This story was produced as part of the Colorado Capitol News Alliance. It first appeared on cpr.org.
A group appointed by Gov. Jared Polis to come up with a framework for implementing a state law regulating “consistent” decision-making by artificial intelligence systems has released a proposal on how to do it.
The AI Policy Working Group released its findings on Tuesday. A 2024 law aimed at creating regulations never went into effect, and the state legislature postponed its implementation date to June 2026 amid disagreements over how it would be designed to balance consumer protections with innovation and workability.
The working group’s recommendations still need to be signed into law at the state Capitol and would require the Legislature to approve this session.
“I am deeply grateful to the hardworking members of the Colorado AI Policy Working Group who unanimously agreed on AI policies to protect consumers and support innovation in our state,” Polis said in a written statement.
Colorado has been grappling with its AI policy since 2024, when lawmakers passed the nation’s first comprehensive law regulating how companies and governments use artificial intelligence to make important decisions about people’s lives. The law does not target deep fakes or fraud, but applies to the way AI is used to assess people on things like college applications, hiring, loans, access to healthcare or insurance. The goal, the bill’s sponsors said at the time, is to prevent discrimination.
The working group sought to tackle the thorniest issues of transparency and accountability. It recommends that an AI system developer should describe key parts of the system to the people who use it to make key decisions, the operators, on matters such as the intended use of the tool, the categories of data used to train the system, its limitations, and instructions on appropriate monitoring and ‘meaningful human review, where appropriate’.
The user of the AI system should also tell the public in plain language what role the AI system played in making a resulting decision. A business, government agency, or school that uses AI to make a resulting decision would also be required to “provide clear and conspicuous notice to consumers.”
When something goes wrong, liability is assigned to implementers and developers based on the role they played in what went wrong, the working group recommends. For example, existing anti-discrimination and consumer protection laws could answer questions such as: Did the operator use the system in the manner it was advertised, configured or contracted for?
The attorney general’s office would establish rules on disclosures an operator must provide to a consumer following an adverse outcome involving an AI system.
The Colorado Technology Association, which participated in the working group, said it voted to advance the framework based on “targeted revisions.”
“We look forward to seeing this progress reflected in upcoming legislation and continuing the dialogue on how we can protect consumers while allowing innovation to flourish,” said CTA President Brittany Morris Saunders.
Democratic Rep. Brianna Titone of Arvada was a sponsor of the 2024 bill. She said the recommendations are a good starting point, but it’s not clear whether a bill with these parameters can make it through the legislative process without significant changes.
“While voting members agreed, there were many caveats to their ‘yes’ votes. It is a meaningful step forward, but only if the proposed bill can stay on this trajectory,” Titone said.
Democratic Senate Majority Leader Robert Rodriguez, the lead sponsor of the 2024 bill, said he appreciates the task force’s work and looks forward to its recommendations.
“The devil is in the details. I have indicated to the governor’s office that my priority is to ensure that they let consumers know that they have access to the decision and the ability to correct it. Those are my core values,” Rodriguez said.
Rodriguez said he wants the law to have teeth.
