Artificial intelligence is set to transform the financial services industry within the next five years, and regulators and the government need to start preparing for it, according to a recent report published by FCA executive director Sheldon Mills (The Mills Review).
The report drew on nearly 140 written submissions from the industry, commissioned research such as a survey of more than 5,000 UK financial consumers, as well as other analysis, research and engagement.
Views from across the financial services landscape found that respondents believed AI will: transform firm operations, evolve the consumer journey, reshape competition and amplify existing market risk.
There is already consumer appetite for AI in personal finance, with roughly 11 million UK adults using it for pre-set financial goals. However, many remain uncomfortable with trust and control of AI.
“Artificial intelligence will transform financial services by 2030,” Mills said. “It creates significant opportunities for consumers, firms and the wider economy.”
However, to ensure the industry is ready for these opportunities, the review outlines several recommendations.
These included: strengthening oversight across the financial services system, building and adopting an AI-enabled supervisory model and adapting regulatory frameworks for AI.
Ashley Alder, chair of the FCA, said: “As is clear in the report, we need to keep pace with a rapidly changing environment and the principles-based, outcomes-focused approach we’ve taken on AI – relying on the Consumer Duty and Senior Managers Regime – has been critical to us doing so.
“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.”
See also: Alpha FMC: Asset and Wealth manager investments need to deliver ‘tangible value’ in 2026
Industry experts have generally welcomed the recommendations of the review, but warned there is still work to be done.
Brian Byrnes, head of personal finance at Moneybox, noted: “Innovation and consumer protection shouldn’t be competing priorities—they need to go hand in hand and the FCA urgently examining whether the regulatory perimeter remains fit for an AI-enabled financial system is the correct first step.”
However, the speed of AI development means that “this work cannot wait”.
“As AI increasingly influences financial decisions, it is essential that consumer protections evolve at the same pace; otherwise, we risk the regulatory equivalent of closing the stable door after the horse has bolted,” Moneybox’s Byrnes noted.
John Allen, director of innovation and operations at the Investment Association, agreed, highlighting that AI use among investment managers will drive improvements in efficiency and personalisation. However, he noted that it must be “deployed in ways that sustain trust”.
See also: Mercer finds AI now used by majority of asset managers in investment process
Marianna Hunt, personal finance specialist at Fidelity International, added: “The rise of AI tools – particularly among younger investors – is changing how people research, plan and make decisions about their financial futures.”
According to recent Fidelity research, almost 22% of UK retail investors have turned to generative AI tools to inform investment decisions, almost 36% of whom are aged between 18 and 34.
Recent research by Quilter and Boring Money also supported this, with nine in ten under 45’s saying they are comfortable using AI to support financial decision-making.
Sam Christopher, proposition director at Quilter, said: “[This] suggests trust and adoption are likely to increase as younger cohorts build wealth and become more active investors.
“As such, there is also a clear opportunity for financial services firms to play a leading role.” For example, Christopher added that hyper-personalisation and the advancement of AI could enable bias or lack transparency, which means regulators and governance should play a role in mitigating this.
The proposition director said: “Nearly half of consumers say they would trust AI more for financial decision-making if the accuracy of the information was guaranteed by a regulated financial provider.
“Firms and regulators should work together to create an environment where innovation can flourish, while ensuring consumers remain protected from the risks of inaccurate information, bias and misuse.”
Fidelity’s Hunt said: “Measures such as the FCA’s new Targeted Support regime will enable providers to offer clearer nudges and guidance at key moments, helping people make better long-term decisions and avoid harmful actions or inactions.
“By combining high-quality guidance with the right regulatory framework, we can help to ensure that people feel confident and informed – whether they are using traditional advice channels or exploring new digital tools.”
See also: FCA ‘reduces burden’ of consumer duty by exempting non-UK clients
Ben Goss, CEO of Dynamic Planner, said the recommendations from the Mills review which enable safe innovation and better outcomes “will be an essential guiding light for financial services businesses large and small”.
“In an era of unprecedented change, AI combined with the human touch can be the catalyst for unlocking access to trusted advice for millions of people and supporting better decisions and outcomes for a wider population.
“The potential of Agentic AI especially, is boundless, not just in reshaping the entire consumer journey but also, used in the right way, in empowering advice firms to create capacity and scale to serve more clients more profitably.”
He added that, according to the FCA, some 20% of adults are already open to AI making decisions for them. In Dynamic Planner’s own Advice 26 study, the firm found that one in ten firms are already servicing clients with agentic AI or chatbots.
“But, with such benefits come risks,” Goss pointed out. “AI should be rigorous, data grounded and tested under regulated conditions. It should also be built around explainable, trusted and repeatable algorithms with clear reasoned and repeatable assumptions, where a given set of inputs always produces a given set of outputs.”
