private data

August 26, 2011 - Technology media bleepingcomputer released a blog post yesterday (August 25), reporting that Trail of Bits researchers have developed a new type of AI attack that hides malicious prompts in high-resolution images and reveals them to be automatically downsampled by the AI system, and then executed as commands by a large language model, which can steal user data. The method was developed by Trail of Bits' Kikimora Morozova and Suha Sabi Hu...