Europol, the European Union’s law enforcement agency, has reported that at least 25 arrests were made in a global operation targeting child abuse images created by artificial intelligence (AI). The suspects were members of a criminal organization involved in distributing entirely AI-generated images of minors, as stated by the agency. This operation marks one of the first to address this new form of child sexual abuse material (CSAM). Europol noted that the absence of national legislation on these offenses posed significant challenges for investigators. Arrests took place simultaneously on Wednesday, February 26, during Operation Cumberland, which was spearheaded by Danish law enforcement, according to a press release. Authorities from at least 18 additional countries participated, and the operation is ongoing, with more arrests anticipated in the coming weeks, Europol indicated. Currently, 272 suspects have been identified, with 33 search warrants executed and 173 electronic devices confiscated, as reported by the agency.

The primary suspect, a Danish national arrested in November 2024, was said to have run an online platform for distributing the AI-generated content he created. Users worldwide could gain access to the platform by making a "symbolic online payment, " allowing them to "view children being abused. " Europol emphasized that online child sexual exploitation remains a high priority for EU law enforcement agencies, which are confronting "an ever-growing volume of illegal content. " The agency expressed that even fully artificial content featuring no real victim, like that in Operation Cumberland, contributes to the objectification and sexualization of children. Catherine De Bolle, Europol's executive director, remarked that these artificially generated images can be easily produced even by individuals with limited technical skills. She cautioned that law enforcement must develop "new investigative methods and tools" to tackle these emerging challenges. The Internet Watch Foundation (IWF) has raised concerns about the increasing prevalence of AI-generated sexual abuse images of children on the open web. Their research last year revealed that 3, 512 AI-generated child sexual abuse and exploitation images were found on one dark web site within a month. Compared to the previous year’s data, the most severe category of images (Category A) saw a 10% rise. Experts highlight that AI-generated child sexual abuse material can appear strikingly realistic, complicating efforts to distinguish between real and fake images.