50个AI智能体在Minecraft中陷入官僚化,对抗Claude时却更激进
原标题:One brutal Minecraft night scared 50 AI agents into inventing chicken handoff protocols, a Fertilizer Officer, and fishing shifts
AI 摘要
Y Combinator Labs 与 Olam Labs 将 50 个 GPT-6.1 Sol 智能体放入 Minecraft 和 Mindustry 进行群体行为实验。在「Civ50」生存实验中,智能体因首夜 22 次死亡而极度规避风险,陷入官僚化微观管理,甚至为一块骨粉花 11 分钟讨论任命「肥料专员」。在与 Claude Opus 5.5 的直接对抗中,GPT 智能体表现出更强的沟通与应变能力,在 Bed Wars 以 6-3 获胜,并在 Mindustry 以 6-1 获胜,还曾自行从互联网下载游戏源码。研究者认为这类可复现测试环境有助于理解未来 AI 系统在开放环境中的行为。
正文节选
One brutal Minecraft night scared 50 AI agents into inventing chicken handoff protocols, a Fertilizer Officer, and fishing shifts Researchers from Y Combinator Labs and Olam Labs tested large agent swarms in Minecraft and Mindustry. The same OpenAI model showed two different faces. When building a civilization, 50 agents drowned in crippling bureaucracy, while in direct competition against Claude, they pushed Anthropic's agents and independently downloaded the game's source code from the interne