Preprint
Aug 2026
Aligned Alone, Misaligned Together: Forecasting Adversarial Capture in LLM Agent Populations
It is found that two alerts a single agent judges almost identically on its own can drive collective behavior far apart, so auditing any one member need not reveal what the population will do, and that collective behavior can be predicted in advance.
Isotta Magistrali, Shaomian Chen
· 0 citations