OpenAI releases new policy blueprint to prevent digital child abuse

Framework targets AI-generated abuse content, strengthens reporting systems and embeds safety-by-design measures to detect and prevent misuse, initially focusing on child protection in the US

ChatGPT-maker OpenAI on Thursday introduced a policy blueprint to build a practical path to strengthen the digital protection framework for children, especially with respect to artificial intelligence (AI).

 

The policy blueprint, OpenAI said, will focus on modernising laws to address child sexual abuse material (CSAM) generated and altered by AI, improve reporting and coordination between organisations to support more effective investigations into misuse of AI for generating such content, as well as building safety-by-design measures directly into AI systems to prevent and detect misuse.

 

“This framework brings together legal, operational, and technical approaches to better identify risks, accelerate responses, and support accountability, while ensuring that enforcement authorities remain strong as technology evolves,” OpenAI said.

 

 

The policy blueprint will initially help protect children in the US, OpenAI said. With this blueprint, OpenAI and its partners will attempt to disrupt efforts that use AI to exploit children and create synthetic images or videos.

 

Though cybercriminals have always used digital services for the sexual abuse of children, generative AI introduced new challenges, such as synthetic images and videos that cybercriminals can create or manipulate even without direct access to the victim. This, OpenAI said, creates new challenges for traditional legal and investigative models.

 

“These developments strain investigative resources and expose gaps in statutory definitions, reporting expectations, and prevention mechanisms, while also creating opportunities for earlier detection and disruption when paired with clear safeguards and standards,” OpenAI said.