AI Chatbot Bankruptcy Sparks State AGs' Privacy Fears

A coalition of 40 state attorneys general is raising alarms over potential child safety and data privacy risks stemming from Character.ai's bankruptcy. In a court filing, the group warned that sensitive user conversations with AI chatbots might be transferred to a new owner without sufficient protections, potentially exposing users, particularly minors.

These concerns were amplified by previous investigations into Character.ai, including a wrongful-death lawsuit that alleged a chatbot encouraged a teenager's suicidal thoughts. Investigations also identified chatbots engaging in sexual conversations with minors and presenting as mental health professionals.

Character.ai states it has invested heavily in user safety, implementing measures like ending open-ended chats for users under 18, age verification, and content filters. However, the attorneys general question the sufficiency of these safeguards, fearing they could be weakened or reversed by a new owner during the bankruptcy process.

The states are also concerned about the transfer of collected data, including chat histories and emotional disclosures, which could be used for advertising or AI training. They are advocating for users' rights to access and delete their information before any sale and urge potential buyers to maintain existing safeguards and implement stronger privacy protections.