IMF warns AI share crash threatens global financial stability

In its semiannual Global Financial Stability Report, the International Monetary Fund warns that the risk of an AI stock crash spilling over into global markets is threatening world financial stability. The report lists an AI crash, rising government borrowing costs and large-scale cyberattacks as the three "salient risks" to financial stability, and says the three form a "plausible catalyst" for a sharp near-term tightening in financial conditions and market turmoil.

The IMF's core formulation is that "the consequences of an AI correction could be global." This means its concern is not confined to U.S. tech stocks themselves, but points to a cross-market transmission: once AI-related asset prices are repriced, the shock could spread to other regions and asset classes through valuations, leverage and financing arrangements.

The first fact the IMF uses to support its concern is valuation and concentration. The "Magnificent Seven" large-cap tech stocks—Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla—account for more than 35% of the U.S. S&P 500, up from 12% a decade ago; the combined market capitalization of these large tech stocks has grown by more than 50% over the past two years.

This concentration means that a repricing of AI-related assets is no longer merely an industry event, but could directly change the trajectory of major U.S. stock indices and transmit to other markets through indices, funds and cross-border holdings.

The IMF specifically points to three rising sources of leverage: companies' own mounting debt, hedge fund investment in the sector, and retail investors borrowing to speculate on AI stocks. The IMF says that if an AI-led market downturn occurs, "these growing sources of leverage would increase the likelihood of forced selling and surging stock price volatility."

This judgment advances the risk from "high valuations" to "high valuations built on borrowed money." The presence of leverage means a price decline could trigger passive deleveraging, and passive deleveraging would in turn push prices down further, creating an amplifying effect.

The IMF notes that "circular financing" transactions are widespread, binding stronger AI companies together with weaker partners in the same industry and potentially intensifying the impact of a stock pullback. According to the report, hyperscalers refer to companies building the underlying infrastructure for AI adoption (including numerous large data centers), including Amazon Web Services, Google, Meta and Oracle.

The quantitative clues the IMF provides are: in the past three years alone, there were $300 billion worth of transactions and another $650 billion in debt between hyperscalers and "weaker AI companies," in exchange for $1.4 trillion in future financial commitments. These figures show that the funding and commitments within the AI ecosystem have formed a mutually nested network; once one link comes under pressure, the stress could transmit back along the transaction relationships.

The IMF warns there is "growing evidence" that successful AI adoption "may not require the scale of investment that 'hyperscale' companies are currently undertaking," and says AI's "potential economic gains may be smaller than expected."

This doubt points directly at the foundation of the current AI investment narrative: if revenue and profit growth are insufficient to support the scale of capital already invested, then the sustainability of valuations and debt will come under renewed scrutiny. The IMF says that weakening or unsustainable AI investment and profitability could trigger a pullback in AI-related asset prices, and that it would be intensified by rising leverage, concentrated exposure and circular financing arrangements.

On geographic transmission, the IMF points out that aside from the United States, Asian markets have the largest risk exposure, while the UK and the euro area may be relatively more insulated, though their investors holding U.S. assets would still be hit.

This judgment translates "global consequences" into an asymmetric distribution of exposure: investors directly holding U.S. AI assets would be first in line, while markets indirectly linked through trade, financial conditions or risk appetite could be affected secondarily.

It should be clear that the IMF's warning is a forward-looking risk warning, not a market event that has already occurred. The report gives no timetable or probability for an AI pullback, nor does it provide the IMF's quantitative estimate of the size of a pullback.

In addition, the evidence comes from a single source only (The Guardian's report on the IMF report), with no cross-verification from other sources; the original IMF report is not provided, and the figures and formulations are all quoted from that article. The specific content and sources of the "growing evidence" cited by the IMF are not explained in the report, and the specific counterparties and structures of the circular financing transactions are not disclosed. The specifics of the two risks of rising government borrowing costs and cyberattacks are likewise not elaborated in the report.