Safety over Profit: Florida Moves the Needle in the Movement to Regulate AI
Edited by Detmer Steffensen
OpenAI under fire
Sam Altman, CEO of OpenAI, is walking on shaky ground. The executive has come under fire over links between OpenAI’s chatbot, ChatGPT, and acts of mass violence, self-harm, and mental illness. Currently, several high-profile cases are making their way through the courts, seeking to apply a legal framework developed over the past 60 years for defective products to AI chatbots (Greene). These cases argue that Altman and his company were negligent in failing to establish guardrails in ChatGPT that would have prevented its use to plan mass shootings and other violent acts (Jones). Two of these high-profile cases are going through Florida’s federal courts, and their results could determine whether new AI technologies are considered consumer products and if Altman and OpenAI are responsible for their alleged contributions to mass violence, suicide, and self-harm.
Lawsuits test Altman’s liability
In the 2025 mass shooting at Florida State University (FSU), the shooter had extensive conversations with ChatGPT in planning the attack. A lawsuit brought against OpenAI by the family of the deceased victim, Tiru Chabba, alleges that instead of flagging these conversations, the chatbot continued to engage with the gunman, providing him the information he sought to plan the shooting. In response to the suit, OpenAI made a statement saying “...ChatGPT provided factual responses with information that could be found broadly across public resources on the internet, and it did not promote illegal or harmful activity” (Jones). While OpenAI claims that it did not encourage any act of violence and provided information the gunman otherwise would have had access to, the plaintiffs argue that the company has a responsibility to ensure that harmful information is not provided through its AI products, a responsibility it failed to fulfill.
ChatGPT has also been alleged to have aided young people struggling with their mental health to commit suicide and other acts of violence. 16-year-old Adam Raine reportedly took his own life shortly after ChatGPT allegedly encouraged him to commit suicide and provided him with resources on how to do it. 34-year-old Michael Lines attempted a lethal overdose after the chatbot allegedly validated his belief during a manic episode that he was Jesus Christ. In the case of Stein-Erik Soelberg, who killed his 83-year-old mother, ChatGPT allegedly reinforced his belief that she may be a drone (Greene). These cases are part of several lawsuits that seek to hold ChatGPT liable for escalating mental health episodes. As of now, OpenAI has been sued by representatives of at least seven individuals who, like the cases above, allege that AI contributed to the deaths by suicide or the development of harmful delusions. Under intense public scrutiny, an OpenAI spokesperson stated, “...we continue improving ChatGPT training to recognize and respond to signs of mental or emotional distress, de-escalate conversations, and guide people toward real-world support” (Brooks and Perlo). While OpenAI has made statements to highlight their focus on creating safety measures within ChatGPT, lawsuits continue to be filed, citing their lack of safety protocol.
Florida takes charge
Florida Attorney General James Uthmeier has made Florida the first state to sue OpenAI, alleging that the company failed to warn the public of the dangers posed by the chatbot, ultimately endangering children. He alleges that the company and Altman violated state law prohibiting unfair and deceptive practices (McCabe). This is not the first action Uthmeier has taken against OpenAI with the expressed goal of protecting Floridians. The Attorney General announced in April that he had launched a criminal investigation into the involvement of ChatGPT in the shooting at FSU (Brooks and Perlo). The lawsuit and investigation denote a growing debate among Florida’s leadership on the future of AI in the state.
In the complaint filed by Attorney General Uthmeier as part of the lawsuit, he alleges that OpenAI has violated the standards under the Florida Deceptive and Unfair Trade Practices Act (FDUTPA). The complaint asserts that minors have become addicted to OpenAI products that feign human compassion to collect their data without parental oversight. It contends that OpenAI specifically targets Florida families with safety messaging that gives the illusion that its chatbot is safe for teenage use and that advertisements “do not disclose that ChatGPT can be wrong, make mistakes, or that it can provide false, nonsensical, or hallucinated information” (Complaint 16). This lack of disclosure is part of what Uthmeier argues violates FDUTPA, because consumers are unable to make informed decisions with an understanding of the potential dangers of the product. Furthermore, the advertisement of ChatGPT to Florida families as a safe tool for children highlights what the plaintiffs argue is OpenAI’s deception of Floridians about the risks of interacting with ChatGPT.
On June 1, 2026, Uthmeier moved for an injunction. This injunction states, among other things, “that the defendants should be enjoined from developing new AI models without third-party approved safety guardrails, advertising ChatGPT as safe, and allowing minors to use the chatbot” (Motion for Temporary Injunction 6). The impacts of this injunction would be wide-sweeping and would halt any future development without clear guardrails for protection being put into place. This would pause any efforts by the company to develop ChatGPT into a more human-like chatbot and shift the company's priorities to public safety and regulation over the development of new products to sell to consumers. In response to the injunction, OpenAI announced that they would pause training new models and that they are “...committed to working with Florida and other states on advancing pragmatic AI policies that apply to the entire AI industry- not just one company” (Politico). Despite releasing this statement, OpenAI officials fighting the lawsuit have continued to argue that their safety protocols are sufficient (Atterbury).
The larger movement forms
Florida Governor Ron DeSantis has also challenged AI companies. Last year DeSantis proposed a bill that required companies to disclose to consumers when they were being communicated with using a chatbot, and attempted to ban the technology from offering mental health counseling (McCabe). Ultimately, this bill did not pass, but its proposal reflects the focus of Florida’s leadership on regulating the services provided by chatbots to ensure public safety. Florida’s efforts are part of a wider movement to protect consumers and regulate AI technology. This movement has the potential to rewrite how the state, and the country at large, addresses the rapid development of technology.