Illinois Governor JB Pritzker has signed the AI Safety Measures Act (SB 315) into law, establishing what advocates are calling the strongest state-level AI safety framework in the United States. The law, signed this week, makes Illinois the first state to require annual third-party safety audits for frontier AI developers.

What the Law Requires

The legislation targets the largest and most capable AI models — known as frontier models — and imposes several key requirements:

- Annual third-party audits: Developers like OpenAI and Anthropic must commission independent safety audits every year, a first-of-its-kind requirement in the nation. - Transparency obligations: Companies must disclose safety testing results, training data sources, and known limitations of their models. - Risk mitigation plans: Developers must file comprehensive plans for addressing identified risks, including bias, misuse, and systemic harms. - Whistleblower protections: Employees who report safety concerns are shielded from retaliation.

A State-Driven National Standard

Illinois joins California and New York in passing comprehensive AI legislation, but its third-party audit mandate goes further than any other state. Lawmakers noted that Illinois, California, and New York together account for roughly 40% of the U.S. economy, creating a de facto national standard through state-level action.

"This law sends a clear message: innovation and safety must go hand in hand," Pritzker said at the signing ceremony in Chicago. "We're not slowing down AI development — we're making sure it happens responsibly."

National Context

The Illinois law arrives amid a wave of state-level AI regulation across the U.S. This legislative season saw more than 30 AI-related bills advance in California alone, while New York passed a kids chatbot safety bill, an AI training data transparency act, and a ban on AI-assisted surveillance pricing before wrapping up its session on June 1.

Missouri and Rhode Island also enacted AI laws this year. Missouri passed SB 1019, which prohibits AI chatbots from offering therapy services without a licensed professional, imposing fines of up to $20,000 for violations. Rhode Island Governor Dan McKee signed three AI measures into law last month, including a therapy chatbot ban and a chatbot self-harm safety measure.

Industry Reaction

The law has drawn mixed reactions. Consumer advocacy groups praised the third-party audit requirement as a necessary check on rapidly advancing AI capabilities. Industry trade groups expressed concern about compliance costs and the potential for a patchwork of state regulations across the country.

With California legislators now on summer recess until August 3 and over 30 AI bills still pending in Sacramento, Illinois has effectively leapfrogged ahead — setting a precedent that could shape AI governance nationwide.

Sources