A WIRED investigation uncovered a network comprising over 400 connected domains and apps, some disguised as mundane services like pet portrait generators or language tuition tools. These apps, which appear to be run by Chinese developers and a Hong Kong registered entity named KeLing Technology Limited, stealthily load web interfaces capable of generating explicit AI sexual images and videos using user-uploaded photos. The network was discovered after researchers investigated thousands of video ads containing explicit deepfake content published on Meta’s Facebook and Instagram.
These ads linked to the deceptive apps available in both Apple and Google’s app stores, allowing users to make payments via Apple and Google Pay. Paul Bouchaud, a lead researcher at the nonprofit AI Forensics, stated, “This is an organized, commercially run operation: One operator behind hundreds of disposable apps and domains, cloaked to pass App Store review and pushed through tens of thousands of Meta ads under fake payer names.” Bouchaud explained that when an application is downloaded, the server responds with instructions on whether to display benign content or load the explicit version, noting that one trigger appears to be downloading the app after clicking an associated ad on Meta’s platforms.
The apps had been downloaded more than 350,000 times collectively, and many were rated suitable for all age groups. The network reportedly published these deceptive apps for about a year, sometimes releasing dozens per week, despite platform policies prohibiting harmful deepfake content. Following WIRED’s inquiry, Apple removed 19 apps from its App Store, including the pet photo app Viefur.
