OpenAI releases a Children’s Safety Policy Blueprint, focusing on three priority areas to strengthen child protection in the AI era

Gate News message, April 8, OpenAI released a policy blueprint aimed at strengthening the United States’ child protection framework in the AI era. The blueprint focuses on three major priority areas: updating laws to address child sexual abuse material (CSAM) that is generated or manipulated by AI, improving reporting and coordination mechanisms for service providers to support more effective investigations, and embedding safety design measures into AI systems to prevent and detect abusive behavior. In developing the blueprint, the United States incorporated feedback from the National Center for Missing & Exploited Children (NCMEC), the Attorney General Alliance, and its co-chairs of the AI working group, North Carolina Attorney General Jeff Jackson and Utah Attorney General Derek Brown, as well as multiple other institutions and experts, including Thorn. The co-chairs of the AI working group, Jeff Jackson and Derek Brown, said the blueprint is an important step toward aligning the tech industry’s child safety practices with law enforcement realities, and that effective protections for generative AI require a multi-layer defense system rather than a single technical approach. Karen White, Executive Director of the Attorney General Alliance, said the alliance is advancing practical solutions in the AI and digital safety space by bringing together forces from across the board.