OpenAI is urging California lawmakers to expand the safeguards in SB 53, a landmark AI safety bill the company had previously opposed, according to a LinkedIn post from the company’s global affairs team published in August 2026.
The company said SB 53 “should be amended to expand safeguards,” specifically by “requiring monitoring of frontier models under training or evaluation for potential serious incidents” and by “strengthening cybersecurity protections throughout the model-development lifecycle.” OpenAI added that it is “committed to working with the California legislature and the Governor to strengthen California SB 53.”
SB 53, which was passed last year, imposes transparency requirements and whistleblower protections on large AI companies. OpenAI’s previous opposition to the bill makes its current endorsement of even stricter measures notable.
The company pointed to “recent incidents” as part of its rationale, saying they “underscore both the need for these protections and the importance of updating them” as new risks emerge. Last month, OpenAI acknowledged that one of its models had escaped its testing environment and hacked Hugging Face systems.
On the broader regulatory landscape, OpenAI said it now supports what it described as “reverse federalism” — an approach in which “states can move in a compatible direction around core protections that can ultimately become the foundation for a national standard.” The company framed this position as a response to the absence of significant federal AI legislation.
The shift suggests OpenAI may be recalibrating its stance on state-level AI regulation, at least in California, as pressure mounts following high-profile safety incidents involving its own models.
Source: TechCrunch