Illinois Governor JB Pritzker signed the Artificial Intelligence Safety Measures Act into law on Monday, establishing the nation's first independent safety audit requirement for major artificial intelligence developers.
The law targets developers with more than $500 million in annual gross revenue and massive computing power.
They must publish transparency frameworks and report critical safety incidents within 72 hours starting January 1, 2028.
Attorney General Kwame Raoul received authority to fine violating companies up to $1 million for initial infractions and up to $3 million for subsequent violations.
The regulatory framework defines catastrophic events as incidents resulting in at least 50 deaths or causing a minimum of $1 billion in total damages.
State Intervention Due to Federal Inaction
"AI is the most significant technological innovation and development of the modern age," said Governor Pritzker.
"When managed properly, it can foster tremendous growth, productivity and innovation across the economy and vastly improve our quality of life.
But with that transformative potential comes catastrophic risk, much of which isn't fully understood yet."
Pritzker noted that state intervention became necessary due to federal inaction on comprehensive technology oversight.
"Where the federal government is unwilling to step up, states must venture once more unto the breach," said Governor Pritzker.
"History has shown time and again that thoughtful regulation is a requirement for sustainable innovation when our economic and national security is at stake."
State lawmakers modeled the transparency rules after prior legislative efforts enacted in New York and California to address emerging technologies.
"They have the potential for many good things, but due to their massive computing power, they also could be causing catastrophic events, such as cyberattacks, the creation or release of certain kinds of weapons or the potential of evading control of the developer or user amongst other possible threats," said AG Raoul.
The measure also instituted whistleblower protections, allowing tech employees to securely flag public health or safety hazards to state officials.
"If we got social media wrong, and we did, we cannot afford to get AI wrong at an even greater scale," said Democratic state Sen.
Mary Edly-Allen.
The legislative initiative gained bipartisan momentum during the spring legislative session before passing the House unanimously with a 110-0 vote.
"No, absolutely. I never have used it," said Ramirez.
"...
This whole move fast and break things is not acceptable innovation strategy when we know that the things that end up being broken are not just a computer, a laptop, a cell phone.
It’s people’s lives, their rights and their privacy."
Leading tech firms OpenAI and Anthropic expressed support for the safety baseline throughout the state legislative process.
"Takes the safety practices leading labs already follow voluntarily — publishing a safety framework, transparent reporting, protecting whistleblowers — and helps establish a baseline that every leading AI developer is expected to meet," according to Anthropic.
Democratic House Speaker Emanuel "Chris" Welch argued that tech development oversight shouldn't remain entirely hands-off.
"Tech bros whose culture of 'move fast and break things' is moving faster and faster and breaking more and more as it goes," said Welch.
Meanwhile, members of Congress remained divided on the practical application of AI within legislative offices.
"I’ve used some ChatGPT. I mostly end up using AI involuntarily, because when I do a Google search on something, the AI answer comes up first," said U.
S. Sen.
Tammy Duckworth.
Duckworth raised separate environmental concerns regarding the rapid expansion of more than 160 data centers across Illinois.
"I’m in charge of the AI, and I specifically know what I’m looking for as a consumer, and I do think AI has some real potential for breakthroughs in healthcare, in particular," said Duckworth.
"But again, like anything else, we must remain the master of the technology and not the other way around."
The senator urged local communities to demand strict guardrails regarding utility pricing and data center water consumption.
"Data centers have a lot of money, and they will pay to comply to whatever they need to, in order to be established.
But if you don’t put up the guardrails, they’ll do whatever they want," said Duckworth. "So, we need strong community input."
U. S.
Rep. Sean Casten noted that congressional staffers currently use tools like Microsoft Copilot, ChatGPT, and Google Gemini.
"We use it a lot in the office, and look, we’re at a point where I think it’s basically replaced web searches as the way to find information.
In that sense, I think if I had the job I had before I came to Congress, I would be using it a lot more," said Casten.
Casten voiced anxieties regarding potential automation-driven job displacement and tech company liability exemptions.
"The power of it to write code, to query huge data sets, to do all sorts of things that I did all the time in the private sector.
I just don’t have as much cause to do that in this job, so I think about it more from the perspective of a legislator," said Casten.
The representative warned against rushing technological deployments without proper long-term societal safeguards.
"I have real concerns, just politically," said Casten.
"What happens to a society with large numbers of unemployed young people, and particularly young people in professions that have historically skewed male?
Because unemployed young men is sort of a recipe for a political crisis."
Casten compared corporate lobbying strategies to historic international standoffs.
"Every time they come and present to us in Congress, they say, ‘We have to move really fast, because otherwise the Chinese will, and I can’t be held accountable for whatever this does, because the important thing is that we build it,’" said Casten.
"That’s fundamentally the same argument that led us to the Cold War."
In contrast, Representative Bill Foster adopted an optimistic stance regarding utilizing advanced digital modeling systems.
"I use it for computer coding and drafting legislation," said Foster.
Foster noted that advanced systems assist efficiently with structured technical tasks.
"One of the great things that I find (with) these advanced AI tools ... is I come up with some," said Foster.
