Colorado's AI Legislation Faces Overhaul with Two Pending Bills: Here's the Crux of the Matter
Representative Lindstedt has introduced a bill, "Concerning Consumer Protections In Interactions With Artificial Intelligence Systems," which underscores the importance of disclosure at the point of interaction for chatbot-based or automated screening tools in employment opportunities.
The Lindstedt bill avoids imposing joint liability on employers, instead focusing on clear disclosures and enforcement through the Attorney General's office. Employers are required to disclose, at the outset of an interaction, that a candidate is engaging with an AI system. This disclosure must include the developer's name, the system's trade name, and contact information for the employer.
The bill sets an effective date of January 1, 2027, giving employers an additional year to adapt. Notably, the Lindstedt proposal would not require ongoing assessments or risk management frameworks, unlike the Rodriguez bill.
Employers that use high-risk AI must conduct annual impact assessments, monitor for discriminatory outcomes, and implement a risk management policy designed to mitigate algorithmic bias. They are also required to publish a statement describing how they use high-risk AI and the safeguards they have in place.
Violations of the Lindstedt bill would constitute a deceptive trade practice, and affected individuals could also pursue claims under the Colorado Anti-Discrimination Act if the AI interaction results in unlawful bias. Enforcement authority rests exclusively with the Attorney General, who may bring actions under Colorado's consumer protection law.
Employers must notify individuals when high-risk AI is used in a consequential decision, provide reasons for adverse outcomes, offer a path for correction, and ensure that meaningful human review is available through an appeal process.
The Colorado AI Act, a comprehensive AI regulatory framework, is scheduled to take effect in February 2026. This Act applies to any AI system deemed "high-risk," which includes tools that affect an individual's access to employment, housing, education, credit, or insurance. Organizations that discover and cure violations and can demonstrate alignment with a recognized AI risk management framework, may assert an affirmative defense.
Meanwhile, Senator Robert Rodriguez has introduced a bill that would repeal most of the existing Colorado AI Act and replace it with a new model focused on transparency and individual rights. The Colorado AI Act includes a narrow carve-out for small businesses that meet specific criteria.
The Lindstedt bill still signals that employers cannot ignore AI compliance. They would need to carefully track where and how interactive systems are used and ensure that required disclosures are delivered clearly and consistently. This development is a significant step towards ensuring transparency and fairness in the use of AI in employment interactions in Colorado.
