Britain expands cyber defences and prepares for autonomous AI incidents

Britain will expand its use of frontier AI to strengthen public-sector cyber security while examining legal safeguards and preparing for serious incidents involving autonomous systems. The programme builds on pilots that have identified and addressed over 1,000 vulnerabilities in government systems and helped detect potential cyber threats.

The government will explore strengthening the Cyber Security and Resilience Bill and the wider regulatory regime to protect essential services, including hospitals and utilities. Areas under consideration include loss of human control over AI systems and how serious incidents and near misses should be identified, reported and used to prevent future harm. Details will be developed with parliamentarians, technical experts and operators.

The National Cyber Security Centre and AI Security Institute are examining how to respond to incidents involving AI agents, which can perform complex tasks with little human supervision. Such incidents could spread faster and more widely than emergencies covered by existing response arrangements. The institute is also evaluating the latest advanced models following work with leading developers and the US Government.

AI Minister Kanishka Narayan urged developers to investigate unexpected behaviour and withhold systems whose risks cannot be adequately controlled. Further support for UK AI verification, which scrutinises claims about development, testing and assessment, will be detailed in the coming months.