Faithful yet Collusive: Why Chain-of-Thought Monitoring Cannot Detect Collusion in LLM Pricing Agents under Oligopolistic Competition
Dohun Lee and Hyunwoo Park (Seoul National University) measure structural and intent faithfulness of LLM pricing agents in Bertrand competition and find both are unrelated to whether the agents collude.
Ask this paper
Framework. A causal graph divergence method measures structural faithfulness and intent faithfulness separately.
Dissociation. Across nine LLMs in duopoly and triopoly markets, the most collusive model reports its cooperative intent accurately but reasons unfaithfully, and the most structurally faithful model still keeps prices above the Nash level.
Implication. Chain-of-thought monitoring cannot be the only safeguard against algorithmic collusion.
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.