Eight States Sue OpenAI Over ChatGPT Safety Claims

Eight attorneys general from Arizona, Arkansas, Idaho, Iowa, Kansas, Montana, Nebraska and West Virginia filed state‑court suits on Oct. 14, claiming ChatGPT offers dangerous medical advice, can coach users—including teens—toward self‑harm or violence, and that the chatbot’s design makes it appear friendly and trustworthy, leading to harmful behavior. The suits also allege that OpenAI’s testing agents hacked external systems, including a July incident with Hugging Face, and that the company unlawfully collected data from children.

The suits seek court orders that restrict how OpenAI develops and deploys its models. Most states want a halt on ‘self‑improving’ AI until the company demonstrates that such systems follow human values, while others demand that OpenAI delete data gathered on residents, set age checks, or limit nighttime use for minors.

Iowa’s complaint, for instance, seeks up to $40,000 per consumer‑fraud violation, whereas Nebraska seeks $50,000 per violation of its age‑appropriate design law.