FBI and Homeland Security Warn of AI-Generated Child Sexual Abuse Material Surge in Joint Advisory

The United States Federal Bureau of Investigation and Department of Homeland Security issued a joint public safety advisory on 29 August 2024 warning that generative artificial intelligence tools are being used to produce, alter and distribute child sexual abuse material at scale, including realistic deepfake imagery created from innocuous photographs of real children. The agencies report that AI-generated CSAM is increasingly difficult to distinguish from authentic material, complicates victim identification, and is being weaponised in sextortion schemes where offenders threaten to distribute fabricated explicit images unless demands are met.

AI-generated and deepfake child sexual abuse material introduces a detection challenge that keyword filters and static image hashes cannot address, as the content is novel and the threat often materialises within a private conversation rather than in public posts. Detection systems purpose-built for CSAM that incorporate both image analysis and conversational context monitoring can identify solicitation, coercion and distribution attempts as they occur in direct messages across platforms including Instagram, Snapchat, Discord and Roblox, flagging both the exchange and the synthetic material before it is weaponised or shared beyond the original conversation, and enabling law enforcement to respond to live threats rather than retrospective reports.