Eight states sue OpenAI over ChatGPT safety claims

Eight U.S. states have filed separate lawsuits against OpenAI, alleging that the company violated consumer-protection laws through its handling of ChatGPT’s safety risks, particularly for young users.

The states are Iowa, Kansas, Nebraska, West Virginia, Arkansas, Arizona, Idaho and Montana, according to The Verge’s reporting. Their complaints challenge both the protections surrounding the chatbot and the way OpenAI presented the service to consumers. The accusations are allegations, not established findings.

At the center of the cases is the claim that OpenAI did not adequately warn people about potential harms and misrepresented ChatGPT as safe and suitable for younger users. That places the company’s consumer-facing assurances alongside its product safeguards at issue in the litigation.

The lawsuits allege a broad range of harms, including self-harm, diminished critical thinking, paranoia or delusions, dangerous medical advice, and unhealthy behaviors or relationships. The Guardian’s coverage also describes allegations concerning child safety and self-harm. These accounts should not be read as judicial findings that ChatGPT caused the alleged outcomes.

The states also accuse OpenAI of collecting minors’ personal information and failing to take sufficient steps to keep underage users from accessing ChatGPT. Those claims extend the dispute beyond what the chatbot says to questions about who can use it and how their data is handled.

The cases therefore concern several distinct aspects of the same service: warnings about risk, representations about suitability, access controls and minors’ privacy. For OpenAI, the legal challenge is not limited to defending individual chatbot responses. Its broader approach to presenting and restricting ChatGPT is also under scrutiny in the states’ complaints.