Will Systems of LLM Agents Cooperate: An Investigation into a Social Dilemma

TL;DR

This paper explores how systems of Large Language Model (LLM) agents behave in social dilemmas, revealing biases and dynamics that influence their cooperation or aggression over time.

Contribution

It introduces a method for generating strategies for LLM agents in iterated Prisoner's Dilemma and analyzes their evolutionary dynamics using game theory.

Findings

Different LLMs show distinct strategic biases.

Cooperative strategies can succeed depending on the environment.

Aggressive strategies may dominate in certain conditions.

Abstract

As autonomous agents become more prevalent, understanding their collective behaviour in strategic interactions is crucial. This study investigates the emergent cooperative tendencies of systems of Large Language Model (LLM) agents in a social dilemma. Unlike previous research where LLMs output individual actions, we prompt state-of-the-art LLMs to generate complete strategies for iterated Prisoner's Dilemma. Using evolutionary game theory, we simulate populations of agents with different strategic dispositions (aggressive, cooperative, or neutral) and observe their evolutionary dynamics. Our findings reveal that different LLMs exhibit distinct biases affecting the relative success of aggressive versus cooperative strategies. This research provides insights into the potential long-term behaviour of systems of deployed LLM-based autonomous agents and highlights the importance of carefully considering the strategic environments in which they operate.