OpenAI Strengthens Security for Frontier Model Training Environment
OpenAI has strengthened containment for frontier model training with offline worlds, defense in depth, and red-teaming by powerful agents. The research environment remains offline; continuous red-teaming addresses