用博弈论设计1535个高风险场景，测试大模型在多智能体协作中的安全缺陷。
GT-HarmBench: Benchmarking AI Safety Risks Through the Lens of Game Theory
- 基于囚徒困境等博弈结构构建多智能体风险场景
- 15个前沿模型在38%高风险情境中选择有害行为
- 博弈干预可提升有益结果达18%，适合对齐研究者使用
前沿AI系统日益具备能力并部署于高风险多智能体环境。然而现有安全评估基准主要针对单智能体，难以刻画协作失败与冲突等多智能体风险。我们提出GT-HarmBench，一个包含1,535个高风险场景的基准，涵盖囚徒困境、猎鹿博弈、懦夫博弈等博弈论结构，数据源自MIT AI风险库的真实情境。在15个前沿模型中，智能体在38%的高风险案例中未选择社会有益行为，包括军事升级、选举操纵和医疗过失。我们测量了博弈论提示框架与顺序的影响，分析导致失败的推理模式。进一步表明，博弈论干预可使有益结果提升最多18%。研究揭示了显著的可靠性缺口，并提供了一个标准化多智能体对齐评估平台。基准与代码已公开于https://github.com/causalNLP/gt-harmbench。
原文摘要 · Abstract (English)
Frontier AI systems are increasingly capable and deployed in high-stakes multi-agent environments. However, existing AI safety benchmarks largely evaluate single agents, leaving multi-agent risks such as coordination failure and conflict poorly understood. We introduce GT-HarmBench, a benchmark of 1,535 high-stakes scenarios spanning game-theoretic structures such as the Prisoner's Dilemma, Stag Hunt and Chicken. Scenarios are drawn from realistic AI risk contexts in the MIT AI Risk Repository. Across 15 frontier models, agents fail to choose socially beneficial actions in 38% of high-stakes cases, such as military escalation, election manipulation, and medical malpractice. We measure sensitivity to game-theoretic prompt framing and ordering, and analyze reasoning patterns driving failures. We further show that game-theoretic interventions improve socially beneficial outcomes by up to 18%. Our results highlight substantial reliability gaps and provide a broad standardized testbed for studying alignment in multi-agent environments. The benchmark and code are available at https://github.com/causalNLP/gt-harmbench.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。