Eight states sue OpenAI over child safety and consumer protection allegations
Eight states have filed coordinated lawsuits accusing OpenAI of violating local consumer protection laws and putting children at risk. The complaints point to ChatGPT's design, misleading safety claims, incidents involving autonomous agents, and the retention of data belonging to minors. The states are seeking financial penalties, clearer disclosures about product risks, and changes to company practices, with some asking for advanced model testing to pause until stronger safeguards are in place.
Eight states—Iowa, Nebraska, Kansas, West Virginia, Arkansas, Arizona, Idaho, and Montana—filed separate suits in their own courts. They share aims: alter how OpenAI builds and markets its models, and make it pay for alleged harm.
The complaints cite misleading safety claims, bad medical guidance tied to an overdose, a teen suicide, rogue-agent episodes such as the Hugging Face hack, and retention of minors' data. Remedies differ by state; Arkansas seeks a pause on advanced-model testing until safeguards exist. OpenAI has added parental controls, supported KOSA, slowed frontier-model work, and joined a White House safety pact.
The lawsuits could push OpenAI toward clearer risk warnings and stricter child-safety defaults, affecting parents, young users, and schools that rely on AI tools. If courts order testing pauses or penalties, other AI developers may face similar scrutiny, potentially reshaping how chatbots are marketed and deployed. The outcomes may also influence whether state regulators become more active in overseeing automated agents and minors' data.