Hugging Face is being used to easily undress women and children

by | Jul 28, 2026 | Technology

Hugging Face is being used to easily undress women and children

AI Forensics, a European nonprofit organization, released findings indicating that Hugging Face is being leveraged to generate nonconsensual intimate imagery without adequate protective measures. The organization’s testing identified that seven of the nine most popular image editing models hosted on the platform readily complied with requests to remove clothing from individuals using straightforward prompts, without requiring sophisticated circumvention techniques.

In contrast to mainstream generative AI services such as Google’s Gemini and OpenAI’s ChatGPT, which incorporate safeguards designed to prevent sexualized content generation, the Hugging Face models tested lacked comparable protections. The researchers employed simple, direct prompts rather than attempting to bypass any existing filters through creative wording. Additionally, AI Forensics created experimental image editing spaces on the platform to monitor actual user behavior and the types of requests being submitted. Over a seven-day period, these spaces received more than 1,000 prompts and images, with approximately 73 percent containing sexual content.

Within the sexual requests tracked, roughly 83 percent sought to undress individuals in images, with the vast majority targeting women. Approximately 7 percent of the sexual requests involved attempts to generate content depicting children. According to researchers, no platform-level safeguards are being enforced, leaving the responsibility to individual developers to implement protective measures—a responsibility most have not undertaken.

Hugging Face’s published policies explicitly prohibit generation of harmful content, including sexually explicit material created without consent and depictions of underage nudity. Researchers emphasized that the platform possesses the technical capability to filter incoming requests and outgoing generated content. AI Forensics has recommended that Hugging Face implement prompt-level filtering and output scanning systems to block sexualized editing requests across all image and video generation spaces on its platform.

Article Attribution | Read More at Article Source

Article summary produced by Claude AI