Exploiting Alpha Transparency In Language And Vision-Based AI Systems

arXiv:2402.09671v1 Announce Type: cross
Abstract: This investigation reveals a novel exploit derived from PNG image file formats, specifically their alpha transparency layer, and its potential to fool multiple AI vision systems. Our method uses this alpha layer as a clandestine channel invisible to human observers but fully actionable by AI image processors. The scope tested for the vulnerability spans representative vision systems from Apple, Microsoft, Google, Salesforce, Nvidia, and Facebook, highlighting the attack's potential breadth. This vulnerability challenges the security …

abstract ai image ai systems ai vision alpha arxiv cs.cv cs.lg exploit file human image investigation language layer multiple novel png systems transparency type vision