The Government of Alberta has been using Claude Code with Opus and Sonnet models since 2025 to review its systems, identify vulnerabilities, and fix them. A team within the Ministry of Technology and Innovation scanned 466 million lines of code in 20 hours, remediated security gaps, and built new tools to enhance system safety. This initiative highlights how government agencies can leverage AI to secure their systems at scale, addressing critical challenges in maintaining secure and reliable public services. Source: anthropic
Alberta’s Ministry of Technology and Innovation manages the systems of all 27 provincial ministries, including approximately 1,280 applications and 3,400 code repositories. Most of these systems have never undergone a systematic security review, and the accumulated technical debt—such as insecure code, unaddressed bugs, and outdated software—amounts to billions of dollars in potential costs. The Ministry’s systems hold highly sensitive information, including tax records, procurement data, and social services case files, making security a top priority. Source: anthropic
The Ministry used Claude Code to assess 466 million lines of government code in 20 hours, identifying issues traditional tools missed. Around 50 agents worked autonomously and in parallel to scan systems for vulnerabilities, weaknesses in infrastructure, and gaps in documentation. The scan covered all repositories, flagged known patterns, and cited exact files and lines for findings. The team estimates that a traditional review would have taken around 6.5 years. Source: anthropic
