Colorado governor signs law delaying and narrowing state AI Act

Colorado Governor Jared Polis signed SB 26-189, substantially rewriting the state’s 2024 AI law less than two years after he signed the original. The 2024 statute, SB 24-205, had required companies to run risk-management programmes and impact assessments before deploying “high-risk” AI systems in decisions such as hiring, lending and housing. SB 189 replaced that framework with narrower disclosure obligations tied to “automated decision-making technology” used to make consequential decisions about consumers, and pushed the effective date — already delayed once from its original date — out to 1 January 2027.

The rewrite followed sustained pressure on the original law. A December 2025 executive order from President Trump had singled out SB 24-205 as “excessive State regulation,” and a large AI developer filed a federal lawsuit that April challenging the statute’s constitutionality, with the Department of Justice intervening in support of the challenge. Law firms tracking the bill described it less as an amendment than as a repeal and replacement, since it dropped the risk-based structure modelled on the EU’s approach in favour of transparency requirements closer to the disclosure-first model several other states had adopted.

The signing was one of several instances in mid-2026 of state legislatures narrowing or delaying binding AI-specific obligations passed a year or two earlier, in favour of lighter-touch rules that stopped short of mandating risk assessments or giving regulators power to block deployments outright.