Internet Watch Foundation Reports 26,385% Increase in AI-Generated Child Sexual Abuse Videos in 2025
Guardii Analysis
The Internet Watch Foundation assessed 8,029 AI-generated images and videos depicting child sexual abuse in 2025, including 3,443 videos representing a 26,385 percent increase on the previous year. Sixty-five percent of videos were classified as Category A under UK law, the most severe level depicting extreme sexual violence, compared to 43 percent of non-AI videos, and the material was found across both dark web and mainstream platforms including chatbots. AI-generated material is now highly photorealistic and difficult to distinguish from real abuse imagery, with fine-tuning techniques enabling bespoke content of identifiable children with minimal expertise.
AI-driven anti-CSAM detection can intercept this material before it reaches distribution channels, but the spike reveals a structural challenge: generative AI allows offenders to manufacture severe abuse imagery at industrial scale without creating direct documentary evidence of contact offending. As a Meta Business Partner, Guardii deploys specialised detection modules that flag AI-generated and deepfake CSAM in direct-message environments across Instagram, Snapchat, Discord and Roblox, identifying synthetic abuse material by analysing the coercive pattern and context of the exchange rather than relying on static image databases that AI-generated content easily evades, and escalating the threat to the appropriate authority in real time.