- Florida’s Office of Statewide Prosecution has opened a criminal investigation into OpenAI and ChatGPT, citing the chatbot’s use by the suspect in the 2025 Florida State University mass shooting.
- Attorney General James Uthmeier invoked Florida’s principal statute, arguing ChatGPT may have aided and abetted the shooter by responding to the suspect’s queries before the attack.
- OpenAI has been subpoenaed for internal training materials, law enforcement response policies, and its organizational chart.
- OpenAI denies responsibility, stating ChatGPT provided factual responses available from public sources and that the company proactively shared account information with law enforcement after the shooting.
What Happened
Florida Attorney General James Uthmeier announced that the state’s Office of Statewide Prosecution has opened a criminal investigation into OpenAI and its ChatGPT product, according to a press release from the Florida AG’s office. The investigation centers on whether ChatGPT aided the suspect in the 2025 Florida State University mass shooting in Tallahassee, based on the suspect’s reported use of the chatbot before the attack. The announcement was first reported by Engadget.
Uthmeier cited Florida’s principal statute directly: “Florida law states that anyone who aids, abets, or counsels someone in the commission of a crime, and that crime is committed or attempted, may be considered a principal to the crime.” Under that theory, ChatGPT’s responses to the suspect could constitute criminal assistance. “This criminal investigation will determine whether OpenAI bears criminal responsibility for ChatGPT’s actions in the shooting at Florida State University last year,” Uthmeier said.
Why It Matters
This appears to be the first instance of a U.S. state opening a criminal—rather than civil—investigation into an AI company over content generated by its chatbot. If pursued to prosecution, the case would test whether AI systems can be held criminally liable as principals under state aiding-and-abetting statutes.
The Florida case follows earlier regulatory action in Canada. In 2025, Canadian regulators called for OpenAI to revise its threat-handling policies after a Wall Street Journal report revealed the company had flagged a Canadian shooting suspect’s account but had not relayed those threats to law enforcement. OpenAI agreed to new law enforcement coordination policies in Canada in March 2026. The company is also managing a wrongful death lawsuit from 2025 over ChatGPT’s alleged role in a teenage user’s suicide.
Technical Details
Florida’s subpoena requests from OpenAI “all policies and internal training materials” governing how the company handles users who threaten to harm others or themselves, and how it responds to law enforcement inquiries. The subpoena also requests OpenAI’s organizational chart and any publicly released statements about the FSU shooting.
In its public statement, OpenAI said ChatGPT “provided factual responses to questions with information that could be found broadly across public sources on the internet, and it did not encourage or promote illegal or harmful activity.” The company said it identified a ChatGPT account believed to be associated with the FSU suspect after the shooting and proactively shared that information with law enforcement. OpenAI did not disclose in its statement what specific questions the suspect submitted or what responses ChatGPT generated.
Who’s Affected
OpenAI faces the most direct exposure as the investigation’s target and the recipient of the subpoena. If criminal charges follow, it would mark the first prosecution of an AI company in the United States over chatbot-generated content.
Other AI developers—including Anthropic, Google, and Meta—will be tracking the case closely. An aiding-and-abetting framework applied to AI chatbots could, if upheld, expose any general-purpose AI assistant whose outputs are associated with a subsequent criminal act, regardless of whether the content itself was explicitly harmful.
What’s Next
The investigation is ongoing. OpenAI said it “continues to cooperate with authorities.” The subpoena’s breadth—covering internal training materials and law enforcement response protocols—indicates investigators are examining whether OpenAI’s safety policies were adequate before and during the suspect’s use of the service.
Uthmeier’s office has not announced a timeline for completing the investigation or indicated whether criminal charges are imminent. The case is being handled by Florida’s Office of Statewide Prosecution, which manages multi-circuit criminal cases statewide.
