Big Tech's Comeback And The Spillover to VCs and Startups — With Eliot Brown

0:00Capital One's tech team isn't just talking about multiagentic AI. They already deployed one. It's called Chat Concierge, and it's simplifying car shopping. Using self-reflection and layered reasoning with live API checks, it doesn't just help buyers find a car they love. It helps schedule a test drive, get pre-approved for financing, and estimate trade in value. Advanced, intuitive, and deployed. That's how they stack. That's technology at Capital One. The truth is AI security is identity security. An AI agent isn't just a piece of code. It's a first-class citizen in your digital ecosystem, and it needs to be treated like one.

0:40That's why Okta is taking the lead to secure these AI agents. The key to unlocking this new layer of protection? An identity security fabric. Organizations need a unified, comprehensive approach that protects every identity, human or machine, with consistent policies and oversight. Don't wait for a security incident to realize your AI agents are a massive blind spot. Learn how Okta's identity security fabric can help you secure the next generation of identities, including your AI agents. Visit Okta.com. That's O-K-T-A dot com.

1:15Welcome to Big Technology Podcast, a show for cool-headed, nuanced conversation of the tech world and beyond. We are joined today by Elliot Brown. He is a reporter at The Wall Street Journal and the co-author of The Cult of We, WeWork, Adam Newman, and The Great Startup Delusion. He's a terrific reporter. We have so much to cover today. We're going to talk about public market valuations of big tech companies and then get deep into Elliot's reporting about how that's spilling over into the private investment world, including SoftBank. Elliot, so glad to have you here. Welcome to the show. Happy to be here.

1:47Great to have you. had you around New Year's. And one of the things I said I love speaking with you about is froth, because you have this internal radar. When you see something not really makes sense in the numbers, you're very good at saying, hey, wait, let's double check here. That being said, let me read you some statistics. Okay, year to date, right? So since we spoke last, here's what big tech is doing. Apple is up 53%. Meta is up 129%. Amazon is up 51%. Microsoft is up 41%. Alphabet is up 34%. And Nvidia is up, this is unbelievable, a whopping 196 % and valued at$1.05 trillion. Wait a second.

2:32You know, this was supposed to be a moment where there's higher interest rates. tech has a pullback on its valuations. And then, of course, there were drawbacks last year. But what's happening this year is somewhat mind boggling. Are we seeing another sort of case of market hysteria or are these numbers actually legitimate? Did the market unnecessarily pull back? Yeah. What were all these people doing? Do they not realize we've got a high interest rate environment? Yeah. What's going on? That's what I've been saying. Yeah. What's the Yeah, I don't have the answer. I mean, basically, if you look at, yeah, like the NASDAQ overall is up like 30 % year to date and is now only like 10 % to 15 % below its 2021 highs when we had like zero interest rate environment.

3:21So what's actually happened? As you sort of pointed out, It's actually really just a few stocks. It is these giant incumbents. And, you know, the market is betting that they are like, you know, essentially AI is going to power these companies to sort of at a new profitable level. So is it irrational? I think in the broad sense, you know, AI is actually as a concept or generative AI is makes a lot more sense as something that would be transformative than basically anything else we've heard out of the venture capital sector in the past 15 years. You can sort of see how it actually would be, could lead to just extraordinary new efficiencies and profit centers within businesses by eliminating jobs and replacing it with code robots.

4:23That said, you look at NVIDIA's valuation, and I'm not actually here to really enlighten with any numbers and more just broad, sweepy statements, but it's trading at something like 30 times revenue, which means the value of the whole company is 30 times a year's revenue, new, let's say, that means profits are sort of minuscule. So it would just take years upon years at sort of a normal company valuation of profit or growth for NVIDIA to sort of pay that back. So that's a long way of saying like, even if NVIDIA just gobbles the entire AI chip market, how do you justify a$1 billion chip company?

5:15um that's probably harder uh but uh who knows um right so here's some concrete numbers i mean nvidia had doubled the sales of amd right amd is the comp not exactly the same company but double the sales amd's market cap is in the 190 billion range and that's a frothy market cap right like amd so yeah yeah that's a good one and so basically nvidia doubled the sales of its most comparable competitor? Five times, more than five times the market cap. How do you, I mean, is that how do you explain that? You know, NVIDIA is a meme stock in the same way that Tesla is a meme stock. Like that's, so what's and it's actually, NVIDIA has sort of always been that way.

6:00It's gotten like the gamers got really behind it and then the crypto people really liked it because their chips were used for crypto. So it's got a lot of retail sort of pushing up this cult. view about the thing. So look, my job isn't to sort of like rate the stock as buy or sell, but it's like, if you're wondering what is up with NVIDIA's valuation, does it completely reflect like a everything is normal environment? It's like, no, that's the result of sort of market hysteria and a frenzy. But in terms of Facebook, Meta, Google, in general, this sort of like big surge in the past six months, I think that is reflective of something that's at least has a lot more potential to be real than crypto or any of the other 12 waves we've seen in the venture capital world in the past decade, like scooters and fake meat and whatnot.

7:03With grilled cheese as a service. The grilled cheese, yeah. Gas. G-Chas. Grilled cheese as a service, yeah. G-Chas? Yeah, definitely. It's a hot new category. You actually do have an example in your book that talks about basically grilled cheese companies becoming tech company evaluations, or commanding tech company evaluations. So we're not just spitballing here. This did happen. So let's talk a little bit, though, about how this now impacts the startup market and private capital. Because there was a moment where it seemed like, okay, startup investors didn't really see exits. M &A was down. The public market was unwelcoming for IPOs.

7:44And you looked at where tech company valuations were plummeting. So it's not like you were pushing towards this exit and you realized that you might need more runway than you did previously. But now, holy crap, there's a real opportunity here, right, where like companies are getting started on AI programs and you can see the public market is really rewarding the faith. So I'm not as deep into it as you are, but I would imagine that this is going to lead or is leading right now to reinvigoration of venture capital and private investment space. Oh, totally. I mean, it's really happened. And it's kind of, it's almost frustrating in that it's impossible to talk to these venture capitalists a lot of the time because, you know, you actually had a moment for one year out of the past 10 where, you know, essentially the bubble had been pierced and maybe they could be introspective about, you know, sort of just why the sector as a whole just is so prone to being frothy and and sort of running like a herd at kind of all these crazy things, because it's like overall and through 2022, I mean, the sector did pretty terrible, even if like venture funds were able to produce good returns here at isolated times.

9:04So that's a long way of saying, yeah, I mean, things have gotten completely off to the races and people are sort of like forgetting like, you know, all that crap that we invested in and put hundreds of billions of dollars into, you know, forget about that. It is an AI frenzy. And again, like that as a concept is probably more rational than getting really excited about mattresses or scooters as some sort of, you know, huge new category, which is what VCs were doing.

9:40because the thing that really does unlock venture capital and venture capital returns is when you have an iPhone or the internet and you open up what they call a platform change. You get this whole new category that you can do. So you could see how AI has the potential to do that. The VCs all really see that. And so valuations have been surging. You have these startups with nothing that are just starting. There was a French startup that raised$118 million last month, and it had been around for five or six weeks. A week. Yeah, exactly. Like overnight. Right. It essentially was just a dude or three dudes, and they get$118 million to go build a company.

10:29That is the peak level of froth. And so then suddenly, like, you know, people are rebranding to be all about AI when when. But if you ask them how Web 3 is going, they probably would not appreciate the question. Right. And so, you know, it's interesting because I said this on CNBC last week and I got a little gruff from it from the commenters, but I'm willing to repeat it here where I said that there's, you know, an iPhone moment. It feels like an iPhone moment for generative AI. There's hasn't been an iPhone product yet. Like, actually, when you think about concrete products that we've seen with this stuff, you have ChatGPT, which actually the interest in ChatGPT is declining if you look at Google search trends, which is, you know, it's not a perfect predictor, but it's roughly interesting.

11:12It's an interesting data point. And then, you know, I'm certainly not using it for as many limericks anymore. Right. Exactly. It's almost like it seems like there is a chance that the novelty has worn off and now we're really waiting for what's going to happen. I mean, Character.ai is another startup that raised a billion-dollar valuation that you could like to speak with Thomas Jefferson or historical figures. And I really haven't seen anybody using their product at all. And it's not charting as far as I can tell. Thomas Jefferson is not that popular these days in case you haven't been paying attention.

11:45Right. No, we have to obviously cancel him. But seriously, do you think I'm off base here saying it's kind of this iPhone moment without the iPhone product? or am I being impatient? I mean, I don't, you know, I'm bad at sort of predicting the future for sure. Yeah, I think that like the jury kind of, or, you know, what smart people seem to say is like the jury's out on whether or not this, let's even assume that generative AI does become a huge thing and put aside the question of whether we're all just getting too excited by a chatbot. That's a good question though. Oh yeah, it's definitely a good question.

12:23Like, I mean, it's - Yeah, but go ahead. It can't be used for very much specifically. You ask how many CEOs of credit or UBS have there been in the past 10 years, it gives you the wrong answer, which is a real life example I did. It made up some quotes from Masa when I was trying to get it to help me on a story. So put that aside and I think then there's a question like, well, does it actually, as venture capitalists would say, do you have a moat? Is there any difference between all these companies that are doing it? And I really don't know enough about the technology. There's a lot of people who would be able to answer that pretty quickly.

13:05But is there anything that distinctive in terms of, if we fast forward a year, will essentially everyone have a pretty similar product to OpenAI? And if the answer is yes, then, or if there's three companies with the same product, then, you know, it's a lot harder to see how you sort of make money off that. And then it's complicated because it really does open up a whole bunch of new potential verticals or sort of areas that you can go into. And it's hard to sort of play that chess out. Now, it's interesting because it does seem like there's a bit of a bifurcation among VCs. And I wonder if this is the case that you hear from the ones that you speak with.

13:48but some are like all in on AI. And we're gonna talk about SoftBank in a minute. That's one of them. But others are like, I'm not even touching this stuff. So Keith Raboy, who is a VC, one of the big Miami crowd, was popping off on the fact that he doesn't care that AI companies aren't starting in Miami because he doesn't want to invest in them anyway. And he said this on Twitter and I said to him, do you think there's no major business opportunity there or just do competitive? And what he says is there's no venture capital returns there in generative AI. Very interesting. I'm kind of curious what you think, you know, what you think in the VCs that you hear about that don't want in on this.

14:23Like, what's their rationale? Yeah, no, that's a pretty interesting thought. I have not explored that in reporting. But I guess the general thing that's happening is, you know, and maybe another way of saying what he's saying is, like, the valuations have gone from, like, zero to a zillion overnight. And that means that, yeah, even if you pick a winner, like say you invest in OpenAI right now and OpenAI does win, they're worth what, like$25 billion? What's the expectation that in 15 years or 20 years, do they go to be a$100 billion company? That would sort of be at normal risk level? Like, there's a pretty high risk that won't happen.

15:10Is it likely to be a$250 billion company because they came up with the first chatbot? Like, I don't know, maybe. But there's not too many$250 billion companies. And so for VCs at sort of that level, they often are looking for like, yeah, a 10x return, maybe 5x for that sort of stage of the company because there's only so many companies that are going to emerge and you hope that one's going to win and sort of pay off the other bets of similar risk. That's sort of how that game works. So that's the way of saying when every company that has.ai after it is suddenly worth over a billion dollars, then it becomes really hard to play the sort of the VC game, which involves spreading 10 bets around and hoping that one does really well up from basically nothing.

16:06So it's hard to do that when you start at a sort of a mature valuation for a mature company. Right. And I think the crucial thing that it's done for venture capital and startups right now is it's giving them all a story, a story to go to LPs and say, even in a time of higher interest magistrates, please trust us with your money. A story to go to venture capitalists and say, you might be sitting on a lot of dry powder and being very careful about where you're going to make your bets because you need to justify to your LPs. But because we're an AI company, that's where you make the bet. And maybe that's sort of the shot on the arm because it did seem all throughout at least 2022 that we were about to head to a really bad place in terms of startup investing and venture capital where it was going to dry up.

16:48It seems like this has been the story that's reinvigorated it or given it reason to continue at previous levels. Yeah, or at least something to talk about on Twitter. I mean, remember that there's X thousand very valuable or companies that raise very large amounts of money out there that would struggle to suddenly convince anyone with a straight face that they're a generative AI company. So, I mean, you have something like a trillion or so of value companies that are not AI companies that were funded by sort of the Ancien regime, and they're still in trouble. That's not changing. But in terms of, yeah, venture capitalists are like, you know, don't ask me about that, that I just spent, you know, the world, the largest fund I've ever raised on.

17:42They're all be like, because now over here, things are just going to be dandy, which could be true. Are you talking about Andreessen Horowitz? No. You know, Andreessen certainly seems to be talking a lot about AI and less about crypto these days. Almost zero about crypto. Yeah, there's not as much coming on the A16Z podcast about Web3 or NFT. I mean, there were numerous episodes of that thing last year about NFTs, or two years ago, I guess. And now it's sort of like crickets on that thing where the fund was trying to convince the world that the entire internet was going to change to this weird tokenized energy intensive thing.

18:28So putting a bow on this, do you think we're going to see some IPOs? Because now the valuations in the public markets are better. Maybe for big tech, but I don't know. Maybe not the Ubers. It really depends. Like the overall, like the tech market isn't good. And if you, you know, compared to six months ago. And so if you strip out Google, you know, the big stocks, it still sucks. So if you were making a random shoe that you sold online to venture capitalists and they gave you lots of money in 2021, I don't think you can IPO tomorrow and get anything close to what you want. But what we are seeing, maybe this sort of segues to another thing you wanted to talk about, is if you do have an AI tie-in, then it's probably a good time.

19:27And so SoftBank, for instance, is planning to or appears to be planning a fall IPO of Arm, which is the chip maker that it bought for$32 billion back in 2016. Right. And it's almost doubled in value since then. So let's jump into the SoftBank. Well, we'll talk about it, but let's jump into the SoftBank topic. So you have a great story out in the journal recently about how Masayoshi-san, who runs SoftBank's vision fund, mentioned AI more than 500 times in quarterly and annual results presentations between 2017 and mid-2022. Yet, out of the 26 generative AI startups valued at more than$1 billion, SoftBank's only invested in one.

20:20I mean, how do you miss that? Yeah, and it even gets a little worse. They basically missed... There's this huge public market frenzy for AI, and then they basically don't have any exposure to that either. So they have like 35 companies that are public, and basically none of them are sort of caught in that wave, largely because it is just these top few big ones. a sort of funny twist of history is that they had like at one point they owned like 4.8 percent of nvidia uh back in 2017 and they spent like four billion on it and uh then they sold it in 2019 and it is up 10 times since then unbelievable how yeah so so talk us through like is it is this just like i mean they knew where it was going they just didn't they i guess they didn't have the conviction to hold NVIDIA?

21:20Although maybe NVIDIA wasn't talking about AI? How do you miss this bad if you think that this is so important? It's painful. If you go back to their slide deck, they showed investors in the public back in 2017 when they were buying NVIDIA. It's like, why are we doing this? And it's like, NVIDIA makes the chips that power the AI revolution or something like that. It's like, so they actually got it right. And then they sold. So NVIDIA, that was a miss. Now, in general, I think what this shows is sort of two things. It kind of shows the craziness about timing in venture capital. So basically, SoftBank has been more or less investing approximately gobs of money per quarter since 2017.

22:12gobs of billions. And then in mid or early 2022, they just shut the spigot off. They're like, oh, well, now that valuations are down 70 % or maybe 50 % by that point, it's like we realize it's a bear market, so it's a bad time to invest. Now, what sort of any academic will tell you, or even a lot of venture capital firms is you do really want to be consistent. I mean, you can go up and down a little, but like a lot of the best ideas because they're so cheap, a lot of the best venture capital wins come at bad times because prices are so cheap. And, you know, then also there's like changes that you don't really see.

23:02So like Airbnb, Uber, Facebook, all of those were sort of born out of venture capital winters. So, anyway, they shut the spigot off. At the very time they shut it off, that's when a lot of these AI companies got funding rounds because VCs were trying to find something new to get excited about. And then OpenAI happened, released ChatGPT. And so then the lemming-like frenzy goes crazy and everything jumps up. And so SoftBank's competitors have spent the past five months just showering all these companies with, or six months with money, and SoftBank has still been sitting there biding its time. And so they missed sort of a chance to get a big, quick return, and then also have missed the chance to get in these companies as valuations have sort of been going up day by day.

23:55So that's one answer. And then the other sort of chunk of it is, they said they were investing in AI, but they weren't. The money was going to companies that convinced SoftBank that they had AI, like WeWork and Compass and Greensill, which was essentially an alleged fraudulent lender. I don't think the alleged fraud has much to do with AI, if I recall the court docs. And what are some other? Oyo Hotels. I mean, basically any company. They said it was every company was about, you know, leveraging AI. And then you look today and maybe they're using AI, but it isn't really reflected in sort of the share price.

24:44Right. There was a line that they had about like a large chunk of their companies were using AI in their operations. And it's like, hold on a second. You can use AI in your operations, but it doesn't mean you're an AI company. For instance, if I use ChatGPT to help me formulate story ideas, I could call big technology an AI-powered company or a company that uses AI in its operations. But I'm not an AI company. I would like to write you a check. Exactly. I mean, there's a huge difference. You can sort of lie to yourself when you have this thesis and you're a lot of money and you're trying to jump out and try to deploy it.

25:22And that kind of seems like what happened with SoftBank. Yeah, I mean, there is a company that is just, they can't really catch a break. But like, I mean, they are a, you know, sort of since 2007 or so, which was the last time they made a really good investment, which was really good. They have. Which was? That was 2006. It was the buyout of the Japanese arm of Vodafone, which has given them billions of dollars to play with, probably tens of billions. They really haven't made many great investments since then. And a lot of what they've done is taken the winnings from their past successes and lost them, making peak of the market investments or they made many peaks of their own with WeWork and just flushing it down the toilet.

26:23So yeah, I think the numbers we use is they basically, they spent$140 billion on at least two funds that were rhetorically fully devoted to AI, and they basically have nothing to show for it. Now, the big caveat there is that ARM, the chip maker I was talking about a minute ago, that's, if you asked analysts six months ago, they would have said ARM would probably IPO for like, who knows,$40 billion. Now they're thinking like 60, 70. So it's actually a pretty bad return over seven years. Like, that's not a good return. It's a bad return. We're certainly below the market and below other chip makers.

27:09But, you know, I didn't make$30 billion in six weeks. There were six months. Like, that's pretty hard. So that's pretty good. Like, that's really helped SoftBank's stock. It's a huge amount of extra money that they can play with. so you know in that sense they did they have a big win but it just wasn't you know sort of the part of the the startup investment frenzy that they were doing how much does soft bank matter like in terms are they that's still the biggest biggest vc firm like do they set the agenda for others depends how much they're going to invest and and right now they're investing nothing so So they don't matter in that sense or virtually nothing.

27:53If they start opening the tap again, then they can really sort of move the market, which we've seen a few times before since 2017 when they launched the first vision fund. And so, I mean, it's just a flood of money that really does change the dynamic, especially depending if they target at a specific area or not. I know you're not going to predict stuff, but just to wrap this section off, it seems like, in a way, private investment has been saved in this moment by the AI narrative and by the stocks that have gone up in the public sector. And perhaps the reports of this era's demise of frothy valuations, of big checks before you have a product, maybe that was premature.

28:40Maybe they're still going to keep kicking as long as they can. Yeah, venture is this, it's like, it is the hurdiest industry I can ever think of. And so you have a lot of money sitting around there. And so then when I think they saw some green grass like spread out of the snow, people just like rushed to it. And so that's what explains a lot of the hyperbolic enthusiasm for AI beyond just the reality that it is really cool to create a limerick out of your email. And this is where I'm going to take Masa's side on this because maybe he does have it right. Like who cares about all these generative AI companies?

29:30The man says he's been investing in AI since 2017. Maybe it is the stuff that, I mean, I know I just gave him crap for this, but maybe it is these companies that put AI into play in healthcare or in construction or even in the restaurant industry and do it in a way that can even help them inch past their competitors. Maybe that's where the real opportunity is. Yeah, I mean, it is legitimately early, you know, as to the question of whether the Vision Fund One is going to make money on that. I mean, they've been selling down a lot of their old holdings. So like they don't own Uber anymore. They're, you know, a lot of them have gone bankrupt for some of them.

30:11So but yeah, like it is very legitimately early. And, you know, there could be a transformation that sort of unlocks one of these older mid-sized money losing companies. But, you know, I would sort of bet like history is more on the incumbents or sort of like the upstarts that win as opposed to the sort of like adolescent or middle age companies. Elliot Brown is here with us. He's a reporter at The Wall Street Journal and the co-author of The Cult of We, WeWork, Adam Neumann and The Great Startup Delusion. In the second half, we're going to come back when we talk about something that's been big and brewing under the surface, which is the risks in commercial real estate.

30:48Very interesting story that Elliot has written recently. And we also have some more stuff to talk about in the second half. So please stay with us. We'll be back right after this. Capital One's tech team isn't just talking about multi-agentic AI. They already deployed one. It's called Chat Concierge, and it's simplifying car shopping. Using self-reflection and layered reasoning with live API checks, it doesn't just help buyers find a car they love. It helps schedule a test drive, get pre-approved for financing, and estimate trade and value. Advanced, intuitive, and deployed. That's how they stack.