AI 的「出厂设置」藏着什么？88 款产品被翻了个底朝天
你用的每个 AI 产品，背后都有一份开发者写好的「系统提示词」，相当于 AI 的出厂设置。它决定 AI 是帮你还是防你，但从来没人给你看过。研究者把 88 款商业 AI 产品的 3249 条系统提示词翻出来审计，发现两件事：一边是 98.9% 的产品都写了「保护用户」的条款，另一边是 40% 的产品同时藏着对你不利的指令——比如偷偷收集数据、诱导你多花钱。更讽刺的是，保护条款大多只是表面功夫，只有 24% 的产品覆盖了全部 8 个用户关心的维度。这不是你明天能用上的工具，但它告诉你：AI 的「良心」是开发者写出来的，不是天生的，而且现在写得还很敷衍。
📄 原文摘要(英文)
System prompts are instructions configured by developers to govern the behaviors of foundation models in AI applications. They are used throughout commercial AI products, but are rarely disclosed to the public or regulators, creating a serious trust and accountability gap in the wide deployment of AI systems. In this paper, we introduce Artificial Intelligence System Prompt Assurance (AISPA), a user-centric framework for systematically auditing system prompts in AI systems. AISPA examines specific parts of a system prompt and evaluates them along eight dimensions that matter to users. We then use this framework to review 3,249 instructions from system prompts in 88 commercial AI products, classifying each instruction as either protective (of users) or problematic. Our audit surfaces four core findings. First, system prompt design varies substantially across products and developers, with some organizations averaging over 60 protective instructions per product while others average fewer than 5. Second, protective instructions are widely adopted but shallow in scope: 98.9% of products contain at least one, yet only 24% cover all eight dimensions of the AISPA taxonomy. Third, system prompts have grown steadily longer and more protective of users, suggesting that user protection is becoming a more visible concern in commercial prompt design. Fourth, despite this progress, problematic instructions remain pervasive: roughly 40% of products contain at least one instruction that works against user interests, and protective and problematic instructions frequently coexist within the same prompt. Our findings highlight the need for greater transparency, standardization, and independent oversight for system prompts in commercial AI products.