{
  "schemaVersion": "1.0",
  "book": "m-411",
  "number": 411,
  "title": "车开出农场之后",
  "authors": [
    "王德生"
  ],
  "sourceEdition": "初版（2026-10-22）",
  "taskVersion": "20261022-m411-unit-v1",
  "parts": [
    "车在农场里绕圈：为什么今天的AI还没变成财富",
    "财富是怎样发生的：从半径、三维、回写到意义",
    "怎样越过门槛：两道门槛、一次换轴、一套判别",
    "意义财富在哪里发生：从一盘红烧肉开始",
    "车开出农场之后：新账本、失败条件与第一道车辙"
  ],
  "tasks": [
    {
      "lesson": 1,
      "problemId": "M411-C01",
      "part": 1,
      "title": "千倍之说",
      "question": "各部门都说自己快了几倍，公司为什么没大几倍？请先猜一个理由，再读。",
      "checkpoint": "核对“千倍之说”要成立所需的条件，和那条直线假设错在哪。失败条件：若未来数据显示AI的好处主要按“谁买得多”而不是“谁改得多”分布，本章要收回。小行动：问一位同事，他的工作被AI加快后，省下的时间去了哪里。",
      "chapterTitle": "第1章　千倍之说",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s7",
      "sourceData": "/books/m/411/unit/sources/chapter-01.json",
      "sourceSha256": "a5374e27d7b6fb57f05516e542d32992a823a9a276cd965cc5e8b951a02dbfbb"
    },
    {
      "lesson": 2,
      "problemId": "M411-C02",
      "part": 1,
      "title": "旧问题的天花板",
      "question": "一件旧事的成本压到零，能多出多少财富？你觉得上限是多少？",
      "checkpoint": "核对“份额加饱和”的说法：天花板不是一堵硬墙，是旧问题在总账里的份额加需求的饱和。失败条件：旧问题单靠提效在钱账上长出数量级的持续增长而没碰到需要的边界。小行动：挑一项你每月花钱的旧服务，估一估它降到零价后你会多买几倍。",
      "chapterTitle": "第2章　旧问题的天花板",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s8",
      "sourceData": "/books/m/411/unit/sources/chapter-02.json",
      "sourceSha256": "30874a2c4fdc92a18e5ccbde9df7145c032e5b65ac79b90154fd4cc2223a21ea"
    },
    {
      "lesson": 3,
      "problemId": "M411-C03",
      "part": 1,
      "title": "绕圈的三笔代价",
      "question": "把周报写得更长、更漂亮，算不算进步？再想想：这样的事做久了，会留下什么？",
      "checkpoint": "核对三笔代价：浪费（原地自扩）、内卷（名次不动的多试几种写法）、土壤压实。失败条件：若系统证据显示省下的时间大多流向新产品与新问题，本章要收回。小行动：数一数你本周有几件事只是把旧东西做得更厚。",
      "chapterTitle": "第3章　绕圈的三笔代价",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s9",
      "sourceData": "/books/m/411/unit/sources/chapter-03.json",
      "sourceSha256": "f54e05528f37f23eb79206cd1cc554a1599b6abe43299362e59a331db6733b97"
    },
    {
      "lesson": 4,
      "problemId": "M411-C04",
      "part": 1,
      "title": "腾出来的去了哪里",
      "question": "AI 省下的人手和工时，你觉得最后去了哪里？先写下三个猜测。",
      "checkpoint": "核对匿占的五类去向：原地自扩、上溯回填、顺流转交、下投、未指认。失败条件：到2033年底，公开项目材料里写明余量接收环节的条目占比远高于一成。小行动：对你单位的一个AI项目，问一句“省出来的人去哪了”。",
      "chapterTitle": "第4章　腾出来的去了哪里",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s10",
      "sourceData": "/books/m/411/unit/sources/chapter-04.json",
      "sourceSha256": "f357933ff5e82b65568b0641a397c54969f012d840b0c0370bc269049f410067"
    },
    {
      "lesson": 5,
      "problemId": "M411-C05",
      "part": 1,
      "title": "名字先锁住了车",
      "question": "同一个新工具，叫“数字员工”和叫“新工种的起点”，会让人怎么用它？",
      "checkpoint": "核对名字怎样决定考核与记账，进而锁住用法；并带走“离开这个工具还能否存在”的新名字检验。失败条件：改了考核口径的单位，往外开的项目占比与只改名字的单位没有差别。小行动：把你手上一个AI项目的名字改成一句话，看它的考核会不会跟着变。",
      "chapterTitle": "第5章　名字先锁住了车",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s11",
      "sourceData": "/books/m/411/unit/sources/chapter-05.json",
      "sourceSha256": "64b776f29ba55ad30b42eee760afa64202d347c69c73b0a562d3d185fefdde19"
    },
    {
      "lesson": 6,
      "problemId": "M411-C06",
      "part": 2,
      "title": "车的本事是远",
      "question": "一辆车最值钱的是跑得快，还是能到原来到不了的地方？请举一个你身边的例子。",
      "checkpoint": "核对“半径比速度更值钱”：新品类从半径外长出来。失败条件：某工具带来的成倍财富全落在旧栏目降本里、没有长出新名字。小行动：找一件“因为交通变了才出现”的职业或店铺。",
      "chapterTitle": "第6章　车的本事是远，不是快",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s13",
      "sourceData": "/books/m/411/unit/sources/chapter-06.json",
      "sourceSha256": "9ed0a568835e317e13dbd8a01f51dadd563c9dbe2c0549c86981ec91e11ece16"
    },
    {
      "lesson": 7,
      "problemId": "M411-C07",
      "part": 2,
      "title": "财富的三维发生",
      "question": "只给AI加速度、不动土壤，会长出什么？想想两家裁缝铺的差别。",
      "checkpoint": "核对三维：D加量、E改写、S显露的关系，以及“被旧E吸收”的说法。失败条件：D加量相近的企业中，事先记下的E改写与后来新品类的出现毫无关系。小行动：列出你所在组织里一条没人想改的“旧规矩”。",
      "chapterTitle": "第7章　财富的三维发生",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s14",
      "sourceData": "/books/m/411/unit/sources/chapter-07.json",
      "sourceSha256": "5320e715d2e6ad745e87e6c22204bd6161903eb81844cb65b0920036efffcb37"
    },
    {
      "lesson": 8,
      "problemId": "M411-C08",
      "part": 2,
      "title": "千倍是复利",
      "question": "如果一个倍数每一步都解释不了，它会藏在哪里？",
      "checkpoint": "核对多轮回写的复利结构，以及“检验增长的构成，不检验千倍这个数”。失败条件：能拿出多个案例，工具一出现旧行业产出就翻了数量级，周边的人与制度都没跟着变。小行动：画一张你所在行业“上一轮增长”的构成图。",
      "chapterTitle": "第8章　千倍是多轮回写的复利",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s15",
      "sourceData": "/books/m/411/unit/sources/chapter-08.json",
      "sourceSha256": "28448eba1850ede41814363b70af21612405f889b2d8b8b262504e7a04c9154c"
    },
    {
      "lesson": 9,
      "problemId": "M411-C09",
      "part": 2,
      "title": "意义度：价值的来源",
      "question": "两件一样难的事，为什么一件更值钱？先凭直觉打个分，再读。",
      "checkpoint": "核对意义度三问（创造、自由、幸福）与“三问占优”的定义，以及降格为“挑问题的规范”的预先声明。失败条件：三问占优的一项五到十年后并未带出更多新名字与已验证条目。小行动：给你手头两个项目各打一次三问分，记下日期。",
      "chapterTitle": "第9章　意义度：价值的真正来源",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s16",
      "sourceData": "/books/m/411/unit/sources/chapter-09.json",
      "sourceSha256": "1ed08c681800a6fac96373502522096a6f6c0611d30fa0cf3ea19923eca65316"
    },
    {
      "lesson": 10,
      "problemId": "M411-C10",
      "part": 2,
      "title": "意义财富与资源天花板",
      "question": "一盘肉的价值怎么翻倍而猪肉一两没多？翻出来的是财富还是故事？",
      "checkpoint": "核对三道检验，尤其是第二道“对这个人”和回写型/在场型痕迹；不得以“新品类天生晚进账”开脱。失败条件：意义财富在钱账与新知识账上长期无痕迹。小行动：想一件你愿意多付钱的“我的”东西，说出多付的理由。",
      "chapterTitle": "第10章　意义财富突破资源天花板",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s17",
      "sourceData": "/books/m/411/unit/sources/chapter-10.json",
      "sourceSha256": "cea595c2589aae288cda4fbacc2c8a3c8349be8921471c4972a2a73212778fcf"
    },
    {
      "lesson": 11,
      "problemId": "M411-C11",
      "part": 3,
      "title": "第一道门槛：工具效率",
      "question": "一件新工具要强到什么程度才算有了旧工具没有的本事？",
      "checkpoint": "核对纽科门机与瓦特机的对照：门槛看的是“知识生产”上的效率。失败条件：只在生成上进步、验证依然昂贵的领域也稳定产出大批被确认的新知识。小行动：写一条“生成快、验证慢”的例子。",
      "chapterTitle": "第11章　第一道门槛：工具的创新效率",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s19",
      "sourceData": "/books/m/411/unit/sources/chapter-11.json",
      "sourceSha256": "f0f7d5a5f16d5fc1071e8b9e6f2344776fa2e17f927e574790c99634206ca9f3"
    },
    {
      "lesson": 12,
      "problemId": "M411-C12",
      "part": 3,
      "title": "第二道门槛：土壤要换",
      "question": "工具有了，为什么还会在旧土壤里空转几十年？",
      "checkpoint": "核对四层土壤：组织、技能、制度、账本与名字，及其在三界的分配。失败条件：土壤未改的一组十年里拿到与改造组相当的回报。小行动：对你的组织，从四层里挑最难动的一层，写下为什么。",
      "chapterTitle": "第12章　第二道门槛：土壤要换",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s20",
      "sourceData": "/books/m/411/unit/sources/chapter-12.json",
      "sourceSha256": "3cde5c92b824a98c4fc2fa5246b115fbdf0edb06a0c09bf66ae2ecfb5df84b66"
    },
    {
      "lesson": 13,
      "problemId": "M411-C13",
      "part": 3,
      "title": "从做事到知道",
      "question": "AI的效率对象从做事换到知道，倍数会乘在哪里？",
      "checkpoint": "核对“交汇处”的定义：老难题被新知识解开，同时逼出前所未有的新问题。失败条件：十年里新行业主要落在纯老难题或不需要新知识的新问题上。小行动：在你的领域找一个“老难题”和一个“新问题”各一个。",
      "chapterTitle": "第13章　从做事到知道",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s21",
      "sourceData": "/books/m/411/unit/sources/chapter-13.json",
      "sourceSha256": "b8888b7a35b439d1a672fefec474d2ee5130cb05ae136065673c81a70d0efcbf"
    },
    {
      "lesson": 14,
      "problemId": "M411-C14",
      "part": 3,
      "title": "生成不等于知识",
      "question": "一夜一千个假设，人类就多知道一千件事吗？",
      "checkpoint": "核对瓶颈在验证，以及这个判断在经济学里早有近亲。失败条件：验证昂贵的领域也稳定带来大批确认的新知识，或领域差异主要随政策体制而不随验证成本。小行动：挑一个AI给你的结论，设计一个能验证它的办法。",
      "chapterTitle": "第14章　生成不等于知识",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s22",
      "sourceData": "/books/m/411/unit/sources/chapter-14.json",
      "sourceSha256": "c9540779b3736e08bfc74261c03793e5c9689f06f6425e1fe3037c8695c3bdad"
    },
    {
      "lesson": 15,
      "problemId": "M411-C15",
      "part": 3,
      "title": "起手选对",
      "question": "同样勤奋用AI，为什么有人越干越窄？第一步从哪里看起？",
      "checkpoint": "核对从E起手与从S起手的区别。失败条件：事先分好类，五年后两类的新增产出落点没有差别。小行动：把你的一个AI计划的立项第一页重写，先写土壤、再写工具。",
      "chapterTitle": "第15章　起手选对",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s23",
      "sourceData": "/books/m/411/unit/sources/chapter-15.json",
      "sourceSha256": "7b9e9296647961dab477cda85868002e26b335bb5d68480e9fb15dfa52db4063"
    },
    {
      "lesson": 16,
      "problemId": "M411-C16",
      "part": 3,
      "title": "三问判别与门槛信号",
      "question": "怎样判断一项AI应用是在开新地还是在绕圈？",
      "checkpoint": "核对三问，以及“两道门槛都越过之后才看得见的信号”。失败条件：新类型问题在多个行业明显增多，“整体未过”的判断就要改。小行动：用三问判别一个你熟悉的AI应用，写出每问的回答。",
      "chapterTitle": "第16章　三问判别与门槛信号",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s24",
      "sourceData": "/books/m/411/unit/sources/chapter-16.json",
      "sourceSha256": "e4058c9d65c6faa0cd6d0f46e90e4b9d8f1d3d9cd28e0753a6775aea984b44ce"
    },
    {
      "lesson": 17,
      "problemId": "M411-C17",
      "part": 4,
      "title": "一盘红烧肉",
      "question": "同一块肉，端给不同的人、不同的日子，分量差几倍。AI能让“我的”变得不贵吗？",
      "checkpoint": "核对“从一盘到千万盘”三机制：支付意愿、新品类、回写，以及“口味固定、看处境”检验。失败条件：同一批人不同处境下评价不随处境系统变化。小行动：回忆一顿对你意义特别的饭，写下当时的处境。",
      "chapterTitle": "第17章　一盘红烧肉",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s26",
      "sourceData": "/books/m/411/unit/sources/chapter-17.json",
      "sourceSha256": "3d93da897a2cd6bca9f28c1af70bcb797e0de131b27c2da180bb88946fe318c2"
    },
    {
      "lesson": 18,
      "problemId": "M411-C18",
      "part": 4,
      "title": "我的药、我的课",
      "question": "药和课一直按平均的人设计，AI读一个人便宜了之后，会在哪里落地、哪里替人走了路？",
      "checkpoint": "核对学习者三问与“把一对一做便宜卖给请不起的家庭记在克里斯坦森一边”。失败条件：控制时间后，学生自己走的路在保留与迁移上没有增益。小行动：写下你学东西时“自己走的那一段”是什么。本章不构成医疗建议。",
      "chapterTitle": "第18章　我的药、我的课",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s27",
      "sourceData": "/books/m/411/unit/sources/chapter-18.json",
      "sourceSha256": "aa33d2ab8114b707c5897caebc00de80d2d27b899b194625ad4bd69e238f275b"
    },
    {
      "lesson": 19,
      "problemId": "M411-C19",
      "part": 4,
      "title": "不可想象的问题",
      "question": "最值钱的新问题在发生前叫不出名字，那我们能做什么？",
      "checkpoint": "核对“土壤策略”：不押注具体问题，而是改土壤、留余量。失败条件：回头查几十年，绝大多数重大新行业在五年前就已被准确命名并按规划长出。小行动：写一件五年前没名字、现在你离不开的东西。",
      "chapterTitle": "第19章　不可想象的问题",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s28",
      "sourceData": "/books/m/411/unit/sources/chapter-19.json",
      "sourceSha256": "bf27da619c754c3842b3dbed158fb533cdb9928b912017eba0bdfd36c19a53a4"
    },
    {
      "lesson": 20,
      "problemId": "M411-C20",
      "part": 4,
      "title": "画像归谁",
      "question": "意义个性化要读你的画像，这份画像归谁？",
      "checkpoint": "核对三项抽查：对照账号比价、推荐品类分散度随使用时长变化、导出与撤回是否生效。失败条件：画像归企业所有，而抽查显示用户并未系统多付、推荐并未越用越窄。小行动：在一个常用App里试一次导出或撤回你的数据。",
      "chapterTitle": "第20章　画像归谁",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s29",
      "sourceData": "/books/m/411/unit/sources/chapter-20.json",
      "sourceSha256": "0c50bf2a42ce02bc074f92d2600e8b04a69d267910142b5fb58c6cd9d605a45e"
    },
    {
      "lesson": 21,
      "problemId": "M411-C21",
      "part": 4,
      "title": "谁来判定意义度",
      "question": "AI让生成方案极便宜后，哪些问题值得解由谁判定？",
      "checkpoint": "核对“判定权改写的两条路”：换判定者、改判定者手里的账。失败条件：把判定权交给按旧收益优化的自动系统的组织，反而解开更多高意义难题。小行动：想一个你所在群体里的“立项会”，写出谁在判、用什么账判。",
      "chapterTitle": "第21章　谁来判定意义度",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s30",
      "sourceData": "/books/m/411/unit/sources/chapter-21.json",
      "sourceSha256": "6a342d88b5ada22035bc7df0b58043691fad3d32da9fce5bffba5ce43661789f"
    },
    {
      "lesson": 22,
      "problemId": "M411-C22",
      "part": 5,
      "title": "新账本",
      "question": "如果财富长到农场外面，用什么账本才看得见它？",
      "checkpoint": "核对两本账：钱账与新知识账（新名字份额加可解问题清单五状态）。失败条件：两项都明显上涨而实际生活与旧账描述毫无不同。小行动：给你的行业列一张“可解问题清单”，每条标五状态之一。",
      "chapterTitle": "第22章　新账本",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s32",
      "sourceData": "/books/m/411/unit/sources/chapter-22.json",
      "sourceSha256": "06f40a52097a739493e26ffaf6159e4aa13279b731b44d3ac5f7cc6e9a863489"
    },
    {
      "lesson": 23,
      "problemId": "M411-C23",
      "part": 5,
      "title": "怎样算错",
      "question": "一本讲未来的书，怎样才能让自己可以被证明是错的？",
      "checkpoint": "核对六条主命题的失败条件与2026—27、2031—32、2036年底的三次评分，以及“期限未到”“新品类晚进账”“意义涨了”不得作为推脱。失败条件：六条中有条款被触发而本书不改口。小行动：给你自己的一个判断写一条“到某年某月若如何就算我错”。",
      "chapterTitle": "第23章　怎样算错",
      "sourceUrl": "https://sdeuniverses.com/books/m/411/text/#s33",
      "sourceData": "/books/m/411/unit/sources/chapter-23.json",
      "sourceSha256": "b83ce4ed48dfd7a83172d5491da32d4f3703f67bc4697e319748fad40fe095fb"
    }
  ],
  "companionDesign": true,
  "realModelTested": false,
  "learningEffectTested": false
}