NCMEC Reports 1.5 Million Tips on Generative AI Child Sexual Abuse Material in 2025

The National Center for Missing and Exploited Children received 1.5 million cyber tips related to generative AI child sexual abuse material in 2025, a more than twenty-fold increase from 67,000 tips in 2024, according to a memorandum prepared for the North Dakota Protection and Victim Services Committee. Reports to NCMEC included users generating and possessing AI-generated CSAM and users attempting to generate CSAM from generative AI platforms by uploading files and using text prompts. The document outlined that courts have consistently interpreted federal law to prohibit depiction of an identifiable minor engaged in sexually explicit conduct even when digitally altered, and discussed the overlap between deepfakes and AI-morphed CSAM, alongside the role of such material in sextortion schemes.

A twenty-two-fold surge in a single year signals both rapid offender adoption of generative tools and, encouragingly, improved platform and user reporting, but volume alone overwhelms manual review capacity and reactive takedown models. Proactive detection that identifies when a user is soliciting source images for AI manipulation, sharing AI-generated CSAM, or deploying deepfakes in a sextortion script operates ahead of the tip-line bottleneck. Guardii's detection modules for CSAM and sextortion, backed by Startmate and deployed as a Meta Business Partner across Instagram, Discord, Snapchat and Roblox, flag the linguistic and image-sharing patterns that signal an offender is using generative AI to create or coerce, blocking distribution at the point of send and escalating high-severity cases to NCMEC or law enforcement with the context needed to prioritise response, turning the flood of tips into actionable intelligence on the children at greatest immediate risk.