E133: Market melt-up, IPO update, AI startups overheat, Reddit revolts & more with Brad Gerstner

But I think that anybody who does an IPO today is going to demand a rate of return into that offering that is a significant margin of safety relative to all deals that were done in 20 and 21

They could have gotten it out for 40 something, but it's a mid 20s billion dollar company.

But I think there's a suggestion here that it may be more about putting the control back in the hands of the user from a data perspective and a monetization perspective. That would be a pretty gangster move

I don't think there's any problems with NVIDIA. I think they're going to continue to perform. I think they'll beat their numbers for the balance of the year.

Certainly the most obvious is demand for chips because all this infrastructure is being built out.

My God, these people that put the money in the seed round should have just bought NVIDIA. Buy some call options so you can make more money.

So I think it'll have a lasting effect in terms of investing in social network or social media type businesses where the users are generating so much of the value and have the ability to kind of communicate with one another and control where value ultimately falls.

In this current version of the Internet, the value is going to go and again, further, further, further erode away from the centralized apps and more towards the individual people

But I would say, as I sit here today, the distribution of potential for them is less than it was before ChatGPT, the distribution of upside.

But its claim is going to come at the expense of search. And so Google may create the best AI in the world, but it's going to have to fight off the cannibalization of core search.

You no longer need an index. You need someone to do the retrieval from the index for you. That's the new service.

This could actually be the reverse of what everybody's thinking. This could lead to higher CPMs and higher cost per click because of intent.

So whoever's putting money in thinking they know what the fuck they're doing, you might as well just light it on fire, go to Vegas and have some fun with it, because you will get more enjoyment from that than you will from making these kinds of investments.

The truth of the matter is almost all those companies went to zero, even though you got a couple bets right. You got the internet right. You got search right. But you didn't have to invest a dollar in search until 2003 and you would have captured 98, 99% of the value.

Constraint makes for great art. Constraint makes for great startups. You need to have pressure on a startup for them to deliver. You cannot give startups five years of runway and expect it's gonna work.

But if that's really where things are headed, then does the $100 million to train models today really make sense if training those same models can be done for $5 million in 18 to 24 months?

You choose, do robots.txt, except we're going to call it AI.txt.

Ultimately, the company loses the value if that data, that content gets extracted and they can't monetize it or capture the value somehow.

they'll get to the right answer, which is give them a healthy rev share. That's the future.

If they spent $400 million in the last couple of years, you could probably assume that doing the same training exercise could be done for five to $10 million 18 months from now to generate the same model.

And all the people that say, I'm sure there's going to be some genius in the comments, but what about open source? And it's like, what about it?

Again, we've said before, most venture investors are probably unprepared for this shock because they've never lived through one.

There's a massive pressure on young, junior partners, principals within these firms to do something.

So we're almost at the end of, I think, the bottoming though.

I'll just say the number of people I interact with on a global basis who talk about ChatGPT versus BARD is like 10 to 1 today.

And the reason there's not going to be a hard landing is you just saw China today basically say, we're going to start to rip in trillions of dollars.

when you think about what these hyperscalers are going to do, they're not gonna spend a billion, they're not gonna spend 10 billion. They'll spend a hundred billion dollars, right, in order to be in this race.

First of all, Jason, I don't think the point about SoftBank needs saving is necessarily true. It's not a company that needs saving, but they've certainly taken some hits.

The one thing I would tell people is do not give up all of your data to one thing who's professing to be at all.

But I do think that there's a radical realization underway that a lot of businesses that are dependent on services that might be replaced by AI don't necessarily have the longevity today.

I would say four out of five startups that we've invested in and the majority of them that we're being pitched on right now are not only talking about AI, which they were all talking about it three years ago, they're actually implementing it now because the tool sets are available.

I think Google is going to be, ChatGPT4, I'm saying it right here right now, I think they're going to beat them because I think that they're better at indexing all this information and understanding it than anybody on the planet. And they have the largest ad network. If they get this done in the next six months, I think it's going to increase the cost per click because they're going to know so much about each user

Because we've said to hyperscalers, you're not allowed to acquire any of these companies. So the unintended consequence of the regulation in Washington is that entrepreneurs and founders and venture capitalists who might otherwise have had a good idea, built something with some traction, they can't find a home for it

And it's following, by the way, a similar cost curve as DNA sequencing, which is actually greater than Moore's law, greater than a 2x cost reduction every 18 months.

If you do those three things, you know, simultaneously, like, you got a good shot at producing really incredible vintage.

We're still trading below the 10-year average for internet and software companies based upon our numbers.

now you have the seven or eight most valuable tech stocks priced to perfection yet again

And so you just have to get prepared for rates just being sticky and inflation being sticky. And I think that that's probably the most reasonable base case for the rest of the decade.