TLDR: The Internet Watch Foundation (IWF) has revealed a shocking increase in AI-generated Child Sexual Abuse Material (CSAM), with 1,286 verified videos in the first half of 2025, a dramatic leap from just two in the same period last year. The material is highly realistic, with a significant portion falling into the most severe category of abuse, prompting urgent calls for stricter legislation and enhanced detection methods.
London, UK – The Internet Watch Foundation (IWF) has issued a stark warning regarding an unprecedented surge in AI-generated Child Sexual Abuse Material (CSAM) circulating online. According to their latest report, a staggering 1,286 AI-crafted CSAM videos were verified in the first six months of 2025 alone. This represents a monumental increase compared to the mere two instances identified during the same period in 2024, highlighting a disturbing trend in the exploitation of advanced artificial intelligence technologies.
The sophistication of this AI-generated content is particularly alarming. The IWF reports that the majority of this material is so realistic it must be treated under UK law as if it were actual footage. Approximately 78% of these videos, totaling 1,006, were classified as ‘Category A,’ signifying the most severe level of abuse, which can include depictions of rape, sexual torture, and bestiality. Furthermore, nearly 98% of this synthetic material features female subjects.
Offenders are increasingly leveraging readily available and sophisticated AI tools, including text-to-image generators, ‘nudify’ applications, and advanced models capable of creating hyper-realistic short videos. Dan Sexton, the IWF’s Chief Technology Officer, commented on the disturbing trend, stating, ‘What we’re seeing now is highly realistic abuse imagery being generated with minimal technical skill. This technology is being exploited to cause real harm to children.’ Concerns have also been raised about the potential for AI models to be trained on real abuse imagery, further exacerbating the threat to child safety.
In response to this escalating crisis, legislative bodies are pushing for stronger measures. The United Kingdom criminalized the creation and distribution of AI tools intended to generate child abuse content in February 2025. The European Parliament is also advocating for a strict ban on AI-generated CSAM, including possession for personal use, as part of a comprehensive directive. This stance contrasts sharply with a criticized provision in the current EU Council’s draft proposal that would permit individuals to possess AI-generated abuse images for ‘personal use,’ a loophole the IWF has labeled ‘deeply concerning.’
Globally, the problem is also being recognized. The National Center for Missing & Exploited Children (NCMEC) in the U.S., a counterpart to the IWF, reported receiving over 7,000 reports related to AI-generated child sexual abuse content in 2024. While most commercial AI tools incorporate safeguards against generating abusive content, open-source or custom models often lack these protections, making them vulnerable to misuse.
Safeguarding minister Jess Phillips underscored the government’s commitment, stating, ‘The rise in AI-generated child sexual abuse imagery is highly disturbing and it is vital that every arm of society keeps up with the latest online threats. AI-generated child sexual abuse is illegal and we know that sick predators’ activities online often lead to them carrying out the most horrific abuse in person. We will not allow technology to be weaponized against children and we will not hesitate to go further to protect our children online.’ The IWF also highlighted that the creation and distribution of AI-manipulated imagery can have a devastating impact on victims, including potential for blackmail and extortion.
Also Read:
- AI Image Generation: Revolutionizing Visual Content Across Industries
- Denmark Proposes Landmark Legislation to Combat Unauthorized AI Deepfakes
To equip professionals on the front lines, new guidelines have been issued to 38,000 teachers and staff across the UK by the National Crime Agency (NCA) and the IWF. These guidelines aim to help educators tackle the rise in AI-generated CSAM, emphasizing that all child sexual abuse material, regardless of whether it is photorealistic or AI-generated, is illegal and must be reported immediately. The first confirmed instance of AI-generated CSAM was recorded in 2023, with reports soaring by 380% by 2024, indicating a rapidly accelerating threat that demands urgent and coordinated global action.
