LONDON / RankWire.AI / – Britain is boosting its national AI security initiatives following government testing that revealed over 1,300 software vulnerabilities. These assessments analyzed hundreds of code repositories utilized across central government agencies. Officials highlighted that advanced AI tools played a crucial role in uncovering cyber weaknesses and pinpointing potential threats. The government now intends to extend this program to cover more public sector systems and includes preparations for serious incidents involving autonomous AI technologies.
The two pilot projects uncovered and remedied 1,352 vulnerabilities within 548 code repositories across 18 government departments. Using sophisticated AI systems, teams conducted scans of software to detect flaws requiring intervention, with departments now prioritizing fixing issues in their most critical systems. Britain also aims to enhance its capacity to deploy frontier AI for cyber defense, with officials stating the program will bolster protection across government digital services and technology infrastructure.
Separately, the AI Security Institute is assessing advanced models to verify if their safety controls function correctly. Artificial Intelligence Minister Kanishka Narayan mentioned that over 100 technical specialists are employed at the institute. Current evaluations include Gemini 4 Argon among other sophisticated AI systems, focusing on model capabilities, security risks, and safeguard effectiveness. Authorities emphasized that independent testing remains a vital aspect of the nation’s AI security efforts.
Expansion of cyber incident protocols
Authorities are also developing procedures to address major crises involving agentic AI systems capable of executing complex tasks with minimal human oversight. The National Cyber Security Centre collaborates with other government entities to establish a comprehensive response framework, considering how to detect issues, contain risks, and support impacted organizations. The work also evaluates how existing cyber response mechanisms can handle failures involving increasingly autonomous technologies.
Britain is reviewing whether the Cyber Security and Resilience Bill should incorporate stricter measures concerning AI-related threats, particularly in cases of serious incidents, near misses, or loss of control. Current proposals already mandate reporting duties for certain regulated operators and digital service providers, with initial cyber incident reports due within 24 hours and comprehensive updates within 72 hours under the proposed framework.
Strengthening AI security measures
Narayan emphasized that Britain requires enhanced cyber defenses, independent technical testing, and domestic expertise in advanced AI. He also indicated that legislation might be needed when existing safeguards fail to address emerging risks. The government is working on additional support mechanisms for AI verification, with plans to release further details in the coming months. Such verification methods are designed to test whether systems operate as intended and to ensure security controls remain effective under challenging or atypical conditions.
These initiatives are part of a broader government review of artificial intelligence and its potential impact on the nation. The Prime Minister’s AI Taskforce is examining issues related to security, public services, economic development, and workforce implications of AI technology. Additionally, Britain has prioritized AI on its agenda for the 2027 G20 presidency. The latest measures aim to address software vulnerabilities, evaluate cutting-edge models, and bolster readiness for serious AI-related incidents within government systems.