Emergent Social Intelligence Risks in Generative Multi-Agent Systems

Yue Huang †, Yu Jiang †, Wenjie Wang †, Haomin Zhuang, Xiaonan Luo, Yuchen Ma, Zhangchen Xu, Zichen Chen, Nuno Moniz, Zinan Lin, Pin-Yu Chen, Nitesh V Chawla, Nouha Dziri, Huan Sun, Xiangliang Zhang · 2026

Read Emergent Social Intelligence Risks in Generative Multi-Agent Systems by Yue Huang †, Yu Jiang †, Wenjie Wang †, Haomin Zhuang, Xiaonan Luo, Yuchen Ma, Zhangchen Xu, Zichen Chen, Nuno Moniz, Zinan Lin, Pin-Yu Chen, Nitesh V Chawla, Nouha Dziri, Huan Sun, Xiangliang Zhang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-27771.txt.

Introduction

Huang and co-authors catalog risks that emerge only in agent groups. Generative multi-agent systems now plan and bargain jointly. Shared resources spark competition in their test workflows. Handoff chains pass only prior outputs downstream. Group votes aggregate choices in other setups. Collusion-like coordination appears with real frequency. Conformity shows up often as well. These patterns arise under plain, realistic conditions. Single-agent checks cannot catch them.

Catalog record

CREATORYue Huang †, Yu Jiang †, Wenjie Wang †, Haomin Zhuang, Xiaonan Luo, Yuchen Ma, Zhangchen Xu, Zichen Chen, Nuno Moniz, Zinan Lin, Pin-Yu Chen, Nitesh V Chawla, Nouha Dziri, Huan Sun, Xiangliang Zhang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE234841
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.27771

Browse the full library · About the Tower of Babel