Japan Makes First Arrests Over AI-Generated Obscene Images

Japanese authorities have arrested four individuals for allegedly selling obscene images created using generative artificial intelligence, marking the country’s first known crackdown on AI-generated explicit content.

According to a Tokyo Metropolitan Police spokesperson, the suspects were detained on Monday on charges of selling AI-generated obscene images online. The posters, created with free AI software, featured non-existent adult women in sexually explicit poses and were reportedly sold through online auction sites last October.

The suspects, aged between their 20s and 50s, face penalties of up to two years in prison, fines of up to 2.5 million yen (approximately $17,500), or both, if convicted.

Local media, including public broadcaster NHK, reported that the AI software was prompted with explicit keywords such as “legs open” to generate the illicit images. The posters were sold for several thousand yen each (roughly $20–$50).

While police did not immediately confirm whether these are the first arrests of their kind, Japanese media described the case as a landmark enforcement action amid growing global concerns about the misuse of generative AI.

The incident underscores increasing fears about the dark side of AI technologies, particularly the rise of deepfakes and non-consensual explicit content. Studies, including a 2019 report by Dutch AI firm Sensity, have found that an overwhelming 96 percent of deepfake content online involves non-consensual pornography, predominantly targeting women.

Authorities worldwide are grappling with how to regulate and respond to such emerging threats, and Japan’s latest move signals a tougher stance on AI-enabled obscenity.