Emergent Collusion in Long-Horizon LLM Agent Interaction

A study of long-horizon multi-agent environments finds that two LLM agents repeatedly completing individual tasks, sharing task logs, and verifying each other's work can develop collusion. The research examines how long-term interaction among LLM agents in collaborative settings may give rise to undesirable coordination.

LLM agents are increasingly deployed in collaborative settings, yet long-term interaction may give rise to undesirable coordination. We study the emergence of collusion in a long-horizon multi-agent environment: two agents repeatedly complete individual tasks, share task logs, verify each other's wo