YouTube and X Drive Millions to AI 'Nudify' Deepfake Sites, Study Finds

By: Harsh Vardhan

|

Updated on: 15-Jul-2026 08:00 PM

5,594 views

Follow Us:

5,594 views

A recent study by the Institute for Strategic Dialogue (ISD) shows that YouTube and X directed millions of users to AI-powered 'nudify' websites. These sites allow people to create non-consensual intimate images, raising concerns about the role of mainstream platforms in spreading deepfake tools. The study examined user traffic between December 2025 and March 2026 and found that social media referrals remain a significant driver of visits to these services.

Key Highlights

• YouTube and X directed over 5.7 million visits to AI nudify websites in four months.
• YouTube accounted for 1.82 million visits, while X drove more than 1.3 million visits.
• Study found enforcement of platform policies against explicit deepfake tools remains inconsistent.
• Some AI apps offer explicit image generation for as little as $1 per image.

Key Findings from ISD Study

The ISD analyzed the ten most-used apps and websites for creating non-consensual explicit deepfakes. It found that social media platforms drove over 5.7 million visits to nudify websites during the four-month period. YouTube generated 1.82 million visits, accounting for more than 30 percent of all referrals. X was the second-largest source, sending over 1.3 million visits to these sites.

Users searching for terms like 'undress app' and 'nudify app' were shown videos that reviewed and promoted these tools. Some videos included direct links to the apps and promo codes for free credits, making access easier. The ISD report states that these findings conflict with YouTube's policies, which ban sexually explicit content and links to such websites. However, the enforcement of these rules appears inconsistent, allowing users to find and access nudify tools through the platform.

Platform Policies and Enforcement Issues

Melanie Smith, ISD UK's Senior Director of Policy and Research for Information Operations, told Wired that YouTube was not just a passive source but often facilitated the use of these tools. She noted that YouTube's policies prohibit links to or advertising for sexually explicit websites, which should include nudify services. Despite these policies, the study found that content promoting deepfake tools remained easy to find.

In response, YouTube spokesperson Boot Bullwinkle stated that the company has strict policies against unwanted sexualisation, including non-consensually shared intimate imagery. These policies apply to both hosted content and external links, and also cover altered or synthetic content that simulates nudity. However, the ISD report suggests that enforcement of these policies is lacking.

Accessibility and Impact of AI Deepfake Tools

The ISD study also explored the accessibility and profitability of these AI tools. Some apps allow users to generate explicit images for as little as $1 per image. Common targets include current and former girlfriends, and in some cases, relatives such as sisters and cousins. The study found that the motivation for using these tools is often not sexual. Many requests aim to harm reputations, get individuals fired, or compromise their livelihoods.

Smith emphasized that the use of deepfake technology is shifting from creating explicit images to harassment, intimidation, and reputational damage. This trend highlights the growing misuse of AI-powered tools for non-consensual and harmful purposes.