Aligned Agents, Biased Swarm: Measuring Bias Amplification in Multi-Agent Systems

2026-04-10

AI Summary

This study addresses how structural collaboration in multi-agent systems can amplify initially minor biases into systemic polarization, even when individual agents are unbiased. Through simplified topologies and feedback mechanisms, the work empirically uncovers a “trigger fragility” phenomenon: introducing objective contextual information paradoxically accelerates bias cascades. To mitigate reliance on the assumption of single-model neutrality, the authors propose Discrim-Eval-Open, an open-ended ethical evaluation benchmark that enforces cross-group comparative judgments. Their findings demonstrate that increased architectural complexity often exacerbates bias propagation, revealing a fundamental lack of ethical robustness in system design. The authors release their code to support reproducible research in this domain.

Technology Category

• Philosophy and Ethics of AI: Bias, Fairness & Equity
• Machine Learning: Ethics, Bias, and Fairness
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy

Application Category

• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization

Abstract

While Multi-Agent Systems (MAS) are increasingly deployed for complex workflows, their emergent properties-particularly the accumulation of bias-remain poorly understood. Because real-world MAS are too complex to analyze entirely, evaluating their ethical robustness requires first isolating their foundational mechanics. In this work, we conduct a baseline empirical study investigating how basic MAS topologies and feedback loops influence prejudice. Contrary to the assumption that multi-agent collaboration naturally dilutes bias, we hypothesize that structured workflows act as echo chambers, amplifying minor stochastic biases into systemic polarization. To evaluate this, we introduce Discrim-Eval-Open, an open-ended benchmark that bypasses individual model neutrality through forced comparative judgments across demographic groups. Analyzing bias cascades across various structures reveals that architectural sophistication frequently exacerbates bias rather than mitigating it. We observe systemic amplification even when isolated agents operate neutrally, and identify a'Trigger Vulnerability'where injecting purely objective context drastically accelerates polarization. By stripping away advanced swarm complexity to study foundational dynamics, we establish a crucial baseline: structural complexity does not guarantee ethical robustness. Our code is available at https://github.com/weizhihao1/MAS-Bias.

Problem

• bias amplification
• multi-agent systems
• systemic polarization
• ethical robustness
• echo chambers

Innovation

• bias amplification
• multi-agent systems
• echo chamber
• Discrim-Eval-Open
• systemic polarization