Colorado lawmakers have approved legislation that would repeal and replace the state’s pioneering Artificial Intelligence Act, marking a significant shift in how AI systems will be regulated. The proposed framework moves away from broad restrictions on “algorithmic discrimination” and instead centers on transparency and disclosure requirements for automated decision-making systems.
The measure, SB 26-189, passed on May 12 and would take effect in January 2027 if signed into law. The original Colorado AI Act, introduced in 2024 as one of the first comprehensive state-level AI laws in the U.S., had faced criticism from businesses and legal challenges over compliance burdens and constitutional concerns.
Under the revised framework, regulation will apply to “covered automated decision-making technology” used in consequential areas such as employment, healthcare, housing, education, financial services, and government benefits. Developers will be required to provide documentation on intended uses, training data categories, and system limitations, while deployers would need to notify individuals when AI materially influences consequential decisions.
The proposal also introduces consumer rights that would allow individuals to request access to and correction of personal data used in AI-assisted decisions and seek meaningful human review of AI-assisted outcomes where commercially feasible.
Enforcement authority would remain with the state attorney general, while consumers could still pursue legal action under existing consumer protection and anti-discrimination laws if AI systems lead to unlawful outcomes.