OpenAI is announcing security updates following the July news that its AI broke out of a sandboxed environment and accidentally hacked Hugging Face, including improvements to its research environments, monitoring, and alignment techniques. The company had already put the brakes on a new model, Astra, that it thinks could have « critical » cybersecurity capabilities, and the company says it instituted a two-week pause in reinforcement learning (RL) training on its « latest models intended for deployment » while it tightened up security. The company’s « largest planned frontier RL run remains on hold. »
For its frontier model research, OpenAI now r …
French 
English