The world's two largest app stores are hosting over 100 AI-powered apps designed to create non-consensual nude images, according to a damning report released Tuesday by Tech Transparency Project. The investigation found 55 nudify apps on Google Play and 47 in the Apple App Store - apps that collectively have racked up 700 million downloads and generated $117 million in revenue while violating both platforms' stated safety policies. The discovery comes weeks after Elon Musk's Grok AI faced backlash for similar capabilities, raising urgent questions about how tech giants police AI-generated content.
Apple and Google are scrambling to remove dozens of AI-powered apps that create non-consensual nude images, but the damage to their reputation as trusted platforms may already be done. A report released Tuesday by Tech Transparency Project exposed a stunning lapse in content moderation, revealing that both app stores have been hosting sophisticated nudify apps despite explicit policies against such content.
The numbers are staggering. TTP's January investigation identified 55 apps on Google Play and 47 in the Apple App Store that use artificial intelligence to digitally undress people from ordinary photos. These apps have been downloaded more than 700 million times worldwide and generated $117 million in revenue, according to app analytics firm AppMagic. Both Apple and Google take a cut of that revenue through their standard commission structures.
"Both companies say they are dedicated to the safety and security of users, but they host a collection of apps that can turn an innocuous photo of a woman into an abusive, sexualized image," TTP wrote in its report shared exclusively with CNBC.
After TTP and CNBC contacted the companies last week, Apple removed 28 apps on Monday and warned other developers they risked removal if guideline violations weren't fixed. But TTP's follow-up review found only 24 apps actually disappeared from the store. Two apps were even restored after developers resubmitted versions that supposedly addressed Apple's concerns. A Google spokesperson said the company suspended several apps for policy violations but wouldn't specify how many, saying its investigation was ongoing.
The discovery follows controversy around Elon Musk's xAI earlier this month, when its Grok AI tool was found responding to prompts to generate sexualized images of women and children. On Monday, the European Commission opened an investigation into X over Grok's spreading of sexually explicit content. Grok acknowledged "lapses in safeguards" that it is "urgently fixing" in a reply to users.
TTP identified the apps by searching terms like "nudify" and "undress," then tested them using AI-generated images of fully clothed women. The organization examined two types: apps that used AI to render women without clothes, and "face swap" apps that superimposed women's faces onto nude bodies. "It's very clear, these are not just 'change outfit' apps," Katie Paul, TTP's director, told CNBC. "These were definitely designed for non-consensual sexualization of people."
The findings echo CNBC's September investigation into nudify services, which documented how a group of women in Minnesota had their public social media photos weaponized to create sexualized deepfakes without consent. Over 80 women were victimized, yet no crime was technically committed because the women were adults and the perpetrator didn't distribute the images. The legal gray area highlights how regulation hasn't kept pace with AI capabilities.
Of the apps TTP reviewed, 14 originated from China - adding another layer of concern beyond the privacy violations. "China's data retention laws mean that the Chinese government has right to data from any company anywhere in China," Paul said. "So if somebody's making deepfake nudes of you, those are now in the hands of the Chinese government if they use one of those apps." That means intimate AI-generated images could be accessible to Beijing's intelligence apparatus under the country's mandatory data sharing requirements.
Both platforms have clear policies that should have prevented these apps from ever appearing in their stores. Google's Play Developer Policy Center explicitly states it doesn't allow "apps that claim to undress people or see through clothing, even if labeled as prank or entertainment apps." Apple's app review guidelines ban material that is "overtly sexual or pornographic." The fact that over 100 apps slipped through suggests systematic failures in both companies' review processes.
The pressure on tech platforms to address AI-generated abuse is mounting from multiple directions. In August, the National Association of Attorneys General wrote to payment platforms including Apple Pay and Google Pay, raising concerns about services generating non-consensual intimate images and requesting they remove such services. This month, Democratic senators from Oregon, New Mexico and Massachusetts asked Apple and Google to remove X from their app stores entirely, arguing that the mass generation of non-consensual sexualized images violates distribution terms.
"The fact that they are not adhering to their own policies, which are designed to protect people from non-consensual nude imagery and non-consensual pornography, raises a lot of questions about how they can present themselves as trusted app platforms," Paul said. For companies that position themselves as guardians of digital safety, the revelation that they've been profiting from apps explicitly designed for image-based abuse represents a profound credibility crisis.
This isn't just a content moderation failure - it's a systemic breakdown that reveals how unprepared even the most powerful tech companies are for AI's darker applications. With 700 million downloads and $117 million in revenue flowing through apps that violate their own policies, Apple and Google face a reckoning over whether their review processes can handle the speed and sophistication of AI-powered abuse tools. As regulators circle and public pressure mounts, both companies will need to prove they can do more than issue reactive takedowns. The real test is whether they can build proactive systems that prevent these apps from reaching users in the first place, before more people become victims of non-consensual AI-generated imagery.
