Illinois has enacted Senate Bill 315, a landmark AI safety law establishing safety standards and transparency requirements for frontier AI developers in the state.

OpenAI and Anthropic supported the bill, which passed the Illinois House 110-0 and coordinated with sponsors to set accountability and baseline safety standards.

The law imposes penalties up to $3 million per repeat violation and assigns enforcement authority to the Illinois attorney general.

Catastrophic events under the measure are defined as incidents causing at least 50 deaths or $1 billion in damages.

Whistleblower protections are strengthened to shield employees who disclose substantial public safety threats or violations.

Audits must be conducted annually by independent third parties, with redacted reports published to show compliance.

The bill borrows elements from New York and California AI risk frameworks to shape its approach.

The act takes effect on January 1, 2027, after bipartisan passage in the General Assembly.

The law targets the most capable AI models using thresholds based on revenue of $500 million and specific computing metrics to establish a national standard amid federal inaction.

Developers must assess and mitigate catastrophic risks, undergo independent audits, report critical safety incidents within 72 hours, and publish a transparency framework detailing safety practices and incident response.

Legislation passed the Senate 52-5 and unanimously in the House, reflecting broad bipartisan support.