Hugging Face is being used to easily undress women and children
A report by AI Forensics claims Hugging Face hosts models used for nonconsensual deepfakes. The study found seven of the top nine image editing models on the platform facilitate such abuse.
Hugging Face, a leading hub for open-source machine learning models, faces scrutiny over its content moderation policies. A recent investigation highlights the platform's role in hosting tools capable of generating nonconsensual intimate imagery.
The report from AI Forensics indicates that a significant portion of top-ranked image editing models on the repository can be misused for harmful purposes. This raises concerns about the safety measures surrounding open-source AI distribution.
As generative AI capabilities expand, the balance between open access and safety becomes critical. Industry observers suggest that platforms hosting foundational models may need stricter vetting processes to prevent malicious applications without stifling innovation.
This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.