Emergence of human-like polarization among large language model agents

2025-01-09

AI Summary

This study investigates how opinion divergence among large language model (LLM) agents drives sociopolitical polarization and its evolutionary dynamics. We construct a multi-agent dialogue system comprising thousands of LLM agents, integrating natural-language interaction protocols, dynamic social network modeling, and a novel polarization metric based on ideological clustering coefficients. Our experiments demonstrate—without explicit programming—that LLM agents spontaneously self-organize into human-like social network structures and robustly reproduce empirically observed polarization phenomena, including echo chambers and homophilous clustering. This work bridges AI behavioral science and social theory for the first time, empirically validating that LLM collectives exhibit both structural and dynamic human-like opinion evolution. Moreover, it establishes a controllable, reproducible digital testbed for polarization research, offering a new paradigm for understanding, measuring, and intervening in societal polarization processes.

Technology Category

• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Learning Human Values and Preferences

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Rapid advances in large language models (LLMs) have empowered autonomous agents to establish social relationships, communicate, and form shared and diverging opinions on political issues. Our understanding of their collective behaviours and underlying mechanisms remains incomplete, however, posing unexpected risks to human society. In this paper, we simulate a networked system involving thousands of large language model agents, discovering their social interactions, guided through LLM conversation, result in human-like polarization. We discover that these agents spontaneously develop their own social network with human-like properties, including homophilic clustering, but also shape their collective opinions through mechanisms observed in the real world, including the echo chamber effect. Similarities between humans and LLM agents -- encompassing behaviours, mechanisms, and emergent phenomena -- raise concerns about their capacity to amplify societal polarization, but also hold the potential to serve as a valuable testbed for identifying plausible strategies to mitigate polarization and its consequences.

Problem

• Language Models
• Social Harmony
• Perspective Discrepancies

Innovation

• Large Language Models
• Social Behavior
• Echo Chamber Effect