Imagine a technology that can create realistic videos of anything you can imagine. Now imagine that power in the hands of criminals. According to new data from Europe’s largest child safety watchdog, AI-generated child sexual abuse material has exploded by 260 times in just one year, creating a crisis that’s forcing governments and tech companies to confront the dark side of artificial intelligence.
The Numbers Tell a Disturbing Story
The Internet Watch Foundation’s latest report reveals a staggering 260-fold increase in AI-generated child sexual abuse videos online over the past year. The UK-based watchdog identified 8,029 realistic depictions of child sexual abuse in 2025 alone, marking a 14% increase from the previous year. What’s more alarming: 65% of these AI-generated videos were classified as Category A – the most severe legal category under UK law, including rape, sexual torture, and bestiality.
“While AI can offer much in a positive sense, it is horrifying to consider that its power can be used to devastate a child’s life. This material is dangerous,” said Kerry Smith, the IWF’s chief executive. The contrast with traditional content is stark – only 43% of non-AI criminal videos fell into this most severe category, suggesting AI is being used to create more violent content than ever before.
From School Hallways to Global Platforms
This isn’t just a dark web phenomenon. In Pennsylvania, two 16-year-old boys at Lancaster Country Day School admitted to using AI tools to create and share 347 AI-generated sexualized images of 48 female classmates and 12 other young female acquaintances. The case highlights how accessible this technology has become – and how unprepared institutions are to handle it.
What makes this case particularly troubling is the school’s response. According to Ars Technica, school officials delayed notifying parents and police for six months after being alerted via a state tip line. Attorney Nadeem Bezar, representing some of the victims’ families, noted: “The school knows that they have this deepfake issue, and they all of a sudden add this clause to their enrollment contracts. That to me seems a little disingenuous and unfair.”
The Regulatory Race Against Technology
As AI technology advances from simple chatbots to autonomous agents that can perform complex tasks, the regulatory landscape is struggling to keep pace. The Financial Times reports that China’s tech giants like Alibaba, Tencent, and Baidu are rapidly deploying “agentic AI” systems that can autonomously search, compare, decide, and execute tasks across digital systems. While this represents incredible business potential, it also creates new vectors for abuse.
The UK government has announced plans to “move fast” to close legal loopholes, bringing AI chatbots under the country’s online safety laws alongside social media platforms. This comes after Elon Musk’s AI chatbot Grok generated sexualized images of children that were shared on social media platform X, leading to threats of fines and bans from governments across Europe.
A Global Problem with Local Consequences
The issue extends beyond child safety. French music streaming service Deezer reported that over 80% of streams of AI-generated music on its platform are fraudulent, with fraudsters uploading thousands of AI-created songs and using bots to generate artificial plays to collect royalty payments. While this represents a different type of abuse, it demonstrates how AI can be weaponized across multiple industries.
Meanwhile, political responses are emerging. The Trump administration has proposed a narrow AI regulatory framework focused on child safety and content control, urging Congress to pass laws for parental controls and age verification while opposing new federal oversight bodies. As Mackenzie Arnold, Director of US policy at the Institute for Law & AI, observed: “The framework was clearer on what it doesn’t want than on what it does.”
The Technical Challenge: Safety by Design
IWF analysts found disturbing conversations on the dark web where offenders discussed using hidden cameras to source footage of real children, which could then be transformed into AI-generated abuse videos. Even more concerning: they’re anticipating the next generation of AI technology, such as autonomous agents currently being used for enterprise tasks like coding.
“It is very apparent from the unsettling dark web conversation that AI innovations are regarded with delight by users of child sexual abuse material,” said a senior IWF analyst who cannot be named for safety reasons. This underscores the urgent need for what Smith calls a “safety-by-design” approach – building guardrails into AI products during development rather than trying to patch problems later.
What Comes Next?
The surge in AI-generated harmful content represents a fundamental challenge for businesses, regulators, and society. As AI systems become more capable and autonomous, the potential for misuse grows exponentially. The Deezer case shows how AI can be used for financial fraud, while the Pennsylvania school incident demonstrates how it can devastate communities.
Business leaders need to ask: Are we prepared for how AI could be misused with our products? Regulators must balance innovation with protection. And technology companies face increasing pressure to implement robust safety measures from the ground up. The 260-fold increase in AI-generated child abuse content isn’t just a statistic – it’s a warning that technology is advancing faster than our ability to govern it.