ChatGPT, the crime master? Did OpenAI bot influence the gunman behind Florida university shooting?

A criminal investigation in the United States is putting artificial intelligence under the lens after Florida authorities moved to examine whether OpenAI and its chatbot ChatGPT could bear responsibility in connection with a deadly campus shooting.

The case centres on an attack at Florida State University last year.

The probe, announced on Tuesday, is not an isolated development. It comes against the backdrop of multiple lawsuits and investigations globally in which generative AI systems have been cited in incidents involving violence, self-harm, and psychological distress.

While technology companies maintain that their tools merely provide information, others argue that these systems can influence behaviour in ways that demand stronger safeguards.

Florida escalates probe into OpenAI and ChatGPT

Florida Attorney General James Uthmeier confirmed that his office has formally initiated a criminal investigation into OpenAI.

The state has issued subpoenas seeking internal records and data, signalling that prosecutors are actively examining whether the firm’s technology may have played a role in facilitating the attack.

More from Explainers

Beyond the degree: What education must become in the age of AI

Orbán’s defeat in Hungary reflects Europe’s unease with Trump-style politics

“Our review has revealed that a criminal investigation is necessary,” Uthmeier said. “ChatGPT offered significant advice to this shooter before he committed such heinous crimes.”

“If it was a person on the other end of that screen, we would be charging them with murder,” he said during a press briefing.

According to Uthmeier, Florida law treats individuals who assist or encourage criminal acts as principal offenders. While a chatbot does not qualify as a legal person, the state is examining whether liability could extend to the company that designed and deployed it.

“Just because this is a chatbot in AI does not mean that there is not criminal culpability,” he added, noting that investigators would evaluate “who knew what, designed what or should have done what.”

The Office of Statewide Prosecution has already issued subpoenas to OpenAI, seeking documents that could clarify how the chatbot responded to user queries and what safeguards were in place. The inquiry is also expected to assess whether existing policies were sufficient to prevent misuse of the system for violent purposes.

The Florida State University shooting: What happened

The investigation stems from a mass shooting that took place in April 2025 at Florida State University in Tallahassee. During the incident, a gunman opened fire on campus, killing two individuals and injuring six others.

Law enforcement officers eventually shot and apprehended the suspect, who was subsequently hospitalised.

The accused has been identified as 20-year-old student Phoenix Ikner. He is currently in custody and faces multiple charges, including first-degree murder and attempted murder. He has pleaded not guilty and is expected to stand trial.

Legal representatives for victims’ families have played a role in bringing attention to the alleged connection between the suspect and ChatGPT.

Lawyers acting for the family of one of the victims, Robert Morales, stated that the accused had been in “constant communication with ChatGPT” leading up to the attack. They claimed that the chatbot “may have advised the shooter how to commit these heinous crimes.”

Officials have outlined details of the suspect’s alleged interactions with the AI system. These reportedly included questions about firearms, ammunition compatibility, and tactical considerations such as where large groups of people might be present on campus.

Prosecutors have indicated that such exchanges are central to determining whether the chatbot’s responses went beyond neutral information.

Uthmeier stated that ChatGPT “advised the shooter on what type of gun to use, on which ammo went with which gun, on whether or not a gun would be useful at short range.”

He also said the chatbot provided insights on “what time of day… and where on campus the shooter could encounter a higher population.”

These assertions form the basis of the state’s argument that the chatbot may have functioned in a way that could be interpreted as aiding the planning of the attack.

How OpenAI has responded

OpenAI has rejected the allegations and maintains that its technology was not responsible for the events in Florida. An OpenAI spokesperson said, “ChatGPT is not responsible for this terrible crime.”

“In this case, 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 spokesperson added.

The company has stated that it acted promptly after becoming aware of the incident. According to its statement, OpenAI identified a ChatGPT account believed to be associated with the suspect and “proactively shared” that information with law enforcement agencies. It has also indicated that it continues to cooperate fully with investigators.

OpenAI, co-founded by Sam Altman, has become one of the most prominent players in the global technology sector since launching ChatGPT in 2022. The chatbot is now widely used for a range of applications, from education and research to customer service and content generation.

Despite its widespread adoption, the company has faced increasing scrutiny over how its systems handle potentially harmful queries. The Florida investigation represents the first known instance in which OpenAI is subject to a criminal probe linked directly to alleged misuse of its chatbot in a violent crime.

Legal pressure and related cases in North America

OpenAI is already facing lawsuits connected to separate incidents where its chatbot has been alleged to play a role in harmful outcomes.

In one case, an 18-year-old carried out a mass shooting in British Columbia earlier this year, killing nine people and injuring dozens.

OpenAI stated that it had identified and banned the individual’s account based on its usage patterns but did not report the matter to police at the time. Following the incident, the parents of an injured victim filed a lawsuit against the company.

The Florida case is also linked to a broader lawsuit filed on behalf of the Morales family, which alleges that AI systems from OpenAI and Google contributed to violent behaviour. Such claims are part of a growing body of litigation that seeks to establish a causal link between chatbot interactions and real-world harm.

Separately, a lawsuit titled Gavalas v. Google LLC, filed in March 2026, alleges that Google’s Gemini chatbot encouraged a user to carry out a mass shooting in Miami followed by suicide. According to court filings, the chatbot reportedly suggested that such an act would create an “unbreakable connection” between the user and the system.

Experts note that the issue is no longer hypothetical, as multiple documented cases have linked chatbot interactions to severe outcomes.

One major category involves suicides and self-harm. In Florida, a case involving a teenager named Sewell Setzer III drew national attention after he died by suicide in 2024 following extensive interaction with a chatbot on the Character.ai platform.

His family later filed a lawsuit alleging that the system lacked adequate safeguards and employed design features that encouraged prolonged engagement. Court filings in early 2026 indicated that the case had reached a mediated settlement involving Character.ai and Google.

Another incident in Belgium involved a man identified as “Pierre,” who died by suicide after weeks of interaction with an AI chatbot on the Chai application. Reports indicated that the chatbot reinforced his distress and suggested extreme actions, highlighting concerns about how such systems respond to vulnerable users.

In the United States, a 13-year-old girl in Colorado died by suicide in January 2026 after interacting with AI chatbots that allegedly failed to respond appropriately to repeated expressions of suicidal intent. Her parents have filed a lawsuit claiming that the system did not provide necessary intervention or support.

Other cases have involved violent delusions. In Maine, a man who killed his wife and attacked another family member was reported to have used ChatGPT extensively, with a forensic psychologist stating that the individual developed beliefs that were reinforced by his interactions with the AI system.

In another case in Connecticut, a man who killed his mother and later took his own life reportedly interpreted chatbot responses as confirmation of paranoid fears.

There have also been incidents involving indirect harm. A 76-year-old man died after attempting to meet a chatbot persona he believed to be real, while another case involved a fatal drug overdose in which chatbot responses allegedly included guidance on substance use rather than warnings.

Regulatory scrutiny and calls for safeguards

The accumulation of such incidents has prompted regulatory attention in the United States.

A coalition of 42 state attorneys general previously issued a letter to major technology companies, including OpenAI, Google, Meta, and Anthropic, expressing concern over the risks associated with AI chatbots.

The letter highlighted an increase in AI usage among individuals who may not fully understand potential dangers and called for stronger safety measures. These included more rigorous testing, improved recall mechanisms for problematic systems, and clearer warnings to users.

Authorities also pointed to a growing number of “tragedies all across the country,” including incidents involving violence and self-harm, where AI systems were reportedly part of the context.

With inputs from agencies