OpenAI pauses development of Astra model over critical cyber capabilities

OpenAI is slowing development of its upcoming Astra model after internal evaluations indicated it reached a "Critical" cybersecurity threshold. The model demonstrated the ability to identify and develop functional zero-day exploits in hardened systems without human intervention.
7 sources
OpenAI built a model so capable at hacking that they paused their own development on it. Then...x.com
The AI model OpenAI won’t release yet — and what it found in testingthenewstack.io
OpenAI puts the brakes on a new model because it’s supposedly too powerfultheverge.com
OpenAI on upcoming model "Astra" (GPT-6): "We're treating it as our first "critical" model for cybersecurity"openai.com
OpenAI by email
Get an email when OpenAI has news
No news that day, no email.
More stories today
- Anthropic's Ultracode coding mode gains industry attention
- Testing the Motion Context node for Stable Diffusion
- Krea2 Turbo BBOX fine-tune uploaded to HuggingFace
- Reddit user shares Minimax H3 character/object V2V swapping template
- ChatGPT accidentally makes photorealistic image mistaken for real photo