UK steps up AI cyber defences as minister warns dependence is a national security risk

The UK government will expand its use of frontier AI to protect public services and examine stronger protections against autonomous systems escaping human oversight, as AI Minister Kanishka Narayan warned that dependence on overseas technology has become a national security issue.

Announcing the measures at Imperial College London on 12 October, Narayan said increasingly capable AI demanded stronger technical expertise, incident response capabilities and domestic industrial capacity.

The package includes wider deployment of AI to identify weaknesses in government systems, consideration of incident and near-miss reporting measures, and work to prepare for failures involving autonomous AI agents.

For the UK’s security and advanced technology industries, the announcement signals growing government interest in AI assurance, cyber protection and the capabilities needed to investigate systems acting with limited human supervision.

AI to strengthen cyber defences

According to the government, two pilots using frontier AI have helped identify and address more than 1,000 potential vulnerabilities across government systems.

Narayan also said the government’s central vulnerability-monitoring service had reduced the time taken to fix identified weaknesses from 50 days to eight. He announced an urgent expansion of frontier AI use to find vulnerabilities before attackers exploit them and improve responses to emerging threats.

“I want cybersecurity discussed in every boardroom in Britain, rather than simply delegated to middle management,” he said.

Alongside this operational work, ministers will explore whether the Cyber Security and Resilience Bill and its wider regime should be strengthened to address AI-driven threats and loss-of-control risks.

Potential measures include identifying and reporting serious incidents and near misses, reducing risks across essential services and enabling reviews of serious AI incidents. Detailed requirements remain to be developed with parliamentarians, technical experts and operators.

Preparing for autonomous AI incidents

The National Cyber Security Centre and AI Security Institute will help develop government responses to incidents involving AI agents, including containment and support for those affected.

Narayan warned that failures could spread rapidly across connected systems, requiring investigative capabilities beyond those traditionally needed by government.

“When that happens, government must be able to establish quickly what an agent was asked to do, what it accessed, why it acted as it did and whether the same failure could occur elsewhere,” he said.

The government also intends to support AI verification, which tests whether claims about systems’ development, testing and assessment withstand scrutiny. Narayan highlighted early research into hardware-enabled verification as an opportunity to draw on UK semiconductor design expertise.

UK capability becomes a security priority

The minister linked these protective measures to a wider argument for retaining domestic technological capability and influence.

“Much of the technology that will shape our economy, our public services, our national security is being developed outside this country,” he said.

Pointing to the government’s Sovereign AI programme and AI Hardware Plan, Narayan argued that attracting international investment must be accompanied by greater UK control over strategically important technologies.

“We want the world’s leading companies investing here, but we will not accept dependence as the price of that investment.”