AI智能体如何运行城镇经济？100个LLM智能体的26周模拟研究
原标题：But How Would AI Agents Run a Town's Economy?
AI 摘要
研究团队在真实博卡拉湖畔地理环境中构建了一个由100个带记忆的LLM智能体组成的封闭、货币守恒的空间经济系统，运行长达26个模拟周（远超同类研究1-2周），共91次有效运行、244万次决策、215亿token。结果显示货币传导在到达企业后停止：12%的旅游需求冲击使企业收入上升4.62，但工资仅变动1.03，仅0.3%的菜单项被重新定价；随机现金转移后96.7%仍被持有，边际消费倾向仅3-4%。研究还发现财富分布在该文献常用时间尺度上近乎冻结，但12周和26周后基尼系数下降，说明研究时长会改变结论；更换底层LLM会改变所有测量结果，而删除智能体记忆则无显著影响。
正文节选
But How Would AI Agents Run a Town’s Economy? Abstract. We placed 100 memory-equipped large language model (LLM) agents in charge of a closed, money-conserving spatial economy on real Pokhara Lakeside geography (earning wages, running businesses, setting prices) and ran this multi-agent simulation for up to 26 simulated weeks, well past the 1–2 weeks typical of agent-society studies. Across 91 validated runs (2.44M agent decisions, 21.5B tokens), the money stops moving, in a specific and measura