FCA chief sets out an AI-era approach to market integrity and supervision

A 22 October speech calls for technology-assisted supervision and stronger cross-agency responses as personal agents and continuous markets change financial services.

The UK's Financial Conduct Authority is preparing for a financial system in which personal AI agents, continuous trading and interconnected technology providers change the pace of both innovation and risk. In a [speech published on 22 October](https://www.fca.org.uk/news/speeches/market-integrity-digital-age), chief executive Nikhil Rathi argued for technology-assisted supervision and more coordinated use of existing powers.

The published text describes agents assessing cases, combining regulatory returns and intelligence, and identifying gaps in large datasets. It also points to cooperation through the Digital Regulators Forum concerning Meta Muse. These are statements of direction in a speech, rather than a new standalone AI rule.

Supervisory information becomes a connected picture

A regulator receives information through many channels: structured returns, case records, firm communications and external intelligence. Each can be useful individually, while a significant pattern may appear only when several are considered together.

AI-assisted analysis offers a way to organise those inputs and direct a supervisor toward material that deserves closer attention. The resulting workflow is different from asking a model to make the final judgment. It can reduce the work needed to locate a discrepancy or assemble a case while leaving the consequential decision with the responsible professional.

That distinction becomes especially important when the question is system-wide. Similar behaviour at several institutions may have a different meaning from the same behaviour at one firm. The analysis needs to preserve both the individual records and the connections among them.

Personal agents change the route to a financial service

The speech also treats consumer-facing agents as part of the emerging market environment. A customer may increasingly describe an intention to an assistant rather than navigate a provider's website directly.

For banks and other financial businesses, that can change the first step of a service interaction. A product may need to explain its terms to an agent, receive a request with delegated authority and return information that remains intelligible to the customer. The commercial interface and the accountability path need to work together.

Those changes can cross existing institutional boundaries. A personal assistant, a financial provider and an infrastructure company may each control a different part of the same customer journey. Understanding the whole interaction requires information from more than one organisation.

Speed makes coordination more operational

Rathi's published remarks emphasise partnerships among regulators, industry and other public bodies. The practical issue is how quickly a relevant signal becomes a shared understanding and then an action within the responsible organisation's powers.

A useful operating model connects the technology with that decision path. An alert needs a recipient, the supporting information needs to be accessible, and the organisation needs to know which response it can take. Better analysis can help shorten the path, but it does not replace the institutional work of assigning responsibility.

For financial firms, the speech is a current indication of supervisory priorities as AI adoption expands. It connects the regulator's own use of agents with a broader view of market integrity, consumer interfaces and shared technology dependencies. The emphasis is on adapting supervision to the market taking shape, while keeping outcomes and the responsible use of existing powers in view.

Sources

- [Market integrity in a digital age](https://www.fca.org.uk/news/speeches/market-integrity-digital-age) — Financial Conduct Authority