FCA issues Mills Review warning AI will transform finance

The Financial Conduct Authority published the Mills Review on July 6, 2026. The report, prepared by FCA director Sheldon Mills, concludes that artificial intelligence will reshape retail financial services by 2030 and that it could amplify risks including fraud, cyber security threats, consumer harm and market concentration.

The review maps how AI may change the way firms operate, how consumers use services and how markets compete across the rest of the decade. It notes potential benefits such as improved access, personalisation and efficiency, and warns the scale and speed of harms could grow where models act autonomously within preset goals.

FCA research cited in the review found about one in five UK adults, roughly 11 million people, are likely to use AI capable of autonomous action. The report also records consumer concerns about trust and control of AI tools and sets out a roadmap for regulators, government and industry to prepare for changes through to 2030.

The review recommends that firms and regulators focus on governance of AI systems, consumer safeguards, testing and resilience against cyber attacks, and monitoring for market concentration created by large AI providers. It calls for clearer expectations for firms using AI, stronger oversight of models that make or influence decisions, and closer coordination between regulators and government on policy and enforcement.

The review follows a Treasury Committee warning in January that the Bank of England and the FCA risked exposing the public and the UK financial system to serious harm by adopting a ‘wait-and-see’ approach. The committee said current regulatory posture could leave consumers and market stability exposed.

Mills writes: “AI will transform financial services by 2030. It creates significant opportunities for consumers, firms and the wider economy.” FCA chair Ashley Alder described the review as highlighting how consumers and firms can gain benefits while managing risks, and pointed to the FCA’s existing programmes to test AI with firms and its use of AI in regulatory work.

The Bank of England has signalled concern about agentic commerce and agentic trading. At the ECB’s Sintra Forum, deputy governor Sarah Breeden said existing technology-agnostic frameworks may not remain adequate as AI capabilities increase and reported the Bank is exploring tools such as trading “kill switches” to stop activity if models behave unexpectedly.

The report notes the FCA has expanded work on AI, including an open testing environment for firms. It sets out actions and expectations for industry and regulators to address the issues the review identifies through to 2030.