Colorado Hits Reset: How SB 189 Repealed the Nation's First AI Act Before It Ever Took Effect

2026-06-28 · Source: Global Privacy Laws & Compliance Frameworks | ComplianceHub.Wiki · Field: Legal & Regulatory — Compliance & Risk Management, Regulatory Affairs & Government Relations · Depth: Intermediate, long

Summary

Colorado's Governor Jared Polis signed SB 189 on May 14, 2026, repealing and replacing the state's original comprehensive AI law, the Colorado Artificial Intelligence Act (CAIA, SB24-205), roughly six weeks before its June 30, 2026, effective date. The CAIA, enacted in May 2024, was the first U.S. state-level cross-sector AI law, modeled on the EU's risk-based approach with duties for developers and deployers of "high-risk" AI systems. SB 189 discards these defining features, eliminating the duty of reasonable care, risk-management programs, and impact assessments. Instead, the new law, effective January 1, 2027, establishes a lighter, transparency-oriented regime focused on pre-use consumer notice, 30-day adverse-outcome explanations, and meaningful human review for "automated decision-making technology" (ADMT). Enforcement is exclusively by the Colorado Attorney General, with a 60-day cure period and no private right of action.

Key takeaway

For Directors of AI/ML overseeing multistate deployments, Colorado's repeal of its comprehensive AI Act underscores the volatility of state-level regulation. You should prioritize building a single AI governance program to the highest standard you face globally, such as the EU AI Act or Connecticut's CART Act, rather than optimizing for each jurisdiction. This approach ensures resilience against legislative changes and maintains audit-ready compliance, even as specific state requirements shift.

Key insights

Colorado's AI regulatory reversal signals fragility in comprehensive state-level AI laws, shifting to a lighter, transparency-focused model.

Principles

• First-of-its-kind laws with long effective dates are unstable drafts.
• Comprehensive AI regulation faces significant political and economic durability challenges.
• Multistate AI governance requires designing to the highest applicable standard.

Method

SB 189 requires pre-use consumer notice, 30-day adverse-outcome explanations, and meaningful human review for automated decision-making technology (ADMT), enforced by the Attorney General with a 60-day cure period.

In practice

• Inventory ADMT systems impacting Colorado individuals.
• Build pre-use consumer notices for ADMT deployment.
• Establish a 30-day adverse-outcome explanation process.

Topics

• Colorado AI Act
• SB 189
• AI Regulation
• Automated Decision-Making Technology
• EU AI Act
• Algorithmic Discrimination
• Compliance Strategy

Best for: CTO, VP of Engineering/Data, Executive, Legal Professional, Policy Maker, Director of AI/ML

Related on AIssential

• Colorado’s new AI governance law — AI governance laws are dynamic, subject to rapid legislative amendment and legal challenges.
• Connecticut's CART Act: Inside SB5, One of the Nation's Most Comprehensive State AI Laws — Connecticut's CART Act establishes a broad, multi-faceted AI regulatory framework emphasizing transparency and supply-chain responsibility over algorithmic discrimination.
• The Attribution Gap: Why Every AI Regulation Leads Back to Identity and Authorization — Enterprises face significant legal and regulatory liability for AI agent actions without verifiable identity and authorization trails.
• Speed vs. Safety: Inside America’s “One Rule” AI Order and China’s Rise as the World’s AI… — The US prioritizes AI speed via deregulation, challenging state laws, while China builds comprehensive, state-controlled AI governance.
• State AI Law Is the Only AI Law. Everywhere It's Crumbling. — State-level AI regulation efforts are broadly failing, resulting in minimal oversight.
• Colorado AI Act: DOJ Steps In As X.AI Suit Pauses — The Colorado AI Act's enforcement is paused due to an X.AI lawsuit and DOJ intervention, highlighting regulatory challenges.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Global Privacy Laws & Compliance Frameworks | ComplianceHub.Wiki.