. Earlier this year he wrote a 0:02:52.200 --> 0:02:55.119 essay for Colossus called AI will Not Make You Rich. 0:02:55.280 --> 0:02:57.359 It came out in September. It seems like ages ago. 0:02:57.720 --> 0:02:59.800 It's very disappointing because I think a lot of people 0:02:59.840 --> 0:03:02.440 really are hoping to get rich on AI, so this 0:03:02.480 --> 0:03:03.840 is a very unwelcome message. 0:03:03.880 --> 0:03:06.760 It's also very much a sort of core odd Lots 0:03:06.840 --> 0:03:10.480 thesis because in the essay he compares and contrasts AI 0:03:10.760 --> 0:03:13.760 with containerization, which is another which facing. 0:03:13.880 --> 0:03:15.800 So let's just get to the guest. Someone I've wanted 0:03:15.800 --> 0:03:18.480 to talk to for a very long time, someone who 0:03:18.520 --> 0:03:21.640 literally is the perfect guest longtime VC started a venture 0:03:21.639 --> 0:03:25.040 investing in nineteen ninety seven. Also a professor at Columbia 0:03:25.040 --> 0:03:28.040 Business School. We're going to be talking to investor Jerry Newman, 0:03:28.120 --> 0:03:31.680 also the co author of a recent book, Founder Verse Investor, 0:03:31.800 --> 0:03:34.640 The Honest Truth about Venture Capital from Startup to IPO. 0:03:35.000 --> 0:03:36.360 Maybe he'll tell us one we'll see. 0:03:36.200 --> 0:03:39.440 You the bringer of excess Halloween candy. So he gets 0:03:39.760 --> 0:03:41.000 he gets brownie points for that. 0:03:41.280 --> 0:03:43.880 Literally the perfect guest. Uh, Jerry, thank you so much 0:03:43.920 --> 0:03:46.000 for coming in. Thrilled to finally have you here. 0:03:46.040 --> 0:03:47.040 Thanks, I'm glad to be here. 0:03:47.160 --> 0:03:48.720 What does that mean? A I won't make you rich. 0:03:48.800 --> 0:03:52.160 AI has made people super rich, and it's making people 0:03:52.240 --> 0:03:53.360 rich every single day. 0:03:53.960 --> 0:03:55.640 You know, as an old mentor used to say, money 0:03:55.680 --> 0:03:58.600 is not money till it's cash. Okay, So is anybody 0:03:58.600 --> 0:04:00.120 really rich yet? Oh? Come on? 0:04:00.440 --> 0:04:03.680 I mean Jensen Wong bought an entire bar in Korea. 0:04:03.760 --> 0:04:05.760 He's like bought beer and fried chicken for everyone. 0:04:05.840 --> 0:04:08.760 He's right, I think it's smart to cash out early. Okay, 0:04:09.080 --> 0:04:10.800 that's what he was doing. Okay, let's say you I 0:04:10.840 --> 0:04:12.720 really believed it, would you be selling at stock now? 0:04:12.840 --> 0:04:15.600 I mean, see what you mean talk about this because 0:04:15.600 --> 0:04:18.400 you obviously have a lot of experience. Actually, that brings 0:04:18.480 --> 0:04:20.160 another line of question that I want to get into. 0:04:20.240 --> 0:04:22.440 But what does that mean it's smart to cash out 0:04:22.480 --> 0:04:24.520 early when in your experience? What does that actually mean? 0:04:24.839 --> 0:04:29.320 So, look, I believe that AI is a revolutionary technology. Okay, 0:04:29.320 --> 0:04:30.640 I'm going to put that on the table. 0:04:30.440 --> 0:04:32.560 Which is important to say because not everyone agrees with that. 0:04:32.600 --> 0:04:34.960 So that's yeah, totally. You know, I'm on Blue Sky 0:04:35.000 --> 0:04:37.839 and nobody agrees with that, but I do. I think so. 0:04:38.279 --> 0:04:41.160 But there's a difference between value creation and value capture. 0:04:41.760 --> 0:04:46.520 So even if AI creates a lot of value for society, 0:04:46.880 --> 0:04:49.039 who's going to get that value? Is it going to 0:04:49.080 --> 0:04:51.880 be the early investors? Is it going to be the court, 0:04:51.920 --> 0:04:53.960 you know, the foundation model companies. Is it going to 0:04:54.000 --> 0:04:57.200 be YEA consumers? You know? They think that's the question 0:04:57.240 --> 0:04:58.040 people need to ask. 0:04:58.440 --> 0:05:00.920 I actually broadly agree with this thesis when it comes 0:05:00.960 --> 0:05:04.080 to AI, but maybe just to clarify the idea here, 0:05:04.240 --> 0:05:08.760 compare and contrast this current AI cycle with maybe previous 0:05:08.800 --> 0:05:11.960 technological breakthroughs. And you know, I mentioned containers. I think 0:05:11.960 --> 0:05:14.400 a lot of people aren't used to thinking about boxes 0:05:14.640 --> 0:05:18.200 as this major advancement in technology, but at the time 0:05:18.440 --> 0:05:19.080 they really were. 0:05:19.680 --> 0:05:22.320 Yeah, I mentioned containerization, and most of my peers think 0:05:22.360 --> 0:05:25.440 I'm talking about doctor. So I'm talking about shipping containers. Right, 0:05:25.440 --> 0:05:26.960 the big boxes they put on ships, and then they 0:05:27.000 --> 0:05:29.400 can move from the ships onto the rail cars and 0:05:29.400 --> 0:05:32.200 now to the back of trucks. And this was a 0:05:32.279 --> 0:05:35.520 revolutionary technology. I mean it changed everything about the way 0:05:35.560 --> 0:05:37.520 we live. I don't remember. I'm probably a little older 0:05:37.520 --> 0:05:39.680 than you all, but when I was a kid, my 0:05:39.720 --> 0:05:42.600 grandmother used to send up oranges from Florida at Christmas time, right, 0:05:42.640 --> 0:05:45.120 because they were rare. You couldn't just go into a 0:05:45.120 --> 0:05:47.920 grocery store and buy them. Now you can buy oranges 0:05:48.040 --> 0:05:51.280 anywhere at any time. People don't really realize how much 0:05:51.279 --> 0:05:54.320 our lives have changed because of shipping containerization, because of 0:05:54.400 --> 0:05:59.400 these global logistics and the globalization of shipping. Now this 0:05:59.480 --> 0:06:03.200 is utionary technology. Who got rich from it? I mean generally, 0:06:03.240 --> 0:06:06.200 if you look at the nineteen sixties, say how many 0:06:06.240 --> 0:06:10.520 people became wildly rich from technological innovation? Can you think 0:06:10.520 --> 0:06:13.920 of anyone? Because I've been asking this question for years. 0:06:14.320 --> 0:06:16.160 There are people who got rich in media and whatnot, 0:06:16.160 --> 0:06:19.440 but there wasn't a lot of technological innovation that made individuals. 0:06:19.240 --> 0:06:22.120 It just start, like twenty years ago, did they have technology. 0:06:21.640 --> 0:06:24.440 Bout the right? I mean, it's I think this is 0:06:24.480 --> 0:06:26.599 the thing. Right, So we talk about computer technology, the 0:06:26.640 --> 0:06:30.920 information and computer technology revolution as technology, but obviously this 0:06:31.000 --> 0:06:33.920 has always been technology. But only at certain times in 0:06:33.960 --> 0:06:36.800 this technological cycle do people seem to make money as 0:06:36.839 --> 0:06:38.159 investors and as inventors. 0:06:38.800 --> 0:06:40.880 Explain more, though, because I mean I could argue that 0:06:41.120 --> 0:06:45.240 Marisk or someone like that got pretty rich off of containerization. 0:06:45.360 --> 0:06:47.880 Like maybe it took a while, but even though the 0:06:47.880 --> 0:06:51.039 shipping industry is highly cyclical, when they are in the 0:06:51.080 --> 0:06:52.840 boom period, they make a lot of money. 0:06:52.960 --> 0:06:56.039 Sure, I mean, the existing shipping companies got very large 0:06:56.279 --> 0:06:58.840 and made a lot of money. They got larger and 0:06:58.839 --> 0:07:02.240 made more money. Is mayor you know who made money. 0:07:02.880 --> 0:07:04.000 Completely new entrants? 0:07:04.120 --> 0:07:06.360 Yeah? So just to me as background, I'm a venture capitalist, 0:07:06.400 --> 0:07:08.200 right or have been a venture capitalist for a long time, 0:07:08.760 --> 0:07:12.560 recently retired, And I think about people investing in making 0:07:12.640 --> 0:07:16.680 money new companies, inventors or entrepreneurs making money, not the 0:07:16.720 --> 0:07:19.440 existing incumbents making money. And I think that people will 0:07:19.440 --> 0:07:22.160 make money on AI. Might be Microsoft making a ton 0:07:22.200 --> 0:07:24.400 of money on AI. It could be AI. You know, 0:07:24.760 --> 0:07:29.320 when containerization shipping containerization came around, Sealand was the instigator 0:07:29.360 --> 0:07:32.600 of this, and the founder of Sealand made money primarily 0:07:32.640 --> 0:07:35.800 because he sold early, right, he sold Sealand to r 0:07:35.840 --> 0:07:37.800 J Nibisco or sorry, it was just r Jr. At 0:07:37.800 --> 0:07:40.400 the time, and they thought they were diversifying, which is 0:07:40.440 --> 0:07:42.200 the big thing. Then paid them a lot of money 0:07:42.200 --> 0:07:43.440 and then they drove it into the ground. 0:07:43.480 --> 0:07:45.560 So what did Sealand do? What was did they? I 0:07:45.600 --> 0:07:47.440 actually don't I'm not familiar with this company at all, 0:07:47.440 --> 0:07:49.000 which I think kind of speaks to your point. But 0:07:49.040 --> 0:07:49.840 what it was Seland? 0:07:49.960 --> 0:07:51.960 Yes, So so it was a truck. It was started 0:07:51.960 --> 0:07:55.000 out as a trucking company. And the founder of Sealand 0:07:55.040 --> 0:07:58.480 was a trucking entrepreneur. And he said, it's silly you 0:07:58.800 --> 0:08:01.600 you go into a port, your truck sits around all day. Well, 0:08:01.880 --> 0:08:03.680 you know, the long charman put a cargo net into 0:08:03.680 --> 0:08:06.400 a container ship, load everything into it, pull it out, 0:08:06.720 --> 0:08:09.000 unload it and then reload it back into your truck. 0:08:09.040 --> 0:08:12.960 This is not efficient. And this obviously it's an obvious idea, right, 0:08:13.040 --> 0:08:14.440 just put it all in a box and you can 0:08:14.480 --> 0:08:15.680 then put that box on a truck. 0:08:15.800 --> 0:08:18.600 The best ideas are always the obvious rights in retrospect. 0:08:18.760 --> 0:08:21.200 But the problem with it was it was a systems problem, right. 0:08:21.440 --> 0:08:23.840 The long charman didn't want it, the ports didn't want it, 0:08:23.960 --> 0:08:26.120 the port authorities didn't want it, the politicians didn't want it. 0:08:26.120 --> 0:08:29.400 Nobody wanted this to happen because this sort of enormous 0:08:29.480 --> 0:08:31.360 change would put a lot of people out of jobs. 0:08:31.360 --> 0:08:33.199 It would up set the existing order, and it did. 0:08:33.320 --> 0:08:36.240 I live in Hoboken, and in Hoboken there's a lot 0:08:36.240 --> 0:08:39.360 of peers that nobody uses except to go running on now, 0:08:39.760 --> 0:08:42.479 because back in the sixties it was a long sharmantown. 0:08:42.840 --> 0:08:45.240 And when I moved there in the early nineties, it 0:08:45.280 --> 0:08:48.240 was empty. It started to gentrify, but because all of 0:08:48.240 --> 0:08:49.959 those people had lost their jobs and moved out. 0:08:50.200 --> 0:08:52.960 And I suppose, like even in the case of marisk 0:08:53.040 --> 0:08:55.000 And I'm sure you know, they obviously have made a 0:08:55.000 --> 0:08:57.920 lot of money because the explosion of global trading volumes 0:08:57.960 --> 0:09:02.360 and containerization is part of it. Like, it wasn't overnight wealth, right, 0:09:02.440 --> 0:09:02.880 it wasn't. 0:09:03.400 --> 0:09:03.840 It was not. 0:09:03.920 --> 0:09:06.880 It was not people got richer, but it was not 0:09:07.040 --> 0:09:09.840 like some bubble get rich quick thing where they suddenly 0:09:10.000 --> 0:09:11.079 cashed in on a new thing. 0:09:11.280 --> 0:09:14.720 Yeah. I mean, I suppose whoever owns mask may have 0:09:14.760 --> 0:09:17.439 gotten richer. Yeah, but it's not like you're going to 0:09:17.440 --> 0:09:18.920 look at the forest four hundred and see all these 0:09:18.960 --> 0:09:22.120 shipping magnates who became suddenly, you know, enormously wealthy. There 0:09:22.120 --> 0:09:22.480 are a few. 0:09:22.640 --> 0:09:25.200 Wait, okay, so if I think about a box, you 0:09:25.240 --> 0:09:26.160 know part part of that. 0:09:26.320 --> 0:09:27.960 We just keep this whole conversation on boxes. 0:09:28.080 --> 0:09:33.840 Actually, well one more question then we will be able 0:09:33.840 --> 0:09:36.320 to discuss ai. But I think about a box, and 0:09:36.400 --> 0:09:39.560 as you say, it's sort of an organizational structural problem, 0:09:39.679 --> 0:09:44.760 Like the box itself is not the huge technological advancement necessarily. 0:09:45.160 --> 0:09:49.760 So what exactly was it about containerization that prevented it 0:09:49.800 --> 0:09:54.040 from being disseminated I guess to new upstarts or new 0:09:54.080 --> 0:09:56.560 companies that could actually use that technology. 0:09:56.679 --> 0:09:59.400 Well it was, so I'm not sure I understand the question, 0:09:59.400 --> 0:10:02.719 because uanization became widespread very quickly, right, But. 0:10:02.720 --> 0:10:05.439 What I mean is like, why was that value seemingly 0:10:05.520 --> 0:10:08.199 captured by incumbents versus startups? 0:10:08.280 --> 0:10:11.719 Right? I think because it disseminated so quickly, Right, it 0:10:12.440 --> 0:10:14.760 was an obvious idea. Everybody who saw it said, Okay, 0:10:14.880 --> 0:10:17.199 we need to do this. Right. Everybody who's already in 0:10:17.240 --> 0:10:19.000 the business said, if this is going to happen, we 0:10:19.040 --> 0:10:20.920 have to do it. We can't be left behind. We 0:10:20.960 --> 0:10:22.880 will be left behind if we don't do it, which 0:10:23.080 --> 0:10:26.040 you know, I think was also obvious. So the reason 0:10:26.080 --> 0:10:27.920 nobody else did it first was it was hard to do. 0:10:28.040 --> 0:10:30.280 It was hard to make happen. And technology has always 0:10:30.280 --> 0:10:34.000 come in these technological systems if they're worthwhile technologies. Right. 0:10:34.040 --> 0:10:37.520 So the personal computer didn't change the world on its own, right. 0:10:37.559 --> 0:10:40.280 It changed the world alongside the Internet, you know, alongside 0:10:40.320 --> 0:10:42.840 a bunch of technologies that formed the system. So the 0:10:42.880 --> 0:10:45.600 hard part here was building the system, not the individual technologies. 0:10:45.920 --> 0:10:48.120 And this is true I think of computers as well. 0:10:48.480 --> 0:10:52.400 The first microprocessors weren't considered revolutionary. Intel didn't consider the 0:10:52.440 --> 0:10:56.000 four thousand and four revolutionary. They considered it evolutionary. The 0:10:56.040 --> 0:10:59.040 engineers have said this, And it wasn't revolutionary until people 0:10:59.040 --> 0:11:01.680 put it to use in ways that they didn't anticipate. 0:11:01.520 --> 0:11:04.320 Actually, can we go. I didn't know that, like, I 0:11:04.360 --> 0:11:07.240 hadn't really thought about that with Intel that at the 0:11:07.360 --> 0:11:10.319 time it didn't feel to them that it wasn't a 0:11:10.360 --> 0:11:12.960 revolutionary technology. That's sort of mind blowing. 0:11:12.880 --> 0:11:15.120 It is, right. This is I think from Michael Malone's 0:11:15.120 --> 0:11:17.080 The Intel Trinity of the book. You know, he interviewed 0:11:17.080 --> 0:11:19.560 a bunch of Intel engineers and he said, you know, 0:11:19.760 --> 0:11:21.760 like they thought they were building a better chip set 0:11:21.800 --> 0:11:25.080 to build pocket calculators or desk calculators. I just say so, 0:11:25.120 --> 0:11:27.240 they had a client, Busycom, who wanted to build a 0:11:27.280 --> 0:11:30.360 better desktop calculator with calculators were big back then in 0:11:30.440 --> 0:11:33.360 nineteen seventy ish, and one of the engineers said, well, 0:11:33.360 --> 0:11:35.520 why do we keep building custom chip sets for each 0:11:35.559 --> 0:11:37.720 different calculator. Why don't we just build a chipset that 0:11:37.720 --> 0:11:40.600 we can customize the software and change what it does. 0:11:40.640 --> 0:11:44.800 And Intel is kind of like eh, and Busycom was like, okay, 0:11:44.840 --> 0:11:46.880 we'll pay for that. And then Busycom actually tried to 0:11:46.880 --> 0:11:48.600 back out and they gave the rights back to Intel. 0:11:48.600 --> 0:11:50.959 So Intel owned the rights to this four thousand and four, 0:11:51.360 --> 0:11:54.040 and then they started selling it and it wasn't Intel. 0:11:54.480 --> 0:11:56.000 You know, Intel believed at the time it was going 0:11:56.040 --> 0:12:00.160 to be maybe used for dedicated hardware, you know, can 0:12:00.200 --> 0:12:02.600 you hardware controllers that kind of thing, not by consumers. 0:12:02.960 --> 0:12:05.319 So it wasn't until people on the outside said, hey, 0:12:05.360 --> 0:12:07.679 you know, I love these IBM mainframes or these deck 0:12:07.720 --> 0:12:09.840 MANI computers. I'd like to have my own, but obviously 0:12:10.320 --> 0:12:12.240 nobody can afford that. Why don't I just try to 0:12:12.240 --> 0:12:15.319 build my own? Right, So there's these kind of outside inventors, 0:12:15.640 --> 0:12:19.439 this permissionless invention, and then it really the real revolution 0:12:19.559 --> 0:12:22.080 didn't happen until this. You know, everybody's like, oh Intel. 0:12:22.280 --> 0:12:24.480 It was the six five oh two where the price 0:12:24.520 --> 0:12:27.640 came down so dramatically that, you know, Steve Wozniak could 0:12:27.640 --> 0:12:30.240 walk into a computer fair you get some for free 0:12:30.280 --> 0:12:31.760 and go home and build up a personal computer. 0:12:31.880 --> 0:12:33.800 It's crazy. I wonder who made the chips for the 0:12:33.840 --> 0:12:36.480 Commodore sixty four computer that I had. 0:12:36.559 --> 0:12:38.959 They were six five o twos. Those were Intel, that's 0:12:39.080 --> 0:12:41.040 you know, they weren't until there were Moss technologies. Oh 0:12:41.160 --> 0:12:43.200 got it, got it, those are the cheap ones. 0:12:43.240 --> 0:12:45.120 Then yeah, wait, what year was that when you had well, 0:12:45.160 --> 0:12:48.360 I had I think I like learned some basic and 0:12:48.679 --> 0:12:52.640 did some coding on it. I would have said maybe 0:12:53.120 --> 0:12:56.319 nineteen eighty eight, nineteen eighty seven somewhere around there, made 0:12:56.320 --> 0:12:59.599 a few I missed those days. It's crazy that I 0:12:59.640 --> 0:13:01.920 didn't be. Can I just say I sometimes when I 0:13:01.960 --> 0:13:04.160 think about my life path, like how did I not 0:13:04.320 --> 0:13:05.640 end up like a tech guy? Because I was like 0:13:05.720 --> 0:13:08.000 very into math. I was like one of those people 0:13:08.000 --> 0:13:10.520 who had a computer when I was six or seven. 0:13:10.679 --> 0:13:13.079 In nineteen eighty seven, you were coding you were seven 0:13:13.160 --> 0:13:13.520 years old. 0:13:13.720 --> 0:13:16.160 Yeah, I got that. My dad got me this magazine 0:13:16.800 --> 0:13:19.040 that just it was very crazy. There was literally just 0:13:19.080 --> 0:13:21.560 sent you pages of code and then you just typed 0:13:21.600 --> 0:13:23.240 it in and you could like make a video game. 0:13:23.400 --> 0:13:24.960 I was doing that at seven. I could be like 0:13:25.000 --> 0:13:27.360 one of those like who's per s gave of a 0:13:27.400 --> 0:13:30.080 computer when he was seven? It anyway, and it's not 0:13:30.120 --> 0:13:30.480 too late. 0:13:30.800 --> 0:13:31.480 It's not too late. 0:13:31.800 --> 0:13:33.920 Yeah, it isn't too late. Well, maybe it is too 0:13:33.960 --> 0:13:52.360 late because now we have AI doing all the coding, right, Okay, 0:13:52.400 --> 0:13:54.680 just so I understand when it comes to I guess 0:13:54.679 --> 0:13:58.880 the advantages of the incumbents to actually monetizing new technology 0:13:58.960 --> 0:14:02.120 or benefiting from new technology is the moat around their 0:14:02.160 --> 0:14:06.240 business the network and their sort of role in the network, 0:14:06.600 --> 0:14:10.000 or is it the vast amounts of cash they have 0:14:10.120 --> 0:14:12.800 and the ability to sort of roll out massive investment 0:14:12.920 --> 0:14:14.280 to capture that value. 0:14:14.559 --> 0:14:17.280 I think it's the latter, right, So anybody can build 0:14:17.640 --> 0:14:20.200 a foundation model, right if you have the money. I mean, 0:14:20.720 --> 0:14:23.440 the technology is not mysterious. It doesn't feel like the 0:14:23.440 --> 0:14:26.040 technology is really changing very quickly anymore. And of course 0:14:26.040 --> 0:14:28.680 I don't have insight into what's happening inside of open ai, 0:14:29.080 --> 0:14:31.040 but looking at it over the past couple of years, 0:14:31.040 --> 0:14:34.640 it's the same thing, but slightly better. It's evolutionary now, right. 0:14:34.960 --> 0:14:38.000 The first part was revolutionary and now it's evolutionary. So 0:14:38.040 --> 0:14:39.760 if you wanted to build one, you could build one. 0:14:39.760 --> 0:14:41.120 And I have friends who are running them on their 0:14:41.160 --> 0:14:45.400 laptops very slowly, but it's possible. So now the question 0:14:45.480 --> 0:14:47.440 is do you have enough cash to build the data centers, 0:14:47.800 --> 0:14:50.280 to buy all the chips, to build something that is 0:14:50.440 --> 0:14:53.040 large enough that when you train it, it does something useful, 0:14:53.360 --> 0:14:57.400 And it's just a question of having that the authority 0:14:57.400 --> 0:14:58.880 in the market to be able to raise that money. 0:14:59.000 --> 0:15:01.480 By the way, speaking of ideas that were sort of 0:15:01.520 --> 0:15:05.200 really obvious that took a while. I'm always blown away that, 0:15:05.280 --> 0:15:07.800 like how long it took them to put wheels on luggage. 0:15:07.880 --> 0:15:10.560 I don't think anyone like made got super rich on that. 0:15:10.560 --> 0:15:11.880 But when I was a kid, I remember like we 0:15:11.920 --> 0:15:14.920 had these big suitcases, and that's the most obvious thing. 0:15:15.240 --> 0:15:17.560 It took a while. Anyway, I don't think anyone goes. 0:15:17.440 --> 0:15:19.960 Technology is still not perfected, as you know, because you've 0:15:19.960 --> 0:15:21.600 been in airports with me and the wheels on my 0:15:21.640 --> 0:15:22.600 luggage are broken. 0:15:22.760 --> 0:15:25.280 But I don't think anyone got like super rich off 0:15:25.320 --> 0:15:27.240 of wheels. That just seems like an that was just 0:15:27.280 --> 0:15:27.800 sitting there. 0:15:27.920 --> 0:15:28.080 You know. 0:15:28.120 --> 0:15:30.600 I want to jump ahead actually in the conversation a 0:15:30.600 --> 0:15:32.800 little bit because I know to forget this point. But 0:15:32.880 --> 0:15:35.520 this is something I've become a little obsessed with, which 0:15:35.600 --> 0:15:38.080 is I've been meaning to ask a VC about this, 0:15:38.320 --> 0:15:40.120 which is that there seems to be this blurring of 0:15:40.200 --> 0:15:44.480 private and public markets in various ways, retail participation in 0:15:44.560 --> 0:15:49.080 private markets, et cetera. For the VC, in my mind, 0:15:49.200 --> 0:15:54.120 I feel like the exit was the IPO or the acquisition, right, 0:15:54.160 --> 0:15:56.960 so you buy do vcs these days have to think 0:15:57.000 --> 0:16:00.440 a little bit more about market timing and selling early. 0:16:00.600 --> 0:16:03.280 So someone who was an early investor in open ai 0:16:03.640 --> 0:16:05.680 or whatever you know, in the past, they might have 0:16:05.760 --> 0:16:08.119 just held or then sell at the IPO or the acquisition. 0:16:08.280 --> 0:16:10.400 It is probably going to happen in an open AI's case, 0:16:10.440 --> 0:16:13.360 they're too big to be acquired. But to vcs and 0:16:13.720 --> 0:16:15.760 in your experience these days, have to think a little 0:16:15.800 --> 0:16:19.360 bit more about this idea of selling early, timing the exit. 0:16:19.920 --> 0:16:22.160 Well, I think vcs always had to think about timing. 0:16:22.680 --> 0:16:24.400 You know, I've done pretty well in VC, and I 0:16:24.400 --> 0:16:27.600 attributed it entirely to being lucky at starting investing at 0:16:27.640 --> 0:16:29.440 the right time. So the first time around, I was 0:16:29.440 --> 0:16:32.240 starting in ninety seven, which any you could make money in. 0:16:32.240 --> 0:16:34.160 The second time around, I started in two thousand and seven, 0:16:34.520 --> 0:16:37.000 which again it was just two thousand and eight. It 0:16:37.080 --> 0:16:38.840 was just an easy time to buy in. 0:16:38.920 --> 0:16:41.600 But I mean, the timing of the sale is that 0:16:41.720 --> 0:16:43.840 something that where in the past it might have been 0:16:43.880 --> 0:16:47.440 automatic exit, now now is not the case. 0:16:47.600 --> 0:16:49.400 Yeah. I wrote this thing about VC in the nineteen 0:16:49.400 --> 0:16:52.160 eighties a long time ago on the blog. You can 0:16:52.160 --> 0:16:54.880 find it. And the thing because nobody talks about the 0:16:54.960 --> 0:16:57.400 nineteen eighties, right, there's plenty of VC in the eighties, 0:16:57.400 --> 0:16:59.360 but nobody talks about it. People talk about the sixties 0:16:59.360 --> 0:17:02.000 and in the nineties. So I was like, all right, 0:17:02.040 --> 0:17:04.560 what happened. Then. One of the interesting things about it 0:17:04.600 --> 0:17:07.040 is there were the IPO windows then where you know, 0:17:07.119 --> 0:17:08.960 nineteen eighty three the IPO window opened a bunch of 0:17:09.000 --> 0:17:10.600 companies on public and then it closed again, and you 0:17:10.640 --> 0:17:12.600 can see it in the numbers when people in public. 0:17:12.640 --> 0:17:15.520 So people always had to think about timing. The IPO 0:17:15.600 --> 0:17:18.000 is obviously the best exit because you want to sell 0:17:18.040 --> 0:17:20.600 to the greatest fool, and nobody's greater fool than the public, right, 0:17:20.680 --> 0:17:24.200 So you look for the IPO window to open. When 0:17:24.200 --> 0:17:26.320 you can't, you have to sell to somebody else. You know. 0:17:26.520 --> 0:17:30.240 Vcs have this problem of their limited fund life. So 0:17:30.600 --> 0:17:33.120 I look at my portfolio. I'm an really early investor 0:17:33.480 --> 0:17:35.720 or have been a really early investor, so I'll look 0:17:35.720 --> 0:17:38.359 at companies and say, oh, I invested ten years ago. 0:17:38.960 --> 0:17:40.480 They're going to have to sell, you know, it's the 0:17:40.920 --> 0:17:43.080 So people look for the IPO window, but if they 0:17:43.119 --> 0:17:44.680 can't find it, they have to sell somewhere else. 0:17:45.000 --> 0:17:47.600 Well, how much of the money flowing into AI startups 0:17:47.640 --> 0:17:50.560 now is just the expectation that a bunch of these 0:17:50.600 --> 0:17:54.520 little companies are eventually going to get bought by larger incumbents, 0:17:54.560 --> 0:17:56.600 and basically you're going to have consolidation and you will 0:17:56.640 --> 0:17:57.359 get that exit. 0:17:58.040 --> 0:18:00.600 I don't think anybody can predict when the IPO opens. 0:18:00.880 --> 0:18:02.840 I mean, I wish I could, but I don't think 0:18:02.960 --> 0:18:06.280 I've never seen anybody even say they could predict when 0:18:06.280 --> 0:18:08.680 the IPO window opens. So I think a smart VC 0:18:08.800 --> 0:18:11.600 invests in a company that can become self sustaining to 0:18:11.680 --> 0:18:14.440