How OpenAI decided Astra was too dangerous to ship open

OpenAI for the first time assigned its GPT-6 Astra model a Critical rating for the ability to autonomously find zero-day vulnerabilities and create exploits. The company paused some work, strengthened isolation and monitoring, then released the model with closed access to the dangerous capability.

• Astra is OpenAI's first model rated Critical for cybersecurity
• Threshold triggered by autonomous zero-day discovery and exploit creation
• Dangerous capability shipped with controlled access, not openly
• Development was paused to strengthen isolation and monitoring