Meta has taken legal action against Joy Timeline HK, the creator of the AI-based nudify app, Crush AI, which allegedly ran thousands of ads across Meta’s platforms. In addition to the lawsuit filed in Hong Kong, Meta is implementing new measures to combat similar applications.
The lawsuit claims that Joy Timeline HK sought to bypass Meta’s ad review process to promote services related to AI nudifying. Despite Meta removing numerous ads for violating its policies, the entity reportedly persisted in placing additional advertisements.
Crush AI employs generative AI to create unauthorized sexually explicit images of individuals and is said to have launched over 8,000 ads for its services within the first two weeks of 2025. According to Alexios Mantzarlis, the author of the Faked Up newsletter, almost 90% of Crush AI’s web traffic came from Facebook or Instagram.
To evade Meta’s ad review mechanisms, Crush AI reportedly set up numerous ad accounts and frequently altered domain names. Many accounts were suspiciously named variations of phrases like “Eraser Anyones Clothes.” At one point, there was even a Facebook page promoting Crush AI’s services.
The issue of moderation extends beyond Meta, as other platforms like X and TikTok also grapple with the challenges posed by generative AI, especially concerning user safety, particularly for minors. As evidence of rising concerns, researchers noted a surge in links to AI nudifying apps on platforms such as X and Reddit in 2024.
In response, Meta stated that it has developed advanced technology to identify ads for AI nudify services, even if they do not depict nudity. The company is expanding its existing systems to quickly detect and remove deceptive ads, having disrupted four separate networks promoting these services since early 2025.
Additionally, Meta plans to collaborate with the Tech Coalition’s Lantern program to share information on AI nudify apps and has provided over 3,800 URLs to assist in combating online exploitation.