OpenAI pauses internal Astra model testing over cyber capability concerns

OpenAI halted internal activities for its Astra model after evaluations showed it could autonomously develop zero-day exploits in hardened systems. The company is implementing new security controls, including sandboxed execution and universal monitoring of Chain of Thought, to mitigate high-risk agentic behavior.
1 source
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- OpenAI-backed Thrive Holdings raises $2B to bring AI to the enterprise
- Open Instruct tutorial covers LLM post-training with SFT, DPO, GRPO
- Twitch streamers can now opt out from training Amazon's AI
- OpenWALDO project launches to create shared, open-source AI training dataset
- MIT Technology Review report: Legacy data systems limit AI agents