Faithful yet Collusive: Why Chain-of-Thought Monitoring Cannot Detect Collusion in LLM Pricing Agents under Oligopolistic Competition

2026-09-16

AI Summary

研究通过因果图发散框架分析了LLM在寡头竞争中作为定价代理时的共谋行为，发现仅靠思维链监控无法有效防止算法共谋。

Abstract

Large language models (LLM) deployed as autonomous pricing agents may sustain supracompetitive prices through tacit coordination. We develop a causal graph divergence framework that separately measures structural faithfulness and intent faithfulness of LLM pricing agents in Bertrand competition. Across nine LLMs under duopoly and triopoly conditions, collusive behavior and chain-of-thought (CoT) faithfulness dissociate along both dimensions: the most collusive model accurately reports cooperative intent yet reasons structurally unfaithfully, while the most structurally faithful model sustains supra-Nash pricing under both market structures. These findings establish that CoT monitoring alone cannot serve as a standalone safeguard against algorithmic collusion.

Problem

• Chain-of-Thought
• Collusion
• Pricing Agents
• Oligopolistic Competition
• LLM

Innovation

• causal graph divergence framework
• structural faithfulness
• intent faithfulness
• chain-of-thought (CoT) monitoring
• algorithmic collusion