LONDON / RankWire.AI / – Britain is scaling up its national AI security initiative after government-led tests revealed over 1,300 software vulnerabilities. The evaluations targeted hundreds of code repositories used throughout central government entities. Officials highlighted that cutting-edge AI tools played a key role in identifying cyber weaknesses and potential risks. The government now intends to extend the initiative across wider public sector systems, including preparations for serious incidents involving autonomous AI technologies.
During the two pilot programs, a total of 1,352 vulnerabilities across 548 code repositories within 18 government departments were identified and addressed. Teams employed sophisticated AI systems to analyze software code and flag issues requiring remediation. Moving forward, departments will focus on fixing flaws detected in their most critical systems. Additionally, Britain aims to boost its capacity to leverage frontier AI technologies for cybersecurity defense. Officials emphasized that this programme aims to bolster protection across government digital infrastructure and services.
An independent AI Security Institute is also assessing advanced AI models to verify whether their safety protocols are functioning as intended. Artificial Intelligence Minister Kanishka Narayan stated that more than 100 technical specialists are employed at the institute. Current assessments include Gemini 4 Argon alongside other sophisticated AI systems, focusing on evaluating model capabilities, security threats, and the robustness of safeguards. Officials reiterated that independent evaluations are a vital element of the nation’s AI security strategy.
Expanding cyber incident response strategies
Preparations are also underway to establish protocols for managing major incidents involving agentic AI systems capable of executing complex tasks with minimal human oversight. The National Cyber Security Centre is collaborating with other government agencies to develop a comprehensive response framework. This includes strategies for identifying issues, containing risks, and supporting organizations affected by such incidents. The initiative also examines how current cyber response systems can adapt to failures involving increasingly autonomous technologies.
The UK government is reviewing whether the Cyber Security and Resilience Bill should incorporate enhanced measures targeting AI-related threats. Considerations include serious incidents, near misses, and cases involving loss of control. Existing proposals already mandate reporting duties for specific regulated operators and digital service providers, requiring initial cyber incident reports within 24 hours, with more detailed follow-up reports within 72 hours under the proposed framework.
Prioritizing AI security safeguards within the national security agenda
Narayan emphasized that Britain needs to strengthen its cyber defenses, improve independent technical testing, and develop domestic expertise in advanced AI. He also indicated that legislation may become necessary if current safeguards are inadequate against emerging risks. Authorities are working on additional support measures for AI verification and intend to publish further details in the coming months. Such verification methods are designed to confirm whether systems operate correctly and to assess if security controls remain effective under challenging or unusual conditions.
These initiatives are part of a broader government effort to evaluate the impact of artificial intelligence on national security and society. The Prime Minister’s AI Taskforce is examining issues related to security, public service delivery, economic development, and workforce adaptation linked to AI. Additionally, Britain has placed AI prominently on its agenda for the 2027 G20 summit. The recent measures focus on addressing software vulnerabilities, assessing cutting-edge models, and enhancing preparedness for serious AI-related incidents within government infrastructure.