TYPO: Instruction-Dense Visual Jailbreaks against Commercial Closed-Source Image-Generation Models

본문 미리보기

arXiv:2607.24897v1 Announce Type: new Abstract: Recent commercial image-generation models can generate high-quality images with readable text (e.g., posters, infographics, and manuals), attracting considerable attention. Yet we first show that this same capability also introduces a previously unreported safety vulnerability: these systems may refuse to generate harmful text directly, yet permit the same content when rendered as text within generated images, i.e., safety alignment does not relia