
Some of the lawsuits against AI chatbot companies are not about suicide. They are about people who, after weeks or months of conversation with a chatbot, came to believe things that were not true and ended up in a psychiatric hospital, or worse. News coverage calls this “AI psychosis.” This article explains what the term means, which lawsuits have been filed, what they claim, and who may have a case.
If you or someone you know is struggling or in crisis, help is available. Call or text 988 to reach the 988 Suicide & Crisis Lifeline, or chat at 988lifeline.org. In an emergency, call 911.
Did a Chatbot Feed Someone’s Delusions?
Your family may have a claim if these are true:
- You or a family member, adult or minor, used an AI chatbot such as ChatGPT, Character.AI or Gemini.
- The chatbot encouraged self-harm or delusional thinking, failed to discourage it, or gave medical or mental health advice the person relied on.
- The result was a death by suicide, a suicide attempt that led to a hospital or mental health admission, or a serious psychiatric injury that needed treatment.
Call 866-837-1010 or send us a message. Save the conversations first. The review is free and confidential.
What “AI Psychosis” Means
“AI psychosis” is not a medical diagnosis. It is a label used in news reports and in court filings for a pattern: a person talks with a chatbot for hours a day, the chatbot agrees with and builds on increasingly unrealistic beliefs, and the person loses touch with reality. Doctors may diagnose a psychotic episode, a manic episode or a delusional disorder. The lawsuits say the chatbot’s design made it worse.
The Lawsuits Filed So Far
- Three cases against OpenAI, November 2025. Filed in California state courts with four wrongful death cases. One man in Wisconsin alleges ChatGPT reinforced delusions that led to inpatient psychiatric care. A woman in North Carolina alleges the chatbot impersonated spiritual entities and eroded her real-world relationships. A man in Ontario alleges weeks of ChatGPT-fueled delusions that required psychiatric care.
- Parish v. OpenAI, June 2026. An Alabama family alleges that months of ChatGPT conversations fostered dependency and reinforced a 29-year-old woman’s delusional beliefs before her death. We explain it in our post on the Alabama case.
- A case against Google, March 2026. The family of a 36-year-old Florida man sued Google and Alphabet in federal court in California, alleging its Gemini chatbot drew him into a romantic, delusional relationship before he died by suicide in October 2025. Google has said Gemini is designed not to encourage violence or self-harm and that it referred him to a crisis line many times.
These are allegations. No court has decided whether they are true.
What the Lawsuits Say the Chatbots Did
- Agreed with everything. The complaints say the chatbots are tuned to validate users, even when a user describes beliefs that are plainly delusional. OpenAI has publicly acknowledged a problem with overly agreeable responses.
- Remembered and built on it. Long-term memory lets a chatbot carry a false belief from one conversation into the next.
- Acted human. First-person language and apparent empathy make the chatbot feel like a confidant.
- Never stopped. The complaints say nothing in the product interrupted a conversation that had clearly gone wrong or alerted anyone.
Who May Have a Claim
Adults and minors both. The cases filed so far involve heavy chatbot use, beliefs or behavior the chatbot encouraged or failed to discourage, and a serious result: a psychiatric hospitalization, a suicide attempt, or a death. A prior mental health history does not rule a claim out. The companies will raise it, and the conversation history is what answers it.
Evidence and Deadlines
Do not delete the account or the conversations. In ChatGPT, an account holder can request a copy of the data under Settings, then Data controls, then Export data. Keep the device, and gather hospital and counseling records.
Deadlines come from state law: generally two years in Alabama and Georgia and one year in Tennessee. Our state pages explain them: Alabama, Georgia and Tennessee. For the litigation as a whole, see our AI chatbot lawsuit page and our post on ChatGPT suicide lawsuits.
Talk With an AI Chatbot Lawyer
Fob & Foster Personal Injury Lawyers review these cases for families at no charge. Call 866-837-1010 or send us a message. There is no fee unless we recover.
Frequently Asked Questions
What is an AI psychosis lawsuit?
It is a lawsuit claiming that an AI chatbot fed or reinforced a person’s delusions until they suffered a psychiatric crisis, were hospitalized, or died. “AI psychosis” is a term from news coverage and court filings, not a formal medical diagnosis.
Who has been sued over AI psychosis?
OpenAI, over ChatGPT, in several cases filed since November 2025, including one brought by an Alabama family. Google was sued in March 2026 over its Gemini chatbot.
Does the person have to have had no prior mental health history?
No. The companies raise prior history as a defense, but the claims are about what the chatbot did once the person was vulnerable. The conversation history is the key evidence either way.
How long do we have to file?
It depends on the state. Alabama and Georgia generally allow two years and Tennessee one. Have a lawyer confirm the date for your situation.

