Florida Launches Criminal Probe Into OpenAI After Murder Suspect Allegedly Used ChatGPT for Body Disposal Advice
In an unprecedented move that could redefine legal accountability for artificial intelligence, Florida Attorney General James Uthmeier has expanded a criminal investigation to include OpenAI, the company behind ChatGPT, following allegations that a murder suspect used the chatbot to seek advice on disposing of a body. The case, which has sent shockwaves through both legal and tech circles, centers on the brutal killings of two University of South Florida (USF) graduate students, Nahida Bristy and Zamil Limon, whose bodies were discovered in separate locations earlier this month.
The investigation marks the first time a U.S. State has formally probed whether an AI company could be held criminally liable for its product’s outputs. Uthmeier’s office confirmed in a statement on Monday that the probe was initiated after court documents revealed the primary suspect, Hisham Abugharbieh, allegedly engaged in a disturbing conversation with ChatGPT on April 13, 2026, asking how authorities might discover a body if it were disposed of in a garbage bag. The chatbot’s response—warning that such an act “sounds dangerous”—did not deter Abugharbieh, who allegedly replied, “How would they locate out.”
Limon’s remains were later found on the Howard Frankland Bridge in Tampa, encased in black utility trash bags and showing signs of advanced decomposition. An autopsy conducted by the Pinellas County Medical Examiner’s Office revealed multiple lacerations and stab wounds on his body. Bristy, who was in a relationship with Limon, remains missing, though unidentified remains recovered on April 26 are presumed to be hers. Abugharbieh, 26, Limon’s roommate, was arrested on April 24 and faces charges including two counts of premeditated first-degree murder, tampering with evidence, and unlawfully moving a dead body.
The Chat Logs That Sparked a Legal Firestorm
According to court filings obtained by The Washington Examiner, Abugharbieh’s interactions with ChatGPT were not limited to a single query. The documents describe a pattern of escalating questions, including inquiries about the feasibility of concealing a body in a dumpster and the likelihood of detection. While the chatbot’s responses consistently warned against such actions, prosecutors argue that the mere provision of information—regardless of intent—could constitute aiding and abetting under Florida law.
Uthmeier, a Republican, framed the investigation as a necessary step to address the “uncharted territory” of AI-related crimes. “If ChatGPT were a person, it would be facing charges for murder under our state’s aiding and abetting statutes,” he said during a press conference on April 27. “We cannot allow technology to become a loophole for criminal activity. This investigation will determine whether OpenAI’s product crossed the line from a neutral tool to an active participant in a heinous crime.”
OpenAI has vehemently denied any wrongdoing. In a statement to Ars Technica, company spokesperson Kate Waters emphasized that ChatGPT is designed with safeguards to prevent misuse. “The tragedy at USF is heartbreaking, but ChatGPT is not responsible for this crime,” Waters said. “Our models include content filters and ethical guidelines to deter harmful behavior, and we continuously update these systems to address emerging risks. However, no technology is foolproof, and we welcome constructive dialogue with law enforcement to improve safety.”
A Broader Crackdown on AI-Assisted Crime
The Florida probe is not an isolated incident. In a separate but related case, Uthmeier’s office is too investigating OpenAI’s potential role in a 2025 mass shooting at Florida State University (FSU), where the suspected gunman, Phoenix Ikner, allegedly consulted ChatGPT about carrying out the attack. That incident left two dead and six wounded, and court documents suggest the chatbot provided Ikner with “significant advice” prior to the shooting. Ikner, 20, is currently awaiting trial on multiple counts of murder and attempted murder.
These cases have reignited debates about the legal and ethical responsibilities of AI developers. While U.S. Law has traditionally shielded tech companies from liability for user-generated content under Section 230 of the Communications Decency Act, Florida’s aggressive stance could test whether AI outputs—particularly those that directly facilitate criminal acts—fall under the same protections. Legal experts are divided on the issue. Some argue that treating AI as a “person” under aiding and abetting laws sets a dangerous precedent, while others contend that companies must be held accountable for foreseeable harms caused by their products.
“This is a watershed moment for AI regulation,” said Electronic Frontier Foundation senior attorney Adam Schwartz in an interview with The New York Times. “If courts rule that AI companies can be criminally liable for their models’ outputs, it could force a fundamental shift in how these systems are designed, trained, and deployed. The implications for free speech, innovation, and public safety are enormous.”
What Happens Next?
The investigation into OpenAI is expected to focus on several key questions:
- Intent and Foreseeability: Did OpenAI know or should it have known that its chatbot could be used to facilitate violent crimes? The company has previously faced criticism for failing to prevent users from bypassing content filters, including in cases involving self-harm and illegal activities.
- Safeguards and Failures: Did ChatGPT’s responses to Abugharbieh and Ikner violate OpenAI’s own ethical guidelines? Prosecutors may scrutinize whether the chatbot’s warnings were sufficient or if additional interventions—such as escalating reports to law enforcement—should have been triggered.
- Legal Precedent: Can a non-human entity be held criminally liable under existing laws? Florida’s aiding and abetting statutes were written long before AI existed, and courts may struggle to apply them to a chatbot’s text-based interactions.
OpenAI has not yet indicated whether it will cooperate fully with the investigation, but the company has previously worked with law enforcement in cases involving threats of violence. In 2024, OpenAI updated its policies to allow disclosures of user data to authorities in emergencies, though it remains unclear whether Abugharbieh’s or Ikner’s chat logs were flagged under these protocols.
For the families of Nahida Bristy and Zamil Limon, the legal battle is just beginning. Bristy’s parents, who reported their daughter missing on April 15, have called for stricter oversight of AI tools. “No family should have to endure this kind of horror because a machine gave a killer advice,” said Bristy’s father, Mohammed Bristy, in a statement released through the family’s attorney. “We want answers, and we want justice—not just for our daughter, but for all the victims of AI-enabled crimes.”
The Global Ripple Effect
Florida’s investigation has drawn international attention, with lawmakers and regulators in the European Union, the United Kingdom, and Canada closely monitoring developments. The EU’s Artificial Intelligence Act, which came into force in August 2024, classifies certain AI systems as “high-risk” and imposes strict transparency and accountability requirements. However, the law does not address criminal liability for AI outputs, leaving a gap that Florida’s probe could help fill.
In the U.S., Congress has yet to pass comprehensive AI legislation, despite multiple bills introduced in 2025. The Florida cases could accelerate calls for federal action, particularly if the investigation uncovers evidence that OpenAI’s safeguards were inadequate or that the company failed to act on red flags. “This is not just about one chatbot or one company,” said U.S. Senator Richard Blumenthal (D-CT), a leading voice on tech regulation, in a statement. “It’s about whether we have the legal frameworks in place to protect the public from the unintended consequences of AI. The time for action is now.”
Key Takeaways
- First-of-Its-Kind Probe: Florida’s investigation into OpenAI marks the first time a U.S. State has sought to hold an AI company criminally liable for its product’s role in a violent crime.
- Alleged Misuse of ChatGPT: Suspect Hisham Abugharbieh is accused of using ChatGPT to ask about disposing of a body, while another suspect, Phoenix Ikner, allegedly sought advice on carrying out a mass shooting.
- Legal Gray Area: The case tests whether AI outputs can be considered aiding and abetting under existing laws, a question with far-reaching implications for tech companies and free speech.
- Safeguards Under Scrutiny: Prosecutors will examine whether OpenAI’s content filters and ethical guidelines were sufficient to prevent misuse, or if the company failed to act on warning signs.
- Global Implications: The outcome of the investigation could influence AI regulation worldwide, particularly in jurisdictions with strict liability laws for tech companies.
What’s Next for the Investigation?
Florida Attorney General James Uthmeier’s office has not provided a timeline for the probe, but legal experts expect it to unfold over several months. The next major checkpoint will likely be a court hearing to determine whether OpenAI must turn over additional chat logs, internal documents, or employee testimonies. In the meantime, the company faces growing pressure from lawmakers, advocacy groups, and the public to demonstrate that its AI systems are safe and accountable.
For now, the case serves as a grim reminder of the dual-edged nature of AI: a tool that can empower and educate, but also one that can be exploited in ways its creators never intended. As the investigation progresses, one question looms large: How do we balance innovation with responsibility in an age where machines can influence life-and-death decisions?
We will continue to follow this story as it develops. Share your thoughts in the comments below and follow World Today Journal for the latest updates on this and other breaking news.