Illinois passes first frontier AI safety law with audit mandate

Illinois has stepped into the gap left by Washington, passing a frontier AI safety bill that could become the country’s most forceful state-level test of how to regulate the most powerful models. SB 315 cleared the Illinois House 110-0 and the Senate 52-5, and Gov. JB Pritzker said he intended to sign it, putting Springfield at the center of a fight over whether states or the federal government will set the rules for advanced AI.

The measure would require large frontier AI developers to create, publish and annually update a frontier AI framework that addresses catastrophic-risk assessment, mitigations, cybersecurity, internal governance, third-party evaluations and internal-use risks. Before deploying new or substantially modified models, companies would have to file transparency reports. They would also face annual independent third-party audits, a requirement that is being treated as a first for any AI legislation in the United States.

SB 315 also adds whistleblower protections and reporting channels for company employees, a detail aimed at surfacing internal safety concerns before they become public failures. If enacted, the law is expected to take effect Jan. 1 and would be administered by the Illinois Emergency Management Agency and Office of Homeland Security, in consultation with the Illinois attorney general.

The bill’s backers argue that Illinois is acting because Congress has not delivered a single national standard, especially after reports that President Donald Trump’s federal AI vetting plan was canceled shortly before the vote. That has left states to fill the vacuum, and Illinois lawmakers are now moving a broader package of AI bills in response. The strategy mirrors earlier moves in California and New York, but with a sharper edge: California signed SB 53, the Transparency in Frontier Artificial Intelligence Act, on Sept. 29, 2025, and New York signed the RAISE Act on Dec. 22, 2025. Illinois goes further by mandating audits, not just transparency.

That extra step helps explain why OpenAI and Anthropic backed the bill throughout the process. The companies are already building safety systems, and the Illinois framework gives them clear, enforceable rules rather than open-ended uncertainty. Supporters also see a chance for the law to become a de facto national standard if other states copy it, especially as lawmakers in Sacramento and Albany have already shown that frontier AI regulation can move through state capitals faster than Washington.