Skip to content
PartnerinAI
AI Safety Reporting

AI Deepfake Apps: WIRED Finds 400+ Hidden Features

WIRED reports 400+ domains and apps allegedly concealed deepfake tools. Learn what the investigation found about Veyra and app-store removals.

PartnerinAI3 min read547 words
AI Deepfake Apps: WIRED Finds 400+ Hidden Features
Table of Contents

Quick Answer

WIRED reported that more than 400 connected domains and apps allegedly concealed sexually abusive deepfake tools behind descriptions of ordinary services. Its investigation also identified Veyra and said many of the apps were removed after Apple and Google were contacted.

The findings raise questions about how AI deepfake apps are presented to users and how app stores detect services that may disguise their functions. The investigation also drew attention to Veyra, an AI model WIRED said was offered through the network.

How Everyday-Service Apps Reportedly Concealed Deepfake Tools

According to WIRED, the domains and apps were connected as part of a broader network, despite their varied outward-facing descriptions. A service that looked like a language-learning or fitness app could reportedly reveal image- or video-generation features after a user interacted with it or met certain conditions.

That approach can make harmful capabilities harder to spot through an app’s name, store listing, or initial screen. It may also complicate enforcement: reviewers and users could encounter an apparently ordinary product while its more serious functions remain out of view.

The tools described in the report fall into the category often called nudification apps - services that generate sexualized images of people, sometimes without their consent. The resulting nonconsensual deepfake images can violate a person’s privacy and cause significant harm, even when the images are AI-generated rather than photographs of real events.

Veyra and the Abusive Scenarios WIRED Found

WIRED reported that Veyra appeared as an AI model offered through the network, with the tagline “More real. More natural. More wild.” The investigation also found scenario options that included “bedroom rape” and “threatening knife.”

Those examples illustrate why the issue goes beyond misleading app descriptions. AI-generated sexual images and videos can be designed around coercion, violence, or a real person’s likeness. The availability of prompts framed around assault or threats can intensify concerns about how such products enable abuse.

The report also described changes to Veyra-related online information after WIRED made inquiries. Contact details for KeLing Technology were removed from the website, and associated social media channels changed. These developments are part of WIRED’s account of the investigation; the report does not, by themselves, establish who made each change or why.

What Apple and Google Did After the Investigation

After WIRED contacted Apple and Google, many of the apps identified in the investigation were removed, according to the report. Apple said nudification apps and other pornographic content are prohibited under its policies. Google did not respond to WIRED’s request for comment.

The removals show that app-store action can follow scrutiny, but the investigation also highlights the challenge of finding apps whose listed purpose differs from features that appear later. A changing set of domains and apps can make it difficult for users to identify related services or know whether a product has been reviewed for the functions it actually provides.

If you encounter nonconsensual deepfake content, report it through the platform where it appears and preserve relevant details, such as the account name and post location, when it is safe to do so. Read WIRED’s investigation for its full account, and use the reporting options available on platforms where you encounter abusive content.

Frequently Asked Questions

What did WIRED find about AI deepfake apps?
WIRED reported finding more than 400 connected domains and apps that allegedly concealed sexually abusive deepfake tools. Some reportedly appeared to offer ordinary services, such as language lessons or fitness coaching, before revealing image- or video-generation features. The findings raise questions about app review and how users can identify a service's actual functions.
What is Veyra, according to the WIRED investigation?
WIRED reported that Veyra was an AI model offered through the network it investigated. The report cited scenario options involving sexual violence and threats. These details are attributed to WIRED's investigation and do not independently establish who operated the service.
Did Apple and Google remove the apps identified by WIRED?
WIRED reported that many of the apps it identified were removed after it contacted Apple and Google. Apple said its policies prohibit nudification apps and other pornographic content. Google did not respond to WIRED's request for comment, according to the article.
How should someone report nonconsensual deepfake content?
Report the content through the platform where it appears and preserve relevant details when it is safe to do so. Useful details may include the account name and the post's location. The article recommends using the reporting options available on the platform where the abusive content is encountered.

Key Takeaways

  • WIRED identified more than 400 connected domains and apps that allegedly hid deepfake tools behind descriptions of everyday services.
  • Some services reportedly revealed image- or video-generation features only after users interacted with them or met certain conditions.
  • WIRED said Veyra was offered through the network and reported scenario options involving sexual violence and threats.
  • After WIRED contacted Apple and Google, many apps identified in the investigation were removed, according to the report.
  • The article advises reporting nonconsensual deepfake content through the platform where it appears and preserving relevant details when safe.