Bank of England deputy governor calls for bespoke agentic AI rules and market kill switches

Bank of England Deputy Governor Sarah Breeden called on June 30, 2026 for bespoke governance frameworks and market-wide circuit breakers to contain autonomous AI agents in financial markets. The call marks a shift from the central bank’s prior view that existing rules were enough. The speech, titled “Agents of change,” was delivered at the European Central Bank Forum on Central Banking in Sintra, Portugal.

The distinctive element is timing and coordination. Breeden’s remarks land six days after the Financial Conduct Authority made a parallel argument, and together they amount to the first time both major UK financial regulators have explicitly called for new rules aimed at agentic AI, the term for software agents that act with limited human oversight.

Breeden’s concern is correlated behavior. “If AI agents respond similarly to the same prompts or triggers, they could amplify volatility in stress,” she said, warning that “relying on a human in the loop for all agent actions is unlikely to be realistic.” Her proposals include enhanced recovery mechanisms letting one bank assume another’s core functions during AI-triggered disruptions, market-wide “kill switches” to halt trading if faulty models cause instability, and bespoke accountability frameworks for agents in markets and payments.

The FCA framed the same problem as a pace mismatch. “Technology is moving much faster than many regulatory paradigms. Legislation will never keep up,” Chief Executive Nikhil Rathi said in a June 24 speech, published by the regulator. The FCA proposed using agentic AI itself as a “first responder” to speed up market monitoring, mandatory accountability frameworks, and a stronger regime governing banks’ dependence on cloud and model providers.

Both interventions are speeches, not rules. No consultation, timeline or draft text accompanied either, so the gap between signaling and enforceable policy remains wide.