Stop Telling Me to Ask an LLM
243 points
• 5 days ago
• Article
Link
人们对把复杂问题一概转给 Claude 或其他生成式 AI 的做法越来越不满。当向经验丰富的专业人士请教细微或棘手的问题时，得到的回答常常是建议去问 LLM 。这种建议往往忽视了一个事实：提问者在寻求人类意见之前，很可能已经尝试过这些工具了。
向专家请教的目的并不是为了获取基础信息，而是为了得到那种只有几十年摸索、犯错与抉择才能积累出的独到经验。就像你更想听某人的私人餐厅推荐，而不是那种千篇一律的线上榜单一样，期望是触及共享的经历或个人口味。提问者往往想要的是资深人士的"伤痕性经验"，想知道在教科书失效或行业共识存在分歧时，专家会信赖什么判断。
被告知去问 LLM 往往让人觉得是礼貌性的敷衍，是一种在不明说没时间或不感兴趣的情况下拒绝交流的方式。的确，人们忙碌，可能无暇处理深度复杂的问题，但这种简单的转移本身并无助益。它暗示提问者只是懒惰或不懂现代工具，却忽略了问题已经超出了模型的局限，仍然需要人类特有的细微判断。
归根结底，机器能处理的查询与需要经过深思的人类综合判断的请求之间存在本质差异。默认建议使用 AI，会扼杀有价值的交流，错失分享来之不易智慧的机会。当问题已经超出搜索引擎或聊天机器人的能力时，以"去问 Claude"作为推辞并不能节省时间，反而剥夺了只有真人才能提供的具体且主观的专业见解。
There is a growing frustration with the reflex to redirect complex inquiries to large language models like Claude or other generative AI tools. When reaching out to experienced professionals for insight on nuanced or difficult problems, the response is increasingly a suggestion to ask an LLM. This advice often ignores the reality that the questioner has likely already exhausted that avenue before seeking human input.
The expectation of asking an expert is not about finding basic information, but about accessing the unique, lived experience that only decades of trial, error, and decision-making can provide. Like seeking a personal restaurant recommendation over a generic online list, the goal is to tap into a shared history or individual taste. The questioner is specifically looking for the "scar tissue" of a seasoned veteran, wanting to know what an expert trusts when textbooks fail or industry consensus is divided.
Being told to ask an LLM often feels like a polite brush-off, a way to decline engagement without admitting a lack of time or interest. While it is understandable that people are busy and may not have the capacity to field deep, complex questions, the redirect itself is unhelpful. It suggests the person asking is simply lazy or unaware of modern tools, failing to recognize that the question has already survived the model's limitations and still requires human nuance.
Ultimately, there is a distinct difference between a query that a machine can handle and a request for thoughtful, human synthesis. By defaulting to the suggestion of AI, valuable exchanges are stifled, and the opportunity to share hard-won wisdom is lost. When an inquiry has already moved past the capabilities of a search engine or a chatbot, the dismissal of "ask Claude" does not save time. Instead, it simply withholds the specific, subjective expertise that only a real person can offer.
139 comments • Comments Link
• 提供已有研究和 LLM 回应具体局限的详细背景，可以把一个看似普通的负担转化为富有成效的协作，从而向专家明确表明提问者已尽过初步努力。
• 将问题重定向给 LLM 常被资深员工当作一种防御性手段，充当对低质量咨询的"工作量过滤器"，以避免在无法带来职业成长价值的事情上浪费有限的时间与注意力。
• 在代码和设计上过度依赖 LLM 会催生被动验证文化，开发者在不理解底层逻辑的情况下直接复制粘贴 AI 输出，最终损害工作质量与主人感。
• 纠正由 LLM 产生的错误或幻觉所消耗的精力，远超过最初的事实核查；这对有经验的从业者是一种"税"，因为他们不得不反复揭穿出现在 Pull Requests 或项目提案中的 AI 错误。
• "Ask Claude"式的回应可能反映疲惫，也可能是一种社交信号，表示接收者并不重视这次互动；这与历史上的"Google it"式敷衍如出一辙，但因为 AI 对专家角色的替代感，影响更严重。
• 资深开发者常因无法满足的时间压力而把 AI 当作守门人以保护专注力，但这可能无意中削弱了以往支撑职业成长的辅导关系和机构知识的共享。
• 依赖 AI agent 解决内部问题存在形成知识孤岛的风险：机器生成的解决方案往往不可追溯、难以被团队其他成员获取，导致重复劳动和计算资源浪费。
• 受经济不稳定与裁员影响，职场正向更愤世嫉俗、交易化的方向转变，短期效率优先于耗时但以人为本的技术讨论。
• 把 LLM 当做教初学者做初步调查的教学工具是合理的，但如果用它来一刀切地回避培养初级人才的责任，就构成职业失职。
• 一些观察者认为对"Ask Claude"式回应的挫败感深层来源于生存焦虑：专家担心多年积累的经验正变得边缘化，自己沦为自动化流程中的盖章者。
围绕在专业环境中使用 AI 的紧张关系，本质是效率与保护人类专业知识之间的摩擦。资深开发者因持续且低效的打扰而疲惫不堪，初级开发者与同事则因把复杂问题简化为自动查询的敷衍回应而感到被疏远。这一动态暴露出职场文化危机：对速度和产出的追求常常压倒对深度理解与指导的需求。总体共识是，LLM 可作为初步探索的有力工具，但无法取代通过人与人协作培养起来的机构知识与批判性思维。 • Providing detailed context about prior research and specific limitations of LLM responses transforms a request from a generic burden into a productive collaboration, effectively signaling to experts that the seeker has genuinely exhausted preliminary options.
• The redirection to an LLM is frequently a defensive mechanism for senior staff, functioning as an "effort filter" against low-quality inquiries that consume limited time and cognitive energy without yielding professional growth.
• Widespread reliance on LLMs for code and design proposals risks creating a culture of passive verification, where developers copy-paste AI-generated output without understanding the underlying logic, ultimately compromising the quality and ownership of the work.
• Addressing misinformation or hallucinations generated by LLMs requires significantly more energy than initial fact-checking, imposing a "tax" on experienced practitioners who must repeatedly debunk AI-generated errors embedded in pull requests or project proposals.
• The "Ask Claude" response can be an expression of fatigue or a social signal that the recipient does not value the specific interaction, mirroring historical dismissals like "Google it" but exacerbated by the feeling that AI is replacing human expertise.
• Senior developers often face an impossible demand for their time, leading them to use AI as a gatekeeper to protect their focus, yet this can inadvertently erode the mentorship and institutional knowledge-sharing that historically defined career development.
• Relying on AI agents for internal problem solving risks siloing knowledge, as these machine-derived solutions often remain untracked and unavailable to other team members, leading to redundant effort and wasted computational resources across organizations.
• There is a perceived shift toward a more cynical, transactional workplace environment driven by economic instability and corporate downsizing, where employees prioritize short-term efficiency over the time-intensive process of human-centric technical discussion.
• Suggesting an LLM is a valid pedagogical tool when used to teach novices how to conduct initial investigations, but it becomes a professional failing when used as a blanket dismissal that avoids the responsibility of nurturing junior talent.
• Some observers suggest the frustration with "Ask Claude" stems from a deeper existential anxiety, where experts fear their years of accumulated experience are being rendered redundant or irrelevant, turning human colleagues into mere rubber stamps for automated systems.
The current tension surrounding the use of AI in professional environments centers on the friction between efficiency and the preservation of human expertise. While senior developers face legitimate exhaustion from constant, low-effort interruptions, junior developers and colleagues feel alienated by dismissive responses that reduce complex technical problems to automated queries. This dynamic highlights a crisis of workplace culture where the drive for output speed often overrides the necessity for deep understanding and mentorship. Ultimately, the consensus suggests that while LLMs are powerful tools for initial exploration, they cannot replace the institutional knowledge and critical thinking fostered through direct human collaboration.