Bank of England Warns AI and Market Leverage Could Amplify Financial Shocks

LONDON — Rapid advances in artificial intelligence and growing leverage in global financial markets could amplify future economic shocks, the Bank of England has warned.

In its July 2026 Financial Stability Report, the Bank's Financial Policy Committee said the UK banking system remained strong enough to support households and businesses during a severe economic stress. However, several global vulnerabilities have become more pronounced since December 2025.

These include stretched valuations in risky assets, pressure in sovereign-debt and private-credit markets and a substantial rise in leverage associated with equity investments. The Bank said hedge funds' global equity prime-brokerage balances had increased by approximately 40 per cent during the previous year.

Leverage allows investors to increase their market exposure using borrowed funds. It can magnify profits when prices rise but also force rapid selling when losses occur or lenders demand additional collateral.

The Bank said equity-market growth has been driven partly by a relatively narrow group of AI-related companies, increasing concentration in major global indices. A reassessment of those companies' future earnings could trigger a decline in share prices that might be intensified by momentum-driven investment and leveraged positions.

AI businesses are also increasingly using public debt, private credit, bank lending and structured finance to fund data centres and other infrastructure. The immediate risk remains contained because outstanding debt is still modest relative to the financial system, but borrowing is growing rapidly.

More complex financing structures may make it difficult to identify where losses would ultimately fall. If expected AI profits weaken, highly indebted companies could face refinancing pressure, while investors exposed through private credit or structured products could experience losses.

Frontier AI also presents growing cybersecurity and operational risks. The Bank said advanced models are becoming increasingly capable of identifying and exploiting software vulnerabilities at greater speed and scale. The same technology may help defenders, but financial institutions will need to improve their ability to detect, prioritise, test and safely repair weaknesses.

Faster patching can itself create operational problems if changes are rushed or poorly coordinated across complex systems. Shared software providers and technology suppliers could also become channels through which disruption spreads across several financial institutions simultaneously.

The Bank is not predicting that AI will inevitably cause a financial crisis. It recognises that the technology could improve productivity, cybersecurity and financial services. Its warning concerns the possibility that investment, debt, market concentration and operational dependence may grow faster than institutions' ability to manage the resulting risks.

The Financial Policy Committee will continue monitoring AI-related borrowing, market exposure and cyber vulnerabilities as part of its assessment of financial-system resilience.