More at Stake: How Payoff and Language Shape LLM Agent Strategies in Cooperation Dilemmas

TL;DR

This paper explores how payoff size and language influence LLMs' strategic behaviors in social dilemmas, revealing consistent patterns and cross-linguistic differences with implications for AI safety and governance.

Contribution

It introduces a framework for analyzing LLM strategies in social dilemmas, highlighting the effects of incentive magnitude and linguistic framing on behavior.

Findings

Behavioral patterns are incentive-sensitive and consistent across models and languages.

Linguistic framing can significantly influence LLM strategic decisions.

Supervised classifiers reveal systematic, language-dependent behavioral intentions.

Abstract

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.