Hundreds of Ordinary-Looking AI Apps Hide Sexually Abusive Deepfake Features
A WIRED investigation has uncovered a network of more than 400 linked domains and apps that pose as everyday services while enabling explicit deepfake generation. Apple and Google have removed most of the apps after being asked about them.
An investigation by the technology magazine WIRED has uncovered a network of more than 400 linked domains and mobile apps that appear on Apple's App Store and Google Play as ordinary services but, under certain conditions, let users generate sexually explicit AI images and videos from uploaded photos. Researchers say the network has been active for about a year, raising questions about how well the platforms enforce their own bans on harmful deepfake content.
How it works
One of the apps, Viefur, presents itself as a pet portrait tool: for digital coins (about $4 buys 130), pet photos can be turned into whimsical portraits. Researchers found, however, that each download sends the server details such as the user's origin and language, and the server decides whether to show the harmless or the explicit version. One apparent trigger is downloading the app after clicking an ad on Meta's Facebook and Instagram. The explicit interface contains dozens of pornographic templates, including scenes depicting violence.
The websites' code also reveals the network itself, since it holds a list of all the domains and apps. Maël Le Touz, a threat researcher at security firm Infoblox, identified almost 500 linked domains.
Scale and those responsible
Analysis by Paul Bouchaud of the nonprofit AI Forensics indicates the apps began appearing in late 2025, with a surge in spring. In September, 27 such apps were live on Google Play, with more than 350,000 downloads combined. One app, DiffAI, was downloaded more than 100,000 times. Researchers link the network to Chinese developers and a Hong Kong-registered entity, KeLing Technology Limited, which did not respond to requests for comment but appeared to erase public links to the deepfake tools after being contacted.
Apple removed 19 apps and said developers changed the content source after approval. Google removed most of the apps before WIRED contacted it. Meta said it banned KeLing Technology Limited from advertising in 2025 and, after receiving WIRED's findings, took action against thousands of accounts and pages and blocked 209 apps from advertising. As the story was being finished, similar ads and apps were still being published.
