FCA demands clear AI financial advice disclaimers
The UK's Financial Conduct Authority has called on AI providers to issue clear warnings on financial advice generated by their models, citing a critical consumer awareness gap and widespread inaccuracies.

The Financial Conduct Authority has mandated that AI providers must issue clear, proportionate and consistent disclaimers for financial advice generated by agentic AI tools. The FCA is calling on general-purpose large language models such as ChatGPT, Gemini and Claude to provide these warnings. Jessica Rusu, the FCA's chief data, information and intelligence officer, made the call during a speech at a Bloomberg event focused on financial technology. The regulator believes warnings delivered directly alongside AI-generated financial information could be more effective than existing consumer awareness campaigns.
The move addresses a growing concern. AI is changing what happens before a consumer ever reaches a regulated firm, as people turn to these models for financial information and advice. The FCA found that AI can widen access to financial tools and information but may blur lines between regulated and unregulated activity and consumer protections. A forthcoming joint FCA and Bank of England survey shows 86% of financial firms now use AI technology.
Consumer awareness gap and AI accuracy concerns
A critical consumer awareness gap underpins the FCA's intervention. Only 40% of consumers correctly recognise that there is no recourse if something goes wrong after acting on advice from a general-purpose AI tool. This leaves a majority unaware of the risks. Independent research compounds these concerns. Saturn, an AI provider in the financial advisory space, reported that AI chatbots are giving out inaccurate financial advice 57% of the time.
Moneybox 19% of consumers would be just as comfortable relying on AI recommendations about where to save as those from an FCA-regulated provider. Four in five less-experienced investors aged 18-40 have used AI to help with investing. Amal Jolly, CEO of Saturn, warned that inaction could lead to consumers suffering serious financial harm. People could make irreversible financial decisions that cost them money and put their families’ financial futures at risk, he stated.
FCA prioritises consumer protection through operational resilience
The FCA requires AI providers to invest in cyber and operational resilience. Jessica Rusu said existing rules requiring major incident reporting should extend to significant AI failures and near-misses, which is becoming a national security and regulatory priority. The FCA warned firms to prepare for significant AI failures, including maintaining fallback arrangements and understanding their reliance on external technology providers.
The regulator published its latest AI update calling on firms deploying AI to do so safely and responsibly to meet Consumer Duty obligations. It also published examples of good and poor AI practice observed during testing with financial services firms. These highlighted concerns about governance, vulnerable customers and operational resilience. Rusu warned some firms were developing AI tools without clearly defined consumer needs, excluding vulnerable customers or introducing automated services without sufficient testing. Some were implementing governance arrangements only after deploying the technology.
Industry response and regulatory balance
Industry participants welcomed the FCA's comments, though some called for the regulator to go further. The challenge is to strike the right balance between embracing AI and managing its risks. Sam Christopher, proposition director at Quilter, said the FCA is right to put consumer protection at the heart of its approach while giving firms room to innovate, test and learn. He added that the response should not become overly prescriptive while technology and understanding develop quickly. Regulation needs to evolve alongside technology rather than trying to anticipate every possible use case today, he stated.
Broadstone senior actuarial director Cormac Bradley said AI has the potential to transform financial services by helping firms cut costs by improving efficiency and delivering better, more personalised outcomes for customers. Firms must ensure that innovation does not come at the expense of consumer protection. Brian Byrnes, director of personal finance at Moneybox, described the FCA’s intervention as a significant step towards improving consumer understanding.
The FCA is taking forward six of the seven recommendations from the Mills Review into the future of AI in retail financial services. This includes work on autonomous AI agents capable of managing investments and executing transactions. The regulator plans to work with industry to develop a framework for autonomous AI systems covering accountability, delegated authority and consumer protection.





