Colorado Workplace Compliance
News & Resources
Business moves differently in Colorado. From Denver’s fast-growing metro area to mountain roads, remote job sites, and rapidly changing weather conditions across the state, employers need to be ready for anything. To succeed in the Centennial State, businesses need resilience, adaptability, and a strong commitment to keeping their people safe.
That commitment includes navigating Colorado’s evolving workforce laws and regulations. The state has built a reputation for progressive labor protections and active enforcement around employee safety, wage transparency, paid leave, harassment prevention, and more. For employers, falling out of compliance can result in costly fines, litigation, and reputational risk. Learn what you need to know to protect your employees and stay compliant with Colorado law.
Stay on top of safety and compliance the right way with this Colorado-specific information, but be sure to seek legal counsel when evaluating how these regulations may directly impact your business. Wherever available, KPA products are updated with the latest government notices and posters for employers.
Colorado HR News
Who: Colorado employers
When: Effective January 1, 2027
Colorado Governor Jared Polis signed SB 26-189 on May 14, 2026, which repeals and replaces the Colorado AI Act (SB 24-205) just weeks before the original statute’s June 30, 2026, effective date. The new law takes effect January 1, 2027, making Colorado the first state to enact a comprehensive AI regulatory framework, though lawmakers substantially revised it from its original form.
The repealed Colorado AI Act imposed broad obligations on AI developers and deployers, including mandatory bias audits; risk impact assessments; extensive disclosure requirements; and an affirmative duty to prevent algorithmic discrimination. The new law eliminates those requirements in favor of a more targeted, transparency-based framework focused on automated decision-making technology used to materially influence consequential decisions.
A federal court issued a temporary restraining order blocking enforcement of the original Colorado AI Act on April 27, 2026, following a lawsuit by xAI (owned by Elon Musk) and a complaint filed by the U.S. Department of Justice. The passage of SB 26-189 supersedes the litigation’s immediate practical impact by replacing the law entirely.
Key Definitions
Automated Decision-Making Technology (ADMT): Any technology that processes personal data and uses computation to generate output used to make or assist a decision concerning an individual. ADMT excludes common infrastructure and security tools, tools used solely to summarize information for human review, and consumer-facing systems not intended for use in consequential decisions.
Consequential Decision: A decision affecting an individual’s access to or eligibility for education, employment, housing, financial or lending services, insurance, healthcare services, or essential government services. Low-stakes or routine processes—such as routine scheduling; administrative routing; customer service triage; workflow management; and advertising and marketing—are explicitly excluded, as are cybersecurity, anti-fraud, and sanctions compliance tools.
Covered ADMT: ADMT used to materially influence a consequential decision, meaning the ADMT output is a non-trivial factor that actually affects the outcome of that decision.
Consumer: Includes employees and job applicants who are Colorado residents.
Obligations Under the New Law
Developers must provide deployers with documentation covering the ADMT’s intended uses, categories of training data, known limitations, and instructions for appropriate use and human review. Developers must also notify deployers of material updates and retain records for at least three years.
Deployers must provide clear and conspicuous notice to consumers before using a covered ADMT in a consequential decision and include instructions on how to obtain additional information. Within 30 days of an adverse outcome, deployers must provide a plain-language description of the decision and the role of the ADMT, explain the consumer’s rights—including the right to access and correct inaccurate personal data—and offer an opportunity for meaningful human review and reconsideration to the extent commercially reasonable. Deployers must also retain compliance records for at least three years.
Risk management programs; impact assessments; ongoing monitoring requirements; and an affirmative duty of care do not apply under the new law—a significant reduction from the prior framework.
Liability and Enforcement
The law splits liability between developers and deployers based on their relative fault. Developers are not liable for downstream uses outside their documented intended uses; deployers remain responsible for their own actions and misuse. Contractual indemnification provisions that shift a party’s liability for its own ADMT-related discriminatory conduct are void because they are against public policy.
The Colorado Attorney General has exclusive enforcement authority. Violations constitute unfair and deceptive trade practices. Prior to initiating enforcement, the Attorney General must provide a 60-day notice and opportunity to cure (where a cure is deemed possible). There is no private right of action. The Attorney General must create implementation regulations before January 1, 2027.
How:
- Identify every AI tool or system your organization uses that could qualify as covered ADMT—particularly any technology involved in hiring, screening, performance evaluation, compensation, or benefits decisions affecting Colorado workers.
- Engage your AI vendors now to obtain documentation on intended uses, training data categories, known limitations, and human review guidance, as developers must provide this information under the new law.
- Create a clear and conspicuous notice process to inform employees, applicants, and other individuals when you are using a covered ADMT in a consequential decision that affects them.
- Develop a post–adverse outcome disclosure workflow to deliver plain-language explanations to affected individuals within the 30-day window required by law.
- Update policies for compliance and ensure that developers and deployers keep related compliance records for three years.
- Monitor the Colorado Attorney General’s implementing regulations, expected before January 1, 2027, for additional specificity on disclosure content, format, and consumer rights procedures.
Additional Resources:
Who: Colorado employers
When: Effective date was originally June 30, 2026; proposed extension to January 1, 2027
On March 17, 2024, Colorado Governor Jared Polis signed the Colorado Artificial Intelligence Act (SB 24- 205), which was originally set to go into effect on February 1, 2026. The law imposes obligations on AI developers and the Colorado employers that use those tools. Requirements include bias audits, risk impact assessments, and extensive disclosures. The law is intended to help prevent AI discrimination in employment decisions and interactions with consumers.
In August 2025, Governor Polis signed the “Increase Transparency for Algorithmic Systems” law (SB 25 B- 004), which delays the effective date of the new AI law from February 1, 2026, to June 30, 2026.
Under the law, employers may not use AI to discriminate against employees through algorithmic discrimination tied to protected classifications. Companies that use or develop artificial intelligence will need to determine whether it impacts:
- Education;
- Employment;
- Government services;
- Healthcare services;
- Financial services;
- Insurance;
- Housing; or
- Legal advice.
Employers with fewer than 50 full-time employees are exempt, but they must still follow the consumer notice and disclosure requirements.
On March 17, 2026, Colorado legislators published a proposed bill titled, “Automated Decision-Making Technology in Consequential Decisions.” That bill is intended to replace the current AI bill and would delay the effective date from June 30, 2026, to January 1, 2027. The new AI bill would remove the bias audit requirements and broad risk impact assessment mandates and replace them with a simpler, more transparent framework. The proposed bill was written by technology industry representatives, consumer advocates, and business groups and was endorsed by Governor Polis. The new framework addresses:
- Automated decision-making technology that influences a consequential decision;
- The exclusion of common AI tools as long as no consequential decisions are involved;
- Notice to applicants and employees when using AI in hiring or employment decisions;
- Disclosure after AI-assisted decisions result in adverse outcomes;
- The right to request a review of a decision by a human if a person receives an adverse AI-assisted decision; and
- Shared liability between developers and employers.
There is no private right of action, and the Colorado Attorney General will have exclusive enforcement authority.
How:
- Take an inventory of and perform bias audits on all of the AI tools you use.
- Create a risk management framework.
- Notify consumers of their rights under the law.
- Continuously update contracts to comply with the law.
- Consult with vendors regarding documentation obligations.
- Track federal developments on the use of AI tools.
- Continue to monitor state-level legal challenges and legislation.
Additional Resources:
SB 24-205 Colorado Artificial Intelligence Act (Delayed Until February 1, 2026)
SB25B-004 Increase Transparency for Algorithmic Systems Act (Delays Implementation to June 30, 2026)
Who: Colorado employers
When: Effective immediately
On April 27, 2026, a federal district court in Colorado ordered the state not to enforce a law governing consumer protections in interactions with artificial intelligence systems (SB-24-205), which was supposed to go into effect on February 1, 2026, but was delayed until June 30, 2026. On March 17, 2026, Colorado legislators published a proposed bill titled, “Automated Decision-Making Technology in Consequential Decisions.” That bill is intended to replace the current AI bill and would delay the effective date from June 30, 2026, to January 1, 2027.
The federal court decision temporarily blocks the state from taking any action to enforce the AI law and protects employers from penalties or investigations while the court decides whether to permanently block the law. The temporary block is in response to a lawsuit filed by xAI, which is owned by Elon Musk, and a complaint filed by the Department of Justice (DOJ).
xAI asserts that the AI law violates the First Amendment because building an AI model is a form of protected speech and compelling redesign of those models to avoid disparate outcomes violates the First Amendment right. They also assert that the law is “unconstitutionally vague” and impedes interstate commerce with its requirements.
The Department of Justice argues that the law violates the Equal Protection Clause of the Fourteenth Amendment by forcing AI models to make decisions based on race, sex, religion, and other protected characteristics. The DOJ also argues that the law requires AI models to advance diversity in a manner that has been ruled to be unconstitutional.
How:
- Consider conducting a bias audit on your your AI tools.
- Communicate with your AI vendors to determine whether they have conducted risk assessments on the AI tools they have deployed on your behalf.
- Continue to monitor Colorado and federal legislation regarding use of AI tools.
Additional Resources:
SB 24-205 Colorado Artificial Intelligence Act
AI LLC and United States of America v. Philip J. Weiser
Executive Order 14365 Ensuring a National Policy Framework for Artificial Intelligence
