AI extrapolates fluency. I extrapolate emotion. AI 外推流畅,我外推情绪
8 Aug 20262026年8月8日
I use AI to draft client correspondence, and I don't forward what it writes. The popular division of labour — the AI handles the technical part, the human handles the human part — is backwards, and I think it's going to be the thing that catches a lot of people out. AI and people fail in opposite directions. That's not a weakness of either; it's the reason the pair works. 我用 AI 起草给客户的信,而且从不直接转发它写的东西。流行的分工——技术归 AI、与人打交道归人——是反的,而且我认为这一点将来会绊倒很多人。AI 和人的失效方向恰好相反。这不是任何一方的缺陷;这正是"这一对"能成立的原因。
Two failure modes, opposite by nature两种失效模式,天然对偶
Watch either side long enough and the pattern is stable. Both failures are extrapolation — filling a gap with something plausible — but they fill different gaps. 把任何一方看得够久,模式都很稳定。两种失效都是"外推"——用一个听起来合理的东西去填空白——只是它们填的空白不同。
| The AIAI | Me我 | |
|---|---|---|
| Fails by失效方式 | Fluency extrapolation. Writing a completed tense for something that never happened. Inventing a duration on someone else's behalf. Producing a comparison that quietly steps on a named person's work — because the sentence flowed that way.流畅外推。把从未发生的事写成完成时。替别人现场估一个工期。写出一句悄悄踩了某个具名的人的对比——只因为句子顺着这么流下来了。 | Emotion extrapolation. Reading enthusiasm as impatience, a delay as going cold, an escalation as a power grab. Filling silence with the worst story that fits it.情绪外推。把热情读成不耐烦,把延后读成变凉,把升级读成夺权。用"最坏的那个说得通的故事"去填沉默。 |
| How the other side corrects it对方怎么纠 | Every "has / will / should" gets checked back against the code and the record before it goes out.每一个"已 / 将 / 应该",发出前都回到代码和记录里核对一遍。 | Evidence before action: go re-read the actual thread, line by line, before responding to a feeling.先查证据再动作:在对一种感觉做出反应之前,逐行回去读真实的对话记录。 |
In practice both columns fire regularly, and both catch real damage. I've caught drafts claiming work was already done that wasn't, tripled time estimates offered on someone else's behalf, and sentences that read as a swipe at a colleague who would have been on that email. Going the other way, I've been talked out of responding to a client's silence as if it were displeasure — and been shown, by going back through the thread, that every part of my read was contradicted by what was actually written. 实际跑起来,两列都会定期开火,而且都真的接住过伤害。我抓到过把没做完的事写成已完成的稿子、替别人报出的三倍工期,以及读起来像在贬低某位同事(而那人就在收件人里)的句子。反过来,我也被劝住过——别把客户的沉默当成不满去回应——然后被带着逐条回读记录,发现我的每一个判断都被白纸黑字反驳了。
One detail matters more than the rest: the AI had the entire context and still missed things. The sentence that stepped on a colleague was written with every relevant file in the window. So the review layer isn't there because one side is under-informed — it's there because every single layer has its own blind spot, and the only reliable fix for a blind spot is a second layer that is blind somewhere else. 有一个细节比其余都重要:AI 拥有全部上下文,依然会漏。那句踩到同事的话,是在所有相关文件都在窗口里的情况下写出来的。所以审查层的存在,不是因为哪一方信息不足——而是因为任何单独一层都有自己的盲区,而盲区唯一可靠的解法,是一个盲在别处的第二层。
The reading time is the product, not a tax理解所花的时间是产品,不是税
Spending two hours understanding a draft I could have forwarded in five minutes looks like waste until you ask what the client is buying. They're buying a person who will put their name on the result. You cannot stand behind something you didn't understand. When they call tomorrow and ask what a particular margin figure means, answering fluently is the deliverable — and you can only do that because of those two hours. 花两小时去吃透一份"五分钟就能转发"的稿子,看上去是浪费——直到你问客户到底在买什么。他们买的是一个愿意把自己名字押在结果上的人。你没法为你没理解的东西担保。明天他打电话来问某个余量数字是什么意思,你能流利地答上来,这本身就是交付物——而你能答上来,正是因为那两小时。
And it capitalizes. Understanding a mechanism once puts it in your head permanently; the next letter that touches it costs nothing. The cost curve for a given client falls on its own as templates and shared shorthand accumulate — so don't estimate the steady state from what a phase boundary costs. 而且它会资本化。一个机制理解过一次就永久留在脑子里;下一封涉及它的信成本为零。随着模板和默契积累,单个客户的成本曲线会自己往下走——所以别用"阶段交界那封信"的成本去估常态。
Here's the part I'd bet on: in two years the market will be full of people forwarding AI drafts. They will fail in exactly the ways in the AI column above — the invented completion, the borrowed estimate, the sentence that steps on someone. Not because their model was worse, but because a forwarder has no second layer. The protocol is the moat, and it's an unusually cheap one. 这是我愿意押注的部分:两年后,市场上会满是转发 AI 稿的人。他们翻车的方式,恰好就是上表 AI 那一列——凭空的完成时、借来的工期、那句踩人的话。不是因为他们的模型更差,而是因为转发员没有第二层。这套协议就是护城河,而且是一条便宜得反常的护城河。
Tiered review: spend where the stakes are分层审查:把力气花在赌注所在处
Reviewing everything at full depth doesn't scale and isn't necessary. Three tiers, chosen by what the letter can cost you: 对每封信都做完整深度审查,既不可扩展也没必要。按"这封信可能让你付出什么代价"分三档:
- S — acceptance, quotes, scope, conflict, first impressions. Understand every detail, however long it takes. With a new client the first several letters are all S, because the first impression prices everything that follows. S 档——验收、报价、范围、冲突、第一印象。吃透每一个细节,花多久都值。面对新客户,前几封全是 S 档,因为第一印象会给之后的一切定价。
- A — substantive technical exchange. Thirty to sixty minutes on three lines only: (1) every commitment — tense, number, date; (2) a scan for anything that steps on a person, absolutizes, or estimates on someone else's behalf; (3) can I restate the core mechanism in my own words? If I can't restate it, I didn't understand it, and it doesn't go out. A 档——实质性技术往来。三十到六十分钟,只审三条线:(1) 每一个承诺——时态、数字、日期;(2) 扫一遍有没有踩人、绝对化、或替别人估时的句子;(3) 我能不能用自己的话复述核心机制?复述不出来,就是没懂,那就不发。
- B — confirmations, status, short replies. Two-minute scan, send. B 档——确认、状态、简短回复。扫两分钟,发。
The restatement test in tier A does most of the work. It's not a proxy for understanding — it is understanding, and it's the only one of the three checks that can't be faked by skimming. A 档里那条"复述测试"承担了大部分工作量。它不是理解的替代指标——它就是理解,而且是三条检查里唯一没法靠扫读糊弄过去的。
What the AI owes the reviewAI 欠这场审查什么
The pair only works if the reviewing side knows where to aim. So every draft comes with a review-focus list attached: the two or three places the AI is least sure about, the complete list of commitments the draft makes, and every sentence that names a person or carries a tone. That single habit cut my review time by roughly a third without reducing what I caught — because the time now lands where the density is. 只有当审查那一方知道该往哪儿瞄,这一对才成立。所以每份稿子都附一份"审查焦点":AI 自己最没把握的两三处、这份稿子做出的全部承诺清单、以及每一句涉及具体人名或带语气的话。就这一个习惯,把我的审查时间砍掉约三分之一,而抓获量不减——因为时间现在打在密度最高的地方。
The rule I'd keep if I could keep only one只能留一条规矩的话,我留这条
Every "X should happen" must have had X actually happen once first. Checklists, protocols, promised behaviours — all of them get written after the thing has been done at least once, never before. Written the other way round they read exactly the same, and they're fiction. This is the same discipline as the rest of my work: don't automate a flow you haven't run by hand many times, and don't claim a result you haven't measured. 每一个"应该发生 X",必须先让 X 真的发生过一次。清单、协议、承诺的行为——全都在事情至少做过一次之后才写下来,绝不在之前。反过来写出来的东西,读上去一模一样,但它是虚构。这和我其余工作里的纪律是同一条:别去自动化一条你没手动跑过很多遍的流程,别去声称一个你没测过的结果。
Written from how I actually run client correspondence — this note is itself a product of the pair described in it, and the draft it started from had three commitments in it I had to go and verify. No specific client, project or person is described. 写自我真实运转客户往来的方式——这篇笔记本身就是它所描述的那一对的产物,而它起源的那份初稿里有三处承诺是我不得不回去核实的。文中不描述任何具体的客户、项目或个人。
Related: the same "make it structural, don't rely on anyone being careful" instinct, applied to a machine — I use AI heavily. The machine still can't hurt you.; and to money — Governing AI agents that touch money. On keeping the communication overhead low in the first place: One email per milestone. 相关:同一种"做成结构性的,别指望谁小心"的直觉,用在机器上——我大量用 AI,机器依然伤不了人;用在钱上——让碰钱的 AI Agent 受治理。至于一开始就把沟通开销压低:每个里程碑,一封邮件。
If a machine you build needs an interface, a device connection, or data that has to land somewhere else — tell me what it's costing you now. You'll get an honest read on whether it's solvable, and usually something running to look at. Start here →如果你造的机器需要一套界面、一个设备连接,或者数据必须落到别处去——告诉我它现在正让你付出什么代价。你会得到一个诚实的判断:能不能解;通常还会收到一个能跑的东西。从这里开始 →