Eight US States Sue OpenAI Over Allegations of Child Harm, Deception

Eight US states have launched legal proceedings against OpenAI. The lawsuits, initiated by attorneys general, allege that the company's generative artificial intelligence models, including ChatGPT, harm children and deceive adult users. These actions contend that OpenAI has violated existing consumer protection statutes within these states.

Connecticut Attorney General William Tong confirmed his state's participation. The litigation addresses concerns about the potential for ChatGPT to expose children to inappropriate content. It also scrutinises the model's accuracy, citing instances of 'hallucinations' or the generation of false information. These issues form the basis of the states' claims regarding consumer deception.

OpenAI faces accusations of failing to implement adequate safeguards to prevent minors from encountering harmful material. The lawsuits further contend that the company has not been transparent about the limitations and potential inaccuracies of its AI outputs. These legal challenges highlight a developing regulatory landscape for generative AI technologies across the United States.

Specific allegations include the generation of non-consensual sexual imagery involving children. This particular claim underscores the severe nature of the child harm allegations. The legal actions seek to establish accountability for content generated by AI models and their impact on vulnerable populations.

The multistate effort aims to compel OpenAI to implement more robust content moderation and age verification systems. They also seek greater transparency regarding the training data and operational mechanisms of its AI. The lawsuits represent a significant move by state authorities to regulate AI development and deployment.

These legal actions raise questions about the extent of platform responsibility for AI-generated content. How will these state-level interventions influence the federal approach to AI regulation?