Multi-agent systems composed of large generative models are rapidly moving from laboratory prototypes to real-world deployments, where they jointly plan, negotiate, and allocate shared resources to solve complex tasks. While such systems promise unprecedented scalability and autonomy, their collective interaction also gives rise to failure modes that cannot be reduced to individual agents. Understanding these emergent risks is therefore critical. Here, we present a pioneer study of such emergent multi-agent risk in workflows that involve competition over shared resources (e.g., computing resources or market share), sequential handoff collaboration (where downstream agents see only predecessor outputs), collective decision aggregation, and others. Across these settings, we observe that such group behaviors arise frequently across repeated trials and a wide range of interaction conditions, rather than as rare or pathological cases. In particular, phenomena such as collusion-like coordination and conformity emerge with non-trivial frequency under realistic resource constraints, communication protocols, and role assignments, mirroring well-known pathologies in human societies despite no explicit instruction. Moreover, these risks cannot be prevented by existing agent-level safeguards alone. These findings expose the dark side of intelligent multi-agent systems: a social intelligence risk where agent collectives, despite no instruction to do so, spontaneously reproduce familiar failure patterns from human societies.
- collusion-like coordination
- conformity pressures
- exclusionary behavior
- unfair or biased allocation of resources
These behaviors occur even when the design contains:
- competition over shared resources
- sequential handoff collaboration
- collective decision aggregation
- other common workflow patterns
Our results suggest that group-level risks are not rare edge cases but frequent and reproducible outcomes of realistic multi-agent workflows.
We therefore call for safeguards that target collective dynamics—monitoring interaction patterns, constraining coordination channels, and auditing allocation outcomes—in addition to agent-level controls.