Illinois Gov. J.B. Pritzker (D) signed the Artificial Intelligence Safety Measures Act last week, requiring the largest developers of frontier models to evaluate and mitigate catastrophic risks, report critical safety incidents within 72 hours, and submit to independent third-party audits. The law imposes fines of up to $3 million for repeat violations and aims to create a de-facto national standard for AI safety that extends beyond the developers to the companies that use these tools.
The act mirrors recent AI bills in California and New York, but it promises a broader reach. While those laws focus on developers, the Illinois measure is designed to influence how businesses across the country deploy high-risk AI systems, effectively setting a benchmark for safe use.
Safety requirements for frontier models
Under the new law, developers of the most advanced AI models must assess and address potential catastrophic risks before deployment. They are required to report safety incidents to the state within 72 hours of discovery. The act also mandates independent third-party audits to verify compliance, ensuring that safety measures are not merely self-reported.
The law targets the "biggest" companies developing frontier models, a category that will be defined by forthcoming regulations from the Illinois Department of Innovation and Technology. The aim is to capture the systems with the highest potential for widespread harm.
Penalties and enforcement
Repeat violations of the act can trigger penalties of up to $3 million. The enforcement framework is designed to push companies to adopt rigorous safety practices, with the threat of significant fines for repeated noncompliance. First-time violations may result in lower penalties, but the law gives regulators the authority to escalate fines for ongoing issues.
A multi-state standard
With Illinois joining California and New York, a patchwork of state AI safety laws is emerging. The Illinois law's focus on companies that use AI, not just developers, could set a de-facto national standard. Businesses operating nationwide may start aligning their practices with the strictest rules to avoid legal risk, making the Illinois act a template for responsible AI deployment.
Why this matters for government professionals
State-level AI regulation is accelerating, and government officials need a clear understanding of the technology's risks and governance models. The Illinois law, along with similar measures, will shape how agencies procure and deploy AI tools. Professionals who work on AI policy can deepen their expertise through resources like the AI Learning Path for Policy Makers.
Your membership also unlocks:
