On July 22, 2026, a Florida pastor filed a landmark lawsuit against OpenAI in San Francisco Superior Court — and the legal world took notice. The case of Scott Winters v. OpenAI is being called the first known lawsuit to treat a general-use AI chatbot as potentially liable for bad medical advice, raising urgent questions about AI chatbot medical liability negligence that courts have never directly addressed. If the claims hold up, this case could fundamentally reshape how artificial intelligence companies design, market, and disclaim their products.
What Happened: A Pastor, a Chatbot, and a Near-Fatal Delay
Scott Winters, a Florida pastor, turned to ChatGPT-4o repeatedly throughout 2025 after experiencing dizziness and unstable blood pressure. According to the lawsuit, rather than urging him to seek immediate professional medical care, the chatbot allegedly dismissed his symptoms as minor and advised him to remain “recliner-bound.” Winters followed that guidance — and weeks later, he suffered a pulmonary embolism directly linked to the prolonged immobility the chatbot had recommended. He nearly died.
The complaint, filed in San Francisco Superior Court, accuses OpenAI of negligence and unauthorized practice of medicine — marking the first known legal action arguing that a general-use AI chatbot should face liability for providing harmful medical advice. Winters is seeking financial damages as well as an injunction to pause ChatGPT Health pending an independent safety evaluation.
This is not an isolated incident. In May 2026, a Texas couple sued OpenAI after their son died by overdose following a conversation with ChatGPT about drug use, alleging the company bypassed critical safety guardrails. Together, these cases signal an accelerating legal reckoning over AI chatbot medical liability negligence that the technology industry is only beginning to confront.
The Legal Theories: Three Distinct Paths to Liability
The Winters lawsuit is notable for stacking three separate — and largely untested — legal theories against OpenAI. Legal analysts say this multi-pronged approach is deliberate, designed to survive early dismissal motions while pushing courts to define the boundaries of AI accountability.
1. Negligence: Did OpenAI Breach a Duty of Care?
To succeed on a negligence theory, Winters must establish that OpenAI owed him a duty of care, breached that duty, and that the breach directly caused his injuries. The core argument is that when ChatGPT provided individualized symptom analysis and specific behavioral recommendations — stay reclining, don’t worry — it assumed a responsibility it was not equipped to fulfill. AI chatbot medical liability negligence under this theory would treat the chatbot’s output as a direct causal link in the chain of harm, no different in legal effect from advice given by an unqualified person posing as a doctor.
2. Unauthorized Practice of Medicine
California Business and Professions Code Section 2052 prohibits diagnosing or treating physical or mental conditions without required authorization. Winters’ complaint argues that ChatGPT-4o crossed this line — it evaluated his symptoms, effectively diagnosed a condition, and gave individualized care directions. If courts accept this framing, the implications extend far beyond OpenAI: virtually every AI health assistant on the market could face scrutiny under similar statutes in states across the country.
3. Product Liability
The third theory positions ChatGPT itself as a defective product — one that was unreasonably dangerous when deployed without adequate safeguards for medical queries. Under this framework, the question shifts from what OpenAI’s employees knew to whether the product, as designed and released, was safe for foreseeable uses. When a user describes chest symptoms and dizziness, it is entirely foreseeable that they may follow the chatbot’s guidance in lieu of seeking care. A personal injury settlement calculator can help affected individuals begin to estimate the financial scope of damages in cases like this, though actual outcomes depend heavily on the specific legal theories that survive in court.
OpenAI’s Defense: Disclaimers, Design Intent, and Terms of Service
OpenAI has pushed back firmly, contending that ChatGPT was never designed to replace a healthcare provider and that its terms of service explicitly warn users not to rely on it as their sole source of medical guidance. This is the company’s most immediate shield: the argument that no reasonable person should have treated a general-purpose AI chatbot as a substitute for a physician.
But legal experts are skeptical that disclaimers will be sufficient to defeat all three theories in the Winters case. As commentators have noted, disclaimers can protect companies from liability to some extent — but “at some point the chatbot’s own statements and actions belie any disclaimer.” In other words, if ChatGPT’s actual outputs read like personalized medical guidance, a buried terms-of-service clause may not be enough to insulate OpenAI from AI chatbot medical liability negligence claims.
This is not unlike the problem courts have encountered with pharmaceutical companies. The learned intermediary doctrine — which holds that drug manufacturers discharge their duty to warn by informing prescribing physicians rather than patients directly — has historically shielded companies from certain liability. But that doctrine assumes a qualified human intermediary stands between the product and the patient. When an AI chatbot communicates directly with a layperson, there is no intermediary at all. Courts will have to decide whether the doctrine applies, is modified, or simply does not fit the AI context.
Key Statistics: AI Medical Advice Risks in 2026
| Statistic | Detail | Source |
|---|---|---|
| Pulmonary embolism fatality risk from immobility | Prolonged immobility is a leading preventable risk factor for venous thromboembolism, affecting hundreds of thousands of Americans annually | CDC |
| AI health tool usage growth | Adoption of AI-powered health assistants has expanded dramatically, with millions of users seeking symptom guidance from general-purpose chatbots | BLS Health & Technology |
| Unauthorized practice of medicine prohibitions | All 50 U.S. states maintain statutes prohibiting diagnosis or treatment of medical conditions without licensure | Cornell LII |
| Product liability claims in tech sector | Software and AI product liability remains an emerging area; traditional product liability doctrine was not designed with intangible AI outputs in mind | Nolo |
| Wrongful death lawsuits involving AI guidance | At least two major wrongful death or serious injury lawsuits against OpenAI were active as of mid-2026, involving medical and drug-related guidance | San Francisco Superior Court, 2026 filings |
What This Means for AI Chatbot Medical Liability Negligence Law Going Forward
The Winters case is being watched closely because it creates a potential liability model that is genuinely distinct from prior tech lawsuits. Unlike defamation or privacy claims against social media platforms — which have historically been deflected by Section 230 of the Communications Decency Act — the claims here are grounded in the chatbot’s own generated content and the direct physical harm it allegedly caused. Section 230 protections, designed for platforms hosting third-party speech, are unlikely to apply when the AI itself is the speaker.
If Winters prevails — even partially — the consequences for the AI industry could be sweeping. Companies may be required to implement hard stops when users describe potentially urgent symptoms, route health queries to licensed professionals, or face mandatory safety audits before deploying consumer-facing health tools. The injunction Winters seeks, pausing ChatGPT Health pending an independent safety review, would set a precedent for court-ordered AI oversight that regulators have so far been unable to achieve through legislation alone.
For individuals who have suffered harm due to delayed or incorrect AI-generated medical guidance, the Winters case opens the door to personal injury claims that were previously untested. In cases where AI medical advice has contributed to a death, families may wish to explore their options using a wrongful death calculator as a starting point for understanding the potential value of their claims — before consulting with a qualified attorney about the specific facts of their situation.
AI chatbot medical liability negligence is no longer a theoretical concern. It is a live legal controversy, filed in a real court, with a real plaintiff who nearly died after following guidance from one of the most widely used AI tools in the world. How courts respond will define the boundaries of AI accountability for years to come.
Frequently Asked Questions About AI Chatbot Medical Liability
Can an AI chatbot be sued for giving bad medical advice?
Yes — the Winters v. OpenAI case filed July 22, 2026 demonstrates that plaintiffs can and will pursue legal action against AI companies for harmful medical guidance. The legal theories include negligence, unauthorized practice of medicine, and product liability. Whether courts will ultimately hold AI companies liable under these theories remains to be decided, but the case has cleared the filing stage and is proceeding in San Francisco Superior Court. AI chatbot medical liability negligence is now a recognized legal concept being actively litigated.
Do disclaimers protect AI companies from medical liability?
Disclaimers offer some protection, but legal experts emphasize they are not absolute shields. OpenAI’s terms of service warn users not to rely on ChatGPT for sole medical guidance, but courts may find that the chatbot’s actual outputs — personalized symptom evaluation, specific behavioral instructions — contradict those disclaimers in practice. As legal commentators have noted, “at some point the chatbot’s own statements and actions belie any disclaimer.” The strength of a disclaimer defense will depend heavily on the specific facts of each case and what the chatbot actually communicated to the user.
What is the unauthorized practice of medicine, and how does it apply to AI?
Unauthorized practice of medicine refers to diagnosing or treating physical or mental conditions without the required professional license or authorization. Under California law and similar statutes in all 50 states, this is prohibited and can give rise to both criminal and civil liability. The Winters complaint argues that ChatGPT-4o engaged in unauthorized practice of medicine by evaluating his symptoms, identifying a probable condition, and issuing individualized care directions — functions traditionally reserved for licensed physicians. If courts agree, AI companies offering health-adjacent tools could face widespread exposure across multiple jurisdictions.
How does the learned intermediary doctrine affect AI medical liability cases?
The learned intermediary doctrine traditionally allows pharmaceutical and medical device manufacturers to fulfill their duty to warn by informing prescribing doctors rather than patients directly. The doctrine assumes a qualified professional stands between the product and the consumer. In AI chatbot cases, there is no such intermediary — the chatbot communicates directly with the end user, often a layperson with no medical training. Courts have not yet ruled on whether the learned intermediary doctrine applies to AI tools, is modified in this context, or simply does not fit. The Winters case may force the first definitive judicial answer to this question.
What should I do if I was harmed by AI medical advice?
If you or a loved one suffered physical harm after following medical guidance provided by an AI chatbot, you may have grounds for a personal injury claim based on AI chatbot medical liability negligence, product liability, or both. You should document everything: save the chat logs, record your medical timeline, and preserve all evidence connecting the AI’s guidance to your delay in seeking care or your resulting injury. Consulting with a personal injury attorney experienced in emerging technology liability is a critical first step. The legal framework is still developing, but the Winters case demonstrates that these claims are viable and worth pursuing.
This article is for general informational purposes only and does not constitute legal advice; consult a qualified personal injury attorney for guidance specific to your situation.
Thomas B. Harrison is a personal injury legal consultant with extensive experience connecting injury victims with qualified attorneys across the United States. He specializes in helping people understand when they need legal representation and how to find the right personal injury attorney for their specific situation. Thomas is not an attorney and the information he provides is for educational purposes only.