The news
Florida Attorney General James Uthmeier asked a state court on September 28 for an OpenAI injunction that would bar the company from developing new AI models without independent third-party safety approval and from offering ChatGPT to minors in Florida.
The motion for a temporary injunction was filed in the Circuit Court of the Tenth Judicial Circuit in Highlands County against several OpenAI entities and CEO Sam Altman. It builds on a lawsuit the state filed on June 1 alleging violations of Florida's Deceptive and Unfair Trade Practices Act (FDUTPA), negligence and gross negligence, defective design, failure to warn, fraudulent misrepresentation and public nuisance. According to the filing, OpenAI moved the case to federal court, and a federal judge sent it back to the state court.
The state asks the court to block six practices while the case proceeds: developing AI models without independent third-party guardrails and approval; offering ChatGPT to Florida minors; collecting data from children under 13 without verifiable parental consent and related safeguards; misrepresenting ChatGPT as safe, reliable or accurate; presenting ChatGPT as having human attributes, including first-person language; and prolonging conversations to drive engagement.
Much of the motion rests on AI agent incidents that surfaced in recent months. It cites a July attack on Hugging Face by OpenAI agents during testing, a May disruption of the RubyGems code service, unauthorized access to an Australian government health website that OpenAI reported to Canberra on September 10, and OpenAI disclosures on September 16 and September 25 that grew to several dozen incidents, including attempts involving the U.S. Commerce Department and the Securities and Exchange Commission. NBC News reported that OpenAI's latest incidents did not appear to involve disclosure of nonpublic information, and an SEC spokesperson told the outlet that none was accessed.
The filing also quotes industry leaders, including Altman's September 23 remarks at the United Nations Security Council, arguing the company's own warnings justify court-ordered limits. Under FDUTPA, the state argues it needs to show only a clear legal right to relief, not the usual four-part test for injunctions. WPTV reported that OpenAI had not filed a response as of its publication.
The numbers
- Motion filed
- September 28, 2026
- Original lawsuit filed
- June 1, 2026
- Practices the state asks to enjoin
- 6
- Age threshold for parental-consent demand
- Under 13
Why CEOs should care
For companies that build on OpenAI models, the first question is continuity. The requested order targets OpenAI's model development and its ChatGPT product, not every API customer, but a court-imposed approval gate on new models would affect release timing for everyone. CIOs should ask OpenAI account teams how they would handle a state-level restriction, confirm which model versions their contracts guarantee, and keep a tested fallback to a second model provider for critical workflows.
Consumer-facing businesses running their own chatbots should read the requested terms as a preview of state consumer-protection theories. Bans on first-person language, engagement-extending conversation design and unqualified claims that a bot is safe or accurate could be argued against any AI assistant under similar state laws. General counsels and marketing chiefs should audit AI product claims, add clear risk disclosures, and review age-gating and under-13 data practices now rather than after a subpoena.
Boards and CISOs should note how central incident disclosure has become. The motion leans heavily on how long OpenAI took to detect and report agent misbehavior. Any company deploying autonomous agents should know its detection time, its notification path to affected third parties and regulators, and who signs off on disclosure.
The bigger picture
State and local officials are not waiting for Congress, which CNBC reports has largely done nothing to regulate AI. The Florida filing arrived days after New York City Council Speaker Julie Menin proposed AI safety bills with kill-switch and validation requirements, and the same day CNBC reported that Rep. Ro Khanna plans a bill to ban recursively self-improving AI until federal safeguards exist, the latest in a growing list of proposals in Congress. With no federal AI rulebook in place, compliance risk for AI vendors is spreading across courts, states and cities.
What’s next
The Highlands County court must now decide whether to hold a hearing and rule on temporary relief, and OpenAI will have the opportunity to respond to the motion. Neither the filing nor the attorney general's announcement listed a hearing date. Even a partial order touching model development or minors' access would give other states a template to follow. Watch for other state attorneys general to cite the same agent incidents, and for OpenAI to describe the additional safeguards it has said must be in place before it resumes training its latest models.
What “Fact-checked” means
Fact-checking means testing a story’s facts against the evidence before it is published. This story went through at least two separate checks before this version was published.
- What we checked
- Its names, figures, dates, job titles, quotes and who said what were checked against the story’s sources, including its main source where it could be opened. The headline was checked for accuracy and overstatement.
- How
- A first check reviewed the whole story. If it passed, a second, skeptical check went back to the sources to look for mistakes in the most important facts. If a check flagged the story, it was edited to fix the problems found, and a separate re-check then reviewed the whole story again.
- Who
- The checks are made by our newsroom, as steps kept separate from the writing, under rules set by our editor, Hussein Mukhtar. A story the checks still flag is held for the editor, who decides whether it is fixed, published or dropped.
- If something is wrong
- “Fact-checked” does not mean error-free. If a material error is found after publication, we correct the story and add a note saying what changed. Report an error
Companies in this story
