UK cyber insurers are beginning to address the exposure created by clients' use of generative AI, though the market has not yet agreed on a standardised underwriting approach. Daniel Winn, a development broker at Jensten London Markets, said the core concern is staff feeding sensitive or regulated information into public large language models outside a client's own environment. Insurers have split in response: some offer full affirmative AI cover up to the policy's total limit, while others cap it, with one or two carriers setting the limit at around £250,000. Winn noted this caution reflects a lack of claims data, and that some underwriters themselves now use generative AI to help quote submissions. Selorm Kofi Domeh, broking manager at Talbot Jones, described a market working through specific exclusions and endorsements rather than a consistent set of questions about how clients use AI, leaving brokers to volunteer material facts. Ethan Godlieb, associate partner at Consilium Insurance Brokers, pointed to regulation as a force pushing toward clearer answers. The EU AI Act entered into force on 1 August 2024, with obligations for general-purpose AI models enforceable from August 2025. Provisions covering high-risk AI systems were originally due from August 2026, but an EU simplification package agreed this year pushed that deadline to December 2027 for standalone high-risk systems.