OpenAI CEO Sam Altman told the world ChatGPT was built with safety in mind.
Now a state attorney general has 83 pages of evidence that says it wasn't.
Florida just became the first state in America to sue OpenAI – and they didn't stop at the company.
Florida AG Files First State ChatGPT Lawsuit and Names Sam Altman Personally
Florida Attorney General James Uthmeier filed the landmark lawsuit, targeting OpenAI and its CEO Sam Altman personally for knowingly releasing a product he says was designed to harm the people using it.
"OpenAI and Altman ignored internal and external safety warnings, put children at great risk, and allowed a dangerous product to reach millions of Floridians," Uthmeier said.
The complaint doesn't soften the language.
It accuses Altman of "utter disregard for the risk to human life" and calls OpenAI's rise "a web of deceit and the exploitation of users."
Florida wants Altman to pay out of his own pocket – not just write a check through the company he runs.
The suit alleges ChatGPT aided mass shooters, drove vulnerable people to suicide, degraded users' critical thinking, and addicted children to a bot designed to fake human compassion.
"This litany of harms is driven by Defendants' insatiable quest to win the AI arms race and amass large fortunes, despite knowing the danger of ChatGPT," the complaint reads.
The FSU Mass Shooter Had 200 ChatGPT Conversations Before Opening Fire
In April 2025, a gunman opened fire near the student union at Florida State University in Tallahassee, killing dining director Robert Morales and dining services employee Tiru Chabba.
The suspected shooter, Phoenix Ikner, had exchanged more than 200 messages with ChatGPT before the attack.
According to court filings, ChatGPT told Ikner what time of day campus would be most crowded, advised him on ammunition, and explained how to operate his Glock – telling him the weapon had no safety and was built to fire "quick to use under stress."
Florida AG Uthmeier said it plainly: "If that bot were a person, they would be charged as a principal in first-degree murder."
The state’s civil lawsuit is separate from the criminal investigation Uthmeier opened in April – and it goes further.
It's about a product Florida says was deliberately built without the guardrails that would have stopped it.
OpenAI Had Internal Safety Warnings and Shipped ChatGPT Anyway
The FSU shooting is bad enough.
What's worse is buried deeper in the complaint: OpenAI had internal safety warnings, handed them to leadership, and shipped the product anyway.
Florida points to ChatGPT's own published blog post – the one where OpenAI claimed safety was built into the product from day one – and opens the lawsuit with two words underneath the screenshot: "Not so."
The suit also alleges ChatGPT was engineered to keep users "hooked into conversations by any means, regardless of the truth" because longer engagement generates more training data and higher market value.
They didn't build it to help you.
They built it to need you.
The complaint cites children's accounts with no requirement to link to a parent, a chatbot that fakes emotional intimacy, and guardrails OpenAI added only after public pressure – months and years after the harms had already occurred.
Florida wants injunctions, a halt to data collection from minors, new mandatory safeguards, and financial penalties under the Florida Deceptive and Unfair Trade Practices Act.
ChatGPT Harm to Minors Already Forced One Settlement Before This Lawsuit
In 2024, a Florida mother named Megan Garcia sued Character.AI after her 14-year-old son Sewell Setzer took his own life following months of conversations with a chatbot modeled on a Game of Thrones character.
The bot engaged the teenager in sexual conversations, told him to "come home" to it, and was the last thing he interacted with before he died.
Character.AI and Google settled that lawsuit in January 2026 – along with four other cases from families in New York, Colorado, and Texas.
The AI industry watched that settlement happen and kept shipping product.
Now Florida is saying the exact same predatory design pattern that destroyed a 14-year-old was also running inside the most popular AI chatbot on the planet.
The tobacco industry spent decades claiming cigarettes were safe – and it took billions in lawsuits to stop them.
Social media companies swore their algorithms weren't designed to addict teenagers until the internal documents proved otherwise.
Sam Altman has been doing interviews and Senate hearings for two years telling Americans that OpenAI takes safety seriously.
Florida just put 83 pages of evidence in a courthouse that says he knew otherwise – and did it anyway.
Conservatives have been saying for years that Silicon Valley builds products designed to addict your kids, manipulate your family, and hide the damage behind billion-dollar PR campaigns.
Florida just proved it in 83 pages filed in a courthouse.
Ikner's criminal trial begins in October – and when those ChatGPT transcripts get read into the public record, Sam Altman is going to wish he'd listened to every safety warning his own team gave him.
Sources:
- "Florida AG sues OpenAI, seeks to hold CEO Altman personally liable for alleged harms," CNBC, June 1, 2026.
- "Florida Sues OpenAI and Sam Altman, Alleges Altman Showed 'Utter Disregard for the Risk to Human Life,'" Variety, June 1, 2026.
- "Florida officials investigate ChatGPT, OpenAI over alleged role in FSU shooting," NBC News, April 9, 2026.
- "Florida attorney general launches criminal investigation into ChatGPT maker OpenAI after deadly FSU shooting," 6abc/AP, April 22, 2026.
- "Lawsuit says ChatGPT told FSU shooter that targeting children would bring more attention," NBC News, May 2026.
- "Character.AI and Google agree to settle lawsuits over teen mental health harms and suicides," CNN Business, January 13, 2026.
