International Monetary Fund warns that AI debt is a bigger systemic risk to financial stability than sky-high equity valuations.

Wall Street's concerns about an AI bubble persist, but the International Monetary Fund warns that the more immediate financial stability threat lies in the mounting debt accumulated by technology companies in their race to build AI infrastructure.

Tobias Adrian, director of the Monetary and Capital Markets Department at the IMF, expressed greater concern about corporate borrowing than an AI valuation bubble during the annual European Central Banking gathering. "What is quite worrisome from a financial stability perspective is that the major tech firms are starting to leverage up themselves," Adrian told Bloomberg's Francine Lacqua, noting there was a "potential maturity mismatch in between the duration of the physical assets and the duration of the debt."

A maturity mismatch occurs when companies finance long-term assets with short-term debt. In this context, the risk is that hyperscalers are pouring massive capital into AI equipment—data centers and chips—that could depreciate before debt obligations are repaid, particularly given the rapidly evolving technology landscape. In worst-case scenarios, financing could dry up before these companies generate sufficient profits to justify their original spending.

Despite investor enthusiasm for AI companies, construction timelines suggest underlying strain. A May analysis from J.P. Morgan found that 60% of data center capacity scheduled for completion by 2027 had not yet begun construction, with another 7% facing delays.

The borrowing binge is substantial. AI hyperscalers including Amazon, Alphabet, Meta, Microsoft and Oracle issued $159 billion in corporate bonds during the first five months of 2026—surpassing their combined borrowing across the previous five years, according to financial services firm Dealogic. Last month, Alphabet announced it was raising $85 billion in equity to finance its AI buildout. NVIDIA also issued $25 billion in corporate bonds last month, its first issuance since 2021, encountering robust investor demand.

Adrian pointed to strong tech earnings as justification for downplaying an AI credit crisis, yet the IMF has raised separate concerns about frontier AI models themselves. In April, Anthropic announced Mythos, a model capable of identifying vulnerabilities in software systems running power grids, major banks and other critical institutions. Anthropic restricted public release, sharing the model only with a dozen trusted corporate partners in the United States.

IMF Managing Director Kristalina Georgieva characterized Mythos as merely the opening act. "What we recognize is that Mythos is just the beginning, there will be more like it," she said last month, according to Politico, warning that advanced AI models could be weaponized by malicious actors to "destroy the financial system" without adequate safeguards.