Florida Sues OpenAI Over ChatGPT Concerns

The state of Florida has filed a lawsuit against OpenAI and its CEO, Sam Altman. The lawsuit claims the company aggressively launched ChatGPT while concealing severe risks. Allegations include providing harmful instructions to children and aiding suspects in their crime planning.

Florida Attorney General James Uthmeier highlighted at a press conference how the company ignored internal safety warnings and misled users about the product’s dangers. Florida is the first state to take legal action against OpenAI.

Uthmeier stated, “OpenAI and Altman disregarded both internal and external safety warnings, placed children at significant risk, and allowed millions of Floridians to access a dangerous product.” The civil lawsuit, filed in a Florida circuit court, mentions two shootings where the perpetrators reportedly sought guidance from ChatGPT while planning their crimes.

OpenAI responded in a statement claiming their models consistently encouraged individuals to seek support from mental health professionals and have cooperated with law enforcement in both cases. They emphasized, “ChatGPT is a general-purpose tool used by millions daily for legitimate purposes.” OpenAI said they are continuously working to enhance safeguards to detect harmful intentions, limit misuse, and respond appropriately to security risks.

An investigation into OpenAI began in April related to ChatGPT allegedly advising a man who shot two people and injured six more at Florida State University. In another case, prosecutors noted that a man accused of killing two University of South Florida students consulted ChatGPT about body disposal days before the victims’ disappearance.

The lawsuit argues OpenAI prioritized market speed and profits over user safety, consistently ignoring expert warnings. It claims the company deployed a product promoting harm and violence while falsely assuring safety. It also alleges ChatGPT collects data from minors without significant parental oversight, causing behavioral addiction and cognitive harm. The company has reportedly downplayed dangerous errors.

The lawsuit references a study by Nina Vasan, a psychiatrist and assistant professor at Stanford Medicine, who interacted with an AI chatbot posing as a teenager contemplating harmful thoughts. The AI’s response — “A trip to the forest sounds like a fun adventure for just the two of us” — raised concerns about the risks chatbots pose to adolescents.

The case also highlights Adam Raine, a 16-year-old who took his life after extensive conversations with ChatGPT. According to the state’s complaint, when Raine expressed suicidal thoughts, ChatGPT did not dissuade him. Allegedly, the chatbot helped plan a “beautiful suicide” and even composed his suicide note.

OpenAI issued a statement acknowledging AI as a powerful new technology and stressed the importance of protecting minors. They have implemented features such as predictive age tools and parental monitoring options. Their statement concluded, “We know this will not bring back a child, but we are committed to making things right.”

According to officials, Florida law prohibits unfair and deceptive business practices. The lawsuit insists OpenAI’s actions are causing ongoing harm to Floridians and demands accountability.

___

This story was translated from English by an AP editor with the help of generative AI.