IMF: Global financial markets face growing risks from AI, bond yields, and cyberattacks

The International Monetary Fund (IMF) has said that global financial markets face growing risks from Artificial Intelligence (AI), bond yields, and cyberattacks.

According to the IMF, as vulnerabilities increase, policymakers should contain leverage, strengthen debt sustainability, and harden cyber defenses.

The Fund said that financial markets have remained resilient amid solid economic growth and strong corporate earnings, while financial conditions have been relatively accommodative. However, there is no room for complacency.

Three key risks have been increasing, each with the potential for broader implications: an artificial intelligence-driven market correction, a sharper rise in sovereign bond yields, and cyberattacks. 

"Vulnerabilities beneath the surface could determine how damaging the manifestation of these risks would be, as we detail in the latest Global Financial Stability Report. An AI market correction could trigger broad deleveraging across asset classes. A bond yield shock could reprice assets and reduce the collateral value of bonds in financial transactions. A cyberattack could disrupt critical operations and weaken confidence. High borrowing, scarce liquidity, and tighter links across markets could amplify the impact into broader financial stress.

"AI has become an important support for the global economy and financial markets, as discussed in the World Economic Outlook. Investment in data centers, semiconductors, and related infrastructure has spurred economic activity, while companies driving the AI boom have delivered exceptional earnings growth. But faltering AI investment and profitability could challenge optimistic expectations and trigger a correction in asset prices linked to the AI industry," the Fund said in a statement.

It added "Growing leverage could be a catalyst for such a correction. Hyperscalers, as the largest data center operators are known, are borrowing more and more, while investors have been leveraging to invest in AI stocks. Circular financing arrangements in the AI sector, involving equity and debt flows between AI firms, suggest that stress at key companies receiving investment could lower revenue forecasts and investment valuations, which in turn would weaken the value of the accumulated debt in those deals. Our review of 14 such transactions since 2023 found about $300 billion in equity investments and $650 billion in debt, in exchange for $1.4 trillion in future revenue commitments."

Below is the full statement...

Financial markets have remained resilient amid solid economic growth and strong corporate earnings, while financial conditions have been relatively accommodative. However, there is no room for complacency.

Three key risks have been increasing, each with the potential for broader implications: an artificial intelligence-driven market correction, a sharper rise in sovereign bond yields, and cyberattacks. 

Vulnerabilities beneath the surface could determine how damaging the manifestation of these risks would be, as we detail in the latest Global Financial Stability Report. An AI market correction could trigger broad deleveraging across asset classes. A bond yield shock could reprice assets and reduce the collateral value of bonds in financial transactions. A cyberattack could disrupt critical operations and weaken confidence. High borrowing, scarce liquidity, and tighter links across markets could amplify the impact into broader financial stress.

Heavy reliance on AI

AI has become an important support for the global economy and financial markets, as discussed in the World Economic Outlook. Investment in data centers, semiconductors, and related infrastructure has spurred economic activity, while companies driving the AI boom have delivered exceptional earnings growth. But faltering AI investment and profitability could challenge optimistic expectations and trigger a correction in asset prices linked to the AI industry. 

Growing leverage could be a catalyst for such a correction. Hyperscalers, as the largest data center operators are known, are borrowing more and more, while investors have been leveraging to invest in AI stocks. Circular financing arrangements in the AI sector, involving equity and debt flows between AI firms, suggest that stress at key companies receiving investment could lower revenue forecasts and investment valuations, which in turn would weaken the value of the accumulated debt in those deals. Our review of 14 such transactions since 2023 found about $300 billion in equity investments and $650 billion in debt, in exchange for $1.4 trillion in future revenue commitments.

Government debt risks

Sovereign bond yields have risen amid renewed inflation pressures from higher energy prices, fiscal concerns, and, in some cases, stronger-than-expected economic growth. It may be the case that bond yields are moving back toward the levels seen before the global financial crisis, when growth and inflation were higher than in the 2010s. At the same time, risks of adverse feedback loops related to higher fiscal imbalances and leveraged trades in bond markets are increasing. In addition, as central banks have been reducing their balance sheets, greater reliance on price-sensitive investors could contribute to further yield volatility.

As government bonds underpin the global financial system, the consequences of higher sovereign yields could extend well beyond public debt. Higher yields could further raise discount rates and depress valuations across other assets, including those that are AI-related, while reducing the value of bonds used as collateral. 

Cyberattacks

As finance becomes more digital and interconnected, cyber disruptions have emerged as a growing source of systemic risk. A major incident could disrupt payments, trading, access to deposits, or other critical financial services. The danger is particularly acute if an attack affects a major financial institution, payment system, cloud provider, or software platform, allowing disruption to spread through an increasingly interconnected financial system. 

Importantly, cyber threats can be hard to contain. In more than 70 percent of reported cases so far this year, software providers had almost no time to patch vulnerabilities before attackers exploited them. That’s more than double the share in 2021, when providers typically had close to 300 days to fix vulnerabilities. Frontier AI models could accelerate both attacks and defenses, increasing the importance of cyber resilience, incident reporting, crisis preparedness, and oversight of critical third-party providers.

Financial fault lines

Underlying vulnerabilities could amplify the impact of these three risks into systemic events. 

Although global banks, the core of the financial system, remain resilient overall under our adverse stress scenario, there could be weak links. Banks with 22 percent of global banking assets would emerge with post-stress capital ratios below regulatory limits, an increase from 18 percent a year ago. Moreover, the links of such banks with nonbank financial intermediaries are deepening, creating additional channels for transmitting stress.

Beyond banking, pension funds and insurers have steadily increased allocations to illiquid assets and derivatives, while relying more heavily on the same asset managers. This increases the risk that at a time of stress, they could face sudden liquidity needs through margin calls and could react in similar ways—for example by selling the same assets—amplifying market moves. Some of these institutions also hold illiquid private credit, another potential amplification channel. 

At the same time, as borrowers have become more leveraged, some direct-lending funds have committed to provide funding on demand. If financial conditions tighten, borrowers could draw on those facilities just as the funds that provide them could be facing their own liquidity pressures. Such dynamics could further reduce credit to companies that need it most, with consequences for the broader economy. 

In emerging and developing economies, bond yields have also risen, although their spreads compared with their advanced economy peers remain fairly compressed. However, recent global shocks have diminished buffers, thereby increasing external vulnerabilities and reducing their ability to absorb stress. 

Policies to safeguard 

With government debt high, inflation pressures re-emerging, and higher borrowing costs, policymakers have much less room to support the economy following shocks than in the past. 

We see three priorities for action:

• Debt needs to be put on a sustainable path.
• Leverage and liquidity mismatches in AI, private credit, and other nonbank financial sectors should be limited, including through sectoral capital and concentration requirements on banks’ exposures to these sectors.
• Cyber defense capabilities need to be strengthened, for example through better incident reporting and resilience testing among financial institutions. 

Emerging and developing economies should also rebuild buffers while financial conditions remain accommodative. Proactive policies are the best way to prepare as risks are rising.