The bill mirrors California\u2019s SB-53<\/a> and New York\u2019s Responsible AI Safety and Education Act<\/a>, which were each signed in late 2025. It establishes new reporting standards for the possibility that the AI model could be used for large-scale harms, such as by providing users assistance in creating a chemical, biological or nuclear weapon or committing cyber-attacks.<\/p>\n Senate sponsor Sen. Mary Edly-Allen, D-Libertyville, said there is an urgency for states to protect against those potential harms.<\/p>\n \u201cWe are not willing to wait for Congress to act,\u201d Edly-Allen said. \u201cThere\u2019s an old saying: Give a man a fish, he eats for a day. Teach him to fish, he eats for a lifetime. Teach AI to fish, though, and it might just empty the whole river trying to figure out how.\u201d<\/p>\n Though the three states only account for roughly 20% of the national populus, lawmakers estimate that they represent roughly 40% of the U.S. AI market, effectively creating a de facto national standard.<\/p>\n Sen. Mary Edly-Allen, D-Libertyville, speaks at a bill signing event on July 6, 2026, for the Artificial Intelligence Safety Measures Act she sponsored in the state Senate. (Capitol News Illinois photo by Maggie Dougherty) The new law requires model developers to publish an AI framework outlining how the developer identifies and assesses \u201ccatastrophic risk,\u201d defined as the likelihood that incidents that could cause death or serious injury to more than 50 people or more than $1 million in property damage.<\/p>\n Developers will also be required to report any incidents that could cause harm to the state within 72 hours of identifying the incident, or 24 hours if it poses an imminent risk for death or serious physical injury.<\/p>\n The bill\u2019s House sponsor, Rep. Daniel Didech, D-Buffalo Grove, said the harms the bill will regulate are not theoretical.<\/p>\n \u201cWe have already seen the first AI-inspired mass shooting. We have already seen AI systems utilized to attack a municipal water and drainage utility,\u201d Didech said.<\/p>\n He also alluded to the example of Anthropic\u2019s<\/a> Mythos model, which the company said was too powerful a cyberweapon<\/a> to release to the public. Anthropic supported Illinois\u2019 bill and had representatives present at the signing on Monday.<\/p>\n \u201cEvery transformative technology in our history, from automobiles to electricity to air travel, has delivered enormous benefits while carrying real risks, and in every case the government responded not by banning the technology and not by taking a hands-off approach, but by building safeguards, so everyday people can trust that these technologies are safe,\u201d Didech said.<\/p>\n Rep. Daniel Didech, D-Buffalo Grove, speaks at a bill signing event on July 6, 2026, for the Artificial Intelligence Safety Measures Act he sponsored in the House. (Capitol News Illinois photo by Maggie Dougherty)<\/p>\n Illinois\u2019 version, similar in most ways to the standards set in New York and California, adds a first-in-the-nation requirement for mandatory annual third-party audits; New York\u2019s version only required a single independent audit at the time when developers became large enough to qualify under the law.<\/p>\n During debate in the General Assembly, the third-party audit provision was a point of contention for some industry stakeholders, including TechNet, a coalition of tech executives across the industry.<\/p>\n \u201cWe remain concerned that Illinois would effectively be requiring private actors to make highly subjective determinations requiring AI safety compliance without established national standards, certifications, or clear regulatory guardrails,\u201d TechNet representative Ninia Linero said in committee May 20.<\/p>\n Read more: Illinois lawmakers pass landmark AI accountability bill<\/a> | Bill regulating powerful AI models advances as advocates say it\u2019s only the first step<\/a><\/p>\n OpenAI and Anthropic both supported the bill on its path through the Illinois General Assembly, and it passed with broad bipartisan support in both chambers; only five<\/a> Republican senators voted against it, and it passed unanimously<\/a> in the House.<\/p>\n Though the large developers pushed for a federal framework rather than what they were concerned would be an inconsistent patchwork of state regulations, Caitlin Niedermeyer of OpenAI\u2019s Global Affairs told the Senate\u2019s AI and Social Media committee in April that OpenAI was open to a coordinated state-driven approach.<\/p>\n \u201cWhile we have been very clear that the federal government remains well-positioned to lead on frontier safety because it has the resources, expertise and institutions, we also strongly actually see a position for both Illinois but also California and New York to really lead in advancing aligned frameworks, which we believe can absolutely help create a de facto national direction of travel,\u201d Niedermeyer told the committee.<\/p>\n What\u2019s on the AI horizon?<\/p>\n Companies that violate it will be subject to civil penalties brought by the attorney general\u2019s office of up to $1 million for the first offense and up to $3 million for subsequent violations.<\/p>\n But lawmakers and advocates say they expect to continue working on the topic of AI in the future. For example, Didech identified medical care and education as likely frontiers needing further evaluation of AI\u2019s public safety risks.<\/p>\n Scott Wisor, policy director for Secure AI, was one of the advocates helping to shape Illinois\u2019 bill. Having more external evaluation of the risks the models pose, and make judgments on when they\u2019re ready for release, would be the next step for further transparency and accountability, he said.<\/p>\n \u201cRight now, the evaluation in this bill is, are you complying with your safety framework? Because suppose you had a safety framework, just like, \u2018We\u2019re going to do A, B, C, and D,\u2019 you do that, the evaluator confirms it, and yet it\u2019s still a risky thing to have out in the world,\u201d Wisor said.<\/p>\n \u201cSo, this is a huge step forward, but I think there\u2019s more we can do,\u201d Wisor said.<\/p>\n Illinois\u2019s law will take effect on Jan.\u00a0 1, 2028.<\/p>\n![\"Mary](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/07\/260706-AI-BILL-EDLY-ALLEN_MD.jpg\")
<\/a><\/p>\n
\nNew guardrails<\/p>\n![\"Daniel](\"https:\/\/www.europesays.com\/ai\/wp-content\/uploads\/2026\/07\/260706-AI-BILL-DIDECH_MD.jpg\")
<\/a><\/p>\n