Bank of England Warns: AI Advances Now a Financial Stability Risk

The Bank of England's Financial Policy Committee flagged on 26 June 2026 that rapid advances in frontier AI—including Anthropic's Mythos model—have lowered both the cost and scale at which malicious actors can inflict operational shocks on financial infrastructure. The FPC cited AI development outpacing expert forecasts, raising cyber and resilience threats the regulator does not yet fully control. The warning arrives as the Committee simultaneously loosens bank capital buffers.