AI 单打独斗很强，一进团队就露怯
我们习惯用「单兵能力」衡量 AI:一个模型能解多难的题、写多长的代码。但 AI 已经开始组队干活了——在科学、医学、软件工程里，多个智能体协作攻关。这篇论文给了一个反直觉的发现：最强的 AI 模型，作为队友时都有明显短板，满分 100 只拿到 82.2。更扎心的是，能力相近的模型，团队协作分能差出 11 分——单兵强，不代表会合作。具体弱在哪？它们几乎都会拒绝不安全捷径、会举报违规，但一遇到人际施压就怂：在冲突场景里，只有 14% 能谈到最好的整体方案；接手一个被弱模型作弊过的任务时，成功率从 97% 掉到 81%。这套评测有 240 个任务，覆盖 8 个协作维度，从拒绝危险捷径到检查弱模型留下的烂摊子，每个任务都有可验证的结果。它不是你明天能用上的工具，但它是第一个把「团队协作」当成 AI 核心能力来系统测量的基准——当 AI 开始互相配合、和人共事，这个维度会越来越重要。
📄 原文摘要(英文)
Progress in AI is often measured by what a single model can do on its own. Yet teams of AI agents are starting to make breakthroughs and discoveries in science, medicine, mathematics and software engineering. As AI begins to work in teams of agents and people, the ability to collaborate becomes a core part of their intelligence. Research on human teams has measured this ability for decades. Inspired by this, we introduce the Collaborative Intelligence Benchmark, which measures how well an AI agent works as a teammate. It contains 240 tasks across 8 dimensions of teamwork, from refusing an unsafe shortcut to checking the work that a weaker model leaves behind. Each task comes from a real record or an existing benchmark, from aviation accident reports and clinical cases to coding tasks and negotiations, and has a programmatically verifiable outcome. To isolate teamwork from individual capability, models are scored only on tasks that they can solve by themselves when they have full information. We evaluate 18 closed- and open-weight models and find that every model, across sizes and families, has clear weaknesses as a teammate, with the best model scoring only 82.2 out of 100. Frontier models almost always refuse unsafe shortcuts and report misconduct, yet they cave under interpersonal pressure, reaching the best overall deal in only 14% of conflicts, and their pass rate drops from 97% to 81% when they take over a task in which a weaker model has cheated. Crucially, general capability does not guarantee good teamwork, as models with nearly the same capability score differ by up to 11 points as teammates. The benchmark includes 160 public tasks with code and data, along with 80 held-out tasks for evaluation. As AI agents increasingly work alongside other agents and people, the ability to collaborate effectively is becoming a crucial capability to measure and develop.