An Experimental Method to Study Opinion Diffusion in Human-AI Hybrid Societies

TL;DR

This paper introduces a new experimental approach to studying opinion formation in hybrid human-AI social networks, revealing that mixed networks reduce polarization compared to human-only groups.

Contribution

It presents a novel experimental method for analyzing opinion dynamics in human-AI hybrid societies, including variations in AI prompt framing effects.

Findings

Hybrid networks had the lowest final polarization.

Human-only networks showed higher polarization and lower neighbor agreement.

Prompt framing influences convergence patterns in opinion dynamics.

Abstract

As artificial intelligence increasingly mediates public discourse, it becomes important to understand how human-AI collectives shape opinion formation, deliberation, and democratic outcomes. We present a novel experimental method for studying opinion dynamics in hybrid human-AI social networks. Participants, human or AI, were embedded in $5\times5$ grid lattice networks and iteratively asked to select and revise statements on a given polarizing topic over eight rounds. We compared three conditions: human-only, AI-only, and hybrid networks with equal proportions of human and AI participants. Hybrid human-AI networks achieved the lowest final polarization while, in contrast, human-only networks exhibited higher polarization with lower neighbor agreement. We also ran additional experiments varying Large Language Model (LLM) prompt framing to explore whether instruction design might influence convergence patterns. Although these early findings are preliminary and cannot yet support broad generalizations, they highlight the potential value of experimental social networks for understanding opinion dynamics in human-AI hybrid societies.