OpenAI has reversed course on California's artificial intelligence safety legislation, now advocating for the state to strengthen SB 53 rather than oppose it. The company previously fought the bill but has shifted its position to support more rigorous safeguards.
The reversal marks a notable pivot in OpenAI's regulatory stance. SB 53, formally known as the AI Safety Bill, originally faced intense industry pushback from major AI developers who argued the proposal was too restrictive and could stifle innovation. OpenAI stood among those opponents. The company's decision to now champion a strengthened version reflects evolving calculations about regulation in the AI space.
OpenAI's new position aligns with a broader trend of large AI labs endorsing selective regulation. The company has previously advocated for federal AI oversight and participated in various policy discussions. By supporting California's safety bill, OpenAI signals that it believes some regulatory framework serves its long-term interests better than an unregulated environment that could invite aggressive federal intervention or fragmented state-level rules.
SB 53 focuses on establishing safety requirements for advanced AI systems. The bill targets large-scale models and requires developers to implement safeguards before deployment. California, as the nation's largest economy and home to most of the AI industry, holds outsized influence over regulatory standards. Tech companies often use California policy outcomes as templates that spread nationally and internationally.
OpenAI's endorsement of a stronger SB 53 likely reflects its confidence in its own safety infrastructure. The company has invested in safety research and red-teaming practices. Supporting stringent requirements effectively raises barriers for competitors with smaller safety teams or fewer resources. Regulation that aligns with OpenAI's existing practices becomes a competitive moat rather than a burden.
The timing connects to broader regulatory momentum. The Biden administration issued an executive order on AI governance in 2023, signaling federal appetite for oversight. Multiple states have introduced AI legislation. Rather than fight every rule, major AI companies increasingly prefer helping shape standards that they can meet while competitors struggle to comply.
California's legislature must weigh OpenAI's new advocacy against its previous opposition. The company's reversal could swing other fence-sitting legislators. Tech industry groups remain divided on how strong AI safety rules should be, and some startups still oppose strict requirements. OpenAI's position carries weight because the company commands significant resources and has built relationships with policymakers.
The shift also reflects OpenAI's maturation as a public-facing company. As the creator of ChatGPT, OpenAI operates under intense scrutiny from regulators, researchers, and the public. Supporting safety legislation helps the company project responsibility and trustworthiness. This matters as OpenAI pursues partnerships with governments and enterprises that demand strong governance records.
SB 53 remains in legislative flux. OpenAI's support for strengthening it will likely intensify industry debate. The company's move demonstrates that AI regulation is no longer a binary fight between "light-touch" and "heavy-handed" approaches. Instead, major players are negotiating over which specific standards become law, betting that favorable rules now prevent worse rules later.