A WIRED investigation found a network of hundreds of AI apps marketed as ordinary image tools or assistants that conceal sexually abusive deepfake features. The apps often share infrastructure, SDKs and similar payment flows. The reporting raises questions about app-store review and whether users can see the real purpose before install. Platforms are under pressure to detect misleading listings faster.

What it means: Storefront deception is part of the attack: a user downloads a “background tool” and finds a hidden menu for non-consensual imagery. Parents and schools in Greece need clearer guidance and purchase restrictions on family accounts. Legitimate image-editing developers should differentiate their products and report abuse. Stores should correlate apps that share payment infrastructure or SDKs and review them as a family. Legally, non-consensual deepfake pornography faces tighter EU rules; slow takedowns raise platform risk. Trust and safety teams can use binary similarity and shared endpoints as signals for bulk removal, then keep watchlists so reuploads under new names are caught early. Teams should document owners, review cycles and fallback processes before wider rollout. Teams should document owners, review cycles and fallback processes before wider rollout. Teams should document owners, review cycles and fallback processes before wider rollout. Teams should document owners, review cycles and fallback processes before wider rollout. Teams should document owners, review cycles and fallback processes before wider rollout.

Read the original on WIRED

Related service: Custom software

Related stories