Illinois now has what several advocates are calling the toughest artificial intelligence safety law in the country. On July 6 Governor JB Pritzker signed Senate Bill 315, the Artificial Intelligence Safety Measures Act, which takes effect on January 1, 2027.
The law reaches the developers of the largest frontier models, the systems trained with enormous amounts of computing power and backed by companies with substantial revenue. Rather than police everyday uses of AI, it targets the handful of firms building the most capable general-purpose systems and asks them to show their work on safety.
What the law requires
Covered developers must publish their safety practices, file transparency reports before deploying a new or heavily modified frontier model, and report serious safety incidents to the state. According to NBC Chicago, companies will have 72 hours to disclose an incident that could cause harm, and 24 hours when there is an imminent risk of death or serious physical injury.
Illinois is the first state to require independent third-party audits of those safety practices, according to the Transparency Coalition, which helped shape the bill. That provision moves the state past the disclosure-only model of earlier laws and puts an outside examiner between a developer's claims and the public.
Penalties and support
Firms that fail to comply face civil penalties reported at $1 million for a first violation and $3 million for later ones, as detailed by the Chicago Sun-Times. The bill drew bipartisan backing in the legislature, and both Anthropic and OpenAI supported it as it moved through the statehouse.
The timing is worth noting. Washington has leaned toward a lighter, voluntary framework for frontier AI, and the European Union recently agreed to delay parts of its own AI Act. Illinois is pulling in the other direction, and because the largest developers sell nationally, a rulebook written in Springfield will reach well beyond the state line.
Commentarii · 0