Meta to Scan Encrypted Messages for Child Sexual Abuse Material
Guardii Analysis
Meta announced it will deploy machine-learning systems to scan end-to-end encrypted messages on Facebook Messenger and Instagram for child sexual abuse material, marking a significant policy shift for the platform. The technology will analyse message metadata and behavioral patterns without decrypting message content, with the company stating the system will be deployed globally in the coming months and flagged content reported to the National Center for Missing and Exploited Children.
Pattern-based detection that operates without decrypting individual messages represents the only sustainable model for protecting children in encrypted environments at scale. Guardii's anti-CSAM and anti-grooming modules analyse the shape and escalation of conversations in real time across Instagram, Snapchat, Discord, Roblox and other platforms, blocking hostile contact before it reaches a child and surfacing acute risk to parents, schools or authorities without creating a feed of message content. Where platforms scan for known material, behavioral detection intercepts novel threats and grooming escalation as they unfold, addressing the harm before abuse is documented.