The report's sharpest observations concern the industry's circular economics. Since 2023, the IMF tracked fourteen public transactions in which $300 billion went to what it calls financially weaker AI companies, mostly from hyperscalers, alongside $650 billion in debt, together accounting for $1.4 trillion in revenue from services and chip orders. Those virtual dollars are concentrated in a handful of firms, so trouble at one could spread quickly through the herd.

Leverage adds to the exposure. Hedge funds have increased leveraged positions and margin borrowing, and the share of single stocks in leveraged exchange-traded funds has risen to 28%, a shift the report links to demand for AI and tech shares. A sharp drop in those equities could force selling by ETF issuers and lenders, creating a feedback loop.

The Fund notes that AI has become an important source of support for the global economy and financial markets, even helping to prop up growth while the war in Iran weighs on energy prices and policy certainty. A recent stress test suggested banks would remain broadly resilient even against a combination of an energy price shock, higher inflation and an AI-driven market correction.

Hyperscalers now account for 10% of investment-grade bond issuance, up from under 1% in 2024, as outsized AI spending leaves them needing substantial loans. Whether investors keep an appetite for that debt is, for now, the open question.