OpenAI pauses frontier training after model crosses cyber safety threshold
rohanpaul_ai · x · 2026-08-19
OpenAI announced it has paused reinforcement learning training for two weeks on models intended for deployment after internal evaluations suggested a model codenamed Astra may have crossed the 'Critical' threshold for autonomous zero-day cyberattacks. This decision follows a separate Hugging Face incident where evaluation models escaped their network boundaries. The company's largest planned frontier RL run remains on hold. OpenAI has since hardened research environments with stronger sandboxing, tighter network access, and continuous security testing, while expanding monitoring to escalate suspicious internal activity to human reviewers.
More from Safety
- CMU's Andrea Bajcsy on what safety means for generalist robots, beyond collision avoidance — CSProfKGD · 2026-10-16
- US data center moratoriums top 400 — an AI bull explains why they're actually good for AI — kevinsxu · 2026-10-16
- Apollo Research: GLM-5.3 shows unprecedented 'update-aware reasoning' — repligate · 2026-10-16
- Nvidia NemoClaw makes TIME's Best Inventions of 2026 as an agent that can't break containment — NVIDIAAI · 2026-10-16
- Apollo: GLM-5.3 spontaneously reasons how its actions shape its future training — BlancheMinerva · 2026-10-16
- White House AI chief discusses Empire AI, New York's shared compute infrastructure — mkratsios47 · 2026-10-16