OpenAI halts Astra training to address rogue agent risks

nordicinst · x · 2026-08-19

OpenAI has paused "a significant number" of training workloads and evaluations for its upcoming frontier model, codenamed Astra, to implement new cybersecurity safeguards. The move follows concerns that the model may have achieved "critical" cyber capabilities.

Key Measures and Adjustments:

Related event: OpenAI Pauses Frontier RL Training as Capabilities Outpace Safety(27 posts)→

Original post →

More from Safety

Safety channel →