Britain is expanding its use of artificial intelligence to detect cyber vulnerabilities and strengthen protections against risks from increasingly capable AI systems.
The government announced the measures on October 12, building on two pilot programmes that have identified and addressed more than 1,000 potential vulnerabilities across government systems. The expanded initiative will use advanced AI tools to help detect weaknesses before attackers can exploit them.
AI Minister Kanishka Narayan called on companies developing advanced models to take greater responsibility for safety. He told AI developers that if they cannot adequately control a system’s risks, they should not release it.
The government is examining changes under the Cyber Security and Resilience Bill framework. These could improve how serious AI-related incidents and near misses are identified and reported, helping authorities learn from failures and strengthen protections.
Officials are also preparing for incidents involving autonomous AI agents, which can perform complex tasks with limited human supervision. The government has warned that incidents involving such agents could unfold more quickly and widely than existing emergency arrangements are designed to handle.
Meanwhile, the AI Security Institute is evaluating advanced models following work with major developers and the US government. The assessments aim to establish their capabilities and test safeguards before deployment.
The measures form part of a wider effort to protect public services and critical infrastructure as AI becomes more capable. The government will work with technical experts, industry and parliamentarians to determine whether further protections are needed.
