UK FCA Builds for an AI-Driven Future

In a speech this week, Jessica Rusu, Chief Data, Information, and Intelligence Officer at the UK Financial Conduct Authority (FCA), outlined how the regulator is approaching AI adoption across financial services while also using the technology to modernize its own supervision.

Rusu said a forthcoming FCA-Bank of England survey finds that 86% of financial firms now use AI, as adoption moves from experimentation toward deployment. The FCA is incorporating six of seven recommendations from its recent Mills Review into ongoing work, including testing and monitoring how firms use AI. It has also published examples of good and poor practice centered on three areas: consumer outcomes, accountability and governance, and operational resilience. Firms should maintain clear accountability for AI systems, ensure appropriate human oversight and escalation, and plan for failures and dependencies.

The FCA is also preparing for more autonomous “agentic” systems that can search products, manage payments and investments, and transact on customers’ behalf. Rusu said these systems will require clearer approaches to agent identity, delegated authority, accountability, execution, and continuous assurance. Significant AI failures and near misses should also be treated as reportable operational events, she said.

The regulator itself is adopting many of the same technologies. Rusu described the FCA’s ambition to become an “agentic regulator,” using AI to gather evidence, connect information from different sources, identify early warning signs, and allow supervisors to focus on higher-risk work. The FCA is developing AI tools to support oversight of around 60,000 firms expected to come under the FCA’s AML supervision in 2028, as well as an agent intended to help supervise roughly 9,000 smaller wholesale-market transaction-reporting firms.

The FCA is also developing machine-readable data collection tools and plans to make greater use of its AI Lab and Sandbox for testing and system-wide simulations. “AI is moving faster than any regulatory framework was designed,” Rusu concluded. “That means regulators will increasingly need to make judgements without precedent, while ensuring that trust remains at the heart of financial services.”