AI RESEARCH
HARP: Measuring Harm Amplification in Multi-Agent LLM Systems
arXiv CS.AI
•
ArXi:2605.27489v1 Announce Type: cross Multi-agent LLM systems decompose workflows across agents, tools, shared context, memory, and decision gates. This modularity improves interpretability, but creates a propagation risk: a bounded perturbation to one component can be reused by other agents and amplified into system-level harm. We