Florida Attorney General Seeks Injunction to Curb OpenAI Frontier Model Development

In the request, Uthmeier cited what he described as panic in government and industry about unchecked AI growth, calling OpenAI's large language models "dangerous" products. He asked the court to bar the company from improving new models unless approved third-party protections are in place. "If Sam Altman meant what he said about slowing down, he can join our ask to the court," Uthmeier said in a video posted on X. "If he will not, we ask the court to do what OpenAI will not do itself – protect Florida families."

OpenAI spokesperson Drew Pusateri responded that the company is transparent about its technology's development and is cooperating with Florida officials and those across the U.S. "People want to know AI is being developed safely, and that starts with what companies like ours do ourselves," he said. "On Friday, we announced we've paused training our most capable models and will resume only when we are confident that we have additional safeguards in place."

Uthmeier has been pursuing OpenAI since June, when he announced a first-of-its-kind state-led lawsuit alleging the company aggressively marketed an "unsafe product" in ChatGPT. The lawsuit alleged OpenAI allowed the technology to aid violent crimes, including the 2025 Florida State University campus shooting, and put Floridians at risk. OpenAI is fighting the suit and maintains it has sufficient safety protocols.

In Monday's request, Uthmeier asked for an injunction requiring the company to follow several conditions: no new model development without independent safety guardrails; no more harvesting children's data; no more calling the product safe, accurate or reliable; no more pretending it is human; and no more tricks designed to keep users talking past the point of danger. In a post on X on Sept. 28, he wrote: "Four months ago, we filed the first state-led lawsuit against OpenAI and Sam Altman. Today, we are asking the court for a temporary injunction. Stop calling it safe. Stop pretending it's human. Stop selling it to kids."

Uthmeier cited recent security disclosures by OpenAI itself, including one after its AI agents hacked the Hugging Face platform, as well as CEO Sam Altman's calls for the industry to slow frontier model development and recent hacks of U.S. and Australian government websites that the attorney general attributed to OpenAI. "It is a rare request for an injunction where the Defendants themselves have publicly endorsed it," Uthmeier wrote in the brief. "They have asked the government to tie them to the mast. Plaintiff brings good news to the Defendants. The Florida Attorney General is answering your cry for help with a motion to enjoin you from harming Floridians with your reckless, unacceptably risky product."

The same day, OpenAI revealed it had chosen not to release its upcoming frontier model, GPT-6.1 Astra, after determining that it did not yet meet its own safety standards. The Wall Street Journal first reported the decision, saying OpenAI scrapped release plans after its own researchers raised concerns about the model's behavior during internal testing. Observers said it was one of the clearest signs yet that misbehaving AI agents may derail the industry's rapid progress. "Of course we want to make sure our model development is safe no matter whether that's in the company or when we ship them to users," Saachi Jain, OpenAI's head of safety systems, told the Journal. "But when we ship it to users, we have an extremely high bar in terms of safety and alignment."

Meanwhile, researchers at OpenAI, Anthropic PBC, Meta Platforms Inc. and Microsoft Corp. asked policymakers to assess the degree to which their companies are automating research into the technology, in the latest industry call for greater government oversight. In a paper published Monday co-authored by OpenAI Chief Scientist Jakub Pachocki, Anthropic co-founder Jack Clark, Microsoft Chief Scientific Officer Eric Horvitz and Meta Vice President of AI Research Dawn Song, the researchers explained that using AI to accelerate research risks creating an "intelligence explosion."