AI 动手前没查证就行动，这才是翻车主因
我们总以为 AI 犯错是因为「不知道」，这篇发现更常见的是「没查就动」。研究者让 AI 用工具去执行真实操作(改配置、发指令、走流程)，用一份「证据账本」逐条核对：它动手前到底有没有先确认必要信息。结果很反直觉：让 AI 判断「该不该做」时它挺靠谱，可一旦真让它动手，它经常在证据没齐时就先干了，或者查了一半就停手。单步操作只要证据齐了基本不会错，但多步流程里，它会把上一步没解决的遗留问题带进下一步。这不是知识不够，是「用证据的方式」出了问题——它没把查证当成行动的前置条件。它不是你明天能用上的，但给了一个更准的视角：AI 出岔子，很多时候不是笨，是急。
📄 原文摘要(英文)
Tool-using agents make consequential changes to external state, yet correct outcomes do not guarantee that their actions were supported by evidence established beforehand. We study where this evidence-to-action chain breaks as agents move from deciding whether to act to executing single actions and dependent workflows. Across ten model-harness configurations, strong static action assessment can coexist with much weaker interactive execution. Failures often begin before execution: agents stop with incomplete investigation or act before required evidence is established. Once required evidence is obtained, single-action execution is usually reliable, while multi-action workflows additionally expose unresolved prerequisites and incomplete execution. For this analysis, we introduce SafeActBench, comprising 656 cases across six operational domains and five protocols that progress from static action judgment and investigated non-action to single- and multi-action workflows. A provenance-bound Evidence Ledger and deterministic trajectory evaluator track what information was established, when actions occurred, and whether downstream dependencies were satisfied. These results show that failures arise not only from missing information, but also from how agents use established evidence when deciding and executing actions.