WIRED finds more than 400 apps and domains concealing deepfake features

A WIRED investigation published November 2 identified a network of more than 400 connected apps and domains that appeared to offer ordinary AI services but could load hidden abusive deepfake features. Researchers traced links from ads on Facebook and Instagram to apps presented as pet portrait tools, language tutors and fitness coaches. The total covers both apps and domains, not 400 separately verified app-store listings.

WIRED linked parts of the network to Chinese-language developers and Hong Kong-registered KeLing Technology Limited. The company did not answer its requests for comment. Apple told WIRED it removed 19 apps after the reporting team shared its findings; it said developers had switched the content supplied to approved apps after review. Meta said it had taken action against 11,600 ad accounts associated with the domains and blocked 209 apps from advertising. These are company-reported enforcement figures.

Apple’s response follows its updated rules against nudify and bait-and-switch apps. This is not a story about consensual adult AI expression. It is about deceptive distribution of tools for nonconsensual imagery — and the gap between a store’s review of an app and what that app later serves.