CFA Institute calls for functional, proportionate AI oversight to safeguard UK retail investors and market integrity

CFA Institute, the global association of investment professionals, has published its response to the Financial Conduct Authority’s (FCA) Review into the long-term impact of artificial intelligence on retail financial services (the “Mills Review”). CFA Institute welcomes the FCA’s technology-neutral approach, while urging greater operational clarity to ensure responsible AI deployment.

In its submission, CFA Institute supports anchoring AI oversight within the UK’s existing principles-based framework, including the Consumer Duty and the Senior Managers and Certification Regime (SM&CR), rather than introducing a standalone AI rulebook. However, it emphasizes that supervisory expectations must be clearer and more practical as AI systems move from assistive tools to advisory functions and, ultimately, autonomous agents.

CFA Institute argues that regulation should follow what AI systems do for consumers, not how they are labelled or constructed. AI-enabled retail interfaces may generate “advice-like” outcomes, such as personalized product steering or portfolio construction guidance, without formally crossing regulatory thresholds. A substance-over-form approach is therefore essential to prevent regulatory arbitrage and ensure consistent consumer protection.

While the Consumer Duty provides a robust foundation, CFA Institute calls for AI-specific articulation of how its four outcomes apply where decision-making is increasingly delegated to automated systems. In particular, the response highlights a risk of automation bias, which may reduce effective consumer outcomes, especially among vulnerable customers.

Firms should be expected to test, monitor and evidence outcomes based on how consumers actually use AI systems in practice, not solely on how they are intended to function.

The submission also identifies a potential governance gap where firms report formal accountability for AI systems yet lack deep operational understanding of complex or third-party models. CFA Institute recommends clearer expectations around what “reasonable steps” and “meaningful oversight” mean under SM&CR and SYSC when AI is deployed in material retail use cases.

It further calls for:

• A proportionate, tiered governance framework aligned to the assistive–advisory–autonomous spectrum
• Clear allocation of end-to-end accountability for consumer outcomes
• Reinforced oversight of third-party AI dependencies and operational resilience risks.

Although retail-focused, the response underscores broader market structure implications, including model concentration, correlated behavior, and third-party dependencies that could amplify volatility in stressed conditions. CFA Institute encourages close coordination between the FCA and the Bank of England, as well as continued alignment with IOSCO and the Financial Stability Board, to reduce fragmentation and support the UK’s global competitiveness.

Finally, CFA Institute stresses that responsible AI adoption depends on developing “hybrid” talent, professionals who combine technological fluency with fiduciary judgement and market expertise. Strengthening professional standards and supervisory capability should form part of the UK’s long-term AI competitiveness strategy.

Olivier Fines, CFA, Head of Advocacy and Capital Markets Policy at CFA Institute, said: “Artificial intelligence has the potential to expand access, improve efficiency and strengthen retail financial services, but only if trust and accountability remain firmly at the center.

“The UK’s principles-based framework is advantageous. The priority now is operational clarity: clear guidance on how the Consumer Duty and SM&CR apply when decision-making is increasingly delegated to AI systems.

“Regulation should follow function, not technological form. Where AI systems effectively shape or execute consumer decisions, protections must apply in substance, not just in label.

“We encourage the FCA to provide practical supervisory guidance by the end of 2026 and to continue close dialogue with industry and international standard-setters. With proportionate safeguards, meaningful oversight and investment in hybrid professional skills, the UK can play a leading role in responsible AI-enabled finance while preserving market integrity and public trust.”