OpenAI discloses six new incidents of concerning AI model behavior and introduces a tracking framework

OpenAI disclosed six reports of unexpected or concerning behavior in its AI models, discovered during training or evaluation over the preceding months. The company also introduced a new framework for tracking, probing, and disclosing instances of model misalignment. The announcement comes amid growing calls from AI industry leaders, including those at OpenAI and Anthropic, for a slowdown in AI development over safety concerns.