OpenAI's Astra model reaches 'critical' cyber threshold

OpenAI says its forthcoming Astra model is the first to reach 'critical' cyber capabilities under its Preparedness Framework, able to independently find and exploit unknown vulnerabilities. It will release Astra broadly 'soon' but limit advanced cyber features to select Daybreak Blue partners. During testing, Astra found two V8 zero-days and used them in an exploit chain with little human help.
How this story unfolded
4 weeks · 1 report · 4 community posts · from Aug 7
OpenAI by email
Get an email when OpenAI has news
No news that day, no email.
More stories today
- Data center spending to hit $31.6T by 2050 on AI boom
- OpenAI explores hiding model 'thinking', raising safety concerns
- Emad Mostaque: Frontier models will one-shot at 10k tokens/sec
- LangSmith adds Messages View for agent debugging
- Notebook collection covers 30 LLM agent memory techniques