Michael Burry, the investor known for predicting the housing crisis, believes the AI bubble could burst sooner than he previously expected. Burry said in a long Twitter post that he is “moving timelines up” because he expects “the bubble in AI may burst sooner than later.” He added that he is “more confident than ever before” that his bearish view will “play out over the next year.” His comments build on his argument that the AI boom is driven by unusually high investment levels, rising debt and large financial commitments by major technology companies.

Burry sees another capital cycle

Burry’s major statement is that the AI boom is part of a larger “capital cycle,” in which companies invest heavily in an area when demand and returns appear strong. As more companies spend, capacity can eventually grow faster than actual demand. He pointed to net investment by S&P 500 companies as a share of US nominal GDP. According to Burry, the figure stood at about 2.07% as of June 30, the highest level seen in nearly four decades apart from the period around the dot-com boom of the late 1990s.

Burry compared the current cycle with the technology, media and telecom boom that ended with the Nasdaq peak in March 2000. He explained that heavy investment can continue even after markets peak. During the dot-com period, capital spending continued to rise even as technology stocks eventually fell sharply. Burry believes the same pattern could happen in AI, with companies continuing to spend heavily even if the financial returns from that spending begin to weaken.

Trillions of dollars tied to AI infrastructure

Burry has also focused on the scale of spending commitments made by the biggest technology companies. He has detailed commitments involving Microsoft, Amazon, Alphabet, Meta and Oracle. These include purchase commitments, future leases, construction projects and other arrangements linked to AI infrastructure. A recent report cited by Burry said Alphabet, Amazon, Meta, Microsoft, Oracle, Nvidia, Broadcom, SpaceX and Advanced Micro Devices had disclosed around $3 trillion in commitments largely connected to AI infrastructure.

ALSO READ

The commitments reportedly include about $1.2 trillion in leases that have not yet started and $1.9 trillion in purchase commitments covering chips, energy and data centre infrastructure. The figure is much larger than the roughly $600 billion in capital expenditure reported by these companies over their latest 12-month periods. Burry’s concern is not simply that these companies are spending a lot. His concern is that they could be locked into expensive infrastructure spending even if demand for AI services does not grow as quickly as expected. Many of these contracts are difficult to cancel. If AI demand weakens, companies could still have to make payments on infrastructure that generates lower returns than expected.

ALSO READ

Depreciation is another concern

Burry has repeatedly questioned how companies are depreciating their AI chips and servers. He claimed some hyperscalers are giving Nvidia chips and other AI equipment useful lives of five or six years, while he believes their economic life could be closer to two or three years. A shorter useful life would mean companies would have to recognise higher depreciation expenses sooner. Burry has estimated that the difference could suppress depreciation expenses by about $176 billion between 2026 and 2028. He has argued that this could make earnings at some companies look stronger than they otherwise would.

He described the practice as “one of the more common frauds of the modern era.” Burry has estimated that the impact could be significant for some companies. His calculations suggest that Oracle’s 2028 earnings could be overstated by 26.9%, while Meta’s could be overstated by 20.8%, if his depreciation assumptions prove correct. These are Burry’s estimates and remain a point of disagreement rather than an established accounting finding.

Burry sees a possible earnings problem

Burry’s concern has expanded beyond the price of AI stocks. He is now looking at what happens to the earnings and cash flows of companies building the infrastructure behind the AI boom. His claim is that companies could face several problems at the same time. AI demand could slow, reducing the returns from new data centres and chips. Higher depreciation could then put pressure on earnings. At the same time, companies that have borrowed heavily or taken on large contractual commitments could face higher financing costs. Burry has described this as a three-way “compression”: demand weakens, earnings fall as expenses catch up and financing becomes more expensive.

Circular financing adds to his concerns

Burry has also questioned the way money moves between AI labs, hyperscalers and chip companies. His concern is that investments and commercial agreements between companies in the AI ecosystem can create the appearance of strong demand, even when the underlying end-user demand may not be growing at the same pace.He worries that the risks are spread across the AI ecosystem. Chipmakers, cloud companies, data centre operators and AI companies are all connected through large investment commitments and financing arrangements. Burry believes this could become a problem if AI demand growth fails to justify the enormous capital being committed today.

What Burry is expecting

Burry’s overall view is that the AI boom may be following a familiar pattern seen in previous investment cycles. Companies are investing heavily because they expect strong future demand. This spending creates more capacity and investors continue to value companies based on expectations of continued growth. The risk, according to Burry, is that the spending continues even after the economics begin to weaken. He expects depreciation, write-downs and financing costs to become more important in the coming years. “I have little doubt the next few quarters will set still higher and higher net investment/GDP marks,” Burry wrote on X.

Disclaimer: This article provides factual analysis only and is not, and should not be construed as, an offer, solicitation, or recommendation to buy or sell securities. Investors must conduct their own independent due diligence and seek advice from a registered financial advisor in the respective jurisdiction.