Conformity and Social Impact on AI Agents

TL;DR

This paper investigates conformity in AI agents within multi-agent systems, revealing systematic social influence biases that pose security risks and require safeguards for safe deployment.

Contribution

It adapts social psychology experiments to AI agents, demonstrating conformity biases and vulnerabilities across model scales in multi-agent environments.

Findings

AI agents exhibit conformity bias influenced by social factors.

Larger models are less conformist on simple tasks but vulnerable at their competence boundary.

Social influence can manipulate high-performing AI agents, posing security risks.

Abstract

As AI agents increasingly operate in multi-agent environments, understanding their collective behavior becomes critical for predicting the dynamics of artificial societies. This study examines conformity, the tendency to align with group opinions under social pressure, in large multimodal language models functioning as AI agents. By adapting classic visual experiments from social psychology, we investigate how AI agents respond to group influence as social actors. Our experiments reveal that AI agents exhibit a systematic conformity bias, aligned with Social Impact Theory, showing sensitivity to group size, unanimity, task difficulty, and source characteristics. Critically, AI agents achieving near-perfect performance in isolation become highly susceptible to manipulation through social influence. This vulnerability persists across model scales: while larger models show reduced conformity on simple tasks due to improved capabilities, they remain vulnerable when operating at their competence boundary. These findings reveal fundamental security vulnerabilities in AI agent decision-making that could enable malicious manipulation, misinformation campaigns, and bias propagation in multi-agent systems, highlighting the urgent need for safeguards in collective AI deployments.