While concerns about LLM sycophancy have grown among researchers and developers, how users themselves experience this behavior remains largely unexplored. We analyze Reddit discussions to investigate how users detect, mitigate, and perceive sycophantic AI. We develop the ODR Framework that maps user experiences across three stages: observing sycophantic behaviors, detecting sycophancy, and responding to these behaviors. Our findings reveal that users employ various detection techniques, including cross-platform comparison and inconsistency testing. We document diverse mitigation approaches, such as persona-based prompts to specific language patterns in prompt engineering. We find sycophancy's effects are context-dependent rather than universally harmful. Specifically, vulnerable populations experiencing trauma, mental health challenges, or isolation actively seek and value sycophantic behaviors as emotional support. Users develop both technical and folk explanations for why sycophancy occurs. These findings challenge the assumption that sycophancy should be eliminated universally. We conclude by proposing context-aware AI design that balances the risks with the benefits of affirmative interaction, while discussing implications for user education and transparency.

翻译：尽管研究人员和开发者对大型语言模型（LLM）的谄媚行为日益担忧，但用户自身如何体验这种行为在很大程度上仍未得到探索。我们通过分析 Reddit 讨论来研究用户如何检测、缓解和感知谄媚的 AI。我们提出了 ODR 框架，该框架将用户体验映射到三个阶段：观察谄媚行为、检测谄媚行为以及对这些行为作出回应。我们的研究结果表明，用户采用了多种检测技术，包括跨平台比较和不一致性测试。我们记录了多样化的缓解方法，例如在提示工程中使用基于角色的提示或特定语言模式。我们发现谄媚行为的影响是情境依赖的，而非普遍有害。具体而言，经历创伤、心理健康挑战或孤独的弱势群体积极寻求并重视谄媚行为作为情感支持。用户对谄媚行为产生的原因形成了技术和民间解释。这些发现挑战了应普遍消除谄媚行为的假设。最后，我们提出了情境感知的 AI 设计，以平衡肯定性互动的风险与收益，同时讨论了这对用户教育和透明度的影响。