All regulations / US Artificial intelligence
Colorado AI Act
Know the rule before it knows you.
The first US state law on high-risk AI and algorithmic discrimination.
What it is
The Colorado AI Act requires developers and deployers of high-risk AI systems (those making or substantially influencing consequential decisions in employment, lending, housing, insurance, education, health and legal services) to use reasonable care to avoid algorithmic discrimination, with documentation, impact assessments, notices and reporting duties.
This applies to you if
- You develop or deploy AI used for hiring, credit, insurance, housing, education or healthcare decisions affecting Colorado residents
- You sell AI tools to US customers in those sectors
What it asks of you
- Developers: documentation and disclosures to deployers; public statements on high-risk systems
- Deployers: risk management programme, annual impact assessments, notices to consumers, right to appeal and correct
- Reporting of discovered algorithmic discrimination to the Attorney General
What it costs to ignore
Enforcement by the Colorado Attorney General as unfair trade practices; rebuttable presumption of compliance for those following recognised frameworks such as NIST AI RMF or ISO/IEC 42001.
Key dates
Effective date postponed to 30 June 2026.
How we help
Not sure which apply to you? Send us three lines about what you sell and to whom, and we will tell you which rules matter and which do not.
Have a regulatory question?
Send a short description of your situation. We reply within one working day, usually with a view on scope and a fixed fee where possible.