Colorado AI Act

Overview

The Colorado Artificial Intelligence Act (SB 21-169) establishes comprehensive requirements for developers and deployers of high-risk artificial intelligence systems. Unlike NYC's LL144, which focuses specifically on automated employment decision tools, Colorado's law covers a broader range of "consequential decisions" where AI is used.

The law places obligations on both developers (companies that create AI systems) and deployers (companies that use AI systems to make decisions about consumers). For employers using AI in hiring, you are likely a deployer under this law.

Scope: What Counts as High-Risk AI?

The Colorado AI Act applies to AI systems that make or substantially assist in making "consequential decisions." In the employment context, this includes decisions about:

Hiring and recruitment
Termination of employment
Compensation and benefits
Promotion decisions
Job assignments or tasks
Performance monitoring and evaluation

If your organization uses any AI tool that influences these decisions—including resume screeners, candidate scoring systems, interview analysis tools, or performance management AI—you are likely subject to this law.

Key Requirements for Deployers

1. Risk Management Policy

Deployers must implement a risk management policy and program that:

Identifies and documents the intended uses of each high-risk AI system
Analyzes potential algorithmic discrimination risks
Implements reasonable measures to mitigate identified risks

2. Impact Assessment

Before deploying a high-risk AI system, you must conduct an impact assessment that documents:

The purpose and intended use cases of the system
Analysis of potential discrimination risks by protected class
Data used by the system and its sources
Metrics used to evaluate system performance
Steps taken to mitigate algorithmic discrimination

3. Consumer Disclosure

You must provide clear notice to consumers (including job applicants) that includes:

That a high-risk AI system is being used in the decision
A description of what the AI system does and how it affects the decision
Information about how to request human review (if the decision is adverse)
How to correct inaccurate data used by the system

Relationship to NYC Local Law 144

If you're already compliant with NYC LL144, you have a strong foundation for Colorado compliance—but there are important differences:

Aspect	NYC LL144	Colorado AI Act
Focus	Automated Employment Decision Tools only	All high-risk AI in consequential decisions
Audit Required	Yes, annual independent audit	Impact assessment (can be internal)
Public Posting	Yes, summary results	No public posting required
Consumer Disclosure	Notice before use	Detailed disclosure + appeal process
Risk Management	Not required	Formal policy and program required

How Paritas Helps

A Paritas bias audit addresses key Colorado AI Act requirements:

Impact Assessment Documentation: Our audit report provides the statistical analysis of algorithmic discrimination risk required for your impact assessment.
Risk Identification: Our three-tier classification (PASS/MONITOR/FLAG) identifies which demographic categories show potential discrimination risk.
Mitigation Recommendations: Our prioritized remediation recommendations provide documented steps you're taking to address identified risks.
Multi-Jurisdiction Coverage: Every Professional and Enterprise audit maps findings against Colorado requirements alongside NYC LL144, Illinois AIVAA, and EEOC guidelines.