A report by the Edinburgh-based Childlight – Global Child Safety Institute says AI-related child abuse material had risen from about 4,700 in 2023 to more than 400,000 in 2025, though Childlight's Prof Debi Fry said improved detection and reporting may explain some of that increase.[1] The images, though fake, can be used to blackmail children for money or more images, the Guardian reported.[1]
Previous studies examined western Europe, according to the Guardian, which reported that the recorded rate of reports involving child sexual abuse material rose by 63% in the US between 2024 and 2025, against 18% in Canada.[1]
The IWF says LoRA, a fine-tuning technique for customising generative models with little skill or money, can produce realistic deepfakes of a specific child from as few as 20 images in as little as 15 minutes.[2]
Childlight's chief executive, Paul Stanfield, said governments are "too often working with an incomplete picture of where harm is happening and who is most at risk", the Guardian reported.[1]