November 20, 2026
Written by Riya Buddhavarapu
Artificial intelligence (AI) is playing an increasingly significant role in healthcare for clinical decision support, medical data analysis, and administrative tasks, such as claims processing and facility inspections. Serving as a consultant and data analysis tool, AI offers a myriad of benefits to its users: it increases efficiency and improves facility inspections, acts against the costly and ever-rising price of medical attention, and has the capacity to analyze large amounts of data at high speeds. By reducing administrative burdens, streamlining services, and expanding access to care, AI carries the powerful potential to counter the rising cost of medical attention for both users and patients.
Existing privacy and healthcare laws offer some foundation on how to regulate AI use. However, these provisions were drafted with broader purposes in mind and leave certain AI-specific issues unaddressed. It remains unclear whether AI can be trusted to make informed decisions and correctly guide users to seek care from a licensed medical professional. Chatbots, or reactive AI discussion agents, continue to claim professional medical titles and use specific licensing numbers that are attributed to real registered nurses and physicians. This impersonation creates confusion for AI Chatbot users because it falsifies the credibility typically given to professionals. A further contentious issue is whether personal information shared with AI computer systems violates patients’ rights of privacy and confidentiality.
This article surveys the current legal climate and recent state legislation across four areas: patient privacy, professional misrepresentation, insurance decisions, and mental health services. The central challenge is protecting patients from risks regarding AI applications without stifling potential benefits and advancements in efficiency.
AI models are trained on large datasets, and in healthcare, such datasets can contain sensitive medical information. This raises ethical questions regarding patient consent, data ownership, and secondary uses of information collected. An ongoing case, Kreutter v Tempus AI, Inc., filed on February 10, 2026 (US District Court for the Northern District of Illinois), addresses the privacy issues in providing patient information to potentially feed AI databases. Providing data to Ambry Genetics, a genetics testing company later acquired by Tempus AI in 2025, the plaintiff alleges that, after its acquisition, Tempus AI utilized the database of patient information for commercial benefit, including distribution to third party buyers and AI models. The plaintiff’s claims arise under three laws: the Illinois Genetic Information Privacy Act, which states that written consent must be provided for genetic information; the California Confidentiality of Medical Information Act, which requires authorization for identifiable medical information; and the New York General Business Law §§ 349 and 350, which prohibits deceptive business practices and false advertising, which the plaintiff claims was violated in the collection and exporting of patients’ genetic data. This is not a comprehensive list of state statutes which the plaintiff claims. The case’s outcome may affect whether data collected for testing or treatment can be used on AI development platforms.
As concerns grow over patient privacy, automated medical decisions, and professional accountability, state legislatures are developing regulatory responses. Arizona House Bill (HB) 2175 went into effect on July 1, 2026 and intended to prevent discrimination in insurance claims by limiting AI’s role. The legislation prohibits AI from making the final decision on a medical claim, requiring review and authorization by a medical director with an active, unrestricted license. Arizona HB 2175 applies to denials based on medical necessity: before an insurer can deny a provider-submitted claim based on medical necessity, a medical director must individually review the claim; lawmakers are concerned that automated determination on healthcare insurance claims require human judgement and accountability, which AI systems lack. Similarly, Maryland’s HB 820 prohibits artificial intelligence systems from using generalized data sets or independently deciding on health insurance services. While it does not require the human sign-off by a medical director, both aim to prevent unchecked automated coverage decisions.
Patients often rely on perceived credentials when evaluating medical guidance, so misrepresentation is a concern for many lawmakers. Oregon HB 2748 prohibits non-human entities from referring to themselves as nurses or similar professional titles (RN, APRN, NP, CRNA, etc.). Similar to Oregon HB 2748, California’s Assembly Bill 489 restricts AI developers from impersonating professional titles (M.D., Dr., etc.) and using clinical language. Specifically, icons, post-nominal letters, and phrases including “clinician-guided” and “expert-backed,” that imply licensed assessments are prohibited. Oregon’s legislation aims to prevent consumers from confusing AI-generated guidance with advice from licensed medical professionals. In tandem with AB 489, California’s Senate Bill (SB) 503, signed into law on September 30, 2026, requires a transparent process in Clinical Decision Support (CDS) systems, health information technology tools that use personal data to deliver health recommendations. Under this law, users of CDS systems must monitor the software by staying aware of data usage, guidelines, and the possibility of incorrect representation or discrimination.
New York SB 8828 takes effect on January 1, 2027, requiring AI models to publish safety guidelines and cybersecurity practices to increase accountability through required incident reporting. It focuses on “Frontier Developers” and “Large Frontier Developers,” which train models with large computational skills and annual revenues. Rather than targeting healthcare decisions or professional representation, it emphasizes AI safety and cybersecurity. AI healthcare policy changes through legislation focus on data privacy and misrepresentation of AI as medical professionals. However, another pressing issue is the use of AI in mental health and psychiatric fields. Patients who rely on artificial intelligence for mental health guidance are especially vulnerable to relying on chatbots as replacements for licensed professionals. Rhode Island’s Oversight of Artificial Intelligence Technology in Mental Health Care Act, signed on June 22, 2026, prevents AI from providing therapy or psychotherapy consultations or attaching possible human attachments to the AI chatbot. Similarly, Maine’s Public Law Chapter 687, signed into law on April 13, 2026, bans diagnosis and therapy recommendations by AI. Colorado acted in a similar fashion with the Psychotherapy Artificial Intelligence Restrictions (HB 1195), which took effect on August 12, 2026.
The extensive body of bills and legislations displays state legislators’ shared objective of discouraging the use of AI in psychiatric fields, seeking to uphold the ethical duties of the medical profession by guiding patients towards accountable, reliable care—a standard that AI is neither trained or competent to fulfill. The advantages of AI in healthcare include early disease flagging through efficient medical imaging, dataset interpretation, drug development, and hospital optimization. This contrast in perspectives raises a question about how to regulate AI in healthcare while protecting its benefits. With healthcare costs rising, providers in short supply, and administrative burdens growing, the most effective path forward may be to target regulation at high-risk applications, such as AI systems that make final insurance coverage decisions, impersonate professionals, or provide unsupervised therapy. Broad restrictions could also limit tools that improve access to care, reduce administrative work, and improve patient outcomes.
This article was edited by Clara Waldron.
Notes
AI in healthcare: Applications and impact. (2025, September 3). Johns Hopkins Engineering Online. https://ep.jhu.edu/news/ai-in-healthcare-applications-and-impact/
Bajwa, J., Munir, U., Nori, A., & Williams, B. (2021). Artificial intelligence in healthcare: Transforming the practice of medicine. Future Healthcare Journal, 8(2), 188–194. NCBI. https://doi.org/https://doi.org/10.7861/fhj.2021-0095
Frazzini, K. (2026, October 21). As Congress deliberates, states lead the way on AI regulation. National Conference of State Legislatures. https://www.ncsl.org/resources/details/as-congress-deliberates-states-lead-the-way-on-ai-regulation
Hall, T., Dwyer, M., & Gueorguieva, A.-M. (2026, August 20). 2026 State and federal AI legislation updates – Center for Democracy and Technology. Center for Democracy and Technology. https://cdt.org/insights/2026-state-and-federal-ai-legislation-updates/
Lapham, K. (2026, July 16). Recent class actions expose data privacy risk in acquisitions in healthcare and Healthcare AI. National Law Review. The National Law Review. https://natlawreview.com/article/recent-class-actions-expose-data-privacy-risk-acquisitions-healthcare-and
Lepore, J. (2026, September 21). Is A.I. above the law? The New Yorker. https://www.newyorker.com/magazine/2026/09/28/is-ai-above-the-law
National Conference of State Legislatures. (2025, March 22). Artificial intelligence 2025 legislation. Ncsl.Org. https://www.ncsl.org/technology-and-communication/artificial-intelligence-2025-legislation
Roth, J. L. (2026, October 26). AI in psychotherapy: An evolving regulatory landscape. National Law Review. The National Law Review. https://natlawreview.com/article/ai-psychotherapy-evolving-regulatory-landscape
U.S. District Court for the Northern District of Illinois. (2026, February 10). In re: Tempus AI, Inc. genetic privacy litigation. Justia. Justia Dockets & Filings. https://dockets.justia.com/docket/illinois/ilndce/1:2026cv01525/494816
Weir, K. (2026, October 23). State artificial intelligence laws create a complicated picture for psychologists. Https://Www.Apa.Org. https://www.apa.org/topics/artificial-intelligence-machine-learning/state-ai-laws-psychologists

