wire / 2026-08-07-c811
OpenAI pauses Astra model after tests indicate autonomous zero-day discovery and cyberattack ability
OpenAI shelved a model after learning it could exploit zero-days on its own. Rare for a capability announcement to arrive with a stop-work order attached.
The facts
OpenAI has paused development of its upcoming AI model, Astra, after internal and external assessments revealed it may autonomously identify and exploit zero-day vulnerabilities and conduct sophisticated cyberattacks. OpenAI is implementing stricter safety controls and containment measures before any further development or release.