Michael Burry, America’s biggest investor, who is best known for predicting and profiting from the 2008 housing market collapse, has now warned that public listings by AI giants Sam Altman-led OpenAI and Anthropic could lead to massive capital destruction, arguing that the companies should be prevented from going public. In a post shared by on social media platform X (formerly known as Twitter) Burry said that markets should ‘tank hard’ in order to stop the planned initial pubic offerings (IPOs) of OpenAI and Anthropic, describing such an outcome as being ‘for the benefit of humanity’. He later expanded on his criticism, claiming the two companies would ‘suck up’ and then destroy ‘TRILLIONS of dollars of capital’. According to Burry, the financial losses would be “the least of the damage they do.”
The financial case behind the warning from Michael Burry
Burry’s latest remarks come as both companies edge closer to the public markets, each carrying enormous capital requirements alongside rapid growth. Anthropic’s confidential IPO prospectus, reviewed by Reuters, revealed at least $518 billion in future cloud, computing, and infrastructure commitments, with the company potentially seeking a valuation above $2 trillion when it lists. OpenAI confidentially filed its own draft registration statement in June, though CEO Sam Altman has said this isn’t the right time to list, and the company is now reportedly aiming for a 2027 debut instead.Burry’s core argument centers on a mismatch he sees between the scale of capital being poured into these companies and how proven their underlying economics actually are — essentially questioning who ends up absorbing the losses if AI revenue growth fails to keep pace with the enormous infrastructure spending required to sustain it.
Not his first AI warning, and not his only bearish bet
This latest broadside is part of a broader pattern for Burry, who has grown increasingly bearish on the AI sector over the past year. His bearish positions have included bets against Nvidia, Palantir, Micron, Oracle, and the Nasdaq 100, built around a thesis that borrowing-driven investment in semiconductors and data infrastructure can’t be sustained indefinitely, and would be especially vulnerable if interest rates rise.
AI labs themselves aren’t entirely dismissive of the risk
Interestingly, OpenAI and Anthropic have both acknowledged, at least to some degree, the dangers posed by their own technology. Anthropic’s confidential draft prospectus reportedly acknowledged that its AI systems could pose catastrophic or existential risk to humanity — though such disclosures are standard practice in IPO risk-factor sections, intended to flag potential risks to investors rather than an admission that these outcomes are likely or imminent.His X posts followed a Substack newsletter he published the same day in which he said he was more confident than ever that his bearish AI thesis would play out within the next year — notably moving up his own previous timeline, which had placed 2028 as his base-case scenario for when the AI bubble would burst.
