The AI Valuation Crisis: Michael Burry's $1 Billion Bet Against Tech Giants

Michael Burry, the legendary investor who famously shorted the housing market before its 2008 collapse, is making headlines again—this time with a massive bet that the artificial intelligence boom has gotten wildly out of hand. In early November, Burry placed substantial put options against two of the market’s biggest winners: Nvidia and Palantir Technologies. His position, starting at roughly $10 million in contracts, could balloon to over $1 billion if these stocks tank as he predicts.

The Scale of the Bet

The numbers tell the story. Nvidia and Palantir combined are worth about $5 trillion, with Nvidia now holding the title of world’s most valuable company. Yet Burry argues this valuation is completely disconnected from reality. He’s betting Nvidia drops roughly 37% to around $110 per share by 2027 (currently near $190), while Palantir plummets to $50 from approximately $200.

After largely staying silent for the past decade, Burry recently closed his hedge fund and launched “Cassandra Unchained,” a finance newsletter on Substack that quickly amassed 171,000 paying subscribers at $379 annually. His central thesis became clear: this isn’t a failure of artificial intelligence itself, but rather a spectacular mispricing of the companies profiting from it.

The Accounting Problem Nobody Wants to Discuss

Burry’s critique goes deeper than simple valuation concerns. He’s flagged troubling accounting practices at Nvidia and its major customers like Oracle and Meta. His analysis suggests these companies are using questionable methods to extend the useful life of chips on their balance sheets—effectively pumping earnings numbers artificially.

More alarming, Burry has documented what he calls Enron-style financing schemes where Nvidia effectively funds its customers’ purchases of its own products. When these deals unwind, he argues, the domino effect could be devastating: stated profits collapse, share prices plummet, future investment dries up, and Nvidia’s growth engine stalls entirely.

Palantir faces different but related pressures. Burry contends the company leans too heavily on government contracts while enriching executives at shareholder expense. Competition from established players like IBM presents another headwind.

The Timing Question Nobody Can Answer

Here’s the catch that even Burry admits: he has no idea when this unwinding happens. Michael Green, chief strategist at Simplify Asset Management and another veteran skeptic of bubbles, notes that Burry’s historical weakness was being early—whether during the dot-com crash or housing implosion. “How quickly does this end?” becomes the critical unanswered question.

Burry himself drew the parallel explicitly, telling podcaster Michael Lewis: “This bubble looks an awful lot like the dot-com bubble, which was not really about the internet at all—it was a data-transmission bubble.” The mechanism differs, but the disconnect between hype and fundamentals feels eerily familiar.

Market Response: Surprisingly Unmoved

Since Burry’s November 3rd disclosure, neither Nvidia nor Palantir have experienced the crash he’s predicting. While both stocks have traded choppily lower, the broader AI rally has largely continued. Nvidia flatly rejected his accusations, issuing a statement: “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.”

Palantir CEO Alex Karp went further, calling Burry “bat-crazy” on live television rather than engaging with the substance of his critique.

The broader market has dismissed Burry’s track record, especially after his catastrophically wrong “SELL” call on January 31, 2023, just before Silicon Valley Bank collapsed. The S&P 500 has surged about 70% since then. But Green offers a counterintuitive observation: market awareness of Burry’s position may actually be backfiring. “Awareness of this has encouraged people to defect and become more convinced that stocks can go to unlimited levels,” he suggests.

What This Reveals About the Market

Whether Burry proves right or wrong, his position highlights legitimate structural concerns about the AI infrastructure build-out. The questions he’s raising—about unsustainable customer concentration, circular funding arrangements, and creative accounting—aren’t easily dismissed, even if his timing remains speculative.

The real debate isn’t whether artificial intelligence matters (it does), but whether the current valuations reflect anything close to sustainable business realities. That’s a conversation the market appears unwilling to have while momentum remains intact.