LONDON / RankWire.AI / – Britain is enhancing its national AI security efforts following government tests that uncovered over 1,300 software vulnerabilities. The assessments involved hundreds of code repositories utilized across various central government departments. Officials indicated that cutting-edge AI tools played a crucial role in identifying cyber weaknesses and potential threats. The government now intends to extend this initiative throughout more public sector systems. Additionally, preparations are underway to handle serious incidents involving autonomous AI technologies.
Earlier this year, the two pilot initiatives discovered and addressed 1,352 vulnerabilities across 548 code repositories in 18 different government departments. Teams employed advanced AI systems to scan the software, pinpointing weaknesses requiring remediation. Moving forward, departments will prioritize fixing issues found in their most critical systems. Britain also aims to expand its capacity to leverage frontier AI for cyber defense purposes. Officials emphasized that these efforts are designed to bolster security across government digital infrastructure and services.
Meanwhile, the AI Security Institute is independently assessing state-of-the-art models to verify whether their safety controls function correctly. Artificial Intelligence Minister Kanishka Narayan stated that over 100 technical experts are engaged at the institute. Current evaluations include Gemini 4 Argon along with other advanced AI systems, focusing on model capabilities, security threats, and safeguard effectiveness. Officials reiterated the importance of independent assessments as a vital element of the nation’s AI security strategy.
Expansion of cyber incident preparedness continues
In recent months, authorities have also been developing protocols for major incidents involving agentic AI systems. These systems are capable of executing complex tasks with minimal human oversight. The National Cyber Security Centre is collaborating with other government agencies to establish a comprehensive response framework. Discussions focus on how to identify problems quickly, contain risks effectively, and support organizations affected by such incidents. The effort also involves reviewing how current cyber response systems can manage failures involving increasingly autonomous technologies.
Furthermore, Britain is reconsidering whether the Cyber Security and Resilience Bill should incorporate more robust provisions for AI-related risks. Key areas under review include serious incidents, near misses, and cases of loss of control. Existing proposals already require certain regulated operators and digital service providers to report incidents within 24 hours, with fuller reports expected within 72 hours under the proposed framework.
AI security measures become a top priority for the UK
Narayan highlighted that Britain must strengthen its cyber defenses, enhance independent technical testing, and build domestic expertise in advanced AI. He also indicated legislation might be necessary where current safeguards fall short against emerging risks. Officials are working on further support measures for AI verification and plan to release additional details in the coming months. These verification methods aim to ensure systems operate as intended and that security controls remain effective under challenging or unusual conditions.
These initiatives are part of a broader government review of artificial intelligence and its societal impact. The Prime Minister’s AI Taskforce is examining security, public service delivery, economic growth, and workforce implications linked to AI. Additionally, Britain has included AI on its agenda for the 2027 G20 summit. The recent steps focus on remedying software vulnerabilities, evaluating advanced AI models, and enhancing preparedness for major AI-related incidents across government systems.