Artificial Intelligence (AI) can greatly benefit manufacturers in the workplace. That said, it should be handled with care. States across the country are attempting to regulate the use of AI in various contexts, from political campaigns to social media, and the workplace appears to be next. Colorado recently enacted the first comprehensive AI legislation regarding the development and deployment of AI, the “Colorado AI Act” (the Act), and its impact will reach various workplaces nationwide. Here is what manufacturers should know about the Act:
- The Act, which takes effect on February 1, 2026, and will become a part of Colorado’s Consumer Protection Act, regulates “developers” and “deployers” of AI. However, manufacturers will likely fall under the latter category.
- The Act requires deployers to use “reasonable care” to avoid algorithmic discrimination from using “high-risk” AI systems. It defines “high-risk” AI systems as any system that “makes, or is a substantial factor in making, a consequential decision,” meaning “a decision that has a material legal or similarly significant effect,” including decisions related to employment. The Act defines a “substantial factor” as one that “assists in making a consequential decision,” “is capable of altering the outcome of a consequential decision,” and is “generated” by an AI system. Thus, AI systems used in human resources practices, such as screening candidates in the recruiting/hiring process, may be classified as “high-risk.”
- Manufacturers may be wondering how they can ensure that they use “reasonable care” when using these kinds of high-risk systems. Interestingly, the Act includes a rebuttable presumption that a deployer is using reasonable care if they:
- Implement a risk-management policy that is “reasonable,” as defined by the Act based on various considerations, and is regularly and systematically reviewed and updated.
- Complete annual impact assessments for high-risk AI systems. The Act outlines various elements that the risk assessment must include.
- Provide several notices required by the Act, including a notice to consumers that AI is being used “to make, or be a substantial factor in making, a consequential decision,” which must be provided to the consumer before that consequential decision has been made. For consumers who are adversely affected by an AI system (for example, denied an employment opportunity), notice must be provided that they have “an opportunity to correct any incorrect personal data” used by the AI system and “an opportunity to appeal an adverse consequential decision.” Manufacturers who employ fewer than 50 employees, do not use their own data to train AI systems, deploy the AI systems for their intended purpose, and make impact assessments available to consumers may be exempt from some of these notice requirements.
- Disclose to the Attorney General the discovery of algorithmic discrimination that the AI system has caused within 90 days of discovery.
- There currently is no private right of action under the Act, meaning the attorney general’s office has exclusive enforcement authority. The Act does outline some affirmative defenses that manufacturers may use if facing an enforcement action.
Manufacturers located in any state that are deploying AI as part of human resources practices, including candidate screening, talent management, performance management, and for other purposes, should consider developing or updating AI policies and frameworks to ensure compliance with this law and the other applicable laws, such as the New York City Automated Employment Decision Tools (AEDT) law. Other states are following suit, and the federal government is also focusing on this issue. We expect the number of states passing such laws to continue rising. In fact, the U.S. Department of Labor just recently released guidance on employers’ use of AI in the workplace, which contains suggested AI principles and foreshadows future federal legislation. Thus, if there is a time to prioritize AI governance – it is now!
