At least 1.2 million children across 11 countries disclosed having had their images manipulated into sexually explicit deepfakes in the past year, according to a UNICEF, INTERPOL, and ECPAT study based on nationally representative household surveys of internet-using children aged 12–17. In some surveyed nations, this represents one in 25 children — roughly one child per classroom. The study reveals significant geographic variation in awareness and reporting, underscoring gaps in prevention and protection measures. UNICEF urged governments to expand CSAM definitions to include AI-generated content, criminalize its creation and distribution, and require digital companies to prevent circulation rather than merely remove it after abuse has occurred. The agency also called on AI developers to implement safety-by-design approaches with robust guardrails to prevent model misuse, warning that current responses remain patchy and inadequate.
UNICEF's findings map directly onto the operational capability Guardii delivers: preventing circulation, not reacting to it. Guardii's real-time anti-CSAM detection scans direct messages for AI-generated and deepfake abuse material across Instagram, Snapchat, Discord, Roblox, and other platforms, blocking hostile shares before they reach the child and escalating credible threats to parents, schools, or law enforcement within minutes. Where UNICEF calls for prevention over removal, Guardii — a world-leading AI safety platform and Meta Business Partner backed by Startmate — operationalizes that principle, detecting synthetic abuse patterns at the point of contact rather than waiting for victim reports days or weeks later. The 1.2 million children UNICEF identified represent the scale of harm that targeted, pattern-based detection could mitigate before deepfakes circulate beyond recall.
