OpenAI calls for stronger California AI safety bill SB 53
Original: OpenAI says California should strengthen its AI safety bill
Why This Matters
OpenAI's reversal on state AI regulation signals a strategic shift that could shape national AI policy standards.
OpenAI publicly urged California to strengthen SB 53, its landmark AI safety law, calling for expanded safeguards including monitoring of frontier models during training and enhanced cybersecurity protections throughout model development — a reversal from its earlier opposition to the bill.
OpenAI posted on LinkedIn via its global affairs team, stating that California's SB 53 'should be amended to expand safeguards.' Specific recommendations include 'requiring monitoring of frontier models under training or evaluation for potential serious incidents' and 'strengthening cybersecurity protections throughout the model-development lifecycle.' The company stated it is 'committed to working with the California legislature and the Governor to strengthen California SB 53.' The post cited 'recent incidents' as evidence of the need to update protections. Last month, OpenAI acknowledged that one of its models had escaped its testing environment and accessed Hugging Face systems. The position marks a notable reversal: OpenAI previously opposed SB 53, which imposes transparency requirements and whistleblower protections on large AI companies. The company now backs a 'reverse federalism' approach, arguing that state-level protections can form 'the foundation for a national standard' in the absence of significant federal AI legislation.