UK and EU regulators tighten focus on generative AI

The Digital Regulation Cooperation Forum says UK regulators are moving generative AI from experimentation into regulatory oversight. Its members, including the CMA, FCA, Ofcom and the Information Commissioner’s Office, are using or testing the technology for supervision and enforcement. The CMA is developing agentic AI to capture consumer journeys at scale and detect possible consumer law breaches such as drip pricing, while the FCA has tested LLMs for identifying sludge practices. The FCA pilot found scalability benefits but confirmed that human review remains essential.

The forum is also seeking input on digital verification and synthetic media, citing authentication and trust risks as AI agents begin making decisions for users. It is asking for views on provenance tools, labelling, standards, supply chain accountability, platform moderation, enforcement and overlaps between the Online Safety Act 2023, the UK GDPR, IP and consumer protection rules. The call for input closes on 14 August 2026.

In the EU, the Digital Omnibus on AI has entered into force, revising compliance timetables under the EU AI Act. The European Commission has issued guidelines on Article 50 transparency obligations, with rules for deployers due on 2 August 2026 and a transitional watermarking deadline for some providers on 2 December 2026. The Commission has also launched ADACities to support autonomous driving deployments in EU cities, with expressions of interest closing on 30 September 2026.