The National Center for Missing and Exploited Children has reported that submissions of suspected AI-generated child sexual abuse material to its CyberTipline tripled in the first half of 2024 compared to the previous six-month period, with the majority involving realistic deepfake imagery of identified minors. NCMEC warns that the barrier to creation has collapsed as generative AI tools become more accessible, and that traditional hash-matching techniques are ineffective against novel synthetic content that has never been catalogued.
Guardii's CSAM detection module is trained to identify both known and previously unseen child sexual abuse material, including AI-generated and deepfake imagery, by analysing visual and contextual markers rather than relying solely on hash databases. Deployed at the message layer across platforms including Instagram, Snapchat and Discord, the technology intercepts synthetic abuse material in real time as it is transmitted, blocking distribution before the target child is exposed and generating an alert that enables rapid escalation to the National Center for Missing and Exploited Children or law enforcement.
