More than 400 interconnected apps and websites have been linked to an operation involving AI-generated sexual deepfakes, raising new questions about how abusive tools can reach users through major technology platforms.
According to an investigation by WIRED, some applications disguised themselves as ordinary AI products while concealing features that could generate fabricated sexual imagery from photographs of real people.
The operation reportedly extended across Meta’s advertising platforms, Apple’s App Store and Google Play, directing users to applications that could unlock the hidden functionality.
Although Apple, Google and Meta keep taking enforcement action against identified applications, accounts and advertisements, the investigation suggests that removing individual listings and advertising accounts may not be enough to dismantle the infrastructure supporting the operation.
How these apps managed to bypass Google and Apple checks
The applications did not necessarily display their abusive functionality during app-store reviews. Instead, developers reportedly used connections to remote servers to control which features appeared to different users.
According to WIRED, some applications transmitted information such as location, language settings, and other identifiers to remote servers. Those servers could then determine which interface to display, allowing an application to present ordinary AI features to some users while exposing hidden sexual deepfake tools to others.
In some cases, installing an application after clicking an associated Meta advertisement reportedly triggered the concealed functionality. That meant that an application could present itself as compliant during the initial review process but behave differently after reaching users.
How Google and Apple’s checks failed to detect the code specifying post-installation behavior, and how these apps managed to change certain components of the code without triggering a recheck, remains unknown.
The investigation also uncovered shared code and infrastructure connecting hundreds of applications and websites. Investigations found that the apps were linked to Chinese developers and KeLing Technology Limited, a Hong Kong-based company. The company has reportedly cut ties with the apps after the report emerged.
More must-read AI coverage
- SS&C Intralinks DealCentre AI vs. Datasite: Which platform is built for the future of dealmaking?
- SS&C Intralinks FundCentre AI vs. Juniper Square: Which platform better supports modern private markets fund managers?
- Why Data, Not Models, Determines AI Success
- The Rise of the AI-Native Factory: How Physical AI Is Transforming Manufacturing
Apple and Google respond as app review practices face scrutiny
Apple and Google have defended their application review processes, pointing to enforcement measures taken against developers distributing sexually abusive AI applications. Apple attributed the problem to developers who allegedly changed their apps’ functionality after approval, replacing reviewed content with explicit material retrieved from external websites.
Google, meanwhile, told Android Authority that the applications identified in WIRED’s investigation had already been proactively removed from Google Play before the report emerged. The company also said it had suspended hundreds of additional applications for violating its policies.
In a statement to Android Authority, Dan Jackson, a spokesperson for Google, said that the company prohibits “apps that contain sexual content or allow users to ‘nudify’ images of people.”
The enforcement actions show that Apple and Google are continuing to remove applications that violate their policies. However, the investigation also raises questions about whether reviewing and removing individual apps can effectively disrupt networks that distribute similar functionality across multiple services and platforms.
Why users need more than app-store guarantees
Repeated reports of applications bypassing app-store checks are a reminder that users cannot rely entirely on Apple and Google’s approval processes to protect them.
Users should watch for suspicious applications, especially those advertised through social media, and report apps that hide their true functionality or promote nonconsensual image generation to their respective app stores.
Reporting suspicious advertisements on platforms such as Facebook and Instagram is equally important, as these channels can direct unsuspecting users to harmful applications. There is also a broader security concern.
If developers can repeatedly bypass app reviews to distribute AI nudify tools, they could exploit the same weaknesses to introduce other harmful functionality, including malicious code designed to compromise users’ devices or expose their data.
For Apple, Google, and other platform operators, the challenge is detecting deceptive behavior that may only appear under certain conditions and identifying connections between related applications, websites, and advertising campaigns.
The case illustrates why app-store protections must extend beyond initial approval, particularly when harmful functionality can be delivered through external infrastructure after an application reaches users.
In other cybersecurity news, suspected cyberattacks disrupted supermarket refrigeration systems across Europe, with attackers remotely raising freezer temperatures to as high as 50°C (122°F), forcing retailers to discard spoiled food and exposing vulnerabilities in internet-connected equipment.