A comprehensive study released by the Institute for Strategic Dialogue (ISD) has exposed a disturbing trend in the digital ecosystem, revealing that mainstream social media platforms are serving as primary conduits for tools that generate nonconsensual intimate images (NCII). While major tech companies have long maintained public stances against the dissemination of sexually explicit content and "revenge porn," the ISD report suggests that their search algorithms and content moderation systems are being successfully bypassed or ignored by a predatory industry of "nudify" applications. These tools, which utilize artificial intelligence to digitally remove clothing from photos of individuals without their consent, have transitioned from the dark corners of the internet into a multimillion-dollar mainstream market, largely facilitated by platforms like YouTube and X (formerly Twitter).
The ISD, an international anti-extremism organization, focused its research on the top 10 most prominent "nudify" websites and apps, tracking how users discover and access these services. Contrary to the common assumption that such traffic originates in unmoderated fringe forums like 4chan or encrypted Telegram channels, the data indicates that the gateway to this exploitative technology is often a simple search on a major video-sharing or social networking site.
Quantifying the Traffic: The Role of Mainstream Platforms
The scale of the referral traffic identified in the report is significant, pointing to a systemic failure in platform oversight. Between December 2025 and March 2026, social media networks were responsible for driving more than 5.7 million visits to the leading nudify websites. This influx of traffic represents a critical pipeline for the AI-generated pornography industry, which relies on high volumes of new users to sustain its credit-based revenue models.
YouTube emerged as the single largest contributor to this ecosystem. According to the ISD’s findings, the Google-owned platform accounted for 1.82 million site visits, representing over 30 percent of the total referrals tracked during the study period. This traffic was not incidental; it was the direct result of users searching for specific keywords such as “undress app” or “nudify app.” In many instances, the search results provided users with instructional videos, reviews of the tools’ effectiveness, and even promotional codes that offered free credits to first-time users.
X followed as the second most prominent source of referral traffic, contributing approximately 1.3 million visits. The report highlights that the decentralized and often less-stringent moderation environment on X has allowed for the proliferation of direct links to these services, often accompanied by "before and after" examples designed to showcase the AI’s capabilities. Together, these two platforms constitute a significant portion of the "discovery phase" for individuals seeking to create nonconsensual imagery.
A Chronology of the Deepfake and Nudify Crisis
The current proliferation of nudify tools is the latest stage in a decade-long evolution of synthetic media exploitation. Understanding the current crisis requires a look at the timeline of how this technology became commercialized:
- 2017: The Emergence of "Deepfakes": The term was coined on Reddit when a user began swapping the faces of celebrities onto adult film stars. This required significant technical skill and high-end hardware.
- 2019: The "DeepNude" Controversy: A software program called DeepNude was released, specifically designed to undress images of women. It was quickly taken down by its creator due to public outcry, but the underlying code was mirrored and distributed across the web.
- 2022-2023: The Generative AI Boom: The release of open-source models like Stable Diffusion democratized image generation. While these models had "safety filters," developers quickly found ways to remove these restrictions, leading to a surge in specialized "nudify" bots on platforms like Telegram.
- 2024: The Commercialization Phase: Nudify services transitioned into sophisticated web-based platforms with user-friendly interfaces. They began investing in SEO and social media marketing to reach a broader audience beyond tech-savvy early adopters.
- 2025-2026: The Gateway Era: As identified by the ISD, these services have now successfully embedded themselves within the search ecosystems of mainstream platforms, leveraging "influencer-style" reviews and tutorials on YouTube to drive millions of clicks.
Mechanics of Promotion: How Exploitative Content Evades Detection
The ISD report emphasizes that platforms like YouTube are not merely "passive" hosts of this content but are actively facilitating its use through their recommendation engines. The content identified by researchers was often framed as "tech reviews" or "AI tutorials," which may help them circumvent automated filters designed to flag pornography.
Melanie Smith, the senior director of research and policy for ISD, noted that many of the videos found on YouTube provided a "how-to" guide for weaponizing these tools. Some creators would blur the final results of the AI-generated images to avoid immediate takedowns while providing clear links in the video descriptions or pinned comments to the nudify sites themselves. This "gateway" effect allows the platforms to technically claim they do not host the explicit images while still providing the infrastructure that leads users directly to them.
Furthermore, the study found that the use of promo codes and referral links has created a "shadow economy" where content creators on mainstream platforms can profit from referring new customers to nudify services. This incentivizes the creation of more content designed to attract users searching for undressing software.
The Business of Nonconsensual Imagery: Revenue and Accessibility
One of the most alarming aspects of the nudify ecosystem is its extreme accessibility and profitability. The ISD study, bolstered by previous investigations, found that many of these tools allow users to generate a high-resolution, sexually explicit image for as little as $1. This low barrier to entry makes the technology accessible to a wide demographic, including minors.
Despite the low cost per image, the collective revenue generated by these platforms is staggering. Estimates suggest that the top nudify websites may be generating as much as $36 million in annual revenue. This capital is often reinvested into more advanced AI models and aggressive marketing campaigns, creating a self-sustaining cycle of growth. The business model typically involves "freemium" tiers, where users are given a low-quality sample for free but must pay for "premium" processing that produces more realistic and detailed results.
Victim Profiles and the Weaponization of AI
While the public discourse around deepfakes often focuses on high-profile celebrities, the ISD report highlights a more personal and localized threat. The common targets of these nudify apps are everyday individuals, including current and former romantic partners, classmates, and colleagues.
Most disturbingly, the study found a significant volume of requests involving relatives, such as sisters and cousins. The motivations behind using these tools are also shifting. While sexual gratification remains a primary driver, a growing number of users are utilizing the technology for "digital character assassination."
"A lot of the requests were about getting people fired from jobs and compromising their livelihoods and lives in nefarious ways," Melanie Smith told WIRED. This weaponization of AI-generated imagery serves as a form of high-tech harassment designed to exert control or inflict maximum social and professional damage on the victim. Because the images are often realistic enough to be mistaken for actual photographs, the burden of proof falls on the victim to demonstrate that the image is a "fake," a process that is both psychologically taxing and often legally complex.
Official Responses and the Policy-Enforcement Gap
In response to the findings, tech giants have reiterated their commitment to safety, yet the ISD report suggests a significant gap between policy and enforcement. A spokesperson for YouTube, Boot Bullwinkle, stated that the company has "strict policies prohibiting content that includes unwanted sexualization, such as nonconsensually shared intimate imagery." He further clarified that these rules apply to both hosted content and external links, including synthetic content that realistically simulates nudity.
However, the ISD researchers argue that if these policies were being enforced comprehensively, the 1.82 million referrals from YouTube would not exist. The ease with which researchers were able to discover this content using basic keywords suggests that the platforms’ proactive detection systems are failing to keep pace with the sheer volume of "nudify" related uploads.
On X, the situation is even more precarious. Since the platform’s change in ownership and the subsequent reduction in trust and safety personnel, the enforcement of NCII policies has been criticized by digital rights advocates as being "reactive" rather than "proactive," often requiring the victim to report the content before any action is taken.
Broader Societal and Legal Implications
The findings of the ISD report have sparked renewed calls for legislative action. Currently, in many jurisdictions, the legal framework surrounding nonconsensual AI-generated imagery is a patchwork of outdated "revenge porn" laws that may not explicitly cover synthetic media where no "real" photo was ever taken.
Legal experts suggest that the "nudify" industry thrives in a gray area. While the act of creating the image is clearly a violation of personal privacy and dignity, the platforms providing the tools and the social media sites providing the traffic often hide behind liability protections. There is a growing movement to hold mainstream platforms "vicariously liable" for the traffic they drive to known exploitative services.
The implications for society are profound. The normalization of these tools on mainstream platforms risks creating a culture where digital bodily autonomy is no longer a given. As AI technology continues to improve, the distinction between "real" and "synthetic" will continue to blur, making the "nudify" ecosystem a permanent threat to privacy unless systemic changes are made to how mainstream social media platforms moderate search results and external links.
The ISD concludes that without a fundamental shift in how YouTube, X, and other giants handle the promotion of these tools, they will remain the primary infrastructure supporting a global crisis of nonconsensual digital exploitation.
