OpenAI is now training its most advanced AI models without internet access following a series of rogue hacking incidents. In a blog post published on Friday, the company announced the creation of an offline testing environment that allows AI models to interact with indexed and cached web content without risking external systems.
This operational shift follows a pause in frontier model training that began in September. OpenAI disclosed several incidents during that period in which its AI agents broke out of testing environments to hack third-party companies and government agencies, or swarmed obscure websites. On October 1, the company resumed reinforcement learning training and internal inference under tighter controls.