UK CMA Says Businesses Are Liable for Their AI Agents and Publishes Agentic AI Research
Sarah
Hotels, rental managers and booking platforms that deploy AI agents for guest service, pricing or refunds should note a clear signal from UK regulators: responsibility stays with the business.
The Competition and Markets Authority (CMA) has published research on agentic AI and consumers, alongside guidance titled "Using AI agents: complying with consumer law." The research page describes agentic AI as a possible "step change in how people use AI" and covers how it is used today, how it might develop and how it could affect consumers. It was updated on 30 October 2026 when members of the Digital Regulation Cooperation Forum (DRCF) published a joint statement on agentic AI.
Key points
- Liability: the guidance says that "if an AI agent you use does something illegal, you are responsible."
- Use cases: the CMA supports AI use for growth and says agents can help with tasks such as engaging customers or processing refunds.
- Expectation: businesses should train their AI agents properly to comply with consumer protection law.
- Regulator coordination: the DRCF statement covers work already under way at the CMA, FCA, ICO and Ofcom, how they will collaborate with each other and with government as personal AI agents grow, and how existing legal and regulatory frameworks apply.
Neither page mentions travel specifically and neither includes figures. The CMA has previously investigated online travel pricing practices, so hospitality businesses adopting AI agents in the UK market have reason to follow the guidance.