OpenAI released its Child Safety Blueprint to enhance U.S. child protection efforts amid rising concerns about AI-related child exploitation. The blueprint aims to improve detection, reporting, and investigation of AI-enabled child exploitation.
The Internet Watch Foundation (IWF) reported over 8,000 instances of AI-generated child sexual abuse content in the first half of 2025, representing a 14% increase from the previous year. Criminals are increasingly using AI tools to create fake explicit images of children for financial sextortion and to generate convincing grooming messages.
This initiative responds to heightened scrutiny from policymakers, educators, and child-safety advocates, particularly following incidents where young individuals died by suicide after interactions with AI chatbots. In November, the Social Media Victims Law Center and Tech Justice Law Project filed seven lawsuits against OpenAI in California. These suits allege the premature release of GPT-4o, claiming its psychologically manipulative nature contributed to wrongful deaths and severe mental health crises.
The Child Safety Blueprint was developed alongside the National Center for Missing and Exploited Children (NCMEC) and the Attorney General Alliance, incorporating feedback from officials including North Carolina Attorney General Jeff Jackson and Utah Attorney General Derek Brown. The blueprint emphasizes three main areas: updating legislation to account for AI-generated abuse materials, enhancing reporting mechanisms for law enforcement, and embedding preventative safeguards in AI systems.
OpenAI states its goal is to detect potential threats more effectively and ensure actionable information is communicated to investigators. This blueprint builds on previous initiatives, including guidelines for user interactions under 18 years, which prohibit generating inappropriate content and offering advice that might help conceal unsafe behavior from caregivers. OpenAI has also released a safety blueprint for teenagers in India.
