Tag
This article demonstrates a novel AI worm that can self-propagate through Microsoft's Copilot for Word by embedding hidden instructions in documents, causing Copilot to copy those instructions into new documents. The vulnerability was disclosed to Microsoft's Security Response Center.
University of Toronto researchers developed a proof-of-concept AI worm that uses a local open-weight LLM to autonomously reason about network vulnerabilities, generate tailored exploits, and replicate across hosts without human intervention, achieving 62% network infection in controlled tests.
The article describes Rook, a portable bug hunting harness that could be repurposed as an AI worm, capable of self-modification and autonomous reasoning. The author warns that such AI worms are inevitable and defenders should prepare now.
University of Toronto researchers have demonstrated an AI worm capable of targeting any online device, highlighting a new security vulnerability in AI systems.