The International Monetary Fund (IMF) has identified three growing threats to global financial markets: a correction driven by Artificial Intelligence (AI), a sharper rise in sovereign bond yields and major cyberattacks.
The Fund said markets had remained resilient, supported by solid economic growth, strong corporate earnings and relatively accommodative financial conditions. However, it warned that policymakers could not afford to become complacent.
In its latest Global Financial Stability Report, the IMF said high borrowing, limited liquidity and stronger links between markets could magnify the impact of any shock and turn it into wider financial stress.
An AI-related market correction could prompt widespread deleveraging, while a bond-yield shock could reprice assets and reduce the value of bonds used as collateral. A cyberattack, meanwhile, could interrupt essential operations and undermine confidence.
AI and rising leverage
AI has supported the global economy through investment in data centres, semiconductors and related infrastructure. Companies at the centre of the AI boom have also recorded exceptional earnings growth.
But weaker investment or profitability could undermine optimistic expectations and lead to falls in AI-related asset prices, the IMF said.
It warned that increasing borrowing could accelerate such a correction. Hyperscalers – the largest data-centre operators – are taking on more debt, while investors are using leverage to buy AI stocks.
The Fund said circular financing arrangements, involving equity and debt flows between AI companies, could leave key firms vulnerable. Stress at those companies could reduce revenue forecasts and valuations, weakening the value of debt built around future deals.
An IMF review of 14 such transactions since 2023 found about $300 billion in equity investments and $650 billion in debt, in return for $1.4 trillion in future revenue commitments.
Sovereign bond yields have also risen because of renewed inflation pressures from higher energy prices, fiscal concerns and, in some cases, stronger-than-expected economic growth. The IMF said yields could be moving back towards levels seen before the global financial crisis, when growth and inflation were higher than during the 2010s.
Higher yields could increase discount rates, reduce asset valuations – including those linked to AI – and lower the value of bonds used as collateral. Reduced central-bank balance sheets and greater reliance on price-sensitive investors could add to volatility.
Cyber and banking risks
The IMF described cyber disruption as an increasing systemic risk as finance becomes more digital and interconnected. A major attack could affect payments, trading, access to deposits or other critical services, particularly if it hit a major financial institution, payment system, cloud provider or software platform.
In more than 70% of reported cases so far this year, software providers had almost no time to fix vulnerabilities before they were exploited. That is more than double the share in 2021, when providers typically had nearly 300 days to respond. Frontier AI models could speed up both attacks and defensive measures.
Under the IMF’s adverse stress scenario, banks representing 22% of global banking assets would have capital ratios below regulatory limits, up from 18% a year ago. Pension funds and insurers are also more exposed to illiquid assets and derivatives, while private credit and on-demand lending could intensify liquidity pressures.
The IMF urged governments to restore debt sustainability, limit leverage and liquidity mismatches in AI, private credit and other nonbank financial sectors, and strengthen cyber reporting and resilience testing. Emerging and developing economies should also rebuild financial buffers while conditions remain accommodative.
