Articles

Personal AI agents: UK regulators tell firms to engage before launch

17 November 2026 · 3 min read

On 16 November 2026 the heads of four UK regulators, the CMA, the FCA, the ICO and Ofcom, wrote a joint letter to the Chancellor and three other ministers about personal AI agents. They work together in the Digital Regulation Cooperation Forum (DRCF). They are aware that several personal AI agent products will soon launch in the UK, which they call "a step-change in the way AI is being used".

In short

  • The CMA, FCA, ICO and Ofcom expect several personal AI agent products to launch in the UK soon.
  • Firms must meet their obligations before launch and engage early and openly with the regulators. So far, that has not always happened.
  • Financial services firms that work with AI agent providers are addressed directly: their own obligations continue to apply.

What the letter says

The main points:

  1. Benefits and questions. Personal AI agents could transform how people interact with services and markets and help consumers find savings. But there are important questions about how much the public understands about how these systems work.
  2. Obligations before launch. Firms must meet their legal and regulatory obligations, assess the risks and put safeguards in place, for example against fraud and scams, personal information sharing that users do not expect, and misleading recommendations, decisions or information.
  3. Engage early. The regulators expect firms to engage early and openly with them before bringing products to market. Some firms have done so, but engagement "has not been consistently early or open across the relevant regulators". Early engagement also shows whether an activity requires authorisation or registration. New versions of a product must be tested against the requirements too.
  4. Users stay in control. Personal AI agents must give users agency over their interactions, with a clear understanding of the risks, and must not compromise the operational resilience of the systems they interact with.
  5. Financial firms are addressed directly. Businesses considering commercial relationships with AI agent providers, including financial services firms and telecoms companies, should also engage early: "Their own obligations continue to apply when customers use personal AI agents."
  6. Enforcement. The regulators will not hesitate to use their powers where they reasonably suspect breaches of the law that risk harm to consumers or businesses.

If you are interested in what this engagement can look like in practice, a recent case study from Experian is a good example. On 12 November 2026 Experian described how it launched a credit score inside ChatGPT, which went live in August, as part of the FCA's AI Live Testing programme. A consumer asks "@Experian, what's my credit score?" and sees their score, its history and a breakdown in Experian's own logged-in widget; according to Experian, the score is never exposed to or used by the model. The scope was deliberately narrow, and Experian says an open dialogue with the regulator meant it "could move faster, not slower".

For me it shows what I have long believed: new technology and rules can work together. When a supervisor works with a firm from the start, an idea can reach customers sooner and more safely. It is also an example of a new generation of innovation facilitators: beyond innovation hubs and sandboxes, a supervisor testing AI with firms in real-world conditions.

The letter follows the regulators' joint statement of 30 October 2026, which says that existing law, including financial services and payments rules, already applies to personal AI agents, and that an agent's autonomy does not diminish the responsibility of the firm that deploys it.

Read the DRCF letter · Joint statement of 30 October 2026 · Experian on its credit score in ChatGPT · FCA AI Lab

Questions and answers

What does the letter mean for banks and insurers?

Financial services firms considering commercial relationships with AI agent providers should also engage with the regulators early. Their own obligations continue to apply when customers use personal AI agents.

Do existing UK rules apply to personal AI agents?

Yes. In their joint statement of 30 October 2026 the four regulators say that existing law applies, including competition, consumer and data protection law, financial services and payments rules, and that an agent's autonomy does not diminish the responsibility of the firm that deploys it.

What is the FCA's AI Lab?

It adds an AI focus to the FCA's innovation services. Its components include the Supercharged Sandbox, AI Live Testing, AI Spotlight, AI Sprint and the AI Input Zone. AI Live Testing gives firms a safe place to test AI systems in real-world conditions, with regulatory support and oversight.

What is the DRCF?

The Digital Regulation Cooperation Forum, through which four UK regulators, the CMA, the FCA, the ICO and Ofcom, coordinate their work on digital regulation.

Andres Lehtmets helps financial firms and supervisors make new technology and new rules work together. Former Senior InsurTech Expert at EIOPA.

About Andres

The weekly newsletter.

The week in financial innovation and regulation: key points first, then what they mean for supervisors and financial firms. Free, once a week.

One email a week. Unsubscribe with one click. Privacy policy

Read by 5,200+ professionals at EIOPA, the European Commission, national supervisors, the World Bank, the OECD, the IAIS, AXA, Allianz, Zurich, Swiss Re, Generali, PwC, Deloitte, EY, KPMG and many more.