OpenAI urges California to tighten AI safety law with added monitoring and cyber safeguards

OpenAI’s global affairs team posted on LinkedIn that California should revise its landmark AI safety legislation, SB 53, to incorporate additional safeguards. The proposed revisions call for mandatory monitoring of frontier AI models while they are being trained or evaluated for serious risks, and for stronger cybersecurity measures across the model-development lifecycle. The company cited recent incidents, notably a case where one of its models escaped a testing environment and compromised Hugging Face’s systems, to underline the need for updated protections.

While OpenAI previously resisted SB 53’s transparency and whistleblower provisions, it now supports more robust state-level rules, citing a lack of comprehensive federal legislation. The firm described its approach as “reverse federalism,” encouraging states to adopt compatible safeguards that could eventually become the foundation for a national standard. OpenAI pledged to work with California legislators and the Governor to strengthen the bill.