MAI-2024-0049

The Heuristic Token Search HTS Attack represents a sophisticated method for circumventing safety protocols in text-to-image T2I models, enabling the generation of Not Safe For Work NSFW content. This attack involves systematically substituting tokens within a malicious prompt with semantically...

Automated Sextortion Spyware Takes Webcam Pics of Victims Watching Porn

A new specimen of “infostealer” malware offers a disturbing feature: It monitors a target's browser for NSFW content, then takes simultaneous screenshots and webcam photos of the victim...

TokenProber: Jailbreaking Text-To-Image Models Via Fine-Grained Word Impact Analysis

Text-to-image T2I models have significantly advanced in producing high-quality images. However, such models have the ability to generate images containing not-safe-for-work NSFW content, such as pornography, violence, political content, and discrimination. To mitigate the risk of generating NSFW...