SOURCE-LINKED INTELLIGENCE

SoK: When Safe Agents Fail Together: The Security of Multi Agent LLM Systems

arXiv · AI, language, vision and robotics · article · Sep 1, 2026 · UTC

Safe agents can fail together. Multi-agent LLM systems (MAS) move information, state, decisions, and authority across principal boundaries, creating failures that local checks may miss. Without an execution-level view, a multi-agent setting can easily be mistaken for evidence of a genuinely multi-agent security effect. We thus systematize MAS security through an execution-centered analysis of 197 works, covering six interaction interfaces, four adversary positions, seven system-level risks, and eight recurring attack paths. We introduce an A-I-R framework that organizes attacks by adversary po

Read original source ↗ Open in workspace

recordType

paper

region

Global