By LEE G. PETRO

As artificial intelligence (AI)‑driven tools proliferate across clinical care, diagnostics, and health insurance operations, the U.S. regulatory landscape is shifting rapidly in 2026. Absent a comprehensive federal AI statute, healthcare organizations face a patchwork of federal oversight (primarily through the FDA, CMS, and HHS) layered with a growing number of state AI health laws.

Federal Framework Shaping AI In Healthcare:

At the federal level, existing authorities and recent policy signals are converging to regulate AI in healthcare without yet creating a standalone “AI Act.” The most important federal AI actors are:

  • FDA: The FDA continues to emphasize premarket review for higher-risk systems (e.g., AI‑assisted radiology tools, ECG or pathology interpreters), while allowing greater flexibility for lower‑risk administrative or workflow tools. January 2026 guidance on clinical decision support software clarified that AI products directly influencing clinical judgment or patient management will likely require FDA review or clearance.
  • CMS and reimbursement: CMS continues to explore reimbursement pathways for AI‑augmented services through coding and coverage decisions in the 2026 Medicare Physician Fee Schedule, signaling that compliant, evidence‑based AI tools can be financially viable. CMS recognizes that many AI tools are iterative, continuously updated services rather than one‑time products, and is developing appropriate payment models.
  • HHS and the National AI policy: The Department of Health and Human Services issued a Strategic Plan for the Use of Artificial Intelligence in Health, Human Services, and Public Health in January 2025, establishing a framework spanning safety, equity, infrastructure, workforce development, and modernizing care delivery. HHS reported significant year‑over‑year growth in the agency’s AI use cases inventory, reflecting a push for operational and clinical automation even as regulatory guardrails are still being developed and refined.
  • White House Executive Order: A December 2025 White House executive order on AI established a national policy framework directing federal agencies to challenge state AI rules viewed as overly burdensome and to work toward a “minimally burdensome” national standard. The first reports mandated by the Executive Order were to be prepared by the Federal Trade Commission and the Commerce Department and released on March 11, 2026. As of the date of this publication, the reports have not been released.

State‑Level AI‑Health Laws In 2026:

In the absence of a comprehensive federal AI statute, states have taken the lead on numerous healthcare‑specific AI issues. By the end of 2025, 21 states had enacted 33 healthcare‑AI‑related laws, with lawmakers in 47 states introducing more than 250 bills regulating AI in healthcare during 2025:

  • Prohibiting AI‑only clinical decisions. Several states require that licensed clinicians retain final decision‑making authority over diagnosis, treatment, and certain utilization‑review decisions. For example, Illinois law (HB 1806) prohibits AI from making independent therapeutic decisions or generating treatment plans in behavioral‑health contexts without review and approval by a licensed professional. Texas and other states have enacted similar rules imposing mandatory human review of AI‑assisted utilization review and administrative decisions.
  • Patient transparency and disclosure. Many new laws require healthcare providers and payors to disclose when AI is used in patient communications, care coordination, or benefit determinations. California’s AB 3030, effective January 1, 2025, requires health facilities, clinics, and physician practices to provide disclaimers indicating when patient communications concerning clinical information are generated by AI and to include clear instructions for contacting a human healthcare provider. California’s AB 489, effective January 1, 2026, extends existing health‑license misrepresentation prohibitions to AI and generative AI systems, barring the use of titles, letters, phrases, or design elements suggesting that a system or its outputs come from a licensed health professional when that is not the case.
  • Behavioral‑health‑specific restrictions. Mental‑health applications have drawn particular scrutiny. Seven states have enacted laws targeting AI‑enabled chatbots, including California’s SB 243, which requires clear disclosure that users are interacting with AI and adds stricter protections for minors. Illinois’ HB 1806 limits AI from making therapeutic decisions or interacting with patients without licensed oversight. These laws generally prohibit unlicensed AI chatbots from representing themselves as licensed therapists and, in some cases, may effectively limit extended autonomous interactions with patients, particularly minors.
  • Insurance and utilization‑review rules. Multiple states have enacted laws governing AI use by insurers and managed care plans, including provisions that bar AI‑only decisions on coverage and utilization review and require that licensed physicians or other clinicians retain final authority. California’s SB 1120, effective January 1, 2025, prohibits health plans from basing medical necessity determinations solely on AI or automated algorithms and requires human clinical review. Maryland enacted similar requirements effective October 1, 2025, and Arizona, Connecticut, Nebraska, and Texas have enacted or are implementing comparable laws. Several additional bills pending in 2026 would require human review of AI‑based claims‑denial algorithms or utilization‑review tools.
  • AI‑transparency and accountability statutes. Broader state AI‑transparency laws now implicate healthcare platforms, especially telehealth and patient‑portal vendors. For instance, California’s AI Transparency Act (SB 942), taking effect August 2, 2026 (after a 2025 amendment (AB 853) delayed its original January 1, 2026, effective date), requires developers or providers of generative AI systems with more than 1,000,000 monthly users to provide tools enabling users to determine whether content is AI‑ Other states are advancing similar disclosure obligations, which may affect patient‑facing chatbots, automated messaging, and clinical‑support tools.

According to the 2026 AI Bill tracker maintained by the National Conference of State Legislatures, there are 177 pending bills in 31 states relating to AI use in healthcare. At the federal level, Senator Marsha Blackburn introduced a new comprehensive AI bill on March 18, 2026, which seeks to codify the December 2025 Executive Order preempting state AI bills. As the kids say these days, until the US Congress slides into the states’ DMs by passing a comprehensive AI law, their relationship status regarding AI use in the Healthcare industry will remain “complicated.”