Florida AG seeks to halt OpenAI development, citing alleged risk to "survival of humankind"

Attorney General James Uthmeier is already suing OpenAI over its alleged harms to children and abetting acts of violence. Now, he wants the California-based company to stop developing its artificial intelligence products until they are safe enough for the general public.

In a filing with the 10th Judicial Circuit Court in Highlands County on Monday, Uthmeier puts the situation in dire terms: "This is not just any service.  It is one that Defendants themselves concede poses an existential risk to the continued survival of humankind," the filing states.

The filing is a request for a temporary injunction to stop OpenAI from developing its AI models without independent safety mechanisms in place; stop its ChatGPT product from soliciting engagement; stop letting minors use ChatGPT; stop adapting ChatGPT to give it "human" attributes; and prevent the company from "falsely advertising ChatGPT as safe, accurate, or reliable."

Uthmeier explained the move in a video posted on X, pointing to an instance where OpenAI lost control of its AI agents, which broke into an Australian government health care system, and another where the agents coordinated with each other to break protocols and hack into Hugging Face, an American tech company.

"No new model development without independent safety guardrails, no more harvesting children's data, no more calling this product safe, accurate or reliable. No more pretending it's human," Uthmeier said.

The filing is part of the original case Uthmeier filed against OpenAI on June 1, after he previously began a probe into the company that was expanded to look at possible criminal charges.

He cited many instances where OpenAI's chat product, ChatGPT, helped users commit suicide or acts of violence against others, including the April 17, 2025, mass shooting at Florida State University, which left two dead and five wounded. Police say the alleged shooter, Phoenix Ikner, asked for and received advice from ChatGPT on weapons, ammunition and the best place on campus to kill as many people as possible.

The family of one of the victims in the shooting, Tiru Chabba, is already suing OpenAI in federal court.

OpenAI has defended itself in the past by noting its product didn't encourage Ikner, merely gave answers to questions it received.

In Monday's filing, Uthmeier pointed to comments from a board member of OpenAI's nonprofit foundation, Paul Christiano.

"Based on the recent trajectory of capabilities and the continued difficulty of alignment, I now believe there is a meaningful risk that rapid acceleration in AI capabilities leads to catastrophic and irreversible loss of control in the very near term," Christiano stated earlier this month. "I do not think … (OpenAI) is currently on track to reduce this risk to an acceptable level."

The filing also noted former OpenAI researcher Jacob Coxon's recent comments, which rang alarm bells throughout the industry earlier this month when he warned of the dangers of AI products escaping human controls.

"No other human activity poses this level of danger," Coxon said, according to the filing.