Michael Burry, the investor who became famous for betting against the housing market before the 2008 crash, is now turning his skeptical eye toward artificial intelligence. In a recent post on his paywalled Substack newsletter Cassandra Unchained, Burry delivered a characteristically contrarian message: the AI bubble will burst eventually, but there's absolutely no way to predict when.
The timing paradox is the interesting part here. Burry acknowledges the bubble might still have plenty of room to expand before it pops. He sees the stock market potentially approaching both a significant upside opportunity and a generational top simultaneously—which sounds contradictory until you remember that bubbles tend to inflate most dramatically right before they burst.
What's driving his concern? Burry points to what he calls "supply-side gluttony"—massive data-center construction projects and multibillion-dollar commitments being made without actual end-user demand to justify them. It's the classic bubble pattern: lots of investment chasing a narrative rather than real revenue.
He reserves particular criticism for Nvidia (NVDA) and CEO Jensen Huang, attributing much of the AI hype to the chipmaker's aggressive marketing efforts. Burry has drawn comparisons between Nvidia today and Cisco during the dot-com crash, which is about as ominous an analogy as you can make in tech investing. Cisco was the infrastructure darling of the late 1990s internet boom, and its stock fell over 80% when that bubble burst.
Interestingly, Burry cautions against acting on this pessimism by shorting stocks or buying puts. Overvalued stocks often carry the most upward momentum, he notes, which means trying to time the top can be financially devastating even if you're eventually proven right about the fundamentals.
Burry launched Cassandra Unchained in November, and his critical stance on Nvidia has been a recurring theme. Whether the AI boom ends up resembling the dot-com bust or proves to be something more sustainable remains to be seen. But if there's one thing Burry's track record suggests, it's that his warnings are worth considering—even if, as he admits, nobody can predict exactly when the music stops.
