A review commissioned by the Financial Conduct Authority has found that AI will transform UK retail financial services by 2030, with around 11 million adults likely to use artificial intelligence capable of making financial decisions autonomously within pre-set goals. The same review warns that AI will amplify fraud and cyber risks, making scams harder to detect and requiring defences to accelerate.
The Mills Review, led by FCA executive director Sheldon Mills, drew on a survey of more than 5,000 UK retail financial services consumers conducted by Yonder Consulting in April, alongside written submissions from firms, academics, and consumer groups. It paints a picture of sweeping change - offering what Mills described as a "once-in-a-generation chance to close the information asymmetries and frictions that have long left people making poor financial decisions."
The opportunity: closing gaps that cost consumers billions
The review identified several structural weaknesses in UK retail finance that AI could help address. Only 9% of consumers use traditional financial advice. Just 30% hold life or income protection. Around £300 billion sits in low-interest accounts earning suboptimal returns. AI-driven tools could lower barriers to entry, improve personalisation, and help people manage money more effectively.
Current AI use in personal finance remains largely assistive. The survey found that 16% of people use AI to support personal finance activities, rising to 23% among those already using AI elsewhere. Users typically deploy it to summarise, explain, simplify, and compare information rather than delegate decisions outright. Adoption is higher in areas like investing, debt management, and tax planning - precisely where consumers have traditionally sought advice.
Some consumers are already willing to go further. One in eight (13%) said they would give AI real-time access to their banking and financial data. Meanwhile, 55% identified at least one possible benefit from AI in day-to-day money management - though 24% said nothing would persuade them to use AI in financial services, citing concerns over data misuse, lack of protection when things go wrong, and concentration of power among large firms.
Fraud and cyber risks enter a new era
The review found that by 2030, AI is likely to make fraud attacks faster, cheaper, more scalable, and more persuasive - while simultaneously making them harder to spot and stop. "Deepfakes, synthetic identities and personalised social engineering are taking fraud and cyber risks into a new era and changing how fraud and cyber-attacks are conducted," the report said. Existing weaknesses can be exploited far more quickly than before.
Defenders will need access to many of the same AI capabilities as attackers. The review stressed that firms, regulators, and their partners must share the right information with those best placed to act, before harm escalates. The same technologies used to attack the system - if deployed defensively - can help protect it.
For finance professionals tracking how AI reshapes the sector, AI for Finance training is becoming increasingly relevant as regulatory expectations and operational demands shift.
A regulatory roadmap and mixed industry reaction
The Mills Review outlined recommendations for the FCA Board, including developing a trusted public-interest AI-enabled financial capability service and strengthening system-wide coordination. FCA chair Ashley Alder said the recommendations "build on work the FCA has been doing - not least allowing firms to test their use of AI with us - and our own use of AI to be a smarter regulator, more efficient and effective." The FCA will publish an AI good and poor practice publication later this year.
Industry reaction has been measured. Katie Horne, banks relationship manager at Flagstone, pointed to intensifying competition in retail banking and the investments firms have made in customer experience. Richard Hogwood, a divorce lawyer and partner at Stewarts, offered a cautionary example: AI might prepare 60% of a pre-nuptial agreement, but "the other 40% that is critical - the detail and nuance needed by that particular couple." He added that a poorly drafted AI pre-nup could fail to protect assets and minimise conflict, ultimately proving expensive.
Why this matters for finance professionals
The review signals that AI will not simply automate existing processes - it will redraw the boundaries of who delivers financial services and how. For professionals across banking, insurance, wealth management, and advisory roles, the next five years will demand fluency in what AI can and cannot do. The FCA's emphasis on consumer oversight, explainability, and the ability to challenge AI-driven decisions suggests compliance and risk functions will need to deepen their technical understanding. Firms that treat AI solely as a cost-cutting tool will miss the structural shift: as the review notes, unequal access to high-quality applications risks widening inclusion gaps, while well-designed systems could radically improve outcomes for those who need most support.
Your membership also unlocks: