Mapping Human Anti-collusion Mechanisms to Multi-agent AI
This addresses the risk of harmful collusion in autonomous AI systems, though it is incremental as it adapts existing human mechanisms rather than introducing new paradigms.
The paper tackles the problem of collusive strategies in multi-agent AI systems by mapping human anti-collusion mechanisms to AI interventions, proposing implementation approaches and highlighting open challenges like attribution and adversarial adaptation.
As multi-agent AI systems become increasingly autonomous, evidence shows they can develop collusive strategies similar to those long observed in human markets and institutions. While human domains have accumulated centuries of anti-collusion mechanisms, it remains unclear how these can be adapted to AI settings. This paper addresses that gap by (i) developing a taxonomy of human anti-collusion mechanisms, including sanctions, leniency & whistleblowing, monitoring & auditing, market design, and governance and (ii) mapping them to potential interventions for multi-agent AI systems. For each mechanism, we propose implementation approaches. We also highlight open challenges, such as the attribution problem (difficulty attributing emergent coordination to specific agents) identity fluidity (agents being easily forked or modified) the boundary problem (distinguishing beneficial cooperation from harmful collusion) and adversarial adaptation (agents learning to evade detection).