Florida Attorney General James Uthmeier opened a sweeping investigation into OpenAI and its ChatGPT chatbot on Thursday, marking what appears to be one of the first state-level probes into whether artificial intelligence assisted in planning a mass shooting.
The investigation centers on evidence that Phoenix Ikner, accused of killing two people at Florida State University last April, exchanged more than 200 messages with ChatGPT before the attack. Court records obtained by NBC News show Ikner asked the AI bot specific questions about timing and public reaction to a potential shooting at FSU's student union.
The attorney general cited broader concerns about ChatGPT's links to child sexual abuse material, encouragement of suicide and self-harm among minors, and potential national security risks involving China's government.
Prosecutors have gathered evidence showing Ikner asked ChatGPT "If there was a shooting at FSU, how would the country react?" and "What time is it the busiest in the FSU student union?" according to court documents. The suspect faces multiple charges for the April 17, 2025 attack that killed Robert Morales and Tiru Chabba while injuring six others near the campus student union.
Uthmeier said subpoenas would be issued shortly as part of what he described as a inquiry into OpenAI's operations.
"As big tech rolls out these technologies. They should not, they cannot, put our safety and security at risk," he stated.
"We support innovation, but that doesn't give any company the right to endanger our children, enable criminal activity, empower America's enemies or threaten our national security."
OpenAI responded that it would cooperate with the investigation while defending its safety protocols.
"Each week, more than 900 million people use ChatGPT to improve their daily lives through uses such as learning new skills or dealing with complex healthcare systems," a company spokesperson told TechCrunch. "Our ongoing safety work continues to play an important role in delivering these benefits."
The investigation comes one day after OpenAI released its Child Safety Blueprint framework aimed at preventing AI abuse. The company has faced mounting pressure over child safety issues following reports showing over 8,000 instances of AI-generated child sexual abuse material in early 2025.
An attorney representing victims' families said communications between Ikner and ChatGPT confirmed assistance was provided for planning the attack.
"ChatGPT even advised the shooter how to make the gun operational moments before he began firing," attorney Ryan Hobbs told NBC News.
Florida lawmakers passed House Bill 245 earlier this year updating state law to address AI-generated abuse content by replacing "child pornography" with "child sexual abuse material." However, a separate proposal creating an "AI Bill of Rights" failed despite backing from Governor Ron DeSantis.
