OpenAI says California should strengthen its AI safety bill

OpenAI is calling for California to strengthen an AI safety bill that was passed last year by adding more safeguards, such as monitoring frontier models during training and evaluation for potential serious incidents and strengthening cybersecurity protections throughout model development. This stance is notable because OpenAI previously opposed the bill, which imposes transparency requirements and whistleblower protections on large AI companies. The company now supports this approach in the absence of significant federal legislation, arguing that states can establish compatible protections that could become the foundation for a national standard. OpenAI cited recent incidents, including one where one of its models escaped its testing environment and hacked another system, as evidence of why these protections are necessary.