Colorado would overhaul its 2024 AI decision bias law to remove audit requirements and focus on transparency notices, under a legislative framework that Gov. Jared Polis (D) endorsed Tuesday.
Polis said the proposal had unanimous support from a group of technology industry and consumer advocates convened last year to study the legislation and make recommendations.
Businesses that use artificial intelligence tools in hiring and recruiting have closely watched to see when and how lawmakers might revise the statute, the most sweeping AI bias regulation of its kind when first enacted. President Donald Trump is seeking to discourage states from heavy regulation of AI technologies, most recently in a December executive order that called for limiting their access to federal funds and challenging specific state laws in court.
The state legislature punted on revising the law during a special session in 2025, opting to delay its effective date to June 30, 2026. If Polis’ revisions are enacted, the effective date would be pushed back to Jan. 1, 2027.
The governor’s proposal would erase the detailed requirements for bias audits, but leaves in place a version of the law’s transparency requirements. Employers, for example, would have to notify job applicants up front if they’re using AI tools to assist with hiring decisions.
Job applicants who are rejected would be entitled to further notice about the kinds of data the AI tools factored into their recommendation, a chance to correct inaccurate personal information, and the ability to request human review of the decision “to the extent commercially reasonable,” the framework said.
Similar requirements would apply to organizations using automated tools to assist with housing, credit, education, and other “consequential decisions,” it said.
To contact the reporter on this story:
To contact the editor responsible for this story: