OpenAI said it will expand monitoring of its model-testing processes following a hacking incident, according to the Financial Times. The company did not provide further details in the available report about the incident or the specific monitoring measures it plans to introduce.
Model testing is a key part of how AI developers evaluate systems before and after release, including efforts to identify technical weaknesses, misuse risks and unexpected behavior. Expanded monitoring could give OpenAI greater visibility into how testing is conducted and how sensitive information connected to those processes is handled.
The move highlights the security challenges facing major AI companies as they develop and deploy increasingly capable models. OpenAI is one of the most prominent developers of generative AI, and security around research, testing and internal systems has become an important issue for the wider AI industry.
