Obscene, Degrading, and/or Abusive Content

Mitigation strategy

1. Implement stringent safety guardrails and filtering mechanisms during the training and development of generative AI models, coupled with robust pre-deployment safety testing by designated bodies, to prevent the system from acquiring or being exploited to generate Child Sexual Abuse Material (CSAM) or Non-Consensual Intimate Images (NCII) at the source. 2. Mandate the adoption of Digital Content Transparency (DCT) techniques, such as provenance data tracking and digital watermarking, to enable the efficient identification and segregation of synthetic content from non-synthetic content, thereby enabling faster detection and removal of harmful outputs post-deployment. 3. Establish clear regulatory and legal frameworks that impose platform-level accountability for the dissemination of AI-generated CSAM and NCII, requiring technology providers to deploy adequate moderation tools, reporting mechanisms, and to enforce user policies that treat synthetic illegal content no differently than traditional illegal content.