2 articles
OpenAI Launches GPT-5.4-Cyber: AI Built for Cybersecurity Defenders with Codex Security Fixing 3,000+ Critical Flaws
April 15, 2026 · low
OpenAI Codex Security: AI Agent Scanned 1.2 Million Code Commits and Found 10,561 High-Severity Vulnerabilities
March 9, 2026 · high