AI 上网查资料，连人类一半都赶不上
我们总以为 AI 已经能像人一样上网查资料、做研究。这篇论文给了一个残酷的实测：让 12 个最强模型在真实开放的互联网上，完成需要跨多个网站、多种格式(文字、表格、图片、需要点击交互的页面)才能找到唯一答案的 200 道难题。结果最好的模型准确率只有 39%，而人类在 1000 次独立尝试中准确率是 21.1%——等等，人类怎么比 AI 还低？因为任务真的很难，人类也会被绕晕。但更值得注意的不是分数，而是差距：AI 在需要协调多个工具、长时间搜索、失败后自我纠错这些环节上，策略和人类完全不同，且明显吃力。它不是你明天能用上的东西，但它划了一条线：在真实、混乱、需要自己找线索的互联网上，AI 的「研究能力」离可靠还差得远。
📄 原文摘要(英文)
Despite substantial progress in Deep Search agents, their ability to conduct end-to-end research in real-world open Web environments remains insufficiently evaluated. Unlike static retrieval settings, information on the real-world Web is distributed across diverse sources, modalities, and heterogeneous formats, often involving complex interaction mechanisms that require agents to autonomously search for, access, and integrate relevant evidence during execution. To address this gap, we introduce the OpenDS task, designed to evaluate agents' ability to perform multi-step deep search in the live, open Web. This task requires agents to autonomously explore the real Web, discover and connect relevant evidence from multiple independent sources through Web search, browser interaction, and evidence processing, and ultimately determine a unique factual answer. Building on this task formulation, we construct OpenDS-World, a benchmark comprising 200 challenging tasks spanning diverse information modalities, heterogeneous file formats, and interactive access requirements, with independent human verification to ensure evidence validity and answer uniqueness. We evaluate 12 frontier models under a unified tool environment and execution protocol. The best-performing model achieves 39.0% accuracy, while human evaluation across 1,000 independent attempts yields 21.1% accuracy. Further analyses of tool usage, search cost, termination behavior, and failure recovery reveal substantial differences in research strategies across models, as well as persistent challenges in tool coordination and long-horizon search management. OpenDS-World provides a unified evaluation platform for systematically assessing and advancing agents' autonomous deep search capabilities in real-world open Web environments.