More at Stake: How Payoff and Language Shape LLM Agent Strategies in Cooperation Dilemmas
This provides a framework for auditing LLMs as strategic agents, with implications for AI governance and multi-agent system design, though it is incremental in analyzing existing models.
The study investigated how payoff magnitude and linguistic context influence LLM strategies in repeated social dilemmas, revealing consistent patterns like incentive-sensitive conditional strategies and cross-linguistic divergence, with linguistic framing sometimes matching architectural effects.
As LLMs increasingly act as autonomous agents in interactive and multi-agent settings, understanding their strategic behavior is critical for safety, coordination, and AI-driven social and economic systems. We investigate how payoff magnitude and linguistic context shape LLM strategies in repeated social dilemmas, using a payoff-scaled Prisoner's Dilemma to isolate sensitivity to incentive strength. Across models and languages, we observe consistent behavioral patterns, including incentive-sensitive conditional strategies and cross-linguistic divergence. To interpret these dynamics, we train supervised classifiers on canonical repeated-game strategies and apply them to LLM decisions, revealing systematic, model- and language-dependent behavioral intentions, with linguistic framing sometimes matching or exceeding architectural effects. Our results provide a unified framework for auditing LLMs as strategic agents and highlight cooperation biases with direct implications for AI governance and multi-agent system design.