AI Selloff Tests Investor Confidence

Author: Bloomberg Television · Watch on YouTube

Video authored by the channel shown, not by IANIA; all rights belong to the creator. Transcript and summary generated automatically; the original-language content is preserved.

Original language: English

Original video: AI Selloff Tests Investor Confidence · Channel: Bloomberg Television · 10:08 · Watch on YouTube

Topics: Enterprise AI applications, AI infrastructure and hardware, Markets and economy, Robotics and autonomous systems

Summary

Barry Ritholtz says recent market jitters reflect uncertainty about which companies will win or lose in the AI boom, rather than doubt that AI can make money. He sees open-source systems and the difficulty of forecasting long-term winners as important challenges, and compares the current buildout with earlier technology transitions such as broadband. Ritholtz is skeptical of confident assurances from NVIDIA CEO Jensen Huang, given his interest in the industry, and argues that investors may be better off owning a broad range of companies than trying to pick winners. He also questions Tesla’s valuation, its product execution, and the possibility that a SpaceX-Tesla combination could support a struggling company.

Key points

Ritholtz sees the AI selloff as part of broader market nervousness after a long rally and uncertainty about the allocation of capital.
The key question is which AI companies will survive and profit, not whether the technology itself can generate money.
He identifies open-source AI systems as a significant competitive challenge.
Strong demand for data-center capacity does not resolve uncertainty over which firms will ultimately succeed.
Ritholtz compares AI’s development with earlier technology shifts, where apparent winners often exited the market.
He says the broadband buildout eventually enabled companies such as YouTube, Netflix, and Facebook, despite earlier infrastructure companies failing.
Ritholtz cautions that Jensen Huang is not a disinterested observer when discussing AI’s prospects.
He says forecasting AI winners is difficult and points to indexing as an alternative to selecting individual companies.
Ritholtz calls the flying-car idea a gimmick and says Tesla’s valuation is difficult to justify given its profits and product challenges.
He raises concern that a merger could use SpaceX’s value to support Tesla, while noting Tesla’s faltering sales and the loss of the $7,500 credit.

Chapters

Start	Chapter	Summary
00:00	Market jitters and uncertainty around AI	Ritholtz attributes the selloff to broader investor nerves and uncertainty about the winners and losers of a transformational technology.
03:23	Jensen Huang and the AI transition	The discussion turns to Huang’s defense of the AI transition and whether his assurances should reassure investors.
06:22	Technology winners and Tesla’s valuation	Ritholtz compares AI with the broadband buildout, then questions Tesla’s prospects, valuation, and potential merger with SpaceX.
08:30	BYD competition and Tesla’s challenges	He discusses BYD’s expansion and subsidies, Tesla’s weakening sales, and concerns about using SpaceX to support Tesla.

Quotes

people are starting to get a little nervous — 00:34

I don’t think there’s a threat that this isn’t gonna make money. — 02:18

every time a new technology, a transformational technology comes along, there are winners and losers. — 05:10

It’s just, you know, own own them all. Let God sort it out. — 06:22

it’s hard to see how Tesla grows into its market cap. — 09:02

Concepts and entities

AI, DeepSeek, CoreWeave, NVIDIA, Jensen Huang, Ritholtz Wealth Management, SpaceX, Tesla, BYD, Barry Ritholtz

Full transcript (original language)

00:00 Now is Barry Ritholtz, the chairman and CIO of Ritholtz Wealth Management, of course, the host of the Masters in Business Podcast. Barry, great to see you as always. Let’s let’s start with Core Weave and the way that you’re looking at what really drove markets over the course of the week. Do do you see it as an outlier or something anomalous, or is it telling you something about where the AI trade is going directionally? So there are so many crosscurrents around the AI trade. It… It’s… We’re always searching for the this is what happened. Hey. It was oil at a a $110 a barrel, and

00:34 that sent yields moving, and Core Weave has a lot of debt. But I think it’s really just a much broader issue, which is this market has had a spectacular run, not only over the past four years, but really since the lows in o nine, and people are starting to get a little nervous, especially when you’re looking at a a a transformational new technology. Market historians know every time a new technology comes along, there’s always winners and

01:07 losers, massive misallocation, and people are starting to guess, hey. Who are the winners and losers of this? Who has too much debt? Where do we need to throttle back a bit? Well, we’ve been talking about this, I feel like, for months now, and it is unclear which aspects of this technology, agentic, generative, task based, token based, are actually going to make Wow. I learn eventually are actually going to make money. So is this a reflection of exactly that? People are… Wanna take a pause and see. They wanna wait to see if

01:42 this technology is actually profitable. Well, that… That’s always a challenge. I think the bigger threat is is from the open source AI systems like DeepSync. I have six different AI apps on my phone and Barry, that’s my nightmare, but okay. So the the fun thing is I ask Claude something and, hey. You’ve used 75% of your, credits for the month, it’s only October 17. So then I’ll hop over to chat,

02:18 and then I’ll hop over to Gemini and work your way down. Perplexity, deep seek. So it’s just a way of allocating harder, more expensive tasks to to higher quality or not as high quality issues. But but back to things like Core Weave, I don’t think there’s a threat that this isn’t gonna make money. I don’t think that’s the problem. This is a giant transformational technology. I I I think it was Cantor Fitzgerald said all of the

02:51 2027 capacity that they had to go out to sell on constructing data centers was already sold out 2027. So it’s not that the funding isn’t there. It’s which of these companies are gonna stick around, who are the winners, and who are the losers. It’s always so hard to tell. Think back to computers. You know, in addition to Apple and Dell, you had IBM and Gateway and HP and go down the list that all those companies look

03:23 like winners. Most of those companies exited that space a long time ago. One more question here on AI. Our colleague Emily Chang was trailing Jensen Huang as he made his way across, Europe talking about the prospects for the growth of of AI, in Europe, and he addressed the concern that we were seeing in the markets, this week. Let’s play a little bit what he had to say to Emily. All of this is gonna pass, and the reason for that is very simple. We’re going through a new platform transition. The computer systems, the computers that we use to run society, countries, and companies today is being reinvented, and we cannot go back. With respect to, some

03:58 of the news in in, it… It’s it’s choppy news. News, as you know, happens all the time, and and… But it won’t change the fundamentals. Let me ask you about the role that Jensen Huang and NVIDIA is playing here, and this isn’t novel new ground. We’ve we’ve talked a bit about this before. I was on air with Dani Burger this week. She talked about the central bank of NVIDIA or the central bank of of Jensen Huang. He’s an investor in Core Weave. He’s he’s playing this outsized role here in the AI universe. And and I wonder if comments like that engender confidence and comfort in you, Barry, as you you hear the worry and speculation about the

04:33 prospects and problems of of circular financing. Is Jensen Huang saying everything’s gonna be fine, grounds for you feeling fine about where things are headed? You know, I I always laugh anytime somebody’s a multibillionaire, and they come out and say, hey. My company and all the revenue and profits we’re generating, don’t worry about the fining or the excess capacity, or do we really need a thousand data centers that are 20 acres each? He is not exactly a disinterested party in in this issue. I don’t disagree

05:10 with him. That’s the ironic part. I I think I think what he is… First of all, there’s few people in the world better situated to to evaluate what’s going on other than the fact that perhaps he’s not the most disinterested objective person in this. All of that said, every time a new technology, a transformational technology comes along, there are winners and losers. There are the secondary effects. There are ups and downs. It’s never a straight line.

05:44 Think back to the build out of broadband and companies like Juniper Networks and Global Crossing and Metro Media Fiber and all those things. Those companies all went belly up, and we were able to purchase the broadband on the other side of bankruptcy for pennies on the dollar. That’s what led to companies like YouTube and Netflix being able to do streaming and Facebook and go down the whole list of everybody that is, taking advantage of broadband, we’re gonna see something similar with AI. It’s just

06:22 all but impossible to guess which companies are gonna be the winners and losers. Forget next year. You have to think five and ten years out. That’s incredibly challenging. The technology changes so fast. There’s a reason why so many people have moved to indexing rather than guess who’s the winner. It’s just, you know, own own them all. Let God sort it out. Alright. Speaking of technology companies, we’ve got SpaceX and Tesla likely to merge. We have

06:53 Well, Elon Musk would hope. Set to merge, trying to merge Perhaps to merge. To merge. What are the odds of that? Do you think it’s a good decision for a bigger conglomerate for Musk? And then what is your take on the flying car? Oh, yeah. Are you gonna buy a Roadster Barry? So the flying car is obviously a gimmick and a stunt and and that’s See? David gave me a hard time. Call me a skeptic. I’m a Barry on this one. Kinda goofy, and look. They’re taking the SpaceX technology and assuming that was actually done, and

07:28 it wasn’t… It didn’t look like there are any wires. But all that said, look. You you look at you look at the the advantage Tesla had five years ago. They were a decade ahead of everybody else. They had enormous, software advantages. They… The over the air updates. The… They they just… Nobody was close to them. The problem is their valuation presumes they’re gonna have, you know, an 80% EV market share that they’re, I don’t know, five

07:59 times the market cap of, Toyota. At the same time, they’re, like, an eighth of Toyota’s profit margin. It it it… The valuation is out of whack. You you look at the… How poorly the company has done introducing, new products. Although this is the year where in 2027, a whole lot of new Tesla products are coming out. The problem is they they failed to do this on a timely basis. BYD is taking over the world in terms

08:30 of EVs. Twice the the the distance, twice the range, double the horsepower, half the price, the whole run of new BYD cars, which you can’t buy in The United States, but are moving into Europe, and Canada just announced a deal, when they were angry Trump to let in… I… It was either 25 or 50,000 cars. The fear is once you open up, the doors… And understand BYD has huge subsidies, free real estate, so the price reflects

09:02 a giant subsidy from the government, and so there is legitimate concerns about product dumping. But all that said, it’s hard to see how Tesla grows into its market cap. And if I’m a SpaceX owner, and I guess everybody who’s an indexer is, you have to worry is is Elon using his own paper from one company to rescue a a company that

09:33 is at risk at faltering. His whole dalliance with with the president and with Doge alienated his core audience at at Tesla. You know, most of the people who were buying Teslas were in blue states and tended to lean left of center, and he basically pissed them all off. Yep. So that’s why Tesla numbers have been faltering. Tesla sales have been faltering plus the loss of the $7,500 credit. It… It’s it’s been a little bit of a a rough road

10:08 the past couple of years. And a good market for those making the bumper stickers that say I bought this before Elon went crazy, which I see in and around Brooklyn and them. Yeah. Barry, thank you very much.