Trump loses more control over AI regulation as Illinois passes landmark law

2026-05-28 · Source: AI - Ars Technica · Field: Legal & Regulatory — Regulatory Affairs & Government Relations, Compliance & Risk Management · Depth: Fundamental Awareness, short

Summary

Illinois lawmakers passed SB 315, the nation's strongest AI safety law, requiring major AI firms to submit public safety plans and annual reports based on independent, third-party safety testing of their frontier models. The bill, which Governor J.B. Pritzker intends to sign, mandates reporting critical safety incidents within 72 hours, or 24 hours for imminent risks of death or serious harm. It also provides whistleblower protections for employees reporting safety risks. Leading AI firms like OpenAI and Anthropic supported the bill, viewing its requirements as a baseline mirroring existing voluntary protocols and a potential strategy to avoid a patchwork of differing state laws. The law, effective January 1, 2027, aims to establish guardrails against catastrophic AI risks, with proponents like Rep. Daniel Didech and Sen. Mary Edly-Allen emphasizing its role in responsible innovation and as a potential roadmap for future federal regulation, despite concerns from groups like Chamber of Progress about exposing sensitive systems to "untested auditors."

Key takeaway

For policy makers considering AI governance, Illinois's SB 315 offers a model for state-level intervention when federal action lags. You should evaluate its framework of mandatory third-party safety testing, incident reporting, and whistleblower protections as a potential blueprint. This approach balances innovation with risk mitigation, creating a "roadmap for responsible innovation" that could inform future national standards and build public trust in AI development.

Key insights

Illinois's new AI safety law mandates third-party testing and incident reporting for frontier models, setting a national regulatory precedent.

Principles

• States are intervening due to federal inaction on AI regulation.
• Industry support for regulation can establish a baseline.
• Independent audits enhance transparency in AI development.

Method

Major AI firms must submit public safety plans and annual reports from independent, third-party safety testing, plus report critical incidents within 72 or 24 hours.

In practice

• Implement third-party safety testing for frontier AI models.
• Establish internal protocols for rapid incident reporting.
• Develop whistleblower protections for AI safety concerns.

Topics

• AI Regulation
• Frontier AI Models
• AI Safety Testing
• State Legislation
• Whistleblower Protections
• Independent Audits

Best for: CTO, VP of Engineering/Data, Director of AI/ML, Policy Maker, Legal Professional, Tech Journalist

Related on AIssential

• Illinois Lawmakers Just Passed America’s Strongest AI Safety Bill — The Illinois AI safety bill mandates third-party audits for frontier AI labs, setting a national precedent for regulatory oversight.
• AI #170: Lack of Executive Order — AI's rapid, broad advancement challenges regulation, human discernment, and societal structures.
• Spooked by Mythos, Trump suddenly realized AI safety testing might be good — Government AI oversight is shifting towards pre- and post-release safety evaluations, but faces challenges in defining standards and ensuring independence.
• A Roadmap for Federal AI Legislation: Protect People, Empower Builders, Win the Future — Effective AI governance requires a balanced federal framework that protects people while fostering innovation and competition.
• State AI Law Is the Only AI Law. Everywhere It's Crumbling. — State-level AI regulation efforts are broadly failing, resulting in minimal oversight.
• OpenAI Backs Bill That Would Limit Liability for AI-Enabled Mass Deaths or Financial Disasters — OpenAI seeks liability shields for frontier AI models causing critical harms, aiming for consistent national regulation.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI - Ars Technica.