UK financial watchdog plans to use AI to automate more regulation work

In short: The UK’s Financial Conduct Authority (FCA) plans to use AI to automate more of its day to day supervision and enforcement work, and it will soon publish findings from tests at financial firms.

What happened

The FCA, Britain’s main financial watchdog, says it wants to become what it calls an “agentic regulator”. In simple terms, that means using AI tools to do more of the routine work, so staff can step in faster and on a larger scale when there are problems.

Jessica Rusu, the FCA’s chief data, information and intelligence officer, is expected to explain the plan in a speech next week. She has said the FCA aims to “automate wherever we can” and “intervene at scale”. She also told lawmakers that AI can be helpful for reading and summarising “unstructured text”, meaning information that is not neatly organised, like emails, documents, and interview notes (like trying to quickly sort a messy filing cabinet).

This comes as more people use “frontier” AI models, such as ChatGPT, Claude, and Gemini, including for money questions. A review led by former FCA executive Sheldon Mills said regulators may need more power to keep up, and raised the question of whether these tools should fall under existing financial rules.

Why it matters

Many people already look online for help with saving, borrowing, or investing, and some are starting to ask AI chatbots for advice. If an AI tool gives bad guidance, it is not always clear what protections you have, or whether you can get compensation, compared with advice from a regulated firm. The FCA’s next steps could affect how closely these AI tools are watched in the UK, and what companies must do before using them in finance.

Source: Financial Times