Making sense of the FCA’s new AI Review in Financial Services

AI is everywhere, changing how the world works, and the Financial Services sector is not immune to its inevitability. From spotting fraud to helping customers manage their money. Now, the UK’s Financial Conduct Authority (FCA) has launched a major review to better understand and anticipate what happens next.

Rather than sitting back and reacting problems when they appear, the FCA is being proactive and attempting to look forward and prepare for the future, and how else AI will play a central role in the industry.

This article will break down this review into clear and simple terms; What it is, why does it matter, and what this means for firms and customers.

What is the FCA doing?

The FCA has started a long-term review into how artificial intelligence might shape the future of retail financial services. For instance, the designing and delivery the products that everyday people use such as banking, insurance and savings.

The review will lead by the FCA’s Executive Director of Consumers and Competition, Sheldon Mills, which has led to the review unofficially being named The Mills Review.

Most importantly to note, this review is not a new rulebook, nor will it be over quickly. The review has purposefully been designed with longevity in mind in order to take in the bigger picture and how fast-moving AI tech could change:

- How firms operate

- How markets compete and grow

- How customers make decisions

- And how regulation should evolve

All with the intent and purpose to answer; “What should safe and fair finance look like in an AI-Driven world?”

But Why Now?

It’s a very good question; AI has been penetrating force for some years now. What makes this review notable is the timing. Regulators have historically moved after markets change, and this review signals an attempt to instead, shape the direction of this change. The FCA have decided that now is the right time, for three simple reasons:

1. AI is already everywhere – Financial Services firms are already using AI to detect fraud, assess credit risk, automate customer service and making improvements to operational procedures. So, the future is now.

2. Rapid improvements to technology – AI is now in a place where it can generate content (not this content!), continuously learn and adapt, and in some cases, act independently. This raises many questions around governance and control.

3. Keeping up – Regulators do not want to fall behind. If they continue to wait, there is a risk of harm being caused to consumers and the market. The FCA knows that they need to prepare early and not react too late.

A question that needs to be asked, is whether regulators can meaningfully keep up with the evolving technologies? If they can, this review could spark the beginning of a more proactive model of financial regulation. However, there is a risk of highlighting that they it’s not possible, and that there is an ever-growing gap between technological capability, and regulatory control.

The Big Four Questions

The focus of the review has been broken into four focused areas to explore:

1. What is the future capability of AI?

Will AI continue to be a helpful assistant, or, will it continue to evolve so that it becomes more independent and makes more of its own decisions. And if so, where does the accountability sit when it inevitably goes wrong?

2. How will it affect markets?

There are so many advantages to AI, that spread the depth to new entrants to the market, to those with the capital to invest to increase computing power. But will this increase competition, or market concentration?

3. What about Customers?

Like the effects of the market, customers could see the benefits of cleaner, quicker and personal experiences when dealing with retail financial firms. However, AI also has the possibility to introduce further bias in decision making, or poor automated service.

4. What about the rules?

Us Risk and Compliance Managers love a rule, or a boundary. It helps keeps things in order and moving efficiently. But the question is being asked to whether the existing set of rules are adequate, or we need brand new AI laws that help protect customers?

Whilst it’s the regulator exploring these questions, the answers go the heart of financial services. In an AI-driven market, those that have that competitive edge, may not be those with the deepest pockets, or even those with the best products or lowest costs. But instead, those that can combine data, trust, governance and accountability with AI-driven autonomy, creating a simplified customer experience. This review, perhaps the potential to signify regulation becoming intertwined with strategy; not just defining it.

How does this fit with the FCA’s wider work

So, this isn’t the start of things, the FCA has already been running regulatory sandboxes where firms have volunteer to test AI tools. The FCA have also been working with industry and tech providers on ensuring responsible innovation and studying how different firms have been using AI in its current state. This review brings all the previous learnings together and continues to look forward.

So, what could change?

Whilst the review has been labelled an ‘exploratory review’, early indications suggest some inevitable impacts:

- Retail Financial firms should expect more focus on governance, accountability, and, you guessed, consumer outcomes. This will be even more prevalent, when AI has been used for decision making.

- FinTech’s and AI builders should expect more regulatory focus when a third-party tech provider is involved. Additionally, attention will be paid to data concentration and systemic risks linked to the use and design of AI models.

- The aim, for customers, is to ensure a simpler, more personalised and accessible financials services.

Those firms that treat AI governance as a core strategic requirement, rather than just a regulatory obligation, are likely to be the innovators for the next era of financial services. We all know too well that strong risk management became a signifier of competitive strength after the global financial crisis; perhaps responsible AI use could be the next signifier. Those who invest now could not only avoid scrutiny from the FCA, but they have the potential actively shape the future.

What’s next?

As I mentioned, this isn’t a quick review, and the FCA has planned to report its findings to their board mid-2026. And the outcome of the review will shape future guidance and supervisory priorities, not create new rules. The release of the review may be seen as the moment regulators begin formally preparing for machine-mediated finance; a system where algorithms and machine learning shape everything.

So, this review is less about today’s compliance checklist, and more about tomorrow’s financial system, shifting supervision closer to technology, and technology becoming inseparable from conduct.

Final thoughts?

Let’s be clear, the FCA is not trying to put a stop on firms using AI. As a member of the RegTech Strategy Group, I’ve spoken with the FCA innovation team, and they are actively encouraging firms to make use of the AI lab that they have developed. The FCA’s AI Lab is a safe, collaborative space where firms and regulators can share ideas, try out AI in real-world settings, and help the FCA better understand the risks and opportunities of AI in financial services.

This shows that they understand that this is now the world that we live in, and to keep up, and keep both consumers and firms safe, they need to evolve. This review is their way to make sure AI develops in a way that is safe, fair, and beneficial for all. It’s about the future of trust in financial services.

By doing this review now, they, and I, hope that this means the industry can avoid bigger problems later down the line, or when it’s too late, and help build a financial services system that people can trust, even as technology drives forward, and become part of everyday life and decision making. Customers are likely to no longer just firms only on price or performance, but on how transparent and responsible their automated decision-making is. The development of ethical technology has the potential to overtake brand value, or customer service, in terms of importance to customers.

Perhaps the question should be; “How do we keep finance human-centric in age of intelligent machines?“