The latest Childlight report has sounded an alarm: reports of AI‑generated child sexual abuse material have surged past 400,000 in 2025, an eight‑fold increase from roughly 4,700 two years earlier. This staggering rise, equivalent to an 8,400% jump, underscores how quickly malicious actors are weaponising deep‑fake technology to exploit the most vulnerable. The data, drawn from tech‑company disclosures to the US National Center for Missing and Exploited Children, paints a picture of a hidden crisis exploding across digital platforms. Experts warn that the speed of this escalation demands immediate, coordinated action from governments and the tech sector alike.
The rise is staggering.
The Shocking Numbers
The Edinburgh‑based Childlight – Global Child Safety Institute analysed more than 400,000 reports filed in 2025, up from about 4,700 in 2023. The material includes synthetic images created by “nudify” apps that transform ordinary photos into abusive depictions, as well as deepfake videos that never existed in reality. Such fabricated content can be leveraged for blackmail, demanding cash or further illicit images from children, turning virtual abuse into real‑world coercion. In the United States, the volume of reports climbed 63% between 2024 and 2025, while Canada saw a more modest 18% rise, highlighting uneven growth across the continent. The International Watch Foundation (IWF) reported that the first half of 2026 alone contained over 6,300 AI‑generated images that meet the legal definition of child sexual abuse – a 40% increase on the previous year’s total.
Geographic Hotspots and Offline Realities
Childlight’s research zeroed in on east Asia and the Pacific, Latin America and the Caribbean, and North America, regions where the surge is most pronounced, whereas earlier studies had focused on western Europe. In North America, almost one‑third of girls and more than one in ten boys experience in‑person sexual abuse before adulthood, including assault, rape and “upskirting”. Globally, an estimated 300 million children endure at least one form of technology‑facilitated sexual exploitation or abuse each year, and roughly one in four will suffer some type of abuse before turning 18. The IWF’s recent assessment of 6,310 AI images in early 2026 – 40% higher than 2025’s full‑year total – confirms that the problem is not confined to any single nation but is a worldwide phenomenon.

Urgent Calls for Policy and Innovation
Prof Debi Fry, Childlight’s global director of data, described the findings as “shocking” and warned that the rapid evolution of AI threatens children’s safety like never before. She urged governments to treat this issue as a preventable public health emergency at the centre of policy, invest in robust evidence‑gathering programmes, and translate intelligence into swift action. Paul Stanfield, chief executive of Childlight, added that millions of children are being exploited while authorities operate with an incomplete picture of where harm occurs and who is most at risk. “
