Meta Expands Nudify and AI Abuse Detection Tools Across Instagram and Facebook

Meta has announced the global rollout of detection technology designed to identify and remove AI-generated nude imagery, including deepfakes created by nudification applications, on Instagram and Facebook, alongside expanded notification systems that alert users when their images may have been manipulated and shared without consent. The company reports that the tools use machine learning to detect synthetic intimate imagery at scale and will automatically blur suspected nude images in direct messages to minors, with additional controls enabling users to block senders and report the content for review.

Platform-level moderation typically intervenes after content is posted or reported, leaving a window in which the victim has already been exposed, and reliance on user reporting places the burden on the child to recognise and escalate the abuse. Guardii closes that gap by detecting the hostile contact before the image reaches the recipient: its anti-sextortion and synthetic-CSAM filters analyse message metadata and conversation dynamics across Instagram, Snapchat and other platforms to identify coercion, solicitation and image-based abuse in progress, blocking or flagging the approach in real time and surfacing the child to a parent, school counsellor or emergency responder when the threat escalates, detecting patterns rather than reading content.