AI-generated child sexual abuse images are up 40% already in 2026, with first-half figures topping the ENTIRE 2025 total

From the Newshound archive · first published 5 October 2026 · last updated 6 October 2026.

The Internet Watch Foundation has assessed 6,310 AI-generated images meeting the legal definition of child sexual abuse in just the first half of 2026, already surpassing the 4,500 assessed across the whole of 2025. The majority showed girls aged between seven and 13, and the IWF is now pressing for binding legislation to force tech companies to build safer AI models. Andy Burnham's government has not signalled that a dedicated AI bill is imminent, despite calls from parliamentarians.

Children have already submitted 420 reports to the 'Report Remove' service about images manipulated to appear explicit — more than in the whole of 2025.

IWF policy chief Hannah Swirsky is calling for 'binding legislation on AI' to compel companies to produce safer models, saying tech firms must bear responsibility.

The IWF and National Crime Agency are now urging parents to keep their children's photos off social media entirely, as photorealistic abuse images proliferate online.