WIRED Finds Hundreds of Hidden Deepfake Sex Apps Traded on Apple and Google Stores

WIRED found more than 400 connected apps and domains quietly pushing explicit AI sexual deepfake tools through ordinary-looking storefronts, with many of the apps available on Apple and Google before the companies removed them after the investigation. The network matters because it shows how easily harmful deepfake products can hide behind routine app listings, paid ads and familiar payment systems.

What WIRED found

The investigation traced a sprawling network of linked websites and apps that appeared designed to look harmless while directing users to hidden tools for generating explicit sexual images and videos from uploaded photos. According to WIRED, the apps presented themselves as things like pet portrait services, language lessons and workout coaching, while a buried web layer handled the sexually explicit outputs.

WIRED says the network was uncovered through analysis of thousands of Meta Facebook and Instagram ads carrying explicit deepfake content. Those ads linked users to apps in the Apple and Google app stores, and the ecosystem also supported payments through Apple Pay and Google Pay.

How the deception worked

The public-facing pitch was ordinary. The hidden function was not.

That split is central to why the investigation is significant. App store reviewers can see the listing, screenshots and stated purpose, but developers can later shift the back-end content source that powers the app. In this case, WIRED says the network used that gap to funnel users toward sexual deepfake generation tools without making the full nature of the service obvious at the storefront stage.

The report says some templates were labeled with violent sexual abuse themes, including “bedroom rape” and “threatening knife.”

“This is an organized, commercially run operation”

That was how Paul Bouchaud, lead researcher at AI Forensics, described the network to WIRED.

Why it matters

This is not just another moderation headache. The story cuts across platform governance, app store review, ad enforcement and the fast-growing market for AI-generated sexual content. It also raises a blunt question for the largest consumer platforms: how much harmful material can pass through their systems when the product looks benign at first glance?

For readers, the practical concern is exposure. A person could encounter one of these ads on a social platform, download what appears to be a routine utility app, and then be routed into explicit non-consensual imagery tools. For platforms, the reputational damage is obvious: even after policy bans, the infrastructure can keep resurfacing in new forms. For regulators, the case suggests that enforcement may need to move beyond one-off takedowns and toward wider scrutiny of the business model itself.

The investigation also points to a commercial reality behind the abuse. WIRED says the apps were not fringe one-offs. They were part of a network that had been publishing for around a year, with dozens of new apps appearing per week at times. That pace suggests a coordinated operation able to replace removed listings quickly and keep the pipeline moving.

By the numbers

How did the network spread so widely?

The short answer is that it used the ordinary mechanics of app distribution and advertising to hide an abusive product at scale. WIRED says the network surfaced through Meta ads, then routed users into Apple and Google app store listings, where the visible branding did not match the hidden functionality.

That combination matters. Meta ads can create demand and push users toward an app. App stores can give the app a veneer of legitimacy. Payment systems can then make the service feel frictionless. In other words, each layer can make the next layer look trustworthy enough to keep users moving.

WIRED says the infrastructure appears tied to Chinese developers and a Hong Kong entity called KeLing Technology Limited. The report adds that Meta says it banned KeLing Technology Limited from advertising in 2025 after a cease-and-desist dispute.

What the app stores did

Apple and Google both removed many of the identified apps after WIRED raised the issue, according to the report. Apple removed 19 apps after WIRED asked about the network and shared associated domains. Google removed most of the identified apps before being contacted, including DiffAI.

DiffAI is notable because WIRED says it had more than 100,000 Play Store downloads before removal. That kind of reach suggests the network was not hidden in a corner of the internet; it was visible enough to attract a substantial audience before takedown.

What Meta says it did

Meta told WIRED it took action against 11,600 ad accounts and blocked 209 apps from advertising on its platforms. The company also said it had removed more than two-thirds of the 3,000-plus associated ads that WIRED shared with it.

“We have strict rules against promoting nudify apps and work aggressively to keep this material off our services”

That was the statement from a Meta spokesperson, according to WIRED.

Still, WIRED says similar ads and apps were still appearing as the reporting was completed. That detail is important: it suggests enforcement can be reactive and uneven even when the companies are alerted, and that bad actors may be able to keep testing new variations faster than platforms can fully clear them.

Background: what are nudify apps?

Nudify apps are services that use AI to place people into fake sexual images or videos, often from uploaded photos. The category has drawn repeated criticism because it can be used to produce abusive or non-consensual material that looks convincing enough to harm the subject’s privacy, reputation or safety.

Apple and Google do review apps, but developers can later change backend content sources. That technical reality creates a gap between what a storefront listing promises and what the app can deliver after installation. Meta, meanwhile, has long faced criticism for allowing deceptive ads, including ads that link to harmful or misleading apps.

The WIRED report places this broader debate into a very specific setting: a large, coordinated network that used app-store legitimacy and paid advertising to deliver content most users would expect platforms to stop earlier.

What the investigation found on Google Play

Google Play emerged as a major distribution channel in the report. WIRED says it found 27 live nudify apps on Google Play in September, with more than 350,000 collective downloads. That figure alone indicates a meaningful audience, not a marginal experiment.

One app in particular, DiffAI, stands out because it crossed 100,000 downloads before removal. WIRED says Google removed most of the identified apps before being contacted, including DiffAI. The company’s removal step still came after the network had already reached users.

The broader point is less about any one app than about the churn. If dozens of new apps can appear each week at the peak, then removing a single listing does not necessarily shrink the network. It may simply clear one name from the shelf while the operation shifts to another.

A closer look at the timeline

The sequence reported by WIRED shows a network that evolved over months rather than days. Here is the chronology drawn from the investigation:

Who is behind it?

That is still one of the central open questions. WIRED says the app and domain infrastructure appears tied to Chinese developers and KeLing Technology Limited, a Hong Kong entity. But identifying the technical and business infrastructure is not the same as mapping every person involved, every shell entity or every payment channel.

Researchers quoted by WIRED argue the operation looks deliberate and commercial rather than amateur. The number of domains, the speed of new registrations and the repeated resurfacing of similar apps all point in that direction.

Yet the report also leaves room for more to emerge. Even with app removals and ad bans, the underlying network could be reconstituted under different names. That uncertainty is part of what makes the story more than a single enforcement event.

What happens next

The most immediate question is whether the network has actually been disrupted or simply forced to rename itself. WIRED says similar ads and apps were still appearing as the reporting was completed, which suggests the removal wave may not have fully ended the activity.

Watch four things in particular:

• whether Apple and Google remove any additional apps tied to the same domains or payment flows;
• whether Meta widens its enforcement against related ad accounts and linked pages;
• whether regulators or law enforcement open broader inquiries into the operators and revenue paths;
• whether the network reappears under new branding, including new model names or app labels.

The commercial logic is straightforward. If one app name gets flagged, another can be launched. If one ad account is shut down, another can be created. If one payment path is blocked, another can be tested. That is why the story is not only about content moderation but about system design and enforcement velocity.

What readers should take away

WIRED’s reporting shows that explicit AI sexual deepfake tools can be distributed through mainstream digital channels at scale, masked by ordinary branding and aided by the reach of major platforms. The companies involved have now removed many of the identified listings and ads, but the investigation suggests that the underlying problem is broader than any single takedown.

For the platforms, the case is another reminder that deceptive AI products can exploit the trust built into app stores and ad systems. For users, it is a warning that a familiar listing or ad is not always what it claims to be. And for lawmakers and regulators, the report raises the question of whether current rules and enforcement tools are strong enough for a market built to move faster than review systems can respond.