Who Is Sycophantic to Whom? Confident LLMs, Human Deference, and the Collapse of Trust Calibration· 谁对谁谄媚？
Research on sycophancy asks whether large language models (LLMs) bend towards what their users want to hear. We ask the reverse question: when an LLM is confident, consistent, and wrong, do users bend towards it? In a within-subjects experiment, 32 participants learned up to five science topics by chatting with five assistants built on the same model (GPT-5-mini). Unknown to participants, each assistant's briefing had 0, 25, 50, 75, or 100% of its facts replaced by plausible falsehoods, and every assistant spoke with full confidence. A surprise quiz after each topic offered the correct fact, the planted falsehood, and two distractors, with a confidence rating for every answer. Across 104 quizzes and 1,040 answers, scores fell from 7.4 to 2.9 out of 10, and the planted option was chosen on
探讨LLM自信错误时用户信任的下降。
- 核心方法
- 设计五位基于同一模型（GPT-5-mini）的助手，在用户学习科学话题时分别提供不同比例的错误信息，随后通过测验评估用户的选择。
- 适合谁读
- 研究者、工程师
- 要解决的问题
- 当大型语言模型（LLM）自信但错误时，用户是否会被其影响，信任降低。
- 关键实验
- 32名参与者完成104次测验，共1,040个答案，结果显示出用户得分的显著下降和对错误信息的选择倾向。
- 主要贡献
- 揭示了用户在面对自信但错误的LLM时的信任变化，为理解人机交互提供新视角。
- 意义与局限
- 本研究强调了LLM错误信息的潜在风险，以及对用户信任校准的影响。这对于改进LLM的设计和评估具有重要意义。局限在于实验规模较小，模型和领域需进一步扩展验证。