Codex Security: now in research preview

OpenAI announced Codex Security in a research preview for ChatGPT Pro, Enterprise, Business, and Edu users with one month free.

Codex Security builds project-specific threat models, prioritizes and validates issues, and proposes context-aware patches to reduce noise and regressions.

In beta, it scanned over 1.2 million commits, found 792 critical and 10,561 high-severity issues, and reduced false positives and over-reported severity significantly.