AI 只能写诗,功夫却在诗外:论输入才是真正的创造 ✍ Zara Zhang🕐 2026-04-23📦 4.9 KB 🟢 已读 𝕏 文章列表 文章指出人们常误以为写作等同于生成文本,实则思考才是创作的核心。AI虽能高效完成输出,但无法替代“诗外”的功夫——即积累的生活阅历、用户洞察与判断力。真正的价值在于决定“做什么”的输入过程,而非“怎么做”的输出环节。 AI内容创作深度思考职业发展方法论思维提升价值判断 # AI can write the poem. But the real work happens before you pick up the pen. **作者**: Zara Zhang **日期**: 2026-04-23T01:22:56.000Z **来源**: [https://x.com/zarazhangrui/status/2047124027370869016](https://x.com/zarazhangrui/status/2047124027370869016) ---  I was recently discussing writing with a friend, and I realized a lot of people have a misconception about what creation actually is. People tend to think writing = generating text, so since AI is great at generating text, AI must be great at writing. But I have a counterintuitive take: when I sit down at my computer and start typing, that's not the beginning of writing. It's the last step. 80% of the writing is already done. It just happened in my head. The idea I'm writing about, I've probably been mulling over for a week or two. I've talked about it with different people, gotten various reactions and moments of recognition, seen other people post related thoughts. Different ideas and experiences collide and intertwine in my head, gradually forming the rough shape of what I want to write. By the time I actually sit down to type, the words flow out. It's easy, maybe 10 minutes and I'm done. So how long did my writing process actually take? The typing part took 10 minutes. But the thinking, the colliding, the forming of a perspective, that part probably took a week. Except nobody considers that time "writing." That's just "working, living." My social media looks like I'm posting constantly, but I actually spend no more than 3 to 5 hours a week making and publishing content. Because shooting videos and writing posts is just the "last mile" of content creation for me. By the time I'm talking into my phone about something, that idea has been circling in my head for days. I might have been thinking about it while brushing my teeth in the morning, thinking about it on my drive to work, bringing it up at lunch with someone... For content creation, "thinking" is 90% of it, and actually "making" the content is only 10%. And because my day job, my learning, and my hobbies are all AI related, I don't need to set aside extra time to do this "thinking." I'm already doing it day and night. Turning thoughts into content is just the easy part at the end. There's an ancient Chinese saying that I really like: "功夫在诗外" (the real craft lies outside the poem): writing a good poem isn't just about the act of writing. It takes a rich life, reading, thinking. AI can do a lot of things now, and everyone's worried about their jobs being replaced. I think every one of us should be asking: what is the actual essence of my work? - Writing ≠ generating text - Building product ≠ writing PRDs - Designing ≠ making mockups - Engineering ≠ writing code AI can quickly generate text, write product specs, make visuals, write code... but these are all the last mile of each job. They're the "poem" part. The part that happens "outside the poem" might be where the real value lives: - A good PM isn't just spitting out specs. They combine deep user understanding with technical judgment to figure out what's worth building - A good designer isn't just making mockups. They combine insight into human nature with business thinking to find the best version of an experience - A good marketer isn't just making videos and marketing assets. They precisely nail down a product's target audience and value prop, and they deeply understand what story to tell which audience. ... When a marketer writes a piece of viral marketing content, behind it there might be 10 customer interviews, years of industry intuition built up over time, their own real experience actually using the product... It's because of all that solid input that the best creative can happen. When I'm vibe coding, I notice AI is great at adding things. It'll pile on all kinds of features and interactions that "seem reasonable," but what it builds tends to be overly busy. AI is really bad at subtracting. Subtraction is an "outside the poem" kind of thing. Subtraction requires you to judge what actually matters, what's actually worth doing, what will survive contact with real users. Now that we have AI, it's forcing us to look past the surface and think about what makes us human. AI is good at producing output. But the output isn't the work. The input is the work. ## 相关链接 - [Zara Zhang](https://x.com/zarazhangrui) - [@zarazhangrui](https://x.com/zarazhangrui) - [1.4K](https://x.com/zarazhangrui/status/2047124027370869016/analytics) - [But the output isn't the work.](https://x.com/karrisaarinen/status/2045257582470983691?s=46) - [Upgrade to Premium](https://x.com/i/premium_sign_up) - [9:22 AM · Apr 23, 2026](https://x.com/zarazhangrui/status/2047124027370869016) - [1,449 Views](https://x.com/zarazhangrui/status/2047124027370869016/analytics) - [View quotes](https://x.com/zarazhangrui/status/2047124027370869016/quotes) --- *导出时间: 2026/4/23 11:27:03*
A AI圈自媒体大神晚餐观点记录 记录了与AI圈自媒体大神的晚餐讨论,主要观点包括每个人都应做自媒体以建立发声渠道,差异化定位和持续高质量输出的重要性,以及自媒体作为低成本获客方式的商业价值。强调独特观点和真实经验在AI时代的稀缺性。 职场 › 职业发展 ✍ 姚金刚🕐 2026-07-23 自媒体个人品牌差异化AI内容创作获客个人成长观点记录职业发展商业原则
小 小红书AI自媒体运营实战:从选题到变现的完整方法论 作者分享了半年运营小红书的经验,强调不追热点、深耕垂直领域的重要性。文章详细拆解了选题逻辑、内容创作(尤其是视频制作)、封面设计及个人IP打造思路,并指出搬运内容的弊端,建议通过“信息转译”和视频形式建立信任,最终实现工具商单、课程等多元化变现。 职场 › 职业发展 ✍ 小奇说🕐 2026-07-09 小红书自媒体AI变现运营视频剪辑个人IP内容创作方法论副业
A AI时代,会写的人越来越值钱 文章阐述了在AI时代,写作仍是内容创作的底层核心能力。作者指出,虽然工具降低了执行门槛,但能够驾驭AI、提供深度洞见和“灵魂”的作者将更具稀缺性。文章提倡将AI视为团队,自己作为统帅来建立个人写作系统,从而在内容泛滥的时代通过高质量的文字表达构建持久价值。 职场 › 职业发展 ✍ 蜗牛King🕐 2026-07-08 AI写作内容创作个人成长职场竞争力AIGC方法论
A AI 时代真正值钱的能力,是看懂人性背后的行为 文章探讨了在工具迭代极快的 AI 时代,真正核心的竞争力并非掌握某项具体技术,而是理解人性背后的行为逻辑。作者指出,无论工具如何变迁,人的欲望、恐惧和身份认同未变。真正值钱的能力是洞察用户为什么停留、相信和付费,以及如何通过共情而非操控来建立信任。 职场 › 方法论 ✍ Roland.W🕐 2026-07-08 AI人性认知职场发展内容创作心理学营销方法论
为 为什么一人公司OPC注定失败 文章通过分析当前“一人公司”的存亡现状,指出其高失败率是创业常态而非特殊危机。核心观点在于AI通过降低试错成本和补全单人能力短板,赋予个体“强化学习”的机会和凸性优势。但文章强调,AI仅是杠杆而非引擎,创业成功仍依赖于个人无法被替代的特定知识积累。失败应被视为廉价的反馈信号,AI让个体具备了高频试错并从中迭代的条件。 职场 › 职业发展 ✍ 花叔🕐 2026-05-31 AI创业一人公司职业发展方法论独立开发者认知提升
在 在中国,FDE 最危险的结局,是变成驻场外包 文章深入分析了 Forward Deployed Engineer (FDE) 在中国落地时的误区。作者指出,FDE 若缺乏场景、边界和验收的定义权,极易沦为懂 AI 的“高级救火队员”或驻场外包。文章通过教育转化、知识库搭建、车企新媒体三个案例,阐述了 FDE 应从解决具体问题转向重构业务系统,并强调了只有掌握场景定义权、边界定义权和方法沉淀权,才能避免被消耗,实现真正的价值。 职场 › 职业发展 ✍ HuangMing🕐 2026-05-27 FDE职场AI职业规划方法论外包Agent边界感职业发展ToB
A AI时代的内容创作方法论:从信息获取到故事讲述 本文作者分享了其在AI领域三年的内容创作经验,总结了一套包含“获取信息、找角度、创作”三步走的实战方法论。文章强调内容的本质是“讲故事”而非简单的“信息搬运”,提倡跨领域的信息摄入(美第奇效应),并详细阐述了如何通过“反差”和“陌生化”寻找独特视角,以及如何利用喜剧和电影的节奏理论优化内容结构。文末还分享了公众号运营的四个核心数据指标,强调正向价值观的重要性。 职场 › 方法论 ✍ 数字生命卡兹克🕐 2026-05-18 内容创作AI运营方法论故事叙述数据分析自媒体职场技能写作技巧美第奇效应
2 2026年,内容创作者最该卖的,根本不是内容 本文指出在2026年AI普及的背景下,单纯的内容创作已失去稀缺性。作者提出创作者应转变思维,不再售卖可被低成本复制的信息,而应转而售卖“判断力”、“陪伴”、“提问能力”、“身份认同”、“筛选”以及“动态更新”。文章强调,个人独特的认知框架和持续的迭代能力才是不可被抄袭的核心护城河。 职场 › 职业发展 ✍ Stanley🕐 2026-05-06 内容创作个人IP职业发展商业模式AI认知升级付费社群创作者经济思维转变
不 不增长就等死:关于消耗型与积累型增长的本质思考 文章指出增长是系统抵抗熵增、保持生命力的基本方式。区分了“消耗型增长”与“积累型增长”,以前者透支未来,后者通过迭代让未来变轻。结合瑞幸与ofo的案例,以及在AI浪潮下个人价值被重新定价的挑战,强调增长的目的在于保留选择权和韧性,而非单纯追求数字。 读书 › 方法论 ✍ Philip Van🕐 2026-05-04 个人成长职业发展认知升级方法论AI增长思维职业规划熵增
未 未来5年AI时代生存法则:能动性者赢 本文引用顶级博主Dan Koe的观点及神经科学研究,指出AI时代真正的核心竞争力并非技术,而是“能动性”。文章揭示了被动是大脑的出厂设置,而成功的关键在于通过习得性控制来打破被动。通过麦肯锡等数据佐证,AI越强,具备能动性、判断力和创造力的人才越值钱。最后提供了训练能动性的五步循环法,帮助读者在AI浪潮中掌握主动权。 职场 › 职业发展 ✍ Vincent|只上干货🕐 2026-04-08 AI职业发展能动性心智成长方法论神经科学Dan Koe
隐 隐性知识理论与内容创作启示 文章回顾了迈克尔·波兰尼从物理化学家转向哲学家的历程,深入解析其核心思想“隐性知识”(我们知道的比能说出的多)。文章阐述了隐性知识的运作机制、对实证主义的挑战,并特别探讨了该理论在人工智能及内容创作领域的应用,强调个人经验与直觉在 AI 时代的不可替代性。 职场 › 方法论 ✍ Cell 细胞🕐 2026-04-05 隐性知识迈克尔·波兰尼内容创作AI个人IP认知方法论哲学
A AI拆解爆款笔记的4个步骤与提示词实操 文章分享了如何利用AI分析155篇爆款笔记的方法论。作者指出内容创作应基于选题、结构、表达三个核心维度,而非凭感觉。具体实操分为四步:收集20篇高赞笔记进行结构分析、重点拆解开头钩子、归纳整理形成内容模板库、结合人工与AI工具进行数据核查。文章旨在通过拆解帮助创作者建立对优质内容的感知,并附带了原创AI提示词。 技术 › 工具与效率 ✍ 熊叔的茅草屋🕐 2026-04-05 AI小红书公众号提示词内容创作运营方法论AIGC