AI in action · Regulation

The FCA Says It Will Become an 'Agentic' Regulator and That 86% of UK Financial Firms Now Use AI. It Is Not Writing New Rules

In a speech to the Bloomberg UK Policy Series on October 19, 2026, the FCA's chief data officer Jessica Rusu said a joint FCA and Bank of England survey due the following week shows 86% of financial firms use AI, that the regulator plans to become 'a more technology-enabled, and ultimately agentic, regulator', and that it will work with frontier AI labs on a framework for 'agentic finance' covering agent identity, delegated authority and accountability. No new AI rules are planned.

Oct 20, 2026

Britain's markets regulator set out on Monday how it intends to supervise an industry that has, by its own count, almost entirely adopted artificial intelligence. Jessica Rusu, the Financial Conduct Authority's chief data, information and intelligence officer, told the Bloomberg UK Policy Series on October 19, 2026 that a joint FCA and Bank of England survey of AI use, due for publication the following week, shows 86% of financial firms using the technology, and that firms are "shifting from AI experiments into operational deployment". The published text of the speech is marked as a draft that may differ from the version delivered.

The most quoted line was about the regulator itself. The FCA "plans to become a more technology-enabled, and ultimately agentic, regulator," the accompanying statement said, investing in technology and staff "to expand operational capacity, gather evidence, connect disparate information sources and identify warning signs of potential harm earlier." The term is not defined further. The context is capacity: the FCA expects around 60,000 anti-money-laundering firms to fall within its remit in 2028 and already oversees about 9,000 smaller firms that report wholesale-market transactions, and it introduced a machine-readable Handbook API in August.

For trading and investing, the relevant section is what Rusu called agentic finance. The FCA will work with firms, other regulators and the frontier AI laboratories on "a trusted framework" covering four things: the identity of an AI agent, the authority delegated to it, who is accountable for what it does, and how it executes. She also advocated "continuous, conditional assurance" for autonomous systems rather than one-off approval, and reporting of significant AI incidents and near misses. On general-purpose chatbots, the FCA asked AI providers for "clear, proportionate and consistent disclaimers" about the limits of model-generated financial information; its own research found that four in five less-experienced investors aged 18 to 40 have used AI to help with investing, while only 40% of consumers correctly understand that they have no recourse if they act on a general-purpose model's advice and lose money.

What the FCA is not doing is as clear as what it is. The speech kept the regulator's remit unchanged for general-purpose AI, asking only that the boundary be "explained more clearly" rather than automatically extended, and said the FCA would build on existing programmes, the AI Lab, AI Live Testing, the Supercharged Sandbox, the IOSCO TechSprint and its Firm Checker, rather than create the new public-interest financial-capability tool that July's Mills Review had suggested. Of that review's seven recommendations, the FCA said six are already in active work. A page of good and poor practice examples was published the same day; it states that the examples "do not create new requirements".

The speech lands three months after Finance Magnates counted at least ten brokers and platform vendors, among them eToro, Robinhood, moomoo, ThinkMarkets and Bybit, that had connected AI agents to live client accounts in the first half of 2026 with no rules written for the practice. The FCA's answer is that the existing rules, the Consumer Duty and the Senior Managers regime above all, already apply, and that it will supervise with agents of its own. Whether a regulator can be agentic faster than the firms it regulates is the question the survey next week will not answer.

Sources: Financial Conduct Authority — Building the future of finance in the age of AI, speech by Jessica Rusu (Oct 19, 2026); FCA press release via EuropaWire — FCA sets out plans for safer AI adoption and agentic finance oversight (Oct 19, 2026); Law360 — FCA Plans 'Agentic' Shift To AI-Powered Supervision (Oct 2026); Finance Magnates — FCA's Landmark AI Review Lands After the Agents Already Arrived (Jul 6, 2026)