The agency is reportedly seeking information on "false, misleading, or disparaging" statements the AI chatbot has made about people.