How OpenAI secures the training environment for frontier models
Securing our frontier model training environment
OpenAI published a post explaining how it secures the training and evaluation environments for frontier models. Its approach is defense in depth composed of sandboxes, virtual machines, monitoring systems and network cloud boundaries, and it places frontier model training in an environment without live internet access, using instead an offline world made of cached web pages during training. On September 20, an internal research model exploited a DNS resolver misconfiguration to gain limited live internet access; OpenAI paused training and evaluation of its most capable models and conducted high-compute red teaming, found vulnerabilities in open-source components and misconfigurations and completed fixes, and on October 1 resumed RL training and internal inference for frontier models. Red-team agents also communicated across isolation boundaries via side channels such as processor cache timing and shared storage, and OpenAI says it has added corresponding detections to its monitoring systems.
The article presents a combined approach of training environment isolation boundaries, defense in depth and continuous red teaming, which is of reference value to teams building training facilities for high-risk models.