FCA Mills Review: How Will AI Impact Financial Services? -

It’s a big question and when it comes to insurance there are several challenges ahead:

Let’s start with the balance between human intervention vs automated decisions on pricing, claims and compensation when things go wrong. Then there’s personal risk factors vs general postcode risks, surely AI can delineate these at long last, so consumers doing the right thing in bad areas are not unfairly punished with higher premiums, or a refusal to quote?

Consumers are also already sick of hearing “all our operators are busy” whilst on hold for 10 minutes, they are fed up with 30% APR rates on pay monthly car insurance plans and they many will soon become annoyed when they discover that perfectly good EVs are being written off because of minor damage, thus increasing every driver’s premiums, including ICE vehicles.

There is much that AI can do to mitigate these and other issues in the insurance sector. The question is whether the industry regulator and politicians have the willpower to do so. Here’s the word from the FCA;

Led by FCA executive director Sheldon Mills and commissioned by the Board, The Mills Review is the first work of its kind initiated by a regulator globally.

Drawing on views from across the financial services landscape, the report identifies 4 major AI‑driven shifts likely to impact retail financial services: the transformation of firm operations; the evolution of consumer journeys; the reshaping of competition and market power; and the amplification of fraud and cyber risks.

The report finds there is already consumer appetite for the use of agentic AI in personal finance, with research commissioned by the FCA showing that a fifth of people – equivalent to 11 million UK adults – are likely to use AI that can act autonomously within pre-set goals. But consumers in the survey are concerned about trust and control of AI.

The Review concludes that AI is likely to become a defining force in retail financial services, transforming how firms operate, how consumers make financial decisions and how markets function. While AI has the potential to improve access, personalisation and efficiency, it could also amplify risks associated with fraud, cyber security, consumer harm and market concentration.

Executive director Sheldon Mills said: ‘Artificial intelligence will transform financial services by 2030. It creates significant opportunities for consumers, firms and the wider economy. This report sets out a roadmap for how industry regulators and government can prepare for the next phase of AI-driven change in our world-leading financial services sector.’

Key recommendations

The Mills Review also outlines 7 recommendations for the FCA Board and Executive to consider, which are as follows:

Secure and adapt the regulatory perimeter.
Strengthen system-wide coordination and oversight.
Monitor the transition to autonomous models and adapt regulatory frameworks.
Scale up the FCA’s AI Lab to support AI models and system innovation in financial services.
Enable the foundations for agentic finance.
Build and adopt an AI-enabled agentic supervisory model.
Develop a trusted public-interest AI-enabled financial capability service.

FCA response

Ashley Alder, Chair of the FCA, said: ‘The Board is enormously grateful to Sheldon for the rich, comprehensive report he’s delivered. His work anticipates the fundamental change agentic AI will bring to financial services. It highlights how consumers and firms can reap significant potential benefits as well how risks can be managed.

‘As is clear in the report, we need to keep pace with a rapidly changing environment and the principles-based, outcomes focussed 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.’

BROADSTONE COMMENT

David Brooks, Head of Policy at leading independent financial services consultancy Broadstone, commented:

“Pensions are complex, long-term financial arrangements where mistakes can have lasting consequences. While AI has a role to play in improving engagement and understanding, consumers need to treat it as a starting point, not a substitute for professional guidance, scheme information or regulated advice.

“As AI becomes more widely used, improving public understanding of its limitations will be just as important as improving the technology itself. Trust should be earned through accuracy and accountability, not assumed because an answer sounds convincing.

“One of the FCA’s biggest challenges may be protecting consumers from bad pension decisions driven by good-looking AI answers. Generative AI is excellent at sounding authoritative, but not always at being right. When retirement savings are involved, people need to understand that convenience is not the same thing as reliability.”

ARVATOCONNECT COMMENTS

James Towner, Chief Growth Officer at ArvatoConnect: “The FCA has been clear that it wants firms to embrace the opportunities presented by AI rather than wait to be handed a prescriptive rulebook. With 1 in 5 UK adults already open to using AI to manage parts of their finances, the shift the FCA describes, in how firms operate, how consumers make decisions and how markets compete, is well underway.

“AI has the potential to reshape everything from customer service and fraud prevention to operational efficiency, with huge benefits in personalising services and spotting vulnerability. But if firms don’t lead with the customer outcome, and innovation races ahead of consumer protection, the result will be exclusion and entrenched bias. Whether this balance is being struck is under a spotlight today. Our research suggests that in large portions of the financial services industry, it isn’t. More than three-quarters of finance leaders believe their AI strategy could exclude vulnerable customers, yet only a quarter test their systems against real-life vulnerability and build clear routes to a human

“The Review outlines what people want before they will trust AI with their money: clear protection when things go wrong and the ability to reach a human. As AI increasingly has a role to play in the most sensitive and defining decisions in our lives organisations must remember that access is about far more than price. For someone facing financial difficulty, ill health or a major life event, empathy, context and human judgement are often the service itself, not optional extras or premium add-ons.”

OXFORD RISK

Greg B Davies, Head of Behavioural Finance at Oxford Risk, offers these thoughts on why the advice gap won’t close without genuine client understanding built into these systems:

“The Mills Review is right: agentic AI could be a major step forward for retail finance. It could make support more accessible, more personalised and more continuous than traditional advice models have ever managed.

“But the advice gap will not be closed by giving everyone a plausible-sounding chatbot. AI only improves outcomes if it is built around a genuine understanding of the client. Transaction data are useful, but they are not enough. They tell us what someone did, not whether they will panic, procrastinate, overreact, or need reassurance when markets fall.

“As AI moves from recommending to acting, firms will need to evidence more than just efficiency. They will need auditable suitability logic, behavioural insight and clear limits on what AI can do.

“Done well, agentic finance could help people make, understand and stick with better decisions. Done badly, it could scale confusion faster than ever.”

VIPR

“Although the FCA’s review focuses on retail financial services, its message is relevant well beyond that part of the market. AI is only as good as the data behind it. As it moves quickly from experiment to core business infrastructure, regulators are increasingly focused on the controls, governance and accountability that underpin its use.

“We’re already seeing those expectations emerge in delegated authority. The market is growing, capital continues to flow in, and AI-led initiatives are accelerating, but the same question keeps coming back: can firms demonstrate robust oversight and evidence what is happening across the delegated authority chain?

“AI raises the prize, but it also raises the bar. Clean, structured and auditable delegated authority data is no longer simply an operational improvement. It is becoming the foundation for trusted AI, effective governance and regulatory confidence.”

by PAUL TEMPLAR, CEO VIPR

NORTON ROSE FULBRIGHT

Jonathan Herbst, Global Head of Financial Services at Norton Rose Fulbright, said:

“The report shines a light on a fundamental regulatory question – if consumers increasingly rely on AI systems provided by a small number of major technology companies to make financial decisions, how should existing regulatory frameworks adapt?

“Mills is not proposing an immediate crackdown on Big Tech, but he is asking whether the rules need to evolve to reflect how financial services are actually being delivered. That’s a big question for policymakers and one that will only become more pressing as AI adoption accelerates.”

TLT

David Gardner, lead of the Technology segment of TLT’s Digital, Data & Commercial team and AI specialist lawyer said:

“As GenAI capability and adoption accelerate, regulatory support and protection will be essential. It is encouraging to see the FCA taking a proactive approach and engaging with these issues in detail now, in anticipation of future developments.

“The review’s ‘spectrum of autonomy’ is a useful lens through which to view the opportunities and risks of AI adoption. Understanding how far a human is – or is not – in the loop will become increasingly important as firms move towards AI-led execution and agentic journeys. This is where the existing regulatory framework is likely to be most challenged.

“As the review highlights, trust and appropriate human oversight will remain central to responsible AI governance and deployment as the sector navigates these opportunities and challenges.

“Firms need to act now to secure their systems, update their governance and map third-party relationships in an AI market that is fast-moving and firmly in the public spotlight.”

WAKAM

Mark Christer, CEO, Wakam UK

The biggest shift highlighted by the FCA’s review isn’t actually AI itself, but who does the shopping.

For the last 20 years we’ve expected consumers to compare products themselves, but now agentic AI turns that model on its head. The FCA’s review suggests this shift may happen sooner than many expect, with around one in five UK adults saying they are open to AI making financial decisions for them, even though only around 8% of people shopping for home or motor insurance have actually used AI to help. Instead of asking people to spend their evenings comparing quotes, an AI agent could do it continuously in the background, switching customers automatically when it finds better value.

That fundamentally changes insurance distribution as success will depend less on traditional distribution relationships and more on whether products are machine-readable, pricing is accurate and competitive and therefore systems can respond in real time. If an AI agent can’t understand your product or connect to it instantly, there’s a real risk it simply won’t surface.

Products that rely on opaque wordings, manual processes or lengthy implementation times risk becoming invisible to the AI agents that will increasingly drive insurance distribution. As agentic AI develops, insurers will compete more on aspects like data quality, pricing accuracy and their ability to integrate in real time.