Government of Alberta uses Claude Code to scan 466 million lines of code for security gaps in 20 hours

11:30 AM · July 14, 2026

The Government of Alberta deployed Claude Code, running Anthropic's Opus and Sonnet models, to review the security of its provincial government systems, according to Anthropic. Using roughly 50 autonomous agents combining automated pattern-matching with detailed manual review, Alberta's team scanned 466 million lines of government code for vulnerabilities in 20 hours — a task Anthropic says would have taken approximately 6.5 years using traditional manual methods. Where vulnerabilities were found, Claude Code generated fixes, wrote automated tests where none existed, and in some cases modernized outdated systems entirely; one example involved a 25-year-old Java-based subsidy portal that was rebuilt from scratch in just four to five days. Alberta has since set up specialized review agents that continuously evaluate applications throughout development, including red-team agents that simulate attacks and blue-team agents that assess defenses, checking each application against roughly 95 security controls per review cycle. The case study is one of a growing number of government cybersecurity deployments Anthropic is publicizing as it pitches Claude Code as viable for the kind of large-scale, high-stakes code review that public-sector IT budgets have traditionally struggled to fund.