Tokyo Police Arrest Man Over AI Sexual Deepfakes of Track Athletes
Tokyo Metropolitan Police arrested a 32-year-old office worker from Himeji on suspicion of using generative AI to create sexual deepfakes of female track athletes and posting them on social media. TBS News DIG and NHK reported the arrest on August 3 in Japan, including allegations involving images of adults and a photograph depicting a person when she was a child.
Tokyo Metropolitan Police arrested a 32-year-old office worker from Himeji, Hyogo Prefecture, over allegations that he used generative AI to create sexual deepfakes of female track athletes and posted them on social media. TBS News DIG and NHK separately reported the arrest on August 3 in Japan, citing investigative sources. The accusations have not been adjudicated.
What investigators allege
TBS reported that the suspect is accused of defamation after allegedly altering high-school track photographs of three women now in their 20s to make them appear nude, then posting the resulting images where other social-media users could view them. The outlet said investigators also arrested him on suspicion of violating Japan's child-pornography law in a separate allegation involving an old sports-day photograph that depicted a girl in junior high school.
According to TBS, a 17-year-old boy in Kagoshima requested that second image after obtaining photographs of female athletes from social media. Police plan to refer the teenager to prosecutors over his alleged involvement. NHK independently confirmed the central allegation that police arrested the Himeji office worker after images of girls and women in track uniforms were altered with generative AI and posted online.
TBS attributed several scale estimates to investigators and the suspect's reported statement: about 300 images allegedly altered, roughly 100 to 200 posted, and approximately 3,800 obscene images and videos found on his phone. Those figures remain law-enforcement allegations, not court findings. TBS also cited National Police Agency data showing 114 confirmed cases in 2025 in which people under 18 were victims of sexual deepfakes, with most victims in junior high or high school.
Why this matters for AI safety
The case shows how a low-cost image-generation service can be used in a wider abuse chain: obtaining real people's photographs, producing non-consensual sexual images, accepting requests, and distributing the results. For teams building image tools, the practical control surface extends beyond prompt filtering to real-person image editing, age-sensitive safeguards, repeat-use monitoring, reporting channels, and rapid evidence preservation. Those are LDS implications from the reported facts; neither outlet identified the service used or described its safeguards.
Key Points
- 1Tokyo Metropolitan Police arrested a 32-year-old Himeji office worker over allegations that he created and posted sexual deepfakes of female track athletes.
- 2TBS attributed estimates of about 300 altered images, 100 to 200 posts, and roughly 3,800 obscene files found on the suspect's phone to investigators and the suspect's reported statement.
- 3The reported abuse chain involved real-person photographs, an alleged request from a teenager, generative-image tooling, and social-media distribution.
Scoring Rationale
A documented arrest involving non-consensual AI-generated sexual imagery, alleged child-victim material, and social distribution presents high safety, privacy, and platform-governance relevance; all allegations remain explicitly attributed.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
