Bank of England Says it Is Testing AI Risks to Financial System

The Bank of England is testing the risks to the financial system caused by AI by conducting scenario analysis and simulations, the central bank said in a letter published by lawmakers on Thursday.

The BoE said it disagreed with an assessment by parliament’s Treasury Committee that it was taking a “wait-and-see” approach to the risks presented by AI, and that it was analyzing how AI investment and adoption were changing the financial system.

It is also working with international counterparts to understand how AI agents might affect trading in financial markets, BoE Deputy Governor for Financial Stability Sarah Breeden said in a letter to the Treasury Committee.

Testing will focus on “herding” behavior that could amplify selloffs during periods of market stress, Breeden said.

AI risks in the financial system came into sharper focus last week with the launch of Anthropic’s Mythos product. Experts say its powerful coding ability could offer new ways to find cybersecurity vulnerabilities and exploit them.

BoE Governor Andrew Bailey said Anthropic may have “found a way to crack the whole cyber risk world open.”

The Treasury Committee criticized Britain’s finance ministry for failing to commit to bringing major AI and cloud companies into the Critical Third Parties (CTP) Regime – which regulates key financial system infrastructure suppliers – before the end of 2026.

“I am pleased to see the Bank of England is grasping the nettle to some extent but I remain perplexed at the apparent inertia shown by the Treasury,” Treasury Committee Chair Meg Hillier, a member of the governing Labour Party, said.

“The powers offered by the Critical Third Parties Regime are sitting unused while we remain vulnerable. I simply cannot understand why this is taking so long. We will continue to monitor this situation closely.”

Treasury minister Lucy Rigby told the committee that the government expects to make initial CTP designation decisions this year, but would not reveal which firms are under consideration in order to protect the integrity of the process.

The BoE’s Financial Policy Committee on April 1 said firms have yet to deploy advanced AI such as agentic tools in ways that pose systemic risk, but warned that those risks could rise rapidly as the financial sector steps up adoption.

(Reporting by Andy Bruce; editing by Andrew Heavens)

Related:

• ECB to Quiz Bankers About Risks of Anthropic’s New AI Model, Source Says
• Bessent, Powell Warned Bank CEOs About Anthropic Model Risks, Sources Say
• Wall Street Banks Try Out Anthropic’s Mythos
• Bank of England Set to Discuss Anthropic’s Mythos With Banks