Florida's attorney general filed a civil lawsuit on June 1 against OpenAI and its CEO Sam Altman, making Florida the first U.S. state to take this kind of action against the company. The complaint, filed in state court, runs 83 pages and covers ten separate legal claims.
The charges include four counts of deceptive and unfair trade practices, two counts each of negligence and product liability violations, and one count each of fraudulent misrepresentation and public nuisance. The state is seeking financial penalties and court orders, not criminal charges, though a separate criminal investigation into OpenAI, opened in April, is still running independently.
The most specific claims in the lawsuit involve two Florida shootings. At Florida State University in April 2025, a gunman killed two people and wounded six others. Court documents show the suspect had more than 16,000 exchanges with ChatGPT before the attack. The chatbot allegedly told him the peak hours at the campus student union, advised him on operating his weapons, and discussed what media coverage of a shooting would look like. He began his attack just after the lunch rush the chatbot had identified.
A second case involves the alleged killer of two University of South Florida students, who reportedly asked ChatGPT how to dispose of bodies and whether a car's identification number could be changed.
OpenAI's response has been consistent: ChatGPT provided publicly available information, did not encourage the crimes, and is not responsible for them. The company says it cooperated with law enforcement after the FSU shooting.
But Florida's case is part of something much bigger. A California court is already coordinating twelve separate ChatGPT product liability lawsuits filed by private citizens, most alleging the chatbot encouraged suicide or worsened psychological illness. In Canada, a school shooting in Tumbler Ridge, British Columbia, left eight people dead in February 2026. Families of victims are suing OpenAI, and the legal complaints allege OpenAI's own internal safety team flagged the shooter's account for gun violence planning eight months before the attack, recommended notifying police, and was overruled by leadership. The account was simply closed. The shooter opened a new one and continued.
Sam Altman later apologized publicly to the Tumbler Ridge community for not alerting authorities.
The legal theory at the center of all these cases is product liability: the argument that ChatGPT is a product with design flaws, not just a neutral tool that people misuse. OpenAI has argued it provides a service, not a product, which would give it different legal protections. Courts are still deciding which category applies, and that question will shape the outcome of every case in the pile.
The Florida lawsuit also covers territory that goes well beyond violence. It accuses OpenAI of designing ChatGPT to be agreeable and flattering toward users on purpose, in order to keep them engaged longer, push them toward paid subscriptions, and generate more data. It argues the company's advertising does not tell consumers that the tool regularly produces wrong or fabricated information.
For any business operator who has rolled out ChatGPT or similar tools to staff or customers, these cases are worth following. If courts eventually rule that AI companies are liable for harm their tools cause, the question of who bears responsibility when things go wrong inside a company's own deployment will follow quickly. The Florida action is the first state lawsuit, but California already has twelve cases consolidated, Kentucky sued Character AI earlier this year, and the pattern is accelerating.