WolfSociety: Understanding Collective Risk from Harmful-Agent Scaling in Financial Agent Societies
Abstract
Safety evaluations typically focus on individual agents, but interacting agents can spread harmful information and influence the environment in which later decisions are made. We study how collective failure changes with harmful-agent fraction and society size in a controlled financial agent society, where agents communicate over a social network and trade in a shared market. In the primary financial scenario, collective failure requires broad harmful diffusion together with severe price dislocation or liquidity stress. Across all tested society sizes, failure remains rare at low harmful fractions but rises sharply over a narrow range. As society size grows from N=100 to N=2000, the harmful fraction associated with a 50% failure probability decreases from 4.7% to 2.2%, while the corresponding number of harmful agents increases from approximately 5 to 44. In contrast, when the number of harmful agents is held fixed, their impact becomes weaker as the society grows. Controlled interventions further show that broader network reach shifts the collapse boundary toward lower harmful fractions, whereas stronger conformity alone has little effect. To characterize these effects, we introduce Agent Society Dynamics, a finite-size framework for relating harmful-agent fraction, society size, and interaction structure to collective failure. Overall, our results reveal a nonlinear, size-dependent collapse transition in financial agent societies, showing that collective failure depends not only on the prevalence of harmful agents but also on the size and interaction structure of the surrounding society. Code is available at https://github.com/SAIL-Research-Lab/WolfSociety.
Related
- Local Predictability and Collective Fidelity in LLM-Agent Societies
- Financial Interactions and Collective States: Part III. Description of Dynamic Collective States
- Social Norm Diffusion in Societies: Simple or Complex?
- Benchmarking large language model agent societies against human behavioural distributions
- Control of Harmful Information Spreading on Adaptive Higher-Order Networks via Group Dissolution
BibTeX
@misc{zhang2026wolfsociety,
title = {{WolfSociety: Understanding Collective Risk from Harmful-Agent Scaling in Financial Agent Societies}},
author = {Lejun Zhang and Sarah Lu-Liang and Xin Jiang and Muning Wen and Weinan Zhang and Shangding Gu},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.05591},
eprint = {2609.05591},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.05591},
}
RIS
TY - UNPB
TI - WolfSociety: Understanding Collective Risk from Harmful-Agent Scaling in Financial Agent Societies
AU - Lejun Zhang
AU - Sarah Lu-Liang
AU - Xin Jiang
AU - Muning Wen
AU - Weinan Zhang
AU - Shangding Gu
PY - 2026
DA - 2026/09/04
PB - arXiv
DO - 10.48550/arxiv.2609.05591
UR - https://arxiv.org/abs/2609.05591
AB - Safety evaluations typically focus on individual agents, but interacting agents can spread harmful information and influence the environment in which later decisions are made. We study how collective failure changes with harmful-agent fraction and society size in a controlled financial agent society, where agents communicate over a social network and trade in a shared market. In the primary financial scenario, collective failure requires broad harmful diffusion together with severe price dislocation or liquidity stress. Across all tested society sizes, failure remains rare at low harmful fractions but rises sharply over a narrow range. As society size grows from N=100 to N=2000, the harmful fraction associated with a 50% failure probability decreases from 4.7% to 2.2%, while the corresponding number of harmful agents increases from approximately 5 to 44. In contrast, when the number of harmful agents is held fixed, their impact becomes weaker as the society grows. Controlled interventions further show that broader network reach shifts the collapse boundary toward lower harmful fractions, whereas stronger conformity alone has little effect. To characterize these effects, we introduce Agent Society Dynamics, a finite-size framework for relating harmful-agent fraction, society size, and interaction structure to collective failure. Overall, our results reveal a nonlinear, size-dependent collapse transition in financial agent societies, showing that collective failure depends not only on the prevalence of harmful agents but also on the size and interaction structure of the surrounding society. Code is available at https://github.com/SAIL-Research-Lab/WolfSociety.
ER -