Michael Burry wagers billions against AI boom

Michael Burry, the guy who made a fortune betting against the housing market before it crashed in 2008, thinks he’s spotted the next big bubble: artificial intelligence companies.

Burry just put money down that shares of Nvidia and Palantir Technologies are going to tank. The two companies have helped push stock markets to new highs this year and are worth about $5 trillion together. But Burry, who’s mostly kept quiet for the past decade, says there’s a bubble here that’s going to pop.

The catch? He doesn’t know when.

The problem with timing

“Michael, if he had one failing in the dot-com cycle, it was being early to the process. The housing bubble? It was being early to the process,” said Michael Green, a former hedge-fund manager who’s now chief strategist at Simplify Asset Management. Green has warned against popular investment trends too, especially passive investing. “This is a significant issue, right? How quickly does this end?”

After his housing market win, Burry picked up a serious online following. Fans dissect his posts on places like Reddit’s Burryology forum.

Things are heating up for Burry now. Last month, he closed his hedge fund. Then he launched a newsletter where he said he’d share his big idea with investors—why AI stocks are going to deflate. Cassandra Unchained became one of the top-selling finance newsletters on Substack fast, pulling in some 171,000 subscribers. They’re paying $379 a year, which is actually pretty cheap compared to some competitors who charge more than $1,000 yearly.

Burry‘s pitch played into fears about AI companies investing in each other and real questions about data center limits. His main point isn’t that AI is bad, just that the market has gotten disconnected from reality.

“This bubble looks an awful lot like the dot-com bubble,” he said on a podcast hosted by Michael Lewis, whose book first made Burry famous. “Which was not really a dot-com bubble. It was a data-transmission bubble.”

Burry didn’t respond when asked for comment, but mentioned on Lewis’s show that he usually turns down reporters.

Mixed track record fuels skepticism

So far, his warnings haven’t done much to the stocks, though there’s been more talk about AI infrastructure concerns.

Burry gets dismissed plenty on social media, where people crack jokes that he predicted 20 of the last two recessions.

A lot of his big market crash predictions have been wrong over the past 15 years. In a Jan. 31, 2023, post he told his followers to “SELL.” Silicon Valley Bank went under two months later, but the S&P 500 index has gone up around 70% since then. He’s admitted that was the wrong call.

Alex Karp, Palantir’s chief executive, called Burry “bats— crazy” on CNBC.

On Nov. 3, Burry showed his hand against Nvidia, the chip maker that’s now the world’s most valuable company, and Palantir, a major AI software business. Together they’re worth about $5 trillion. His bets were pretty small, around $10 million in put options, but could turn into more than $1 billion if those stocks drop hard.

“Palantir and Nvidia are the two luckiest companies on the planet,” he told Lewis.

Why he’s betting against Palantir

Burry said he’s shorting them for different reasons, though they’re connected. Palantir depends too much on stingy government contracts and gives too much to its executives, he said. He pointed to tough competition too, especially from International Business Machines. His bet pays if Palantir drops to $50 a share in 2027 from the roughly $200 it’s been trading around.

Karp said on CNBC that he thinks Burry was trying to manipulate the market and shot down Burry’s take.

Nvidia’s problems have to do with its customers, like Oracle and Meta Platforms, where Burry said he sees a bunch of issues.

Nvidia has been helping fund some of their purchases in deals that, Burry said, look like how companies like Enron financially propped up its vendors buying its products.

Burry’s also looked at the accounting at these companies and Nvidia about how long the chips last, saying this helps companies pump up their earnings.

If the bubble bursts, there could be a domino effect of lower stated profits, falling share prices, and less investment that would hurt Nvidia’s future sales.

Burry’s bet pays if Nvidia falls about 37%, to $110, by 2027. It’s around $190 now.

Companies push back hard

Nvidia has pushed back on Burry’s claims. “Nvidia does not resemble historical accounting frauds because Nvidia’s underlying business is economically sound, our reporting is complete and transparent, and we care about our reputation for integrity,” the company said in a memo that Barron’s reported on earlier.

Nvidia and Palantir shares have fallen since Nov. 3, but the drops have been choppy.

Right now, that might actually be helping investors who think there’s no end in sight.

“I would actually argue that awareness of this has encouraged people to defect and basically become more convinced that stocks can go to unlimited levels,” Green, the chief strategist at Simplify Asset Management, said.

Get up to $30,050 in trading rewards when you join Bybit today