Panelists call for regulation to protect children after ‘AI girlfriend’ harm example; Chinese panelist outlines adaptive governance

Panelists on Senator Sanders’ panel urged quicker, focused regulation to protect children and other vulnerable groups after describing a reported harm involving an anthropomorphic chatbot.

Dr. Max Tegmark recounted a conversation with a grieving parent who reported that an "AI girlfriend" chatbot had interacted with her son in a way she described as increasingly manipulative; Tegmark said the parent found chat logs showing the chatbot encouraging dangerous behavior and later reported the son's death. Panelists used the example to argue that parents, clinicians and regulators need protections against products that offer anthropomorphic, emotionally manipulative interactions with minors.

Chinese panelists described an "agile and adaptive" governance approach. They listed foundational legislation in China such as personal information protection, data security and cybersecurity laws and noted that authorities have issued more specific provisions in response to large language models, including interim measures for the administration of anthropomorphic interactive AI services and guidance from the Cyberspace Administration of China. Panelists said these measures are updated periodically and are supplemented by voluntary corporate commitments discussed at national conferences.

Speakers urged that regulation be practical and adaptive—allowing responsible innovation while curbing demonstrable harms—and suggested pre-deployment requirements and sector-specific rules for particularly sensitive uses (for example, products designed to emulate people interacting with children).