UK FCA launches “Mills Review” on AI — what it could mean for everyday banking by 2030

January 27, 2026
UK FCA launches “Mills Review” on AI — what it could mean for everyday banking by 2030

LONDON, Jan 27, 2026, 17:17 GMT

  • UK watchdog has kicked off a comprehensive review into advanced AI use in retail finance, with executive director Sheldon Mills leading the charge
  • FCA is calling for feedback from both industry and consumers by Feb. 24, with recommendations slated for its board in summer 2026
  • Lawyers report that UK companies remain cautious about customer-facing AI, though interest is growing

Britain’s Financial Conduct Authority has launched a long-term review to assess how advanced artificial intelligence might transform retail finance and impact consumers. The regulator’s board expects recommendations by summer 2026. 1

The FCA is moving quickly to address the rapid change in tools firms rely on for selling, pricing, and servicing products—ranging from chatbots to systems that operate with minimal human involvement. The regulator is also keeping an eye on the competitive landscape, concerned that larger firms and tech providers could outpace smaller players as they scale up.

This comes as UK lawmakers pressure regulators for clearer rules on AI in financial services, while firms debate how deeply senior managers must grasp the systems they approve. A parliamentary report last week called on regulators to drop their “wait and see” approach and issue guidance this year on applying consumer protection rules to AI, according to a Reuters report. 2

The review aims to look ahead to 2030 and beyond, seeking input on four key themes: the potential evolution of AI, its impact on markets and competition, the effects on consumers, and how regulators might need to adapt. The FCA has set a deadline for feedback on Feb. 24 and plans to release an external report once its board reviews the recommendations.

“AI is already reshaping financial services, but its impact could stretch much further over time,” Mills said, with the FCA positioning the review as a forward-looking effort rather than a set of new rules. The goal, he added, is to back innovation while ensuring AI is used in retail finance in a “safe and trusted” way.

Mills pointed out the wide range of technologies the FCA aims to monitor, including “agentic AI systems” — software capable of planning and carrying out tasks for users — as well as neuromorphic computing, which relies on brain-inspired chips, and emerging quantum capabilities. He connected this focus to the broader move into digital finance, like blockchain-based “smart contracts” and tokenisation, where physical assets get transformed into digital tokens.

The FCA has, for now, held back from introducing AI-specific regulations, opting to rely on its current framework. This includes the Consumer Duty, which mandates firms to achieve positive results for retail customers, alongside rules on operational resilience and the Senior Managers and Certification Regime that holds individuals personally accountable for major decisions. 3

The watchdog has flagged practical risks to watch for, including AI-driven fraud and financial crime. Deepfakes and “synthetic identities”—fake personas pieced together from various data sources—might make spotting impersonation and account takeovers tougher. Plus, automated systems could amplify poor decisions at scale.

The FCA is investigating if market power might shift from traditional financial firms to major tech and AI companies, and what “good outcomes” should look like as consumers hand over decisions—like budgeting and borrowing—to AI agents.

Tom Callaby, a partner at law firm CMS, told Reuters that AI adoption in UK customer-facing retail finance remains “limited” so far. He noted, however, that firms are increasingly exploring new applications. Callaby cautioned the FCA against closing off adjustments to its regulatory stance, pointing out that unclear, non-specific guidance has been a barrier for many companies.

The review is a marathon, not a sprint. It might not resolve the pressing question many firms have: how regulators will judge practices when AI shapes eligibility, pricing, or product design. Plus, the FCA’s choice to avoid AI-specific rules risks creating patchy standards throughout the market.

Right now, the consultation is ticking down, with responses expected by Feb. 24 and a report set for the FCA board in summer 2026. The main focus: will this process push the FCA to offer clearer guidance on how current consumer and governance rules fit when “the system” takes a bigger role in decision-making?