Harm without limits: AI child sexual abuse material through the eyes of our analysts
IWF analysts' report on AI-generated child sexual abuse material assessed during 2025, centring frontline-analyst perspectives and offender-community observations. Documents a step-change in AI-generated CSAM volume and severity and the tooling (including fine-tuning) that enables realistic abuse imagery.
Publisher
Internet Watch Foundation (IWF)
Published
24 Mar 2026
Added
3 months ago
DOI
—
Key Findings
- 8,029 AI-generated images and videos assessed as realistic child sexual abuse in 2025, including 3,443 AI-generated videos (up from 13 in 2024)
- 65% of the AI-generated videos were assessed as Category A (most severe); 97% of subjects were girls
- Documents AI services and tools that let offenders produce realistic abuse imagery of a specific child from a small number of source photos
Methodology Notes
NGO analyst-observation report (published 24 March 2026). Based on IWF analysts' assessments of reported content during 2025; methodology is expert triage against UK legal categories rather than peer review. iwf.org.uk bot-blocks automated fetchers; report and figures verified via the IWF research landing page and news release.
Sources
IWF AI CSAM Report 2026 (research page)(opens in a new tab) (primary)
Report PDF(opens in a new tab)
IWF news release(opens in a new tab) (24 Mar 2026)
Archived snapshot (Wayback Machine)(opens in a new tab) — preserved against link rot
Tags
Cite This
APA
Internet Watch Foundation (IWF). (2026). Harm without limits: AI child sexual abuse material through the eyes of our analysts. https://www.iwf.org.uk/about-us/why-we-exist/our-research/how-ai-is-being-abused-to-create-child-sexual-abuse-imagery/
Related Insights
Deepfake Nudes & Young People: Navigating a New Frontier in Technology-facilitated Nonconsensual Sexual Abuse and Exploitation
Thorn · 3 Mar 2025
Sexual Extortion & Young People: Navigating Threats in Digital Environments
Thorn (with Burson Insights, Data & Intelligence) · 24 Jun 2025
Global Threat Assessment 2025: Preventing Technology-Facilitated Child Sexual Exploitation and Abuse
WeProtect Global Alliance (with Columbia University) · 11 Dec 2025
Uncensored AI in the Wild: Tracking Publicly Available and Locally Deployable LLMs
Future Internet (MDPI) · 18 Oct 2025