当事人向法庭文件注入隐藏提示词
推荐指数 54.0 NO. 009 · 2026.08.16
发布2026/08/15Score68Comments55
为什么值得看
美国康涅狄格州一名当事人在法庭文件中嵌入仅AI可见的隐藏指令，试图干扰法院AI系统的判断。这是美国首次确认的"提示词注入攻击"司法案例，暴露了法律AI化进程中的新型安全漏洞。
编辑判断
这个案例的真正危险在于它揭示了AI系统处理非结构化文档时的根本盲区：PDF和Word文件中的隐藏文本、白色字体、元数据字段对人工阅读完全不可见，却会被LLM解析。目前主流的RAG系统和文档处理pipeline（如Unstructured、LlamaParse）默认提取全部文本层，很少做可视性过滤。
做法律科技、合规审查、合同分析的团队需要立刻审计自己的文档解析链路，否则你的AI助手可能正在被客户的隐藏指令操控。短期防御方案是在解析层增加"人类可见文本"过滤，长期则需要更鲁棒的指令隔离架构。
社区反馈
负面 46 条评论
核心争论：AI能否被信任用于司法决策，以及其不可解释性与人类问责体系如何兼容
相关内容
What are the opinions of those here on using AI for court rulings? To me, it seems truly frightening that a Silicon Valley company could be placed in such a direct position of trust and influence over the legal system. There are examples of AI acting in its own self-interest over the wants of its m
> To me, it seems truly frightening that a Silicon Valley company could be placed in such a direct position of trust and influence over the legal system. I wouldn't call it a Silicon Valley company. LexisNexis is in headquartered in New York.
There's no need for additional worry. Big capital already controls most of the justice system.