FCA says AI adoption in finance has arrived, oversight has not

According to the Financial Conduct Authority, 86% of financial firms now use AI, based on a joint survey with the Bank of England due to be published the week after the speech was given. The figure comes from a speech by Jessica Rusu, the FCA's chief data, information and intelligence officer, delivered at Bloomberg's UK Policy Series on 19 October 2026.

Rusu's point is not that AI adoption is coming. It is that it has already happened, faster than firms have built the accountability to match it. She traces the shift from firms experimenting with AI to firms deploying it, and from deployment to what she calls agentic finance, systems that act rather than just recommend, citing Meta's Muse as an example already live.

The gap the speech describes

The FCA commissioned the Mills Review in July to look at how AI could reshape retail financial services by 2030. Rusu says some of what that review anticipated is already arriving, ahead of the 2030 horizon it was written for. That timing gap is the real subject of the speech: a regulator responding to a review whose predictions are overtaken before the response is finished.

Her description of the FCA's own next move matters here. She says the regulator's job now is not to consult, study and report progress, but to test, assure, monitor and watch what industry is actually deploying. That is a shift from writing guidance to checking behaviour directly, and it is an admission that guidance alone has not kept pace with 86% adoption.

What the speech asks of firms

Rusu sets three conditions for firms using AI: put consumers first, keep accountability clear, and plan for when systems fail. She frames these as the dividing line between good and poor AI practice, though the speech text does not yet detail which firms fall on which side.

The scale context she gives is global rather than domestic: AI spend worldwide is expected to pass $2.6 trillion this year, and ChatGPT has passed a billion active users. Those numbers are not about UK financial services specifically, but they set the pace against which an 86% adoption rate among firms makes sense.

The harder question is what accountability looks like once a system is agentic rather than advisory, when it acts on a customer's account rather than suggesting an action to a human who then decides. The speech names that shift as already underway. It does not yet say, in the text available, how a customer would know which kind of system is making a decision about their money, or who they would hold responsible if it goes wrong.

For now, the clearest fact in the speech is the adoption number itself: most firms are already using AI, and the regulator's response is moving from writing about it to watching it happen.