Colorado Rewrites Its AI Law Before It Takes Effect

Colorado Rewrites Its AI Law Before It Takes Effect

Colorado Rewrites Its AI Law Before It Takes Effect

https://www.forbes.com/sites/alonzomartinez/2026/05/15/colorado-rewrites-its-ai-law-before-it-takes-effect/

Publish Date: 2026-05-15 08:00:00

Source Domain: www.forbes.com

Colorado lawmakers voted to replace the state’s original AI law before it ever took effect, shifting the compliance focus for employers using automated decision-making tools.

getty

Colorado became the first state in the country to enact a sweeping artificial intelligence law aimed at regulating how automated systems influence consequential decisions, including employment. Now, before that law ever takes effect, state lawmakers have voted to replace it.

Senate Bill 26-189, which passed the Colorado Senate in a 34-1 concurrence vote following House amendments, now awaits Governor Jared Polis’ action. If enacted, the law would take effect January 1, 2027, repealing and replacing much of Colorado’s original Artificial Intelligence Act, Senate Bill 24-205, with a materially different framework focused less on governance programs and more on how automated decision-making technology is actually used in decisions about people.

For employers using AI or automated decision-making tools in hiring, promotion, compensation, or other employment workflows, this represents a meaningful shift in how Colorado approaches AI compliance.

A Law That Never Quite Settled

Colorado’s original AI law was ambitious by design. When lawmakers passed Senate Bill 24-205 in 2024, the state positioned itself at the front of a growing national effort to regulate artificial intelligence in high-stakes contexts. The law imposed broad obligations on both developers and deployers of “high-risk” AI systems, including formal risk management programs, impact assessments, public disclosures, and measures designed to mitigate algorithmic discrimination.

For employers, that framework raised immediate operational questions. What exactly qualified as a high-risk AI system? How should employers evaluate third-party hiring technologies they did not build? What level of technical visibility would be necessary to complete annual impact assessments or maintain defensible governance programs?

Even Governor…

Source