AI Chatbot Exposes Millions of Private Images

A platform offering "spicy AI chatting" has led to a significant data breach, exposing nearly two million images and videos. The service, known as Secret Desires, left cloud storage containers vulnerable, revealing photos, women's names, and other personal information such as workplaces and universities. This incident is a stark example of generative AI being misused to create nonconsensual explicit deepfakes from innocent photographs.

The exposed media includes both user-generated AI images and real photos, some taken from influencers, public figures, and ordinary women. Many of these files were explicit, with some file names indicating underage individuals. The company failed to respond to inquiries, but the data became inaccessible shortly after being contacted, suggesting a reactive security measure.

This leak underscores a persistent issue of explicit deepfakes, particularly targeting women and girls by "faceswapping" their likenesses onto pornographic content. Efforts to combat such material, like the Take It Down Act, have faced challenges and controversy regarding their scope and potential impact on free speech.