JOINT RELEASE: Legislation to Establish Guardrails for AI in Healthcare Signed Into Law

DENVER, CO – Yesterday, the Governor signed legislation sponsored by Senators Kyle Mullica, D-Thornton, Judy Amabile, D-Boulder, and Representatives Gretchen Rydin, D-Littleton, and Javier Mabrey, D-Denver, to ensure patients’ continued access to mental healthcare provided by a licensed human professional. 

“No AI-generated algorithm can replace the expertise, nuance, and connection that human healthcare professionals utilize to treat their patients,” Mullica said. “With this new law, we’re establishing necessary guardrails to ensure proper access to quality care for those who need it most.”

“Without input or oversight from a licensed professional, AI chatbots can be mistaken for legitimate therapy. This practice is dangerous for patients, which is why we passed this law to ensure Coloradans are protected and informed,” said Rydin. “This new law establishes reasonable protective measures on AI-use in mental and behavioral health care, including prohibiting the use of AI to independently provide therapy to a patient. Licensed professionals are still allowed to use AI for administrative purposes, but clinical treatment and psychotherapy must be administered by a licensed provider.”

“As policymakers, we cannot let chatbots, several of which are currently facing major lawsuits due to wrongful and horrifying deaths, replace certified mental health providers,” Amabile said. “Some AI models serve as bad actors claiming to offer low-cost care – but this new law puts guardrails in place to ensure patients receive the quality, human care they deserve.”

“Colorado patients deserve access to real, human-centered care,” said Mabrey. “This timely law sets up important guardrails for AI use in mental and behavioral health care. AI chatbots are biased, unlicensed tools and they should not be used for therapy and treatment recommendations without oversight and transparency. With this bill now law, we’re protecting patients while still allowing providers to take advantage of AI tools for administrative tasks.”

HB26-1195 sets standards in clinical settings, limiting the use of artificial intelligence (AI) to administrative tasks with oversight by a licensed professional. Providers must disclose the use of AI for supplementary support, such as recording or transcribing meetings. To ensure patients receive legitimate behavioral healthcare, this new law makes sure that psychotherapy is human-delivered by a licensed professional, such as a social worker, psychologist or addiction counselor. 

To protect consumers and ensure access to quality care, this legislation will prohibit AI chatbots from being marketed to patients as equivalent to a licensed psychotherapist or counselor. AI chatbots would also be barred from implying patient input is covered by commonplace confidentiality protections like HIPAA. 

In 2025, researchers at Stanford University recommended that Large Language Models (LLMs), which power AI chatbots, “should not replace therapists.” Additionally, researchers concluded that “LLMs express stigma toward those with mental health conditions and respond inappropriately to certain common (and critical) conditions.” 

Top AI companies, including OpenAI, Google, and Character.AI, are all facing lawsuits from families after AI chatbots recommended suicide to a person seeking behavioral health advice or support. Last year, parents of children who died by suicide testified before Congress, stating AI chatbots discouraged their teens from seeking support.