TLDR: Across the United States, states are actively legislating to regulate the use of Artificial Intelligence (AI) in healthcare, particularly focusing on mental health services and prior authorization processes. Illinois has banned AI for therapeutic decision-making without licensed oversight, while California requires human review for AI-driven coverage denials. Over 250 AI-related bills have been introduced nationwide in 2025, reflecting growing concerns over patient safety, job preservation for healthcare professionals, and the potential for biased or harmful AI applications. Lawsuits against major insurers highlight the scrutiny on AI’s role in denying care.
Legislative bodies across the United States are increasingly focusing on the regulation of Artificial Intelligence (AI) within the healthcare sector, with a particular emphasis on mental health services and prior authorization procedures. This surge in legislative activity reflects a growing concern over patient safety, the integrity of clinical decision-making, and the potential impact of AI on healthcare professionals.
Illinois has emerged as a frontrunner in this regulatory push. Earlier this month, Governor JB Pritzker signed the Wellness and Oversight for Psychological Resources Act into law. This landmark legislation prohibits the use of AI for mental health or therapeutic decision-making without direct oversight from licensed clinicians. The act specifically bans individuals, corporations, and other entities from providing, advertising, or offering psychotherapy services to the public unless conducted by licensed professionals. Violations of this act carry significant civil penalties, up to $10,000. While restricting autonomous AI in therapy, the law does permit AI for administrative and supplementary support for behavioral health professionals. The stated intent behind this legislation is to safeguard patients from unregulated AI products, preserve jobs for qualified behavioral health providers, and protect children from the potential dangers associated with AI chatbots in mental health. Mario Treto Jr., secretary of the Illinois Department of Financial and Professional Regulation, emphasized this, stating, “The people of Illinois deserve quality health care from real, qualified professionals and not computer programs that pull information from all corners of the internet to generate responses that harm patients.”
Beyond mental health, states are also targeting AI’s role in prior authorization. In January 2025, a California law took effect, prohibiting payers from making coverage decisions based solely on artificial intelligence algorithms. This law mandates that any denial, delay, or modification of care must be reviewed by a physician or other qualified healthcare provider. Maryland has also introduced requirements for patient-specific datasets in AI-driven decisions and mandates reporting when AI contributes to adverse determinations. Other states, including Georgia, Texas, Rhode Island, Minnesota, and Florida, have introduced similar bills in 2025 aimed at restricting AI use in prior authorization decisions.
The broader legislative landscape indicates a significant trend. More than 250 bills targeting AI in healthcare have been proposed in state legislatures nationwide, with 22 enacted so far in 2025 across 40 states. These legislative efforts are driven by concerns over the potential for AI chatbots to engage in harmful conversations with vulnerable individuals and the risk of users unknowingly revealing personal information.
Legal challenges are also mounting against insurers regarding their use of AI. UnitedHealth Group and Humana are both facing lawsuits alleging the wrongful use of AI algorithms, specifically the naviHealth tool, to deny post-acute care for Medicare Advantage patients. While an Optum spokesperson stated that naviHealth is used as a guide and not for sole coverage decisions, Humana maintains a ‘human in the loop’ decision-making process when AI is utilized.
Federal oversight is also evolving. In 2024, the Centers for Medicare & Medicaid Services (CMS) issued guidance to Medicare Advantage plans, clarifying that while algorithms can support coverage decisions, they must comply with agency requirements and cannot be the sole basis for denials. A proposed Medicare Advantage AI Rule in April 2025 further aims to regulate AI use in prior authorizations. Dr. Mehmet Oz, a prominent figure, has even suggested that CMS should leverage AI programs to oversee insurers’ use of AI, identifying and preventing inappropriate denials in real-time.
In addition to AI restrictions, states are also pursuing broader prior authorization reforms. ‘Gold carding’ programs, which allow providers with high approval rates to bypass prior authorization for certain services, have been adopted or amended in states like Arkansas, Colorado, Louisiana, Texas, West Virginia, and Wyoming. Other reforms include removing prior authorization for specific drug classes, such as those for mental health or cancer, and extending prior authorizations for continuity of care.
Also Read:
- Human-Centered AI Market Surges Globally, Driven by Ethical Integration and User-Centric Solutions
- Artificial Intelligence Reshapes Cybersecurity Defenses Against Sophisticated Threats
This comprehensive legislative movement underscores a collective effort to ensure that the integration of AI into healthcare enhances, rather than compromises, patient care and professional standards.
