Illinois is showing the nation how to govern AI

With a stroke of his pen this week, Gov. JB Pritzker ensured that Illinois will help lead the way in crafting new guardrails for advanced AI models, giving state officials new tools to help ensure that the technology benefits local residents while protecting them from potential risks.

The opportunities are enormous. Advanced AI can help create new jobs across Illinois, open new doors for students from the suburbs of Chicago to the communities of southern Illinois, accelerate research and discovery at the state’s world-leading public and private universities, and improve health care for people throughout the state. Illinois is right to embrace that promise while putting strong, workable safeguards in place.

At OpenAI, we appreciated the opportunity to engage with Illinois policymakers as the bill took shape. The final law takes an important step forward: it requires OpenAI and the other largest developers of the most advanced AI systems to publish safety frameworks and transparency reports explaining how they assess and mitigate potentially catastrophic risks. Even more significantly, Illinois goes further than most state laws by requiring annual independent third-party audits to verify that companies are complying with the bill’s requirements.

Illinois’ law builds on approaches to advanced AI safety recently adopted in California and New York, but strengthens that emerging model with a more rigorous accountability requirement. This is reverse federalism in action: states moving toward a shared framework for advanced AI safety, one state at a time, while building the foundation and momentum for federal action.

What will this mean for the people of Illinois?

First, state officials will have a clear line of sight into the most serious safety incidents involving advanced AI models. Under the new law, developers must report critical safety incidents to the Illinois Emergency Management Agency and Office of Homeland Security and the Attorney General after determining that an incident has occurred. If an incident poses an imminent risk of death or serious physical injury, it must be quickly disclosed to an appropriate public authority.

Second, Illinois will not have to rely on companies simply asserting that they have complied with the obligations in the bill. The independent auditors required under Senate Bill 315 must have relevant technical expertise, must be free of financial conflicts and cannot be paid based on the outcome of their audits. Their findings must be shared with state officials, and a high-level summary of the audit findings must be made public.

Third, the law gives Illinois a continuing role as the technology evolves. Beginning in 2029, state officials must produce annual reports with aggregated information about critical safety incidents, relevant advanced AI safety developments and any recommended updates to the law. That creates an ongoing feedback loop, with Illinois learning — in real-time — from what is happening in the real world as the technology evolves.

The law also protects employees who are closest to the relevant risks and may be the first to spot serious dangers. They will have protected ways to report violations or significant public safety threats, including through an anonymous internal process and the Attorney General’s Workplace Rights Hotline.

This is a serious and balanced approach. It targets the largest developers and the highest-consequence risks. It supports innovation while requiring real accountability, including penalties of up to $1 million for a first violation and $3 million for subsequent violations. And it empowers Illinois officials to protect the public as advanced AI systems become more capable.

Thank you to the Illinois lawmakers who drafted and supported this legislation, and to Gov. Pritzker for signing it into law. Under their leadership, Illinois has written a playbook that other states — and ultimately policymakers in Congress and the White House — should follow so all Americans can benefit from the protection of a harmonized federal framework.

• Ann O’Leary is vice president of global policy at OpenAI.