Ben Thompson Says Capital Mismatch Endangers AI Buildout

Ben Thompson称资本错配危及AI建设 随着庞大的资本支出超出短期营收回报，AI热潮面临重现1870年代铁路恐慌的风险，迫使科技巨头转向债务和股权市场来资助电力与算力基础设施。 Ben Thompson分析了当前AI热潮更广泛的经济与地缘政治影响，将其与1870年代铁路泡沫和1990年代末电信扩张等历史基础设施建设相提并论。他探讨了为何美国对中国的绝对AI主导地位可能会破坏稳定、消费级AI在商业变现上为何与企业软件存在根本差异，以及为何资本而非算力可能成为制约AI增长的瓶颈。讨论还详述了大型科技公司的战略布局、针对Nvidia的定制芯片竞争，以及大宗商品经济学如何驱动硬件、内存和电力格局的发展。 美国在AI领域的绝对主导地位将给全球地缘政治稳定带来风险，因为掌控优势AI若能直接转化为军事主导地位，可能会促成诸如摧毁TSMC等激进行动。 (2:16) 开源AI模型虽消除了前期研发开销，但在大规模运行时仍会产生巨大的持续推理成本。 (5:00) 庞大的AI基础设施建设类似于1870年代的铁路泡沫，当时资本枯竭的速度快于直接回报实现的速度，但却留下了基础性的实体资产。 (8:28) AI能力在编程或游戏等有界且可验证的领域能够顺畅扩展，但让模型适应复杂、无法验证的人类现实环境仍是一个未解的挑战。 (15:00) AI基于使用量的可变定价，让习惯了成本可预测、按席位付费的SaaS企业买家产生了预算阻力。 (23:10)

Ben Thompson argues that the surging artificial intelligence boom faces a capital duration mismatch and commodity market dynamics reminiscent of 19th-century infrastructure bubbles. While massive buildouts of compute and electrical power will yield durable economic assets, near-term capital constraints and marginal inference costs will disrupt software pricing, semiconductor supply chains, and Big Tech profit margins. Capital Duration Mismatches and Historical Infrastructure Cycles (8:28) Thompson highlights a widening duration mismatch between immediate capital outlays and revenue realization. Tech companies have rapidly moved down the capital spectrum, exhausting free cash flow to tap debt markets, equity offerings, and structured deals with pension funds. Comparing this to the 1870s railroad bubble, Thompson notes that while financial markets temporarily "ran out of money," the physical tracks remained to drive long-term GDP growth. Similarly, Google's high-margin Search engine resembles See's Candies funding BNSF Railway—a lower-margin, capital-intensive infrastructure project with a much higher absolute profit ceiling. Even if hardware investments face write-downs, Thompson argues financial bubbles leave durable assets behind. Just as the dot-com crash left dark fiber networks that enabled Google and modern internet infrastructure, current AI spending is driving behind-the-meter natural gas plants and nuclear restarts, permanently expanding baseline electrical capacity. Commodity Economics and Semiconductor Supply Chain Bottlenecks (30:08) Thompson contends that current compute shortages stem from underinvestment in 2023 and 2024 rather than instantaneous demand. Projected capital expenditures—estimated at $800 billion this year and $1.3 trillion next year—will only materialize as online compute around 2028 and 2029. Likening data centers to container shipping, he explains that high fixed capital costs compel operators to run at full capacity and slash prices to marginal operating costs once scarcity subsides. Meanwhile, supply chain bottlenecks are reshaping competitive leverage. TSMC's conservative fab expansion intentionally avoids long-term overcapacity, shifting foregone revenue risk to Big Tech while creating an opening for Intel and Samsung logic foundries. Similarly, memory manufacturers' supply discipline resembles strategic deterrence: by restricting supply, they incentivize customers like Apple to engineer algorithmic workarounds and source Chinese alternatives, threatening to neutralize memory makers' pricing power by 2035. Inference Marginal Costs and Consumer Monetization Models (20:00) Aggregation Theory historically relied on zero marginal distribution and transaction costs, but AI disrupts this model because inference—particularly reasoning tasks via test-time scaling—imposes direct variable costs. In enterprise software, token-based metering breaks Microsoft's traditional per-seat bundling by forcing corporate buyers to audit variable monthly usage. In consumer software, Thompson argues subscriptions like OpenAI's $20 monthly fee face rigid consumer price resistance, as consumers do not pay for personal productivity tools. Instead, advertising represents the natural consumer monetization model. While pure-play model builders face subscription ceilings, Google and Meta already capture substantial AI revenue by optimizing their ad marketplaces—a verifiable domain where automated A/B tests continuously validate model outputs against user conversions. Big Tech Architectural Defenses and Disintermediation Risks (45:00) Incumbents are adopting divergent strategies to defend existing moats. Amazon acts as its own "first and best customer," incubating proprietary Trainium and Graviton silicon inside managed cloud services before external commercialization. Apple leverages its hardware aggregator status to deliver on-device AI, offloading cloud inference expenses while facing cultural tension between deterministic hardware manufacturing and probabilistic AI models. Microsoft has largely bypassed unconstrained frontier training costs, generating $40 billion in quarterly free cash flow by adopting Lou Gerstner's 1990s IBM playbook. By packaging third-party models into an enterprise middleware harness, Microsoft captures recurring software spend, though it remains vulnerable to AI-native interfaces like Cursor or Claude that bypass legacy enterprise user interfaces. Power Delays, Custom Silicon, and Nvidia's Margins (67:29) Nvidia's reported margins are partially sustained through indirect balance-sheet risks, such as taking equity stakes in neoclouds and guaranteeing customer compute purchases, which Thompson views as implicit price cuts. Furthermore, power constraints took longer to become an operational bottleneck than Nvidia anticipated, granting hyperscalers runway to optimize internal alternatives like Google's TPUs and Amazon's Trainium. As custom hyperscaler silicon matures and CUDA lock-in weakens among efficiency-focused AI labs, sustaining Nvidia's record margins will become increasingly difficult. Thompson characterizes himself as a "reluctant accelerationist," maintaining that society cannot halt technological progress without stagnation. However, whether frontier AI can expand beyond bounded, verifiable domains into open-ended human workflows before market capital is exhausted remains an unresolved question.

科技分析师 Ben Thompson 在访谈中指出，当前 AI 热潮正在逼近资本投入与实际收入兑现的时间差极限，科技巨头已从消耗自由现金流快速转向举债和股权融资，呈现出类似 1870 年代铁路泡沫中“全球暂时耗尽流动资金”但留下持久物理基建的演进轨迹。他认为，AI 产业扩张不仅受制于半导体供应链的滞后性与推理边际成本，还面临企业端软件打包逻辑动摇与消费端过度依赖订阅模式的商业模式挑战。 地缘博弈与半导体供应链风险 (0:00) Thompson 指出，若美国在 AI 领域取得绝对优势并直接转化为军事霸权，从博弈论角度看，中国可能的最优选择是摧毁台积电（TSMC）。他批评硅谷认为美国能迅速实现供应链本土化是“虚妄幻想”（magical thinking），从晶圆代工到专用硬件的全面脱钩在成本上极其沉重，苹果将部分 iPhone 产线移至印度实质上是在支付高额“保险”，而非真正与中国制造彻底脱钩。 同时，台积电采取高度保守的产能扩张策略以规避长期产能过剩风险，实质上将未实现利润与供给短缺风险推给了下游科技巨头。但 Thompson 认为，这种供不应求反而在商业上迫使超大规模云服务商承担试错成本，转向扶持英特尔（Intel）和三星（Samsung）的代工业务，无意中缓解了美国过度依赖台湾代工的地缘安全风险。存储芯片制造商对扩建产能的谨慎态度虽然构筑了当下瓶颈，但也在倒逼客户通过算法创新降低内存依赖或引入替代供应商，中长期来看或将削弱其自身的议价权。 资本约束与大宗商品化周期 (8:28) 针对 AI 基础设施的融资趋势，Thompson 强调当下的巨额资本开支（预计今年达 8000 亿美元，明年达 1.3 万亿美元）与产生规模化收入之间存在周期错配。科技企业耗尽内部现金流后，正以极快速度转向债券市场与股权工具（如谷歌发行股票，英伟达通过结构化交易引入养老金和保险资金）。他将此比作 1870 年代的铁路泡沫：尽管短期资金耗尽并触发市场恐慌，但建成留存的物理基建在随后的数代人中持续推动长期 GDP 增长；类似地，当下 AI 建设即便发生资产减计，其促成的高等级电力与电网扩容也将转化为整体经济的长期能源红利。 然而，数据中心正受制于大宗商品市场的经济规律。Thompson 以集装箱航运和航空业为例指出，数据中心的核心成本是账面折旧，边际服务成本极低；巨额固定资产一旦落成，运营商绝不会任其闲置，只会在供应过剩时以覆盖边际运营成本的价格竞相杀价。目前各大厂商在产能稀缺期计算的高回报率，随着 2028 至 2029 年大批算力集中上线，可能面临产能过剩与价格崩跌的剧烈周期反噬。 推理边际成本与商业变现困境 (20:00) 以零边际分发与交易成本为基石的“聚合理论”（Aggregation Theory）在 AI 时代遭遇了挑战。Thompson 分析称，虽然简单的消费级查询边际成本极低，但需要模型深度推导的“测试时扩展”（test-time scaling）使得推理计算直接转化为昂贵的可变成本。这种经济学特性的转变动摇了企业级软件的商业逻辑：微软等传统巨头长期仰赖按人头固定包月（如每人每月 50 至 100 美元）的稳定预算模式，而加入基于使用量的 Token 计费后，企业客户被迫按月复核浮动支出并严审工具效能，削弱了传统一站式全家桶软件的议价空间。 在消费级市场，Thompson 认为硅谷盲目推行每月 20 美元的订阅模式重蹈了 Dropbox 当年的覆辙。他指出普通消费者既不愿为软件持续付费，也不关注个人生活生产力；消费级 AI 的终极变现路径应当是广告模式，通过广告主承担算力成本，将商业化与用户真实参与度深度绑定。此外，AI 在编程、数学以及广告转化等存在明确反馈机制的“可验证领域”成效显著，但在法律、商业战略等缺乏清晰闭环反馈的复杂现实世界中，其自动化落地依然充满阻力。 巨头策略分化与芯片生态博弈 (45:00) 在硬件与云端竞争中，Thompson 认为英伟达虽然维持高毛利，但通过入股新型云服务商（neoclouds）并担保算力采购，实际上吸收了客户的信贷与资产负债表风险。长期来看，拥有更低资本成本的谷歌（TPU）和亚马逊（Trainium）正将自研芯片转为低成本大宗商品对外提供服务，加之大型语言模型更关注计算效能而非底层软件黏性，CUDA 的生态壁垒在大型开发者中正被削弱。为此，英伟达正复制英特尔当年的企业级策略，转向技术迁移能力有限的普通企业及主权云客户。 不同科技巨头的防御策略也出现显著分化：微软正执行类似 1990 年代 IBM 的“中间件策略”，不盲目单赌前沿模型训练，而是通过构建企业级合规与治理中间层锁定客户；亚马逊依托“自身即首选最佳客户”的飞轮，先在内部业务打磨自研芯片与 AI 架构再推向外部；苹果则凭借硬件聚合者地位控制消费入口，试图押注端侧本地运行来规避云端推理成本，但其在确定性硬件制造上的精细文化，可能会在面对概率性 AI 与环境计算演进时产生战略摩擦。 Thompson 自称为“勉为其难的加速主义者”（reluctant accelerationist），认为技术停滞不可取，但人类制造繁文缛节与摩擦的能力同创新能力同样漫无止境。目前全美电力瓶颈显现的速度慢于各方早前预期，为超大规模云厂商研发自主芯片争取了喘息窗口；这场由数百上千亿美元资本驱动的 AI 盛宴能否平稳着陆，最终取决于算力基础设施能否在金融资本耐心耗尽之前，在广泛的实体经济中兑现真正的生产力回报。

## Background 该内容由知名商业与投资播客平台 Colossus 旗下的《Invest Like the Best》节目发布于 2026 年 8 月 18 日，由主持人 Patrick O'Shaughnessy 对话著名科技分析博客 Stratechery 的创始人 Ben Thompson。节目简介指出，本次对谈全面拆解了 OpenAI、Nvidia、Intel、Apple、Microsoft、Google 与 Amazon 等科技巨头的竞争格局，并从 1870 年代铁路泡沫和现代集装箱航运的历史切入，探讨中美 AI 竞赛的博弈陷阱以及资本流动性对当前 AI 狂潮的刚性约束。 ## TL;DR Ben Thompson 在本期节目中提出了颠覆性的逆向视角，认为美国在 AI 竞赛中取得绝对单边胜利可能诱发灾难性地缘风险，维持落后 6 到 9 个月的动态均势反而是最优解。他深入剖析了当下 AI 浪潮的本质：当前最严苛的瓶颈不是算力或电力，而是类似 1870 年代铁路建设的资本期限错配，科技巨头在耗尽自由现金流后正通过密集发债与股权增发向下汲取长期流动性。与此同时，他运用大宗商品与边际成本模型，阐明了消费级 AI 必须回归广告变现的宿命、台积电转嫁产能风险如何戏剧性地拯救英特尔，以及超大规模云厂商自研芯片对英伟达非自然高利润率的潜在威胁。即便 AI 资本泡沫发生剧烈断层，这场军备竞赛所逼出的电网与能源基建，也将像当年的暗光纤和铁路一样，成为沉淀给人类社会的永久财富。 ## Atmosphere & style 整场对话呈现出高密度、冷峻且充满战略张力的高智商对谈氛围，两位长期观察技术周期的行家里手在思想交锋中展现出极强的宏观掌控力。Ben Thompson 运用其标志性的框架推演与历史隐喻娓娓道来，将原本喧闹的硅谷叙事剥离出冷静的博弈论与资本账本，既无盲目狂热也不流于消极看空。整期内容听来如同一堂精密展开的商业沙盘推演，在层层剖析巨头暗战与宏观断层的同时，透着一种“不情愿的加速主义者”直面技术洪流时的清醒、克制与苍凉感。 ## Bulletin - `00:00` 本·汤普森提出逆向视角：美国单方面彻底赢得 AI 竞赛极其危险，若将对手逼入绝境，博弈论下的极端应对可能直接演变成破坏台积电，当前维持领先 6-9 个月的动态均势反而更为安全。 - `07:00` AI 建设的核心硬约束不是电力或算力，而是资本期限错配：类似 1870 年代铁路泡沫，巨头短期内耗尽自由现金流与债务容量甚至转向股权融资，面临商业回报无法覆盖短期天量 CapEx 的资本断层风险。 - `30:00` 科技行业普遍忽视大宗商品规律，将短期算力稀缺误认为永久护城河；英伟达超额利润率依赖向新兴云厂商兜底包销的变相降价循环，而台积电则通过保守扩产将产能过剩风险完全转嫁给下游客户。 - `22:06` 消费级 AI 商业化的唯一归宿是广告，单靠订阅制存在明显天花板，且开源模型仅免除研发成本而未降低边际推理成本；企业级软件则正遭遇用量计费破坏传统人头费捆绑模式的严重挑战。 ## Key takeaways - `00:00` 美国在 AI 领域的绝对单边胜利并非最优解，过度垄断优势可能打破恐怖平衡并诱发针对台积电的地缘灾难，维持落后 6-9 个月的追赶格局反而是最健康的博弈均衡。 - `03:30` 脱离亚洲制造业生态的本土建厂充斥着“魔法思维”，去中国化在和平时期是一份成本极高且缺乏商业竞争力的保单。 - `06:15` 开源模型免除的是前期研发与预训练沉没成本，单次推理的边际成本分文未减，中小机构自建集群的调用成本反而往往高于头部闭源 API。 - `07:00` AI 基础设施建设面临严重的期限错配危机，科技巨头正在沿着资本结构向下汲取流动性（从 FCF 到发债再到股权融资），市场资本可能先于商业闭环耗尽。 - `11:34` Google 增发股权的逻辑类似于巴菲特收购 BNSF 铁路：AI 虽是低利润率业务，但资本容量无限，企业思考重心已从追求高百分比利润转向吞吐超级体量的绝对利润。 - `14:23` AI 模型在代码、数学等有明确短回路的可验证领域已具备实用价值，但在开放复杂、缺乏即时反馈的不可验证领域，其通用泛化能力仍未得到证实。 - `20:20` 聚合理论在 AI 时代受到推理边际成本两极分化的冲击：简单事实查询成本极低，但基于思考时间扩展（Test-Time Scaling）的复杂推理边际成本呈线性飙升。 - `22:06` 微软的用量计费模式破坏了传统的员工人头费捆绑优势，迫使企业客户按月审查 Token 开销并反思每个软件组件的真实价值，动摇了 SaaS 商业惯性。 - `25:40` 消费级 AI 唯有拥抱广告才能跑通无限变现飞轮，直接向消费者收取订阅费受制于价格弹性，且普通用户天然抗拒为提高生产力付费。 - `30:00` 当前算力短缺是大宗商品式的供需周期错配而非需求无限，科技巨头砸入的高额固定成本将逼迫算力价格在未来过剩时被打压至边际成本。 - `35:00` 台积电坚守极其保守的资本开支策略以规避晶圆厂三十年折旧风险，实质上将产能过剩的致命风险完全推卸给了承受算力饥渴的下游巨头。 - `42:08` 台积电的产能制约与巨头错失的巨额利润，反向扭转了商业逻辑，迫使西方大厂克服磨合阵痛真金白银扶持英特尔与三星代工，免费获得了地缘供应链对冲。 - `44:51` 亚马逊的核心优势在于充当自研技术的首位最佳客户（First Best Customer），借助自身托管云服务与电商体系掩盖芯片早期缺陷，完成低阻力迭代。 - `48:15` 苹果作为终极聚合者通过掌控终端入口转移了高昂推理成本，但若执念于 iPhone 为中心而忽视环境智能演进，恐陷入类似当年微软错失移动互联网的思维陷阱。 - `54:27` 微软正在复制 90 年代 IBM 的中间件剧本，将变动不居的基础模型封装为企业级稳定服务以避免客户流失，但其核心 UI 界面正面临任务级 AI 智能体的架空威胁。 - `01:02:15` Meta 与 Google 的广告系统是全球最高效的 AI 验证机器，广告竞价 A/B 测试驱动的微小预测力提升可直接转化为数十亿美元净利润，为其高额模型研发提供造血支撑。 - `01:08:24` 英伟达通过注资新兴云厂商并承诺包销算力维持了超高毛利表象，这种资产负债表外安排实质构成了变相降价并揽下了巨大的违约风险。 - `01:12:45` 美国能源供给弹性的超预期释放推迟了电力瓶颈，缩短了英伟达的能效护城河保护期，同时也为云巨头自研芯片的追赶赢得了关键窗口；最终沉淀的庞大电力基建将成为打破能源匮乏的永久遗产。 ## Detailed walkthrough ### `00:00` 重新审视中美 AI 竞赛：完全胜利背后的博弈陷阱 对谈开场，Patrick O'Shaughnessy 提到距离两人上次交流已经过去了很久。彼时世界尚未掀起 AI 浪潮，讨论重心还是聚合理论（Aggregation Theory）。如今技术格局天翻地覆，Patrick 抛出的第一个问题便是：如果美国在人工智能竞赛中取得“决定性胜利”，究竟意味着什么？ Ben Thompson 提出了一个极具颠覆性的逆向视角：美国彻底赢得 AI 竞赛可能并不是一件好事，甚至对全球来说非常危险。硅谷眼下流行着一种乌托邦式的幻想，认为一旦美国完全掌控前沿 AI，就能实现军事霸权、重振本土制造业并解决所有现实供应链难题。但在博弈论框架下，如果中国意识到美国即将在 AI 和国家安全层面建立起压倒性的绝对优势，其最优的博弈应对策略可能直接演变成“炸毁台积电（TSMC）”。硅谷某些前沿实验室不断鼓吹“不惜一切代价击败对手”的叙事，却忽视了这种将对手逼入绝境所诱发的灾难性地缘风险。 ```mermaid flowchart TD accTitle: AI 绝对领先引发的地缘博弈风险 accDescr: 展示美国单方面垄断 AI 优势如何反向推高地缘冲突风险的逻辑链条 subgraph US["美国单边胜出假设"] A["前沿 AI 绝对领先"] --> B["获取不对称军事与国家安全优势"] end subgraph Reaction["博弈论最优对抗反应"] B --> C["破坏核心硬件咽喉节点 (如台积电)"] C --> D["全球半导体与算力供应链归零"] end US -.->|"打破恐怖平衡"| Reaction ``` ### `03:30` 制造业脱钩的“魔法思维”与现实成本 针对“美国能否通过本土建厂规避台积电断供风险”的设想，Ben 直言这充斥着不切实际的“魔法思维”。芯片制造、执行器（actuators）以及各类工业前驱原材料的供应链深度嵌套在亚洲生态中。除非爆发真正的军事冲突，否则西方在和平时期根本无法真正摆脱对中国制造业的依赖。 这里的核心阻力在于经济规律： - **竞争劣势**：如果竞争对手继续从中国采购高性价比零部件，而你选择成本高昂的美国本土替代品，企业将在商业竞争中处于绝对劣势。 - **天价保单**：供应链去中国化就像买一份极其昂贵但平时看似用不上的商业保险，只要没有被逼到悬崖边缘，没有企业愿意主动承担这种沉没成本。 - **苹果案例的局限**：即便白宫对苹果施压，促使其将部分 iPhone 产线迁往印度，这也仅属于边缘维度的产能分散，远谈不上真正切断与中国工业体系的纽带。 ### `05:00` 理想的动态均势：硅谷的稻草人与“六到九个月”差距 在 Ben 看来，硅谷大模型叙事之所以极度执着于“必须击败中国”，很大程度上是因为金融市场需要一个完美的“假想敌”（bogeyman）来支撑 AI 资产的高估值。与之相对，过去几年美国在政治监管层面的应对往往反其道而行之，试图通过加强自上而下的审查与管制来对抗外部竞争。Ben 认为美国的真正优势在于开放度、包容性与自下而上的敏捷创新，而不是模仿对手的集权控制。 谈及全球 AI 竞争的终局形态，Ben 认为当前类似台海局势的“现状维持”反而是最健康的动态均衡（Dynamic Equilibrium）： - **第一梯队（Frontier）**：OpenAI 与 Anthropic 稳居全球前沿，主导突破性研究。 - **追赶梯队**：Google 动向变幻莫测，Meta 与 xAI（Grok）在身后紧追不舍。 - **中国阵营**：凭借极强的工程化能力与快速蒸馏（Distillation）技术，持续保持落后第一梯队约 6 到 9 个月的身位。 这种均势既为西方提供了战略安全感，又避免了因单方绝对垄断而激化极端地缘冲突。真正的变数在于：当 AI 开始被用来写代码、优化架构并训练下一代 AI（即递归自我演进）时，领先者的优势可能会迎来指数级飞跃，届时后发者是否还能依靠模型蒸馏维持这几个月的距离将充满悬念。 ### `06:15` 开源模型的“免费”认知偏差与边际成本真相 行业中普遍存在将“开源模型”等同于“免费使用”的严重误区。Ben 尖锐地指出，开源模型仅仅免去了前期的研发（R&D）与预训练沉没成本，但在实际落地与生产环境中，每一次调用所产生的边际推理成本（Inference Marginal Cost）分文未减。 甚至在很多工程实践中，开源自建部署的单次响应成本反而显著高于采用顶尖闭源 API 的方案。原因在于： - 前沿头部实验室能够借助超大规模请求对算力栈进行极致优化，并在数据调度、KV Cache 压缩和硬件利用率上建立深厚的工程壁垒。 - 领先的闭源厂商如果能够率先利用 AI 工具重构自身的软硬件技术栈，其服务边际成本（Cost to Serve）在结构上反而会远低于自行架设开源集群的中小机构。 ### `07:00` 前沿黑盒化与资本结构的极限透支 针对未来的发展拐点，Ben 表达了对信息透明度与资本流动速度的双重隐忧。由于近期对模型安全与系统风险的恐慌，许多前沿实验室正在缩紧模型发布节奏。当公众与产业界逐渐无法洞察真正的技术边界究竟推进到了何处，社会便容易陷入一种基于弱化版本推断全局的“虚假安全感”。 更严峻的危机潜伏在宏观资本层面：AI 产生的商业投资回报率（ROI）与疯狂膨胀的基础设施资本支出（Capex）之间存在致命的时间错配。科技巨头正在以惊人的速度沿着资本结构向下汲取流动性： ```mermaid flowchart LR accTitle: AI 资本支出融资层级下移路径 accDescr: 展示科技巨头从使用自由现金流到发行债券、增发股票，乃至动用养老基金的资本耗尽过程 A["自由现金流 (FCF) 耗尽"] --> B["公司债市场快速饱和 (仅耗时约1年)"] B --> C["巨头重回股权融资通道 (如 Google)"] C --> D["设立数千亿美元级载体 (汲取养老金与保险浮存金)"] ``` - **自有资金见底**：巨头首先烧光了日常经营积累的充沛自由现金流。 - **债务市场承压**：科技巨头几乎在短短一年内就打满了债券市场的常规容量。 - **股权与另类资金入场**：不仅大型科技公司重新动用股权工具融资，NVIDIA 等关键产业链核心甚至开始筹划规模高达 5000 亿美元的融资结构，试图撬动主权基金、养老金及保险浮存金（Insurance Floats）等长期非流动性资本接盘。如果底层模型产生的商业化现金流无法迅速完成正向造血闭环，整个资本飞轮将面临巨大的失速风险。 ### `10:00` 1870 年代铁路泡沫与 AI 的资本期限错配 本·汤普森（Ben Thompson）指出，当下 AI 的巨额建设如果发生断层——自由现金流无法及时跟上资本开支——市场可能面临剧烈的崩塌。但即便泡沫破裂，AI 绝不会消失，技术进步也不会停滞，就像当年的互联网泡沫或 1870 年代的铁路建设浪潮一样，虽然短期极具破坏性，但从整个人类历史尺度来看，最终都深刻重塑了文明底座。 与过去大规模基建相比，当下的 AI 投资占 GDP 的比例正在迅速逼近甚至超越历史上最大规模的建设期。汤普森强调，人们现在常把目光投向 19 世纪的铁路热潮，其核心启示在于**期限错配（Duration Mismatch）**：修建铁路并实现盈利是一个长达数十年甚至半个世纪的漫长过程，但短期内却必须通过持续发行债务和股权来支付天量账单。1870 年代铁路危机爆发的根本原因，不是别的，正是“全世界把钱花光了”。在当下，人人都焦虑于算力够不够、电力够不够，但最近端最严苛的硬约束，其实是**资金够不够**。讽刺而美妙的是，当年那些铁路在破局之后不仅撑起了美国西部的繁荣，至今仍在向全美输送 GDP。 ```mermaid flowchart TD accTitle: 铁路与AI基建的资本期限错配模型 accDescr: 对比高额短期建设开支与漫长盈利回报期带来的资本枯竭压力 subgraph CapitalMismatch ["资本与时间错配 (Duration Mismatch)"] A["短期重资产投入<br/>(AI训练/电力/算力/铺轨)"] -->|"需海量流动性支撑"| B{"资金流动性缺口"} B -->|"现金耗尽风险"| C["流动性崩塌 / 资本枯竭"] B -->|"长线复利兑现"| D["数十年持续产出 / 支撑全社会GDP"] end ``` ### `11:34` 伯克希尔的铁路启示：从绝对利润看 Google 的增发逻辑 一个颇具象征意味的现实细节是：伯克希尔·哈撒韦（Berkshire Hathaway）来自铁路业务的真金白银，如今正通过持股流入 Google。汤普森谈到近期 Google 破天荒进行股权融资（Equity Issuance）的举动，直言当时感到非常震撼——作为坐拥印钞机般搜索业务、账面现金充裕的科技巨头，Google 竟然选择增发股权稀释老股东，这看似反直觉，但拉平到伯克希尔的商业逻辑来看却豁然开朗。 沃伦·巴菲特（Warren Buffett）曾收购喜诗糖果（See's Candies），那是典型的高利润率业务（Percentage Profit 极高），但由于无法进行大规模再投资，绝对利润（Absolute Profit）的上限被锁死；而巴菲特后来收购伯灵顿北方圣太菲铁路（BNSF），其利润率虽然比喜诗糖果差得多，但资金容量极其庞大。BNSF 单年创造的自由现金流，就轻松超过了喜诗糖果有史以来所有利润的总和。当资本体量达到某种临界点，企业思考的重心就会从“净利率百分比”转向“绝对美元规模”。 Google 传统的搜索业务正是当代最完美的“喜诗糖果”——零边际成本、无与伦比的聚合平台，极高利润率却难以吞吐当下激增的超量资本；而 AI 是一个能够吞噬数千亿美元、吞噬所有白领脑力劳动乃至未来机器人形态的浩瀚赛道。尽管 AI 的利润率注定远低于搜索，但其绝对利润空间却可能是后者的数倍甚至数十倍。Google 先是耗尽经营自由现金流，又发售数百亿美元债券，最终走向股权增发： - 增发固然让现有股东所占的股份百分比变小，但他们持有的是一块**绝对体量大得多（Astronomically Larger）的超级蛋糕**的份额。 - 伯克希尔对 Google 的注资不仅是一笔投资，更代表着一种经营哲学的映射：从追求纯粹轻资产高毛利，迈向拥抱大资本、超大绝对规模的工业级底座。 ### `14:23` 可验证领域 vs. 不可验证领域：汤普森的“不情愿加速主义” 在被问及自己在纯技术层面对 AI 的狂热程度（"AI pilled"）时，汤普森给出了一个兼具极度乐观与高度审慎的双重视角。他坦承，自己目前仍对 AI 的通用泛化能力（Generalizability）存疑： - **可验证领域（Verifiable Domains）**：在代码编写（Coding）和数学（Math）这类有确定标准答案、即时反馈测试的场景中，大语言模型展现出了令人震撼的天赋，"回想一年前人们还在纯手工写代码，简直不可思议"。 - **不可验证或长反馈周期领域（Unverifiable / Long-loop Domains）**：象棋和围棋之所以被 AI 攻克，是因为它们属于规则已知且有边界的“可知领域”（Knowable and Bounded）；然而在开放、无边界、难以快速验证真伪优劣的复杂真实世界中，尚未看到 AI 能自主开辟未知领域的扎实凭据。面对实验室技术人员所谓“以前人们也觉得下棋解不开”的说辞，汤普森直言这是概念偷换。 ```mermaid flowchart LR accTitle: AI应用场景与可验证性象限 accDescr: 展示可验证领域与不可验证领域的区别及经济价值差异 subgraph Verifiable ["可验证领域 (明确规则/短验证回路)"] V1["代码编写 (Coding)"] V2["数学与基础算法"] V3["规则确定性流程"] end subgraph Unverifiable ["不可验证 / 慢反馈领域"] U1["开放式战略构想"] U2["长周期医疗疗效"] U3["缺乏痕迹的情感与复杂意图"] end Verifiable -->|"模型现状已达实用级"| M["已具备庞大商业经济价值"] Unverifiable -->|"需捕捉更底层思维轨迹 (如脑机接口)"| N["通用泛化能力仍待验证"] ``` 互联网文本只是人类思想的最终输出形态（Distillation），屏幕上的 Reddit 发帖并未记录打字背后的具体思考轨迹和情感纠葛；如果像 Neuralink 这样的脑机接口真能捕捉到思考过程的痕迹数据，或许模型泛化能力才能跃迁。但汤普森话锋一转指出，即便大模型就此止步不再变聪明，其经济潜力也已经无比庞大，因为全社会相当大比例的工作岗位本身就属于“规则清晰、拿指令干活”的类确定性事务。他将自己定义为**“不情愿的加速主义者”（Reluctant Accelerationist）**——我们无法倒退，被卡在半途更会两头受困，唯有硬着头皮向前推进。 在谈到垂直领域落地时，汤普森重点提及医疗与法律： - **医疗领域**：这是最大的机遇也是最难的深水区，如果能合规解除数据壁垒，将机器学习应用于全量医疗病历，人类在极其迅速的时间内能发现的新疗法将不可估量。 - **人性与官僚阻力**：尽管人类创造全新需求和工作机会的能力是无限的，但人类制造“繁文缛节（Red Tape）和泥潭障碍”的能力同样没有上限。经济中有相当一部分就业岗位，本质上是人们互相制造冗余流程来让自己忙碌的产物。 ### `20:20` 聚合理论在 AI 时代是否依然适用？ 对话切入 Ben Thompson 在 2015 年提出的标志性商业分析框架——“聚合理论”（Aggregation Theory）。这一理论曾精准预言并解释了上一代移动互联网巨头（Google、Meta、Uber、Netflix 等）的崛起与统治地位。面对主持人提问这一理论在当前 AI 时代是否依然成立，Ben Thompson 坦言自己内心也反复权衡。 批评者认为，聚合理论的核心基石是**零边际成本（Zero Marginal Costs）**，而这在 AI 时代似乎破裂了。在 Web 2.0 时代，零边际成本体现在分发与交易两端： - **分发与发现**：很多创业者抱怨自己“缺乏分发能力，必须给 Google 买量”，Ben 指出这是混淆了概念——“你有自己的网站，你的问题不是缺乏分发渠道，而是缺乏需求”。在内容充斥的丰裕世界中，真正的瓶颈是“发现”（Discovery）。谁解决了特定领域的发现问题，谁就能聚合并掌控需求，形成飞轮效应。 - **交易成本为零**：Google 和 Meta 可以无缝扩展到全球，不仅因为用户端扩张无需边际成本，更因为变现端（广告主）同样实现了全自动化。绝大多数广告主从未与两家公司的销售团队说过一句话，所有竞价投放完全由计算机自动撮合，形成了“零边际服务成本”的完美商业机器。 在 AI 时代，推理成本（Inference Costs）是实打实的服务器与算力开销。但 Ben 强调，关于“AI 推理成本到底有多高”，答案取决于具体的用户场景，其成本分布展现出极大的两极分化： ```mermaid flowchart TD accTitle: AI推理边际成本的两极分化 accDescr: 展示了轻度日常用户与重度推理扩展用户在边际成本上的巨大鸿沟 subgraph LowEnd["低边际成本端（消费级替代）"] A["日常简单查询<br>（食谱、事实问答、Google 搜索替代）"] B["单次前向传递（Forward Pass）<br>成本极低，接近常规网页服务"] A --> B end subgraph HighEnd["高边际成本端（思考时间扩展）"] C["复杂推理与数学定理证明<br>（Test-Time Scaling 思考时间扩展）"] D["模型持续思考数分钟/数小时<br>边际成本随思考时间线性飙升"] C --> D end LowEnd -.->|"成本差距数个数量级"| HighEnd ``` - 对于绝大多数将 ChatGPT 当作 Google 替代品或生成家常菜谱的普通用户，其推理服务成本其实极其低廉，仅比提供一个传统网页稍微高出一点点。 - 另一极端则是依赖**思考时间扩展（Test-Time Scaling）**的用户。以往模型能力的提升仅靠预训练时堆叠参数和数据，而现在的模型在生成答案时可以通过思考链进行深度推理。“你让它思考几天、几周甚至几个月来解决数学定理，每一秒的思考都在直接消耗真金白银的边际成本”。这两种用户的服务成本完全不在同一个宇宙。 --- ### `22:06` 微软定价两难：人头费与用量计费的冲突 这种推理成本的剧烈分化正在重塑企业级软件（Enterprise SaaS）的商业模式。以微软最新推出的企业套件（如传闻中的 E7 套餐，定价每人每月 100 美元并叠加上限外用量计费）为例，Ben 认为微软正陷入一种极为微妙且危险的境地。 - **微软传统的“打包哲学”（Bundling）**：微软商业模式之所以极其稳固，在于它为企业提供一站式服务。单个组件（Word、Teams、SharePoint）或许不是市场上最好的，但它们打包在一起售价单一、彼此兼容。对于企业而言，这种费用直接绑定在“员工人头数”（Headcount）上——HR 每招募一个员工，就顺理成章地加上这笔 50 或 100 美元的软件预算。这种支出在心理认知上类似于一次性资本支出（CapEx），“企业做出一次决定后，就再也不用去多想”。 - **用量计费（Usage-based Pricing）打破认知闭环**：一旦引入用量计费，企业每个月收到账单时就必须审视“我们这个月到底用了多少 Token、花了多少钱”。这带来三重严重冲击： 1. **与企业预算机制脱节**：大多数企业每年只做一次固定预算，根本没有按月动态重新评估软件用量配额的管理流程。 2. **迫使客户重新评估每个组件的价值**：当企业每月都要对着账单细看各项花费时，就会下意识质问：“我花这么多钱到底买了什么？微软的这些独立 AI 功能真的好用吗？我们是否应该把这些工作负载分拆出去，转用其他更垂直便宜的工具？” 3. **微软的无奈之举**：重度使用 Token 的 AI 员工消耗的计算成本远超每人每月 100 美元的定额费用，微软若不按用量收费就会被极少数超级用户彻底吸干利润。微软既要让这部分重度员工的用量由企业埋单，又竭力希望普通员工维持固定包月，但这种“要求客户动脑子思考成本，却又不能让他们思考太多”的平衡，正在动摇微软最核心的商业惯性。 --- ### `25:40` 硅谷的周期性遗忘：为何消费级 AI 必须拥抱广告？ 针对轻度消费级 AI（如菜谱查询、日常问答）尚未跑通超级商业模式的现状，Ben 表达了强烈的观点：他对此“感到沮丧，但毫不意外”。消费级 AI 唯一的归宿必然是**广告变现**，但硅谷每隔十年就必须把关于消费者的基本常识重新学一遍： 1. **消费者极度抗拒为软件付费**； 2. **消费者根本不在意提高生产力**。 Ben 以当年 Dropbox 的发展历程作为教科书般的镜鉴： - 早期 Drew Houston 打造了体验惊艳、无缝同步的 Dropbox，一心想做一家纯粹的消费级公司，甚至在与 Steve Jobs 会面并拒绝苹果收购时被讽刺“你只是一项功能，不是一家公司”（Jobs 随后在苹果体系内打造了 iCloud Drive 实现该功能）。 - Dropbox 在消费端迎来爆发式增长后陷入长达两年的停滞。普通用户拿到免费额度后根本不愿付费升级。Dropbox 不得不停下脚步，自底向上彻底重构整套底层架构——加入权限管理、组织架构与合规管控，转型卖给企业客户。企业之所以愿意按人头掏钱，是因为企业支付员工薪水，提高员工的生产力能产生明确的投资回报率（ROI）；而普通消费者下班回家后只想瘫在沙发上看 Instagram Reels 视频，压根不想继续提高什么生产力。 在 AI 领域，OpenAI 正在以百倍于 Dropbox 的规模重蹈覆辙——初期试图单纯依靠向消费者收取 20 美元/月的订阅费（Subscription），虽然用户体量惊人，但订阅天花板显而易见。不仅如此，随着 Anthropic（Claude）在企业端编程等高价值生产力场景的强劲竞争，OpenAI 面临着两头受压的尴尬，不得不仓促补课企业级市场，同时迅速向广告功能靠拢（如引入 Conversions API、零售商集成追踪等基础设施）。 Ben 批评了硅谷长期以来对广告业务的自命清高与道德洁癖。在过去七八年中，硅谷工程师往往觉得“做广告很脏”，顶尖人才不愿从事广告系统建设；而 Ben 在 Stratechery 创立早期之所以能获得大量读者，恰恰因为他是极少数愿意客观分析广告本质的科技评论家。 Ben 总结了消费级产品结合广告变现的无敌商业飞轮： - **无价格弹性的无限变现**：如果是向消费者直接收取订阅费（如 Netflix），每次上调价格都面临消费者流失和抗议的“价格弹性”约束。 - **由广告主承担提价成本**：在广告驱动的模式下，消费者的黏性与使用深度越强，广告库存与转化价值越高，为涨价买单的是广告主而非用户，其变现空间在理论上是无限扩展的。如果 OpenAI 在 ChatGPT 刚刚爆火时就毫不犹豫地全面拥抱广告体系，如今 Google 和 Meta 恐怕早已陷入远比当下严峻得多的生死危机。 ### `30:00` 算力短缺错觉与大宗商品市场逻辑 两人探讨了消费级 AI 变现的现实困境与当前资本市场的激进狂欢。Ben Thompson 指出向用户收费极其困难，而提供免费服务却轻而易举，OpenAI 迟迟未全力拓展广告商业模式令人费解。与此同时，全球科技巨头正在掀起惊人的资本开支（CapEx）浪潮——今年预计支出达 8000 亿美元，明年更是预计飙升至 1.3 万亿美元，且未来还将继续攀升。市场上充斥着一种论调：所有新安装的算力几乎在上线瞬间就被消耗殆尽，Twitter 上的多头们整天高呼“我们算力严重不足”。 然而 Ben 直言这是一个严重的“时间错配”（Timing mismatch）。当前所谓的算力短缺，本质上是因为 2023 年和 2024 年产业链早期的投资不足，而不是因为需求真正无穷无尽。更为关键的是，台积电（TSMC）在 2023、2024 和 2025 年都降低了自身的资本开支增长率；而芯片晶圆厂（Fab）的建设周期比数据中心还要漫长得多，这意味着算力供给在未来几年可能面临更严峻的瓶颈。巨头们今天砸入的数万亿美元，其最终成果要到 2028 年甚至 2029 年才会完全转化成上线的算力。尽管 Andy Jassy（亚马逊 CEO）和 Satya Nadella（微软 CEO）在财报电话会议上安抚投资者称“我们只是在盖数据中心外壳（Shells），只有在看到确切需求时才会采购 GPU”，但 Ben 认为这纯粹是一套对外讲的说辞（"a lot of BS"）。因为一旦沉没成本砸下、机房外壳建好，没有任何公司会眼睁睁看着巨额资产闲置——他们必须被迫采购硬件让其运转。 ```mermaid flowchart TD accTitle: 科技行业差异化逻辑与大宗商品市场逻辑对比 accDescr: 对比苹果代表的高利润软件生态逻辑与海运航运代表的边际成本定价逻辑 subgraph Tech["科技行业传统逻辑 (苹果模式)"] T1["高额固定资产 / 研发投入"] --> T2["差异化软件与生态闭环"] T2 --> T3["强大定价权 (50% 硬件毛利)"] end subgraph Commodity["大宗商品市场逻辑 (航运/算力模型)"] C1["巨大固定资产投入 (买船 / 建数据中心)"] --> C2["极低边际运营成本 (燃油 / 运营)"] C2 --> C3["无论亏损必须满载运转"] C3 --> C4["价格被边际供给打压至边际成本"] end ``` Ben 尖锐地指出，科技行业长期享受差异化红利（以苹果为典型代表，依靠生态系统、专属软件与开发者网络形成 50% 的毛利率），导致整个行业普遍缺乏对**大宗商品市场（Commodity Markets）**底层规律的认知： - **边际供给决定价格**：在大宗商品市场中，产品同质化，服务成本和边际成本决定了一切。 - **航运业经典隐喻**：买一艘集装箱货轮的大头是折旧成本，而航行的边际成本极低（仅有燃油、船员薪资和港口费）。因此船东无论如何都必须让船满载出航，甚至不惜将集装箱运费打压到仅覆盖边际成本的水平。账面上虽因折旧计提产生巨额会计亏损，但只要现金流覆盖边际成本，船只就不会停运。 - **繁荣萧条与造船滞后**：疫情期间由于运力短缺，单个集装箱运价从日常的 3000–4000 美元暴涨至 1.7 万–1.8 万美元，航运公司暴赚；然而新船制造需要两年周期，当所有公司在暴利刺激下同时下注造船，新产能集中释放之时，运价便会瞬间雪崩。 - **算力估值的脆弱性**：目前的 AI 算力就像疫情期间的集装箱。多头们是基于“绝对稀缺时期”计算投资回报周期（Payback period）；一旦未来大量数据中心和算力集中涌入市场，从稀缺转向过剩，这一测算将彻底崩溃。 ### `35:00` 存储器周期的残酷历史与台积电的风险转嫁 针对主持人提问“芯片制造商（如台积电与存储芯片大厂）在扩产上的保守态度，是否能有效阻止行业陷入毁灭性的过度建设（Overbuild）”，Ben 从存储芯片（Memory）的惨烈历史展开了深度剖析。历史上存储器行业充斥着大量玩家，每当繁荣期到来、内存出现短缺，高昂利润就会诱使全球资本进场扩产；但因为建造晶圆厂成本动辄数十亿美元（先进制程如今更需耗费昂贵的 EUV 光刻机），产能无法即时上线。当过剩产能最终涌入时，价格出现暴跌，大量公司遭遇灭顶之灾。 韩国三星（Samsung）正是吃透了这段历史，凭借惊人的胆魄、纪律性与资金实力，采取“逆周期投资”（在行业萧条期疯狂逆势砸钱扩产，在下一轮繁荣期以压倒性产能清洗对手），一举彻底击垮了日本存储芯片产业。经过数轮血腥洗牌后，全球存储芯片市场最终整合收缩至仅剩三大巨头（三星、SK 海力士、美光）。这三家公司达成了某种不言而喻的默契——极力避免重蹈历史过度建设的覆辙，这也导致它们在此轮 AI 爆发初期对需求转向的反应相对迟缓。 ```mermaid flowchart LR accTitle: 存储行业供给制约演进与应对 accDescr: 展示存储巨头产能受限如何迫使下游客户研发替代方案并最终削弱巨头壁垒 M1["三大存储厂商扩产保守"] --> M2["短期供应紧缺与暴利"] M2 --> M3["下游客户受到制约"] M3 --> M4["算法优化削减内存需求"] M3 --> M5["苹果游说引入中国存储"] M4 & M5 --> M6["长期护城河被绕过 / 削弱"] ``` Ben 将当前的存储芯片巨头比作**霍尔木兹海峡（Strait of Hormuz）与伊朗**： - **威慑的失效陷阱**：封锁霍尔木兹海峡作为一种悬而未决的威胁手段最为有效；一旦真正实施，虽然短期内能造成剧痛并赚取暴利，但阿联酋、沙特等国必然会通过修建陆路管道、开拓新港口等方式彻底绕开该要道。到 2035 年，即便再次封锁也将失去任何战略威力。 - **存储厂商的自掘坟墓**：当前存储厂商的保守扩产正让自己成为行业集火的巨大靶子。下游科技巨头正在全力推进算法层面的革新，头号研发目标就是“如何大幅减少内存消耗”；苹果等公司甚至在积极游说引入中国存储芯片。通过制造人为稀缺，存储厂商正在倒逼整个生态建立规避它们的替代路径。 相比存储行业，台积电（TSMC）的垄断格局更为极端——在先进制程领域几乎是独一无二的独苗（尽管英特尔与三星仍在苦苦追赶）。Ben 揭示了一个科技巨头尚未完全看清的底层现实：**台积电巧妙地将产能过剩的致命风险完全转嫁给了科技巨头**。 - **三十年寿命的重资产包袱**：台积电最恐惧的是产能过剩。建一座晶圆厂需要预想其运行 30 年，一旦过度建设，系统将在数十年内背负高昂的折旧与闲置成本，因此其管理层在扩产决策上天然极度保守。 - **张忠谋（Morris Chang）的传奇逆周期决策**：在文化基因上，张忠谋曾展现过与三星如出一辙的逆周期魄力。2000 年代后期张忠谋退休后，新管理层遭遇 2008 年全球大衰退，慌忙削减资本支出。张忠谋随后强势重返台积电、撤换管理团队，并指出“iPhone 刚刚发布，这是我们见过的最大历史机遇，我们必须进攻而不是撤退”。台积电顶着大衰退逆势大举投资，最终奠定了其在先进制程领域独霸全球的王座。张忠谋也是半导体历史上绝无仅有的“总统山（Mount Rushmore）”级传奇人物。 - **资本与时间的鸿沟（Air Gap）**：Ben 总结道，尽管他对 AI 的经济潜力和长远社会影响力深信不疑，但他最大的担忧在于行业正在面临一场“断层（Air gap）”危机——巨额资本正在以不可持续的速度疯狂涌入，但真正的应用回报未能及时规模化变现；在商业闭环跑通、产生足够现金流以支撑下一阶段发展之前，市场很可能会先一步耗尽可用资本。 ### `40:00` 晶圆制造的保守主义与风险转移：台积电与“史上最长牛鞭效应” 台积电（TSMC）开创的无晶圆厂（Fabless）代工模式是整个现代科技产业的基石，而其创始人张忠谋（Morris Chang）毫无疑问是科技史上最伟大、最具影响力的商业领袖之一。然而，受制于文化与半导体制造的重资产属性，台积电骨子里一直保持着极度谨慎与保守的投资策略。当台积电对外声称“我们不想承担过度建设产能的风险”时，风险并未凭空消失，而是被推转移给了下游。 这表现为全行业面临的算力短缺：如今几乎每一家大型科技巨头都意识到，如果手头能有更多算力，就能赚取巨额利润；巨头们当前错失的天量营收与利润，正是台积电将风险转移给下游的直接体现。台积电在 2020 至 2022 年间为了 5G 建设进行了一轮大规模晶圆厂扩产，随后便放缓了脚步；尽管 ChatGPT 在 2022 年底问世并在 2023 年引发 AI 投资狂潮，台积电的增长率在 2024 年和 2025 年却依然出现放缓，直到 2026 年才重新提速。 半导体制造业一直极其警惕“牛鞭效应”（Bullwhip Effect）——担忧终端需求被层层放大，当位于供应链末端的代工厂砸巨资扩产时，需求泡沫却已破裂。然而，AI 带来的算力短缺构成了“史上最漫长的牛鞭效应”，需要被构建的算力规模极其庞大。过去亚洲的芯片制造商耗费了很长时间才真正确认这一轮需求的持久性，台积电董事长兼首席执行官魏哲家（C.C. Wei）直到最近的一两次财报电话会中，才开始前所未有地大篇幅畅谈 AI 的应用场景。从工厂决策到实际晶圆产能落地，中间存在数年的漫长交付时滞。 --- ### `42:08` 算力饥渴拯救英特尔：地缘政治“保险”的经济学逆转 台积电的保守与全行业的算力饥渴，直接改写了英特尔（Intel）和三星（Samsung）逻辑代工业务的命运。早在 2013 年，市场上就有人呼吁英特尔必须转型建立芯片代工业务，放弃单纯的芯片设计思维，因为先进制程制造本身是一个规模极其惊人的巨大市场。然而整个 2010 年代，英特尔沉浸在云计算爆发带来的服务器芯片红利中，股价节节攀升，直到 2020 年前后才猛然意识到自己不仅制程落后，而且完全缺乏代工服务所需的客户服务导向、企业文化以及成熟的基础 IP 模块生态。 在地缘政治上，西方科技巨头完全依赖距离中国大陆仅 60 多海里的台积电，存在巨大的供应链安全隐患。但过去要求巨头向英特尔购买这种“地缘政治保险”并不符合商业理性： - 科技巨头如果选择让英特尔代工，客户自身需要承担巨大的磨合痛苦去教导英特尔； - 最大的受益者反而是英特尔，因为英特尔借此学会了如何与外部伙伴合作； - 与此相对，台积电服务卓越、交付稳定，因此在常态市场中，巨头没有任何经济动力去寻找英特尔。 ```mermaid flowchart TD accTitle: Compute Scarcity and Foundry Diversification accDescr: Diagram illustrating how AI compute shortage forces Big Tech to overcome friction and subsidize Intel to build a TSMC alternative. subgraph Bottleneck["台积电保守投资与产能瓶颈"] TSMC["台积电控制资本开支<br/>规避扩产风险"] --> Shortage["尖端算力严重匮乏<br/>大厂错失天量利润"] end subgraph EconomicShift["商业逻辑逆转"] Shortage --> Motivation["痛感转移：错失利润的代价<br/>远超培育二供的磨合成本"] Motivation --> Partner["大厂被迫承受痛苦：<br/>扶持英特尔/三星代工"] end subgraph Outcome["地缘保险兑现"] Partner --> IntelResurrect["英特尔获得顶级客户<br/>补齐代工IP与组织能力"] Partner --> Hedge["免费获取去台积电化的<br/>地缘安全对冲"] end ``` 治愈高价的良药正是高价本身，台积电的产能制约逼出了自己的竞争对手。当算力不足导致的潜在利润损失大到无法承受时，科技巨头们终于愿意咽下磨合阵痛，真金白银地协助英特尔与三星打通尖端工艺。解决对台积电地缘依赖的唯一路径，并不是依靠补贴说服大厂购买预期收益为负的“政治保险”，而是创造一个规模大到无可估量的算力需求，迫使巨头出于自身商业利益主动扶持替代者，从而让整个西方世界“免费”获得了供应链安全对冲。 --- ### `44:51` 大厂中最具优势的底牌：亚马逊与“内部头号客户”飞轮 在全美前十至十五大科技公司中，亚马逊（Amazon）在 AI 与技术周期中的布局架构尤为稳健。亚马逊的核心竞争力在于：**他们永远是自己新业务的第一位、也是最优秀的客户（First Best Customer）**。他们能够凭借自身海量的内部业务提供初始规模，将一项技术彻底打磨成熟后，再将其打包输出给全市场。 这种飞轮效应在亚马逊的历史上屡见不鲜： - **云计算（AWS）：** 与流行流言不同，AWS 绝非出售“闲置的服务器算力”；事实上，亚马逊零售业务（Amazon.com）耗费了很长时间才完成向 AWS 的迁移。AWS 诞生的初衷是为了砍掉无休止的跨部门沟通协调会，将底层计算抽象为纯粹的 API 接口。亚马逊意识到，如果零售业务需要这种即插即用的计算服务，那么外部所有开发者同样需要。 - **物流基础设施：** 早期亚马逊完全依赖 UPS、FedEx 和美国邮政（USPS），随后亚马逊在内部重金自建物流网络，如今该网络不仅支持自身电商，还作为第三方配送服务面向外部商家全面开放。 - **自研芯片（Graviton 与 Trainium）：** 自研 CPU（Graviton）和 AI 专用芯片（Trainium）早期版本性能十分糟糕，换作其他芯片厂商早已死在研发周期中。但亚马逊拥有诸如 Redshift 数据库等海量托管云服务（Managed Services），外部客户在购买数据库算力时根本感知不到底层使用的是哪款芯片。亚马逊利用内部托管服务的海量需求吸纳了这些初期表现粗糙的芯片，为自研芯片提供了极高的出货量与迭代试错空间。随着硬件快速演进，Graviton 性能登顶，Trainium 如今不仅能驱动 Anthropic 的前沿 AI 研发，还能直接对外商用。 ```mermaid flowchart LR accTitle: Amazon's First Best Customer Flywheel accDescr: The iterative loop showing how Amazon incubates internal infrastructure into dominant external cloud and hardware products. subgraph Incubation["内部需求与冷启动"] A["自建内部基础设施<br/>(AWS/物流/自研芯片)"] --> B["充当首位最佳客户<br/>提供自给自足的规模"] end subgraph Masking["隐藏粗糙度与加速迭代"] B --> C["以托管服务包裹底层<br/>(如Redshift跑自研芯片)"] C --> D["在真实生产负载中<br/>低阻力快速迭代技术"] end subgraph Monetization["外部商业化输出"] D --> E["产品达到领先水平<br/>(如Trainium支撑Anthropic)"] E --> F["全面向外部第三方开放<br/>开辟全新独立商业航道"] end ``` 这种模式延伸到了 AI 应用层。亚马逊正利用 AI 重构其庞大的客服呼叫中心网络与购物体验，生成的对话系统能够切实解决退换货与物流诉求。亚马逊在实体世界（仓储、物流、履约）筑起的资本重墙，几乎完全免疫于纯模型层 AI 的颠覆；模型技术只会赋能其实体业务，而其自生自发孵化新业务线条的架构能力，让其核心护城河比任何同行都更为深邃。 --- ### `48:15` 苹果的聚合者护城河与潜在的“微软式陷阱” 相比积极军备竞赛的巨头，苹果（Apple）在生成式 AI 热潮中似乎始终作壁上观，但其目前的处境可以说是“与其聪明，不如走运”（Better be lucky than good）。作为经典的聚合者（Aggregator），苹果牢牢掌控着数十亿终端用户的访问入口与注意力，这赋予了它决定性的谈判议价权——在商业生态中，拥有最终客户的人决定规则，AI 供应商必须求着接入苹果生态，而非反向制约苹果。 苹果在 AI 时代拥有得天独厚的防御与变现优势： - **推理成本的转移：** 大多数普通消费者并不追求极端的生产力突破，他们只需要一个真正好用、能听懂人话的 Siri。未来这些功能将高度落地在设备端（On-device AI），这意味着苹果甚至不需要承担高昂的云端推理成本与数据中心电费，因为这些成本直接由用户自身的手机电池和家庭用电所分摊。尽管现阶段为了驱动高阶模型，苹果仍需借助谷歌云（Google Cloud）和英伟达（NVIDIA）芯片，但端侧化的大方向不可逆转。 - **物理硬件的护城河：** 制造精密的实体手机、建立遍布全球的零售店（Apple Store）和分销物流极度困难。智能手机堪称人类消费电子史上最完美的形态：小到能放进口袋，大到能看视频玩游戏，接管了人类全部的数字化生活与娱乐，连电视机如今都降级成了手机的附属配件，短期内没有任何硬件形态能取代手机。 然而，苹果的隐忧在于其是否会落入当年“微软对待移动互联网”式的思维陷阱： - 微软当年并未错过移动化，他们早在 Windows Mobile 时代就极早布局了智能手机；但微软犯下的致命错误是**将手机视为“一台小型 PC”**，执念于 PC 必须是数字宇宙的中心，手机只是 PC 生态的延伸附属物。 - 苹果当年能颠覆行业，正是因为史蒂夫·乔布斯（Steve Jobs）打破了 Mac 的中心论，意识到手机必须是独立演进的核心计算平台。 - 面对如今 OpenAI 等机构描绘的环境智能（Ambient AI——在家庭或环境中用户只需自然发问即可获取服务）愿景，苹果最具落地优势；但若苹果因为 iPhone 的巨大成功而产生路径依赖，坚信手机必须永远是数字生活的绝对中心，且自身缺乏顶尖的前沿基础大模型（Frontier Models），就极有可能重蹈当年微软的覆辙。 ### `50:00` 确定性之王与概率性浪潮：苹果的克制与前沿五强的“宗教狂热” 面对环境计算、云端以及全知全能的生成式 AI 是否正在动摇苹果生态护城河的质疑，Ben Thompson 认为苹果选择加倍押注自身的核心强项是完全合理的。本质上，现代 AI 是一种概率性的事业（probabilistic endeavor），而苹果则是“确定性之王”（king of deterministic ideas）。作为一家顶尖的物理硬件公司，iPhone 一旦出货就必须做到无可挑剔，一旦翻车将面临数十亿乃至上百亿美元的惨重损失。苹果历史上从未发生过任何一次 iPhone 召回事件，这种对供应链的严酷把控、极致的审慎决策和对物理工业精度的执念，与训练大语言模型所需要的探索逻辑与容错文化完全相悖。企业应当专注于自身擅长的事，因此对于苹果不去亲自深耕底层前沿模型、继续打磨终极硬件终端的做法，反而是最理性的选择。 在审视当前角逐前沿模型的五大阵营——OpenAI、Anthropic、Google（Gemini）、xAI（Grok）与 Meta 时，各家的竞争态势与组织心力截然不同。OpenAI 与 Anthropic 承载了最高的风险，却也具备最大的上行空间。绝对不能低估“信念的力量”，历史表明最石破天惊的变革往往由宗教般的热情所驱动。硅谷当下的两座“宗教殿堂”中，OpenAI 如同主流教会，每周雷打不动做礼拜；而 Anthropic 则更像全情投入的福音派信徒——“他们发自内心地坚信自己正在创造上帝”，再加上缺乏成熟现金牛业务庇护，为了生存必须拼死跑通商业化的饥饿感，构成了极其强大的战斗力。 相比之下，成熟巨头的心态迥异。Google 只要搜索引擎业务不迅速消亡就算守住阵地，凭借雄厚的云基础设施与自研 TPU，继续跟进是顺理成章的防守反击；而 Meta 之所以能杀入前沿梯队，纯粹是马克·扎克伯格（Mark Zuckerberg）个人意志的极致体现。换作任何职业经理人掌舵，手握极其暴利的数字广告印钞机，都会选择安于现状，但扎克伯格却以纯粹的创始人力量（founder energy）推倒重来、重新组建顶尖团队，这种近乎疯狂的执行力令人赞叹。至于马斯克的 xAI 与 SpaceX 提出的“轨道太空数据中心”构想，概念虽具差异化，但其商业逻辑存在软肋：即便未来地球上的电力与土地资源耗尽，太空算力也完全可以运行第三方模型（正如其向 Anthropic 开放算力一样），是否有必要耗资数十亿美元自研模型仍存疑问。 - **确定性 vs. 概率性**：物理硬件的容错率为零，AI 的本质则是概率分布；要求苹果转型为顶尖 AI 研发者无异于强行扭转其底层基因。 - **硅谷的宗教隐喻**：OpenAI 对应主流教会（Mainline Protestant），Anthropic 对应狂热的福音派（Evangelicals），两者皆怀揣创造“神”的终极信念。 - **创始人意志的溢价**：扎克伯格主导的 Meta 是大厂中唯一跳脱出“守成心态”、从零搭建全新前沿模型团队的异类。 --- ### `54:27` 微软重演 90 年代 IBM 剧本：中间件防线与底层颠覆危机 与狂热押注前沿模型的巨头相反，微软选择了一条极为理性的守成之路。微软不争夺最尖端的训练前沿，上一季度依然豪掷 100 亿美元分红、创造充沛的自由现金流。他们的策略是充当模型中介，打造让企业客户构建应用的中间件（Middleware）与平台底座，将各大底层模型彻底“管道化”（disintermediate）。这套战术与 20 世纪 90 年代郭士纳（Lou Gerstner）拯救 IBM 时的剧本如出一辙。垄断的代价往往是组织能力的退化，70 年代如日中天的 IBM 到了 90 年代因缺乏竞争压力而臃肿平庸；郭士纳看透了这一点，意识到把 IBM 拆分成各个平庸的细分组件毫无意义，其最大价值正是“庞大本身”。面对奔腾而来的互联网浪潮，全美传统企业渴望上网却无所适从，IBM 顺势组建起庞大的咨询顾问大军，开发中间件将客户旧式的传统大型机（Mainframes）与新兴的网络服务无缝连接，从而为 IBM 赢得了长达三十年的生命周期。 ```mermaid flowchart TD accTitle: 90年代 IBM 与当代微软的企业中间件策略对比 accDescr: 展示两家巨头如何通过中间件抹平新技术锯齿，绑定传统企业并实现商业闭环 subgraph IBM_Playbook["90年代 IBM 破局之道"] A1["传统遗留系统<br/>（企业大型机 Mainframes）"] A2["IBM 咨询顾问团队 + 中间件层<br/>（抹平锯齿 / 降低门槛）"] A3["新兴互联网浪潮<br/>（Web 网站与电子商务）"] A1 --> A2 --> A3 end subgraph MSFT_Playbook["当代微软 AI 战术体系"] B1["企业旧有生态<br/>（Windows / Office 历史兼容性）"] B2["Copilot / Azure 平台中介<br/>（统一治理 / 模型解耦）"] B3["前沿 AI 模型阵列<br/>（OpenAI / 开源开源模型等）"] B1 --> B2 --> B3 end ``` 微软正在将这套战术复刻到 AI 时代：它扮演企业最值得信赖的老朋友，承诺用户既不用交出核心商业机密，又能平滑拥抱 AI，将日新月异、变动不居的底层模型封装在稳定的企业级接口之后。中间件固然会“磨平锋利的边缘”，提供的是最大公约数式的中庸体验，但它契合了企业软件市场最古老的销售真理——通过制造对其他供应商“锁定”的恐惧来实现自我锁定（正如 80 年代甲骨文以逃离 IBM 锁定为口号推销关系型数据库，最终自身却成为锁死客户的垄断者）。因此，微软的大规模资本开支主要聚焦于满足真实客户推理需求的确定性算力，而非高风险、缺乏确定性的前沿预训练集群。 然而，这一精妙战术的背后潜藏着极其严重的生存威胁。用户至今使用微软产品的根本原因往往只是因为“以往就在用”，但如果 AI 能够直接接管所有非结构化信息、文档与工作流，人类为何还需要复杂的 Word 文档、Excel 表格乃至电子邮箱收件箱？软件行业过去将“记录系统”（Systems of Record）视为不可撼动的护城河，理由是迁移数据过于繁琐枯燥，但处理繁琐任务恰恰是 AI 的看家本领。更致命的是，微软真正的命脉在于“人机交互界面”（UI）。当 Claude Coworker、Codex 这类原生具备任务执行能力的智能体横空出世时，它们就像一支支利箭，精准刺向微软赖以生存的交互界面中枢。微软的战略固然理性，但在这种近乎绝望的系统级颠覆面前，其防线依然风雨飘摇。 - **中间件哲学**：不追求最前沿的模型智能，而是将前沿的不确定性封装为企业级稳定服务，赚取确定性的工程溢价。 - **锁定悖论**：企业软件的销售话术总是“帮客户避免被第三方绑定”，其最终目的却是建立更隐蔽的自我绑定。 - **UI 护城河的瓦解**：一旦 AI 智能体直接穿透至任务执行层，微软积累数十年的操作系统与办公软件界面层将面临被架空的风险。 --- ### `01:00:00` 数字化企业的隐忧与内容经济学：从 YouTube、Netflix 到 Meta 的零成本奇迹 从宏观视角来看，生成式 AI 的崛起正在对所有纯数字化软件公司构成普遍威胁，Meta 亦无法独善其身。Meta 的核心商业资产是“用户注意力时间”，只要 AI 应用抢占了更多原本属于社交刷屏的时间，其底盘就会受到侵蚀。虽然类似 OpenAI Sora 这样的视频生成产品尚未真正引发社交网络格局的剧变——搭建一个黏性极高的社交网络门槛极高，且模型生成视频的算力成本高昂——但在内容供给的底层成本结构上，各大互联网巨头之间存在着本质差异。 媒体与平台的内容经济学可以清晰地划分为三个层级：Netflix 处于资本风险最高的一端，必须在制作周期提前预付数十亿美元的巨额内容制作成本，寄希望于后续订阅收入实现回本；YouTube 则开创了更为健康的模式，从创立之初就坚持与创作者共享广告收入，虽然无需预付成本，但根据播放量按比例分成的机制构成了其商业模式中不可忽视的持续边际成本；而 Meta（涵盖 Facebook 与 Instagram）则创造了不可思议的商业奇迹——它坐拥海量的用户注意力并榨取极其惊人的广告收益，却在平台的内容生产成本上支出为“零”。如果未来 AI 生成内容走向常态化，依靠极度廉价的推理成本批量制造内容，这种变革对于本就需要承担创作者分成成本的 YouTube 而言可能演变为优化财务结构的契机，但对于原本享受着海量免费内容供给的 Meta 来说，其背后的注意力生态与变现平衡却面临着全新的未知变量。 | 平台 / 公司 | 内容获取成本模式 | 资本风险与边际成本特征 | |:--- |:--- |:--- | | **Netflix** | **前置预付（Upfront）** | 承担高额先期制作资本风险，边际用户播放成本极低，依赖订阅规模摊薄固定资产。 | | **YouTube** | **后置分成（Revenue Share）** | 无需先期预付，但内容分成构成显著的持续边际成本，随广告总盘子同步膨胀。 | | **Meta (Instagram / FB)** | **零成本 UGC（Zero Cost）** | 内容生产边际成本完全为零，享受全球最高毛利的注意力变现红利。 | ### `01:00:00` 社交网络与娱乐产品的鸿沟：Meta、TikTok 与真实人际连接 在探讨 Meta 的内容成本时，Ben Thompson 指出，如果由 Meta 亲自用 AI 生成内容，其利润率结构实际上比现状更糟糕——因为平台目前获得的用户生成内容完全是“免费”的。不过，在更宏观的层面上，Meta 正处于一个极佳的战略位置。当未来世界充斥着无所不在的 AI 交互时，人类对于“真实人际连接”的渴望反而会变得更加强烈，这恰好让 Meta 有机会回归其作为社交网络的立足之本。 回顾 Meta 过去的产品发展史，他们犯下的最大错误源于对产品形态本质的认知偏差。Meta 本质上是一家社交网络公司。当初 Snapchat 崛起时，Meta 能通过 Instagram Stories 等功能彻底遏制 Snapchat 的增长，是因为 Meta 意识到 Snapchat 是个极其出色的产品形态，而他们直接将自己的社交关系网络覆盖上去，用庞大的网络效应压制了对手。然而面对 TikTok 时，Meta 却出现了盲区： - 业界普遍将 TikTok 归类为“社交网络”，但 TikTok 根本不是社交网络，而是一款纯粹的“娱乐产品”。 - 在 TikTok 上，关注谁无关紧要；用户看到的内容取决于算法对观看行为的即时反馈。 - TikTok 的底层洞见在于：将优质内容的供给局限在用户的个人社交圈子内，本身就是一种人为的约束。TikTok 致力于从全网搜集最顶尖的内容呈现给用户。即便绝大多数 UGC 内容质量极差（可能优质率微乎其微），但在庞大的基数效应下，全网顶级内容的绝对数量依然庞大。 Meta 此前一直坚持根据用户的社交好友关系流分发内容，因而被 TikTok 侵蚀了巨大的市场份额。后来 Instagram 和 Reels 经历的转型，实际上正是 Meta 承认了自身必须打造一款面向全网抓取内容的“娱乐产品”，而传统的社交网络功能则逐渐退守至群聊等私密场景中。 ```mermaid flowchart TD accTitle: TikTok与传统社交网络的分发模式对比 accDescr: 对比传统社交网络基于人际关系的分发约束与TikTok基于全局优质内容算法池的绝对数量优势 subgraph SocialNet["传统社交网络 (Meta 原生模式)"] A["个人好友关系链"] --> B["好友发布内容"] B --> C["受限于人际圈的分发"] end subgraph Entertainment["全网娱乐产品 (TikTok / Reels)"] D["全球用户生成海量内容"] --> E["算法全局筛选最高绝对值内容"] E --> F["即时兴趣匹配与个性化娱乐"] end ``` --- ### `01:02:15` 广告即验证机器：AI 模型在精准变现上的飞轮效应 谈到 AI 模型最直接的变现渠道，Ben Thompson 认为当下产生最大增量商业回报的既不是 OpenAI 也不是 Anthropic，而是 Google 和 Meta 的广告引擎。尽管目前多数底层基础设施仍基于早期的非 LLM 架构，但大语言模型正在加速渗透。AI 领域追求的是“可验证的领域”（Verifiable Domains），而在广告生成中，验证标准极其直接冷酷——这幅图或这段文案到底能不能促成转化与购买。 Meta 拥有任何其他科技公司都无法企及的优势：其广告竞价市场本身就是一个由全球数十亿真实用户驱动的“验证机器”。 - 广告平台每天在无数广告素材上执行海量 A/B 测试。虽然绝大多数投放都是毫无价值的无效展示，但这构成了全球最大规模的连续验证闭环，反向驱动其内容生成与图文模型的迭代优化。 - 传统的广告匹配逻辑相对原始：提取用户特征向量与广告特征向量，通过嵌入（Embeddings）计算相似度来匹配。 - 引入 LLM 后的核心质变在于“预测力”。LLM 能精准预测用户接下来可能最渴望看到什么，并主动检索或生成相应内容。由于 Meta 的广告体量极为庞大，匹配效率只需提升几个百分点，就能直接沉淀出数十亿美元的净利润。单凭这一项投资回报，就足以支撑 Meta 在最前沿的基础模型训练上不惜代价地投入。 --- ### `01:03:45` 广告的社会价值与扎克伯格的沟通盲点 然而，Meta 面临的一大硬伤是他们始终不擅长讲述广告的价值故事。Ben Thompson 直言，Mark Zuckerberg 与 Sam Altman 存在同样的偏见——内心深处并不喜欢广告。但客观而言，Meta 坐拥全球最卓越的广告业务，其商业模式具备深厚的社会正外部性： - 互联网的真正魅力在于“长尾利基市场”（Niche Market），全球一定有人需要某种极度细分的产品，核心痛点是“供需如何相遇”。 - Meta 的广告机制就像精准雷达，帮助创业者的产品找到那些原本不知道自身需求、但买到后极其满意的受众。 - 在这一闭环中，中小创业者拓展了新业务，消费者获得了满意的商品和海量免费的娱乐内容，股东与平台也获得了充沛回报。 然而，在过去二十年里，Zuckerberg 很少正面阐述广告对实体经济和创业生态的促进作用，而是长期将这块业务交给 Sheryl Sandberg 打理，自己退居幕后不愿沾染其中的“脏活累活”。Sandberg 在任期间，每季财报电话会都会例行讲述小企业主借助广告成长、改变命运的具体案例；但在她离职后，Meta 的叙事出现了巨大断层。整家公司给人的感觉仿佛对靠广告赚钱感到羞愧，急于证明自己是在做智能眼镜或顶尖 AI，却忽视了广告本身的伟大力量。 这种公关沟通与叙事的缺失，让 Meta 在面对监管和外界批评时付出了沉重代价： - 当苹果单方面推行 ATT（应用追踪透明度）政策时，Meta 几乎毫无招架之力。Ben Thompson 评价 ATT 是科技史上最恶劣的反竞争行径之一——苹果打着隐私旗号摧毁了整个移动互联网原有的商业闭环，同时却自建广告网络并进行自身的追踪，还推出极具误导性的公交车偷听广告来妖魔化行业生态。 - 听证会上 Tim Cook 指责其他公司“贩卖数据”，而事实上 Meta 从不卖数据，数据是其安身立命的核心资产。 - 由于 Meta 没能在公众心智中树立起广告的正向形象，加之此前在 Oculus 与元宇宙业务上累计烧掉了上千亿美元（且饱受投资人质疑），导致当管理层如今宣布需要为 AI 基础设施投入巨额资本开支时，极难说服华尔街给予信任。 --- ### `01:06:40` 大宗商品改变世界：智能与算力究竟意味着什么 话题转向英伟达与 Jensen Huang（黄仁勋），以及硅谷对大宗商品（Commodities）认知的匮乏。针对“算力”乃至“智能”未来是否会彻底沦为大宗商品的讨论，Ben Thompson 提出了深刻的经济学反思： 在科技语境中，人们往往推崇高溢价、高差异化的专有产品，看轻无差异的大宗商品。但互联网历史上最震撼的奇迹恰恰是“免费分发”——带宽本身就是一种高度标准化的大宗商品。正是因为在边际成本上带宽等同于免费商品，世界上任何人掏出手机都能连接全人类的信息库。大宗商品往往比差异化产品具有更大的全行业影响力，因为差异化产品必然伴随价格弹性与支付意愿门槛，其总体潜在市场（TAM）受到天然限制（例如苹果绝不可能靠卖硬件设备覆盖全球每一个人，但 Google 的免费大宗化服务可以）。如果未来“机器智能”最终也被压制成随处可得的大宗商品，其所释放的社会变革能量将远超任何高利润率的封闭产品。 --- ### `01:08:24` 英伟达的“非自然”利润率与隐秘的价格战 聚焦到英伟达当前的统治力，Ben Thompson 明确表示，英伟达维持的高利润率处于一种“非自然”状态。虽然表面上看，各方竞争对手都在涌入芯片赛道，而英伟达依然能够为单颗芯片标出天价并维持超高毛利率，但其背后正充斥着类似电信泡沫时期朗讯（Lucent）风格的“循环融资”（Circular Financing）： - 英伟达积极向各类新兴云厂商（Neoclouds）注入股权资本，并签署兜底协议，承诺包销直至 2030 年的算力。 - 新兴云厂商凭借英伟达的兜底背书，大幅拉低了自身的资本成本，进而得以向公开市场融资、采购更多的英伟达芯片。 - 资本成本降低的核心逻辑在于**英伟达主动承担了底层违约风险**。正如金融学铁律所揭示的：风险从未消失，只是转移了阵地。承担风险必须付出对价。 如果 AI 需求永远指数级扩张，英伟达便能通吃全部上行收益；但一旦市场算力供给出现结构性过剩，各大超大规模云服务商（Hyperscalers）自给自足不再外购算力，缺乏客户承接的新兴云厂商将把滞销算力扔还给英伟达。英伟达被迫为无人问津的算力买单，将引发严重的亏损。 从整体现金流折现（DCF）和风险期望值的视角审视，这种兜底注资实质上构成了“变相降价”。虽然账面标价和财报毛利率依然光鲜，但降价已经通过极其离奇的资产负债表外安排隐匿发生了。英伟达通过眼花缭乱的金融运作维持着高价表象，然而在长远周期里，真正的终极威胁依然来自掌握庞大内部需求与自主研发能力的超大规模云服务商，特别是 Google 和 Amazon。 ### `01:10:00` 超级云服务商的自研芯片与英伟达的真正威胁 超大规模云服务商（Hyperscalers）正在积极推动自研芯片的对外输出。Google 已经达成协议，将大约 20% 的 TPU 算力出售给 Anthropic；在最近的财报电话会上，亚马逊 CEO 安迪·贾西（Andy Jassy）也基本确认，AWS 最终将向外部客户销售 Trainium 3 甚至 Trainium 4。研发定制芯片需要极其庞大的 R&D 投入，将其作为大宗商品（commodity）对外售卖，不仅能为这些公司提供长期绑定客户的杠杆，更能摊薄巨额资本开支。客户并不是因为 Trainium 有不可替代的独特性才去选 AWS，因此这种外销完全不会削弱其云平台的吸引力。 这正是英伟达（NVIDIA）面临的重大挑战。超大规模云服务商拥有核心竞争力——庞大的规模与远低于新兴云服务商（Neoclouds）的资金成本（Cost of Capital）。新兴云公司可能会毫无顾忌地大批量采购英伟达芯片；埃隆·马斯克旗下的公司（如 xAI、SpaceX）对外宣称只买英伟达是因为它最顶尖，但更底层的逻辑是英伟达芯片的通用流动性（fungibility）最高。虽然随着基础模型不在意底层硬件，CUDA 的护城河被显著削弱，但如果你打算采取“自建大量算力并对外出租、必要时收回自用”的商业策略，英伟达仍然是最容易出租变现的资产。 ```mermaid flowchart TD accTitle: 芯片市场的两极分化竞争 accDescr: 展示超大规模云厂商与英伟达在不同客户群体及护城河上的竞争策略 subgraph Hyperscalers["超级云厂商 (AWS / Google)"] Cost["极低的资金成本与规模效应"] ASIC["自研定制芯片 (TPU / Trainium)"] Commodity["作为大宗商品外售，摊薄 R&D 成本"] Cost --> ASIC --> Commodity end subgraph NVIDIA_Strategy["英伟达 (NVIDIA)"] Fungible["最高通用流动性 (Fungibility)"] TargetMarkets["下沉锁定市场: 主权云与政企客户"] Fungible --> TargetMarkets end Commodity -.->|"侵蚀通用需求"| Fungible ``` - **类比 Intel 的历史教训**：AMD 之所以在超级云厂商中击垮 Intel，是因为这类大客户拥有极强的工程实力与极大规模，只要有更划算的替代方案，他们愿意投入工程师资源抹平架构微差；但 Intel 稳固的基本盘是政府与传统企业，因为这些机构既缺乏超级云公司的研发资源，也不具备大规模议价能力，只会照常采购惯用产品。 - **英伟达的渠道转移策略**：早在 2023 至 2024 年间，英伟达就开始大力推销“主权云”（Sovereign Cloud）和各类企业级套件。这些看似免费的预装模块只能在英伟达硬件上运行，其核心目的就是构筑软件锁定，避开与超级云厂商的技术死磕，扎根到对硬件芯片缺乏权衡能力的企业与政府市场。 --- ### `01:12:45` 能源瓶颈、泡沫遗产与电力充裕的未来 从能源供需角度来看，彻底的电力受限反而有利于英伟达。若电力成为绝对短缺的刚性瓶颈，所有算力买家就必须极度追求每一度电产出的 Token 效率，而英伟达在能效表现上依然处于领先地位。然而，过去几年英伟达遭遇的最大变数，恰恰是美国能源供应上线速度远快于预期。无论马斯克利用表后天然气发电机快速并网的模式被各地效仿，还是西德克萨斯州利用天然气、甚至多座核电厂被重启，美国能源体系展现出了极强的供给弹性。 每一次资本泡沫如果破灭，最重要的是必须给社会留下长久运转的物理资产（lasting asset）。2000 年代互联网泡沫破裂后，地下铺设的海量暗光纤（dark fiber）留存了下来，Google 当年正是通过近乎免费的价格大量收购 WorldCom 等破产公司遗留的光纤网络，搭建起全球搜索帝国的骨干网；1870 年代铁路泡沫破裂后，北太平洋铁路留下的铁轨基础设施最终化为伯克希尔·哈撒韦旗下 BNSF 持续产生的现金流。AI 泡沫若破灭，GPU 会迅速折旧贬值，数据中心寿命也有限，但由此被逼出来的庞大能源基建将成为人类永久的财富。如果这场泡沫留给世界一个电力无限充裕的时代，将彻底打破人类自古以来的能源匮乏假定。 - **能效与护城河赛跑**：能源短缺成为瓶颈的时间被大幅推迟，这直接剥夺了英伟达原本期待的护城河保护期。随着充沛电力提供了更充裕的时间窗口，亚马逊得以打磨下一代 Trainium，Google 也让 TPU 的能效不断逼近英伟达，使英伟达高昂的超额利润率在未来变得更难以维持。 - **泡沫价值衡量标准**：真正决定 AI 繁荣长期经济效益的并非芯片库存，而是围绕清洁能源与电网扩建沉淀下来的物理现实资产。人类长期受制于能源紧缺思维，AI 军备竞赛倒逼的电网大扩张或将引领社会步入前所未有的能源丰富期。 ## Notable quotes > `00:00` 炸毁台积电（TSMC） — Ben Thompson > `10:00` 全世界把钱花光了 — Ben Thompson > `14:23` 回想一年前人们还在纯手工写代码，简直不可思议 — Ben Thompson > `20:20` 你有自己的网站，你的问题不是缺乏分发渠道，而是缺乏需求 — Ben Thompson > `20:20` 你让它思考几天、几周甚至几个月来解决数学定理，每一秒的思考都在直接消耗真金白银的边际成本 — Ben Thompson > `25:40` 你只是一项功能，不是一家公司 — Steve Jobs > `30:00` 我们只是在盖数据中心外壳（Shells），只有在看到确切需求时才会采购 GPU — Andy Jassy / Satya Nadella > `42:08` 治愈高价的良药正是高价本身 — Ben Thompson > `35:00` iPhone 刚刚发布，这是我们见过的最大历史机遇，我们必须进攻而不是撤退 — 张忠谋 (Morris Chang) > `48:15` 与其聪明，不如走运 — Ben Thompson > `50:00` 他们发自内心地坚信自己正在创造上帝 — Ben Thompson ## Mentioned Ben Thompson, Patrick O'Shaughnessy, Stratechery, OpenAI, Anthropic, Google, Meta, xAI, Grok, TSMC, Intel, Samsung, SK Hynix, Micron, NVIDIA, Apple, Microsoft, Amazon, AWS, Berkshire Hathaway, BNSF, See's Candies, Warren Buffett, Steve Jobs, Mark Zuckerberg, Andy Jassy, Satya Nadella, Morris Chang, C.C. Wei, Lou Gerstner, Sheryl Sandberg, Tim Cook, Elon Musk, Dropbox, Drew Houston, Netflix, YouTube, TikTok, Snapchat, WorldCom, Neuralink, SpaceX, Claude, ChatGPT, Gemini, Trainium, Graviton, TPU, CUDA, ATT ## Video description My guest today is Ben Thompson, the founder and author of Stratechery. Ben is one of my favorite business thinkers and I love talking to him about everything happening in markets and technology.&nbsp; We go through every important company, including OpenAI, Nvidia, Intel, Apple, Microsoft, Google, and Amazon. We also discuss why he thinks it would be dangerous for the United States to win the AI race outright, what container shipping and the railroads of the 1870s tell us about the buildout, and why the binding constraint on all of this may be capital rather than compute.&nbsp; Please enjoy my conversation with Ben Thompson. For the full show notes, transcript, and links to mentioned content, check out the episode page ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠here⁠⁠⁠⁠⁠.&nbsp; ----- Become a Colossus member to get our quarterly print magazine and private audio experience, including exclusive profiles and early access to select episodes. Subscribe at ⁠colossus.com/subscribe⁠. ----- ⁠Ramp’s⁠ mission is to help companies manage their spend in a way that reduces expenses and frees up time for teams to work on more valuable projects. Go to⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ ⁠ramp.com/invest⁠⁠ to sign up for free and get a $250 welcome bonus. ----- Trusted by thousands of businesses, ⁠Vanta⁠ continuously monitors your security posture and streamlines audits so you can win enterprise deals and build customer trust without the traditional overhead. Invest Like the Best listeners get a special offer of $1,000 off Vanta when you go to ⁠vanta.com/invest⁠.&nbsp; ----- WorkOS⁠ is the infrastructure B2B and AI-native companies use to sell to enterprise. It covers everything enterprise security requires: SSO, SCIM, RBAC, Audit Logs, AI governance, and more. Trusted by 2,000+ fast-growing companies, including OpenAI, Anthropic, Cursor, and Vercel. ----- Rogo is the AI platform for finance. They're building agents for Wall Street that are trained to understand how bankers and investors actually do work: from diligence and modeling, to turning analysis into deliverables. To learn more, visit rogo.ai/invest. ----- ⁠Ridgeline⁠ has built a complete, real-time, modern operating system for investment managers. It handles trading, portfolio management, compliance, customer reporting, and much more through an all-in-one real-time cloud platform. Visit⁠ ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ridgeline.ai⁠. ----- Editing and post-production work for this episode was provided by The Podcast Consultant. Timestamps: (00:00:00) Welcome to Invest Like The Best (00:02:16) Winning the AI Race With China (00:08:28) Timing, Capital, and the Railroads (00:11:34) Berkshire, Google, and Absolute Profits (00:14:23) Verifiable and Unverifiable Domains (00:20:20) Aggregation Theory in the AI Era (00:22:06) The Real Cost of Inference (00:25:40) Why Consumer AI Needs Advertising (00:30:08) Compute Shortages and Commodity Markets (00:35:46) Memory Cycles and Boom Bust Dynamics (00:42:08) TSMC, Intel, and Where Risk Goes (00:44:51) The Best Setups in Big Tech (00:52:14) The Frontier Model Contenders (00:54:27) Microsoft's IBM Playbook (01:00:45) Meta, Attention, and Advertising (01:07:29) NVIDIA, Commodities, and Power [Source](https://traffic.megaphone.fm/CLS4584299095.mp3)

[0:00] Ben Thompson: 炸毁台积电（TSMC） [10:00] Ben Thompson: 全世界把钱花光了 [14:23] Ben Thompson: 回想一年前人们还在纯手工写代码，简直不可思议 [20:20] Ben Thompson: 你有自己的网站，你的问题不是缺乏分发渠道，而是缺乏需求 [20:20] Ben Thompson: 你让它思考几天、几周甚至几个月来解决数学定理，每一秒的思考都在直接消耗真金白银的边际成本 [25:40] Steve Jobs: 你只是一项功能，不是一家公司 [30:00] Andy Jassy / Satya Nadella: 我们只是在盖数据中心外壳（Shells），只有在看到确切需求时才会采购 GPU [42:08] Ben Thompson: 治愈高价的良药正是高价本身 [35:00] 张忠谋 (Morris Chang): iPhone 刚刚发布，这是我们见过的最大历史机遇，我们必须进攻而不是撤退 [48:15] Ben Thompson: 与其聪明，不如走运 [50:00] Ben Thompson: 他们发自内心地坚信自己正在创造上帝

[5:00] Ben Thompson: The talk about them being free is bizarre to me because it's marginal costs... You're not paying necessarily the R&D to create the AI, but you're definitely paying the inference to run it. [20:00] Ben Thompson: The hard problem is not distribution, it's discovery. So the companies that solve discovery in their domain come to dominate that market. [30:08] Ben Thompson: In a commodity market, the price is set by the marginal supplier. Cost of service is all that matters. [35:46] Ben Thompson: The issue with the Strait of Hormuz is it's very effective. It's more effective if you don't use it because then it's always hanging out there... Turns out it worked... But they're not going to let this happen again. [42:08] Ben Thompson: The cure for high prices is high prices... The way to solve the geopolitical problem of security dependence on TSMC is to come up with a compute use case that is so massive that everyone is economically incentivized to bring other people up to speed.

[0:00] Ramp is the only platform built to make your finance team leaner, faster, and better, saving businesses 5% annually on average so you can stay focused on growth. Ramp customers grow revenue 3.2 times faster than the average American business. Visa, Vercel, Cursor, Stripe, Notion, 11Lab, Shopify, and 70,000 other businesses all now run on Ramp. Mine does too, and so should yours. Learn more at ramp.com slash invest. OpenAI, Cursor, Anthropic, Perplexity, and Vercel all have something in common. WorkOS. To achieve enterprise adoption at scale, you have to deliver on core capabilities like SSO, SCIM, RBAC, and audit logs. Instead of spending months building these mission-critical capabilities yourself, you can just use WorkOS APIs to gain all of them on day zero. That's why so many of the top AI teams you hear about already run on WorkOS. WorkOS is the fastest way to become enterprise-ready and stay focused on what matters most, your product. Visit workos.com to get a single prompt into finished client-ready work using your firm's own templates, context, and standards. Send Felix an email like, take these comments and turn them for me, or update my tracker with the context of these emails, and Felix sends back finished PowerPoint decks, Excel models, and sourced research. Felix works the way your team already does, delivering work quickly and accurately around the clock. Learn more at rogo.ai slash Felix. Hello and welcome, everyone. I'm Patrick O'Shaughnessy, and this is Invest Like the Best. This show is an open-ended exploration of markets, ideas, stories, and strategies that will help you better invest both your time and your money. If you enjoy these conversations and want to go deeper, check out Colossus, our quarterly publication with in-depth profiles of the people shaping business and investing. You can find Colossus along with all of our podcasts at colossus.com. Patrick O'Shaughnessy is the CEO of Positive Sum. All opinions, Thank you. So Ben, if you can believe it, how long it's been since we last did this. The world was very different, no AI at the time. We talked about aggregation theory mostly. I thought a fun place to begin since the world has changed so much. is to hear what you think it would mean for the U.S. to win the AI race. I think it would be very problematic for the U.S. to win. Let's say we take the most sort of fantastical scenario where if you control AI, you basically your military is better than anyone else. Somehow it fixes their manufacturing, all these things that I don't think AI is necessarily going to do because they sort of deal with the real world. But in this world, what is the... Game theory optimal response of China. To blow up TSMC. Game theory can get very sort of convoluted and complex. To me, this one actually isn't that complicated. There is a fundamental disconnect that I have with a lot of the rhetoric coming out of Silicon Valley coming out, I think, of one of the labs in particular, where if we get to a place where. we have a meaningful superiority in terms of a military national security perspective. I think that's very dangerous for the world. But in that state, how much does it extend beyond TSMC being blown up? Because in that state, I would assume we figured out how to build fabs here in the U.S., to some degree and are less reliant on that one choke point. I think there's a little bit of magical thinking there. which I just invoked in terms of manufacturing and whether it be fabs, whether that be actuators, all these precursors. I think the degree to which we are dependent on China is underappreciated and is not something that is going to be fixed outside of a conflict. Just because fixing so many of these things is going to be dramatically dumb. If your competitor is sourcing from China and you're going to start sourcing or getting things from the US, you're going to be at such a disadvantage, relatively speaking, that you're just not going to do it. So you do it when you have literally no choice. And that works for very big headline items. You can browbeat Apple to move some of their iPhone manufacturing to India, for example. But even that is a good example because Apple is not – truly moving out of China. They're diversifying to an extent, but It would just cost so much. It's like paying an insurance policy. That if you don't have to pay it and it's astronomically expensive, you're just not going to pay it. It's one of those sort of hypotheses that I just have a hard time even grokking because the only world I see where we truly – pull out and have no dependency on China, such that if they want to blow up Taiwan, who cares? There's no impact on us. [5:00] It seemed pretty fantastical to me. And I think there's a bit of... facing reality in this regard that, you know, is not present in these conversations. Put yourself in their shoes. What do you think the motivations are? Everyone can use a good bogeyman. I think from the AI trade perspective, nothing works better than we have to beat China. And I do think we need to... We need to be competitive. I despair at the extent to which over the last few years in particular, so many of our responses, particularly from a political perspective, has been to try to be like China. I think we should be going the other direction, more openness, more innovation, less top-down control, less restrictions on politics. speech and things along those lines. America succeeds by being on the leading edge and by leading into that. You said probably the U.S. being purely dominant in AI is not the right end state for the world. What is your ideal equilibrium for how this goes worldwide? There's a bit where AI right now is kind of like the Taiwan situation. in that the current status quo actually doesn't seem so bad. The question is, how sustainable is it? But maybe it's sustainable for longer than we think. The way I think about it right now is I think OpenAI and Anthropic are clearly on the frontier. Who knows what's happening with Google and then Grok and Meta are chasing them. Meanwhile, the Chinese are very capable, very smart, and also definitely distilling these models to sort of stay about six to nine months behind. And, you know, It feels like a pretty good equilibrium that I think is generally favorable to the U.S., Now, the question is, how long can it stay this late? And there's lots of questions on there. Like, can the Chinese actually pull ahead? I'm still a little skeptical for various reasons, whether it be from chips or Getting to the waiting edge in that last six to nine months is very difficult. It's going to be instructive how Meta and Grok do in terms of actually catching up, especially as we get to the world of AI improving itself, using AI to make the AI better, which I think is definitely a real thing. I think you see a real acceleration from both OpenAI and Anthropic recently, which was sort of theorized and it seems to be coming true. And to the extent that's true. Can you actually catch up? And I think the other question about this, by the way, is to what extent does that apply to cost to serve, to marginal costs? If you can apply AI to optimizing your stack, to figuring things out, to analyzing all the data, is your cost to serve structurally lower than anyone else? This is the thing about the open source models. The talk about them being free is bizarre to me because it's marginal costs. cost per answer is significantly higher. Everyone referring to these as free, it feels like in the narrative, it's in people's head that free is free. Now I can use AI for free. No, you can't use AI for free. You're not paying necessarily the R&D to... create the AI, but you're definitely paying the inference to sort of run it. So right now, I kind of like where we are. And the pushback would be, that's right now. It's not going to stay that way, which I think is fair pushback, but I don't know this way longer than we think. If you could know anything about the future of how this will go to be more confident in like where the equilibrium will end up, what is it? Is it like the length of the S curve? Like how far up the S curve we are? At some point, these things presumably will level out. Maybe not. What would be the thing you'd want to know that would give you a better sense of what the future might look like? I am concerned that with the scare around... people freaking out about Mythos and this hugging face incident, that the actual... implication of that is not that we reduce the these dangers, but we just stop releasing stuff. We on the outside... start to lose any sense of like where exactly what is actually the frontier and where it is. there becomes sort of a false sense of security situation Because right now everyone's basing their understanding of mythos on fable. But how good is Fable actually relative to Mythos? That's a gap. is only going to, I think, increase over time. So I think that's a real question that I'm not sure about. This question of the recursiveness and AI sort of making itself better, does that lead to some sort of takeoff? And at the end of timing questions in lots of different ways. I'm worried about the timing mismatch in terms of the actual return on investment producing enough revenue to fuel investment. We're working our way down the capital curve. We started with free cash flow. The speed with which the tech companies blew through the debt markets is kind of incredible. It took like a year. And now Google is issuing equity. NVIDIA is putting together these. This $500 billion thing. This $500 billion thing to tap into like pension funds and insurance floats and things like that. What's after that? Where does the money come after that? Well, ideally, we actually flip back. [10:00] to free cash flow funding this. But If there's a gap there, if we don't get there soon enough, then – we could have a big blow up. But at the same time, Even if we have this blow up. The AI is not going away. It's not going to stop improving. It's going to keep sort of progressing in a way that we look back on the dot com era. We look back on the railroad era. We look back on whatever bubbles through history, ultimately immaterial in terms of the broad scope of humanity, even if they were immaterial. Very devastating. What can the railroads teach us to think? It's not the last bigger build out, right? In terms of percentage GDP. I think we might be bigger at this point, or it's like it was the biggest in the ballpark. The railroads had a real duration mismatch. To build a railroad and make money off it was a decade or multiple decades long endeavor. Whereas you had to issue money to pay for it in the short term. And the world ran out of money, right? And I think that is probably the aspect. I think that's why people reach for the railroads. because everyone talks about, are we going to have enough? Compute. Are we going to have enough electricity? Maybe the nearest term questions, are we going to have enough money? Which is kind of a bizarre thing to think about. That's what happened in the 1870s. The world just ran out of money. The funny thing is the railroads kept, operating and they expanded the West. Their contributions to GDP was astronomical. They're still contributing to GDP. Railroad money. is what's going into Google right now for Berkshire Hathaway. It's very funny. It's quite literal. Berkshire Hathaway has this problem. To me, this NVIDIA deal is very much Paired with the Google equity issuance, which I thought was shocking when it happened. Why is it shocking? Because it's Google. They can't raise money. Like, why are they issuing equity? Why are they reducing their upside if they believe so strongly in this? But the Berkshire comparison is interesting because to a rough approximation, they make they have seized candies famously, right? tremendously high margin business. The problem with a lot of high margin businesses is the percentage profit you can make is very high, but the absolute profit you can make is no reinvestment. That's right. You're just accumulating cash. The brilliance of the BNSF railway thing was basically they took the seized candy profits and said, here's another industry whose margins are way worse. But the absolute dollar amounts are so large that those way worse margins result in absolute profits that are much larger. BNSF in 2025 or something, the amount of free cash they threw off in one year was more than seized candies that thrown off its entire lifetime. a very high margin business. I think there's an aspect from Berkshire Hathaway where the Once your capital gets so large, you start operating in a world of like absolute numbers as opposed to percentage numbers. And the reason why I thought that was so interesting, that story, is – it seems to capture where Google itself might be going. And so it is very symbolic for them to invest in Google. Google has this unbelievable high-margin business business, of search. One of the most perfect, beautiful business models of all time. And the purest aggregator of them all, like scales in every direction, doesn't have to invest any money to do it. Everything's zero marginal cost. It's amazing. Meanwhile, there's this AI opportunity, which requires just astronomical, it's just incinerating cash. But you can imagine if AI is... intelligence and it's TAM is basically all white collar work. And eventually with robotics, more everything, potentially the absolute profits available here, even if the margins are lower, is so much larger that will we look back in Google search with seized candies? It feels like that's what's happening in that world. You use all your free cash flow. They've done that. You tap the debt markets to the tune of hundreds of billions of dollars. They've done that. You issue equity. What does an equity issues do? It dilutes your interest in your interest, your shareholders. So you have a smaller percentage of the pie. Well, you have a smaller percentage of an astronomically larger pie. At the end of the day, no one's going to be complaining. It was very symbolic, Berkshire being the symbol of that equity issuance. In that, are they actually not just an investor in Google, but a model for Google and where they're going? I'm curious, setting aside the commercial and competitive components of this, like you're describing, How AI pilled on a pure technology would you say you are relative to AI? other people thinking about the space. I have a view that is both super bullish and less bullish in some respects. Okay. So – I am not fully convinced about the generalizable argument. AI is clearly incredible at coding. It kind of blows my mind that people were doing this a year ago, like actually – [15:00] writing out code, He's very good at math, obviously, but the obvious riposte is that these are sort of verifiable domains. What is the evidence or where is the compelling evidence of, Being very good at verifiable domains queenly translates to being very good at sort of unverifiable domains or domains that have very long sort of verification loops. I think that's still a little bit to be determined. And it's interesting because I raised this question and there are some people in the labs that were on a panel and I was kind of annoyed at the answer because the answer took me for an AI bear. Oh, well, people thought we couldn't solve chess or we couldn't solve gold. We solved those easy enough. And I'm like, no. I thought we could solve chess. I thought we could solve gold because they're knowable domains. Scale was the answer to both of those, but also both of those were bounded. What is the go-to example that's not chess, that's not go, that is genuinely in balance? a new space that's sort of an unknowable space where it's doing things that were not possible. That is sort of the... I'm not fully convinced since. However, AI trains... Not a rough approximation trained on all the data on the internet. All the data of the Internet... That's distillation. It distilled all of the end state of human thought. The actual typing on Reddit, it doesn't have the traces. It doesn't actually have the thought, the emotion or whatever that went into typing that comment or typing, writing that essay. say, Neuralink, whatever. What if the actual path from Neuralink is actually capturing the traces of human thought? that actually dramatically expands the capabilities of these models. In this world, my concerns about verifiability is like, well, we solve verifiability by getting more data. My sense is that a huge number of jobs, a huge amount of economic activity. does not exist in these domains that I'm not convinced that AI is good at. Actually, there's a lot of people in the world who are kind of like sentient AIs to a certain extent. They operate very well in verifiable domains. They're given jobs. They do them. And it's almost like a somewhat pessimistic view of humanity to a certain extent. But I think that market is so huge and so large that if the models did not improve at all from where they are right now, the economic opportunity is actually massive. I wrote an article a while ago. There's the whole like accelerationist movement. What I call myself was a reluctant accelerationist. I think we need to push forward because... We can't go back. And the worst thing we can do is get stuck where we are. So I'm very AI-filled in terms of its impact on technology. the economy. It's sort of upside in terms of monetization. I'm not sure about the timing. What would be like the gradient towards it? Imagine law or medicine where I don't know whether or not you would consider those verifiable like law is like a code. of some sort medicine. We have a certain state understanding of things. I mean, I think medicine is by far one of the biggest opportunities. It's both of the biggest opportunities and also one of the most challenging ones because of all the regulations and all the access. Like if you could turn an AI, turn machine learning onto all the medical records, I think the number of discoveries and improved treatments we could come up with in a very rapid amount of time would be unbelievable. So that is a very optimistic view. On the flip side, that going to happen, right? I think the optimistic frame I put on humans is our capacity to create needs is sort of unlimited. So I think we'll do a very good job of creating new opportunities and job serving the fullness of time. The sort of more pessimistic way to put it is our ability to create red tape and muck is also fairly unlimited. How much of our economy is actually we've managed to create more and more jobs that is just make busy and make slow to a certain extent. Vanta automates security and compliance for over 16,000 fast-moving companies like Ramp, Cursor, and Harvey, keeping them audit-ready around the clock. It's the number one agentic trust platform, and it now helps companies like yours watch for the risks that show up between audits, across your vendors, your AI tools, and your whole environment. Every new tool your team signs up for, every vendor that turns on AI features, is an opportunity for something to go wrong, and most security programs weren't built for AI's pace of growth. works like a 24-7 GRC engineer in the background, finding issues, drafting fixes for you, and cutting vendor assessment time by up to 50%. Whether you're a fast-growing startup or a global enterprise, Vanta helps you earn and prove trust. Invest like the best listeners get a special offer for $1,000 off at vanta.com slash invest. Ridgeline is the first end-to-end system of record with embedded AI for investment management firms, running portfolio accounting, reconciliation, reporting, trading, and compliance on one unified platform. Firms are moving off legacy technology and onto [20:00] Ridgeline because of how far ahead Ridgeline's AI features are compared to anything else in investment management software, which is why I believe that firms that come out ahead in the AI era will be the ones running on Ridgeline's unified platform. If you're serious about your firm's AI strategy, Ridgeline should be part of that conversation. You can request a demo at ridgeline.ai. you If I go back to the early 2010s, maybe the aggregation theory was stewing in your brain, and then you published it in 2015. I think it's fair to say like that theory, that idea, maybe you can just quickly remind people what it is. defined the winners and losers of that era of technology. I'm really curious how you're thinking about what theory or principles will define this era of winners from like a financial perspective and market cap perspective. I go back and forth even just on the question of aggregation theory itself. How much does that apply in the current era? Just the same thing. Yeah, yeah. Yeah, because like a pushback that people have is one of the key components of aggregation theory is zero marginal costs. And zero marginal costs. shows in lots of ways. The one that I focused on the beginning was distribution. And people say, oh, I don't have distribution. I have to pay Google for ads. I go, no, you have a website. Your problem isn't that you have distribution. Your problem is you don't have demand. And you're paying for demand when you're paying for ads and things on those because the aggregators control demand. And they control demand because in a world of abundance, you're going to have a the hard problem is not distribution, it's discovery. How do you actually find what you're interested in? So the companies that solve discovery, in their domain come to dominate that market. They get a virtuous feedback loop, that sort of aggregation theory in a nutshell. And the other thing is transaction costs. There's no transaction costs. Google can scale to the whole world and it can scale to the whole world, not just on the user side, but also on the monetization side. The vast, vast, vast majority of advertisers on Google or Meta never talk to someone at Google or Meta. They just go up and they buy ads. It's all done by computers. The perfect business. And those computers, from a business perspective – cost zero dollars. AI, obviously, that changes significantly. Inference costs are real. But then again, how real are they? They're real right now. I don't know. Are they? Depends on the company, but they're way more real than those prior examples. Well, like if you look at gross margins. For sure. But you have this incredible spread. So you have people, I think the vast majority of people who are using ad today are using it as basically a Google substitute. or like a recipe maker, or whatever it might be. And my suspicion is that the cost to serve those people is – Extremely low and low in the basically similar to serving them a web page. I would imagine it's marginally higher, but not that much higher. Then you have on the other extreme people who are actually leveraging test time scaling. It used to be we just scale by making the models bigger and bigger. Now you can scale. As far as time, how long do you think about the answer? Well, you could think about the answer... for days or weeks or months, that is directly marginal cost. Every second longer you're thinking is costing more money, which speaks to like, We think about AI and inference as this one question that's I was pushing back on you, but actually the marginal cost question for the different user, the user using free chat GPT and the user trying to solve a math theorem. They're not even remotely in the same universe. I think you see this challenge... actually in the enterprise in a very interesting way. Microsoft recently, they are shifting their enterprise plan. So they come out with like an E7 plan. $100 per user per month. That includes some amount of usage. But then they also... are charging for usage on top of that. I think this is kind of a fraught position for Microsoft to an extent, because the positive way to think about Microsoft is... They do everything you need as a business. Every individual component might not be the best. But you get it all for one price, and they all mostly work together. And if you're particularly a small or medium sized business or even a large enterprise, there's real value in that. That's right. It makes life easy. The moment you start having to think about how much you're paying, you're It's not just that that's a new decision, number one, that is untethered from headcount. Microsoft got the benefit is when you were hiring a new employee, you would think about the cost of that employee and baked in the cost of that employee is $100 a month or $50 a month for their license. It was kind of a thoughtless process. revenue stream from Microsoft. Now, if you think about usage, you have to think every single month, how much do I want to spend? That introduces two problems. Number one, most companies aren't set up to do this. They make budgets like once a year. This idea we're going to be thinking about through our budgetary allotment on like a monthly basis doesn't compute. There's an aspect where they're used to thinking about CapEx decisions or one-time cost. And there's a bit where I'm talking about this employee, like the loaded cost of employee. it's not CapEx, but it's kind of like CapEx. It's like you make the decision up front and you don't think about [25:00] anymore. This isn't sort of already made. But if you're thinking about usage, you're doing it again. The final thing is, if you're every month looking at your Microsoft bill and how much did I use, You start thinking about what am I paying for? How good is each of these products? Should I actually just start thinking about and spraying this out? And I think they had to do it because that extreme of user who uses a ton of tokens and is actually leveraging AI costs way more to Microsoft than $100 a month. They can't support them. But they want to hold on to... this set cost for the vast majority of employees who can fit in that because they need to ask their customers to think a little bit. for those extreme employees, but they don't want them to think too much because that breaks the model in very surprising ways. Are you surprised at all that the recipe builder user that is very low cost to serve? that there hasn't been a great business model that's emerged around them just yet. Google and Facebook are sort of business perfected in this prior era. They haven't seemed to figure this out at all. I am frustrated, but not surprised. This is obviously a market that should be supported by advertising. That is why advertising is always the consumer business model. Consumers don't want to pay. There's two things to understand about consumers that Silicon Valley has to relearn about every 10 years. Number one, consumers do not want to pay for software. And number two, consumers do not care about being productive. We went through this in early SaaS. The canonical company for this, in my mind, is Dropbox. So Dropbox unbelievable product, like especially when it first came out in business school. I was one of the first people who used Dropbox. And that went off like crazy. I have so much storage still, like my free Dropbox, because I gave out my code to like so many people. So Drew Houston makes this amazing product so easy to use, just absolutely seamless. He was very clear about this. He wanted to build a consumer company. And there's that famous story of him meeting with Steve Jobs. Apple was interested in acquiring Dropbox. And I go, we want to build a company. And Steve's, you know, you're a feature, not a company, which that plain Jane just files sync. Apple did make a feature as far as like sort of iCloud Drive and what Dropbox did. They grew very fast. And then they have like a two-year lull. And in that two-year lull, what they had to do was basically completely rebuild the app from the bottoms up. Because... Not enough consumers are going to pay for it. Enterprises could see the value they would pay. But if you want enterprise, you need permissions. You need control. You need someone else to be able to set all these sorts of things. And their app wasn't even created to do that at all. So they had to rebuild the whole thing and realize the only way we're going to make money is by selling to companies. Why do companies pay? Because companies are paying employees. To the extent they can make their employees more productive, they're getting a greater return on their investment. It's the complete inverse of – A consumer, a consumer is like, I spent all day working. Why do I want to come home and be more productive? I want to sit on the couch and watch reels. But you see that with AI. And you also have this overarching just skepticism. of advertising. I've gotten so much traction on Chatechery by being an advertising appreciator. And I go back and read my early articles about advertising that were kind of directionally correct, but also like. We're not very good at all, but I got so much traction doing it because I was the only person writing about advertising in a world of everyone want to have a blog and Twitter. No one wants to talk about advertising. But even now, there's in Silicon Valley, this sort of embarrassment about the fact that the Valley is in many respects. monetized by advertising. And particularly during the last sort of eight years, there was a Facebook's icky. The best engineers don't want to go work on this problem. And so you literally had OpenAI replaying the Dropbox story, but at like 100 exercise, being like, no, we're going to sell... subscriptions to consumers. They did. They sold a lot. but they didn't sell enough. If you're going to be in the consumer market, you have to be doing advertising. They're doing advertising now. It's a little weird. They finally pivoted to doing advertising. At the same time, they're like, oh crap, we need to go for the enterprise because... Anthropic is kicking over in. So I'm not quite sure what they're doing there. They have been rolling out ad features very rapidly. Things like Cappy and the connections with retailers. So, you know, if a person went through so you can do all the tracking and things like that. I'm very interested to see how that goes. There's a bit where have they leaned into advertising immediately as soon as ChatGPT was a hit? I think they would have a killer app product right now. I think that Google would be in much bigger trouble. I think Meta would be in much bigger trouble, because if you have this flywheel, the thing about advertising with consumers, is your ability to monetize the consumer - Goes up as the bond goes up, yeah. - It's infinite. Because the advertiser is bearing the price increase. So there's zero elasticity issues. If you're charging consumers a price, if you want to raise the price, like Netflix, this is their problem with the subscription plan, how much can they raise prices before consumers rebel and [30:00] Yeah. drop a tear or give up the service entirely. Charging people money is hard. giving people things for free is easy. And it's very frustrating that, OpenAI did not pursue this sooner. I know you've been spending time with some of the big money firms and sources of capital. What is your sense of their appetite right now and how they're thinking about the future? Because I think this year it's going to be 800 billion or something that we're going to spend in CapEx next year. It's supposed to be one point three trillion, I think, is the current estimate. It's going to keep going up from there. We're burning through all the compute that gets installed. basically immediately. It's such a strange circumstance that we can use the capacity right away as soon as it's online. Well, that's the thing, though. So there's a few timing mismatches that are happening right now. We can't use it right away. All the bulls on Twitter is always like, we don't have enough compute, we don't have compute. Well, we don't have enough compute because there was insufficient investment made in 2023 and 2024. Which, yes, absolutely. And by the way, if you think there's not enough compute, TSMC decreased their rate of growth in 2023, in 2024, in 2025. Our shortage of compute is going to get worse in the next few years. Because a fab, the lead time is even greater than a data center. Today, when we say there's not enough compute, It's not like all the money that the companies are putting in today is Manifest in compute. No, it all manifests in compute in 2028 and 2029 on the calls. You have both Andy Jassy inside Adela are out there saying, look, we're just building data centers like these are the shells. We might not use them now. Maybe we'll use them in the future. And we only buy GPUs when we know there's demand for them. That is a great story to tell. I'm not sure that I think is a lot of BS because the reality is if you've built the shell, that money is sitting there. You're not going to let it just sit there. If you invested a fixed cost, and this is the whole logic of commodity markets, I think tech in general doesn't understand commodity markets. Tech is by and large focused on if I produce a highly differentiated product and that differentiation could be like, software, it could be a network in terms of developers, it could be a social network sort of thing where peer to peer, where I'm highly differentiated than my ability to charge higher prices provides sort of my profit margin. So the... Class example is like Apple. They have their ecosystem and they have their software and they have third party and all those sorts of things. And so they can charge 50 percent margins on their iPhone. Everyone looks at Apple as like the ideal business model. That's how you run a business. But in a commodity market, the price is set by the marginal supplier. Cost of service is all that matters. That's right. I had a good friend in Taiwan. who is in shipping, fascinating industry. It's kind of like the airlines too, another industry that I love to look at. You buy a ship and the cost of that ship is depreciation. your marginal cost is actually quite low. It's the fuel to run the ship and the cost of the crew and like your port fees. Not that much. What that means is you are going to run that ship. As full as you really cost. No, you're going to run it no matter what. And you're going to bring down the price of a container as low as it needs to be to cover your marginal costs. Now, your paper losses in this situation might be very large because your accounting loss includes depreciation. But the depreciation is an accounting segment. You already paid the money. You're going to run that ship at whatever the market will bear in the container. The beauty of the container, it is a pure commodity. The cost of the market is going to be the marginal cost. Now, if it gets low enough, at some point, people will exit because their marginal costs, they're actually losing money on a shipment, not just paper money, but like actual real money. They will exit. But then the supply is diminished, so then the price will go back up, and you get this interplay of sort of coming in and off. But then let's say the market's very high, like it was during COVID. It's like, wow, we're making so much money right now because there's not enough supply. There wasn't enough supply of ships. So containers went from usually being like $3,000, $4,000 to $17,000, $18,000. The amount of money that these shipping companies made in a very short amount of time was insane. What happens, though? Well, imagine if we had more ships, right? The problem is it takes two years to build a ship. If everyone makes this decision simultaneously, you suddenly have a lot of ships, price plummets, et cetera. Where we see this is in components, in memory in particular. Memory, very famous for boom and bust cycles, people entering the market late. But to what extent are data centers working? going to be memory makers. where right now everyone can see we don't have enough compute. So everyone's like, we absolutely have to be investing because there's so much money to make. And look at our payback period. The problem is you're measuring your payback period in a time of scarcity. Is that payback period going to hold in a time of abundance? And the sort of the bold say, there's never going to be time of abundance. AI, we're going to be short forever, which maybe we will be. My concern is even if that's right, we could still have an air gap. in that there's [35:00] much money going into it right now And not enough has come online to actually make sufficient revenues to handle the situation where we run out of capital. I believe in AI. I think it's a real thing. I think the economic impact is going to be astronomical. I think all the concerns about societal impact are very real and are going to come to bear in a major way. You can believe all that and still be worried about. Are we going to make the bridge to this actually generating the level of returns necessary to continue to feel this sort of going forward? Can you zoom in on TSMC and the component makers where fabs are involved? And so far, at least my understanding is that they've been quite conservative in their willingness to expand capacity, build new fabs, meet the market's demand with similar growth. which they have not done, if that just really limits this whole thing and prevents us from getting one of these giant overbuilds, We can talk about a few different ones. We'll start with Memory. Memory used to have tons and tons of Memory Makers. Every time there'd be a boom, Memory Makers would reenter the market New countries have come in, like Taiwan used to have like a memory market, but you would get these exact dynamics. If there's a shortage of memory, there's so much money to be made, you can't bring capacity on immediately. The same as shipping, the same as what we're seeing right now, that would spur the... People come in the market, you get too much capacity, prices would plunge, and people would just get blown out. Because the issue is the upfront cost for these is so large. Just like buying a ship, like building a fab is even more so. And memory now, like the leading edges of memory are using things like EUV machines. So the costs are getting into the billions of dollars for these lines. What happens is every time with these boom and bust cycles, some people would enter, more people get washed out. You go through these famous historical moments for these memory cycles. Companies just get blown out. is how Samsung sort of took over memory. was they saw it as an opportunity and they had studied history and they realized that actually the way to take over the market is to invest into downturns so that you're ready when the next cycle comes around, which requires a ton of guts and a ton of discipline and a ton of money. But they did that and basically wiped out the Japanese. That's when the South Koreans generally took over the market in a major way. But it got down to three. And the problem is three, about let's not make the mistakes of the past. And we're not colluding, but we all are on the same page about let's not do that. And I think that dynamic sort of ran head on to the current moment where it just took a while for them to realize, no, there is a secular shift in society. memory demand that didn't exist for a very long time. I think the memory solution will be solved eventually. The other risk they run is Apple's lobbying to get Chinese memory. What is the number one focus of like algorithmic changes? How can we use less memory? I think the memory makers probably screw themselves in the long run by creating such a massive target on their back. I've analogized memory makers to Iran. The issue with the Strait of Hormuz is it's very effective. It's more effective if you don't use it because then it's always hanging out there as something you could do. Now they did it. Turns out it worked. But the UAE, Saudi Arabia, they're going to build pipelines. They're going to build new ports. They're not going to let this happen again. It's very painful right now. But say Iran wants to close the straighter moves in 2035. It's not going to have any effect because it will have been built around. My concern for the memory makers is they might have done the same thing. No one's going to let themselves get in this situation again as far as memory goes. TSMC is arguably worse because there's only one. There's one company on the leading edge. Obviously, Intel and Samsung are trying to get there. It's the same thing. All markets carry risk. And a lot of the question is, Who ends up holding the risk? What I think a lot of the tech companies didn't fully appreciate is the extent to which TSMC has offloaded risk onto the big tech companies. And the way they've done that is the risk that TSMC is worried about is overcapacity. If we build too much, it's not just that we built too much. And we have all these fixed costs that are not being fully utilized. But if we build a fab, we expect that fab to run for 30 years. We've like baked in too much capacity into the system for years and years and years. So they are very biased towards being much more conservative. There's a little bit of a culture component to this, too. One of the most interesting TSA stories, it's kind of a Nellis, that Samsung story, was Morris Chang retired in like the late 2000s. New leadership took over. There was the Great Recession. And so they pulled back their planned spending. He comes in, fires everyone. And he's like, the iPhone just launched. This is the biggest opportunity we've ever seen. We need to be investing, not cutting. And they invested through the Great Recession and through that downturn. That's what laid the foundation for them taking over sort of leading edge semiconductors in that time. Morse Chang is a one of one on the Mount Rushmore. [40:00] more in my mind, of the greatest and most impactful tech executives of all time. The entire fabulous model is so critical to what tech is and what it does. And also just the guts to do that at that time, particularly in... someone who lived there, a culture that doesn't necessarily tend to make those sorts of bets. TSMC, they were... Pretty conservative, to be totally honest. So what happens, though? Where'd the risk go? TSMC is like, we don't want to take the risk. Risk doesn't disappear. It just moves. The risk is right now where you have... Every single big tech company realizes if we had more compute, we could be making more money. So there's lots of foregone revenue and foregone profits. That is the manifestation of the risk that TSMC handed off to them. Risk doesn't disappear. It just gets handed off. And sometimes that risk doesn't manifest in losing money. It manifests in not making money. And there is money not being made right now because. What happened was they were very excited about 5G. They did a big wave of like investment, expanding their fabs in around 2020, 2021, 22. And they're like, oh, yeah, we're good. Like I said, 2024, Chats of Beauty is out of 2022. Big thing in tech in 2023. In 2024, their growth rate went down. In 2025, their growth rate went down. In 2026, it's up now. It was very funny because I was writing about this a while ago. And then I think it was like one or two earnings calls ago. Suddenly, CeCe Way, the CEO and chairman, is talking about like, Use cases for AI, the whole earnings call in a way he never had before. This is why the lawmakers are scared. Usually there's like a bullwhip and they're worried about being at the end of the bullwhip where the demand happens and it works its way down the chain and they're at the end. And then they double down. It's already too late. They're wasting all their money. And I think the thing with AI is. If it's a bullwhip, it's like the longest bullwhip of all time. There's still so much to be built. And it just took a while for Asia to get the message where these sort of companies are. I think they've by and large gotten it. But them getting the message, it then takes several years for that to actually materialize. Do you have a sense for how long you think it will take given the extreme shortage of compute? The interesting thing is what this means for Intel and Samsung's logic business. I've been writing about the... problem of this dependency on TSMC for years. One of my first articles in 2013 was exhorting Intel. Well, I say you have to build a fab business. You're not going to be a desire anymore. There's a huge business in manufacturing chips. I thought I was late writing it then. Their stock goes to the moon throughout the 2010s as they're riding the sort of cloud wave. And it wasn't until... 2020, where they finally realize we fell behind. By the way, there's this huge opportunity. We're totally unprepared for it. We don't have a customer service mindset or culture organization or all the IP building blocks and all these things that TSMC has. And they need a customer. They need customers to help them actually build a real foundry business. So I would write about this as a problem. I write about the China issue. Like you're dependent on a company that is 60 miles offshore of our greatest geopolitical opponent who thinks it's theirs. So these are big problems. That's where I came to appreciate this insurance issue for a big tech company to go to Intel and say, Intel, you make our chip. And by the way, the biggest benefactor of this is going to be you because you're going to learn how to work with a partner. And the biggest benefit. pain is going to be us because we're going to have to figure out how to work with you. We could just go to TSMC. They are awesome. They are so great to work with. We know they're going to do a good job. It just never made rational sense for anyone to go work with Intel. That was their fundamental problem. In an unchanging world, TSMC would just win forever. But this is where TSMC, in some respects, made the same mistake as the money makers, made the same mistakes as I ran, if I can continue the analogy. Because they didn't invest the last few years, going to be so acute. big 10 companies that were foregoing so much revenue and so many profits because we don't have enough compute. We will go through the pain of getting Intel up to speed, of getting Samsung's logic up to speed. The scarcity is what ultimately saved Intel. I expect at some point. that they're going to announce some major partner for the first time. It's going to be a big deal. But ultimately, TSMC brought it on themselves. It's the cure for high prices is high prices thing. We're going to route around them. There's all these things as like an analyst sitting on the side. You can write these things. And it was as I sort of warned, no one's going to pay insurance that they don't need to pay when that insurance expected value is negative. The way to solve the geopolitical problem of security. dependence on TSMC, is to come up with a compute use case that is so massive that everyone is economically incentivized to bring other people up to speed. And then we get the sort of geopolitical insurance for free. If you think about the, let's say, top 10 or 15 technology companies, which ones do you [45:00] most interesting setups today for their business. The answer is always Amazon. The reason Amazon is so compelling is the extent to which they build for them. They are their first best customer. They provide the scale to get basically anything off the ground, which they then sell to other people. AWS is the most obvious example. AWS, contrary to sort of popular thought, was not spare Amazon capacity. Actually, it took a long time to get Amazon.com onto AWS. What it drove was the understanding that we can't be having so many meetings. Like we need to have just compute that you can plug in purely API surface. You don't need to talk to anyone. It's just there. And oh, by the way, if we do that for our internal retail teams, we could do that for anyone. Yeah. Turns out the retail is so big, we have to start with everyone else. AWS actually started serving external customers before it served internal ones, but now it serves them all. You got other products like, say, the logistics where – It was the opposite. Right now we're using external providers for logistics, UPS and FedEx and USPS. We need to build this up ourselves. And now they built it up themselves, they're offering to third parties. Other people can use their delivery services. You see this in market after market. They're talking about some of their AI products or their chip products. What's the beauty of the Graviton or the Tranium, particularly the early versions? The early versions were terrible. But if you're on Amazon and you're using some of their managed services, like, say, the Redshift database service, they don't tell you what the processor is underneath that. You're just buying a managed service. So they can put all their crappy processors underneath the services they're selling, and that gives them the volume and the capacity to iterate them and get better. And they get to the point where they can actually sell them externally. Because they were the first best customer for Graviton, Graviton got better. Because they were the first best customer for Tranium, Tranium got better. And now Tranium is obviously running Anthropic and AI products. We'll see if any of them take off. They have call center software. Their call center or their customer experience is going through AI. Haven't tried it. Moving back to America, I've been buying lots of stuff. Every summer, I'd buy lots of stuff in a very brief amount of time. Sometime in like the last year or so, you can go on and you're clearly talking to a chat bot, but the chat bot does a great job. And it actually does take care of the problem. So you could see that actually starting to work in that regard. But they're building up these AI services for their own business that they're going to make broadly available. And some of them will work. Some of them won't. It's such an elegant approach, given they have so many investments in the real world. Their core business feels so impervious to AI. for the model version of AI. It will benefit from AI, but their moat feels deeper than anyone as far as their core business and their ability to just sort of generate new business lines organically. is very compelling. What about Apple? They've sat this whole thing out, it seems. It feels like it might be a situation that better be lucky than good, to a certain extent. Apple has their whole ecosystem, At the end of the day, they do own access to customers, so they can get suppliers. This is the classic aggregator play. If you own access to customers, you're going to get a lot of money. suppliers come to you, not the other way around. So they can get suppliers for their AI as needed. And by the way, To the extent it's true that people don't want to be productive, they just want a sort of a chatbot. Not only can they serve them a chatbot and finally getting a Siri that works, but you can see a future where this absolutely can work on device. And they actually don't even need to pay for inference costs either because they're using the customer's electricity. I don't think we're quite there. There's a reason they're using Google Cloud and NVIDIA chips. But you can certainly imagine a future where that's the case. And they're in physical goods. Actually making phones is hard. Having retail, having distribution for physical goods. They're more insulated. The smartphone is so perfect. It's small to fit in your pocket. It's big enough to watch basically anything on it. You can run your whole life on it. All your entertainment is there. When we talk about customers who want to be entertained, the TV is now an accessory. It's all on your phone. I don't see anyone taking over the phone. The question is, is the phone always going to be the center? Or is there a bit where, particularly in the home, these are where OpenAI's efforts here are very interesting, where you want sort of an ambient AI, where you just talk to the AI and it tells you what you need. Apple is the best position to provide that. But can they provide that without having leading-edge models. Can they provide that if they're so phone centric or is it like a Microsoft situation? Microsoft didn't miss mobile. They were very early to mobile. The problem is their mobile was a small PC. They assumed the PC would always be the center and their phones were going to be Something that was off that Apple realized, no, we need to reset. The phone is not going to be accessory to the Mac. The phone is going to be the phone. The iPod helped them realize that and going with Windows and all that. But will they fall into a Microsoft like track? Assuming the phone's so good, it's always going to be the center. [50:00] And then let's figure out around it. Or is this finally the time when actually... ambient, the cloud, just in general, AI being everywhere. it can manifest through your phone, it can manifest through a device, can manifest on your computer, is actually better. is actually disruptive to them. I think it's possible I also think it's totally valid for Apple to double down on what they do. The other thing about the AI stuff is, On what basis should we expect Apple to be good at this? At the most crude level, AI is this probabilistic endeavor. Apple is the king of deterministic ideas. products, a physical product. You ship that iPhone, you ship it once. And it's gotta be good. If it's bad, it costs you billions and billions and billions of dollars. Apple's never had an iPhone recall. It's amazing that care and decision making and diligence and fierceness in terms of your supply chain and making hard decisions is very, very different than everything goes into like making great AI. I'm generally prefer companies to do what they're good at. So from my perspective, I'm fine with Apple not doing AI. I want them to keep making great devices. Of the five potential frontier AI winners, so OpenAinthropic, Gemini, SpaceX AI, Grok, and Meta, which of those firms do you think has the most interesting setup? opening and anthropic obviously are the riskiest, but also have the biggest upside. Never discount, number one, the power of belief. They think they're creating God. The most impactful things in history have usually been fueled by religion. The two religious organizations in Silicon Valley are... OpenAID is kind of like mainline. They go to church every Sunday. They're sort of like evangelicals. That's anthropic. They're all in. It is core to their belief. That goes a long way. The fact you need to make a business work for you to survive goes a very long way. Google just needs search to not die too quickly. Meta has the huge advertising business. In a world where Meta was run by anyone other than Mark Zuckerberg, they would not be on the weaning edge. That is one of the purest manifestations of founder Twitter. energy for better or for worse. Their business is so amazing. You see them just easily sort of doubling down on that. Google, there's a bit where they had Google Cloud, they have TPUs, they've been doing research in this. It makes sense why they're pursuing this. Meta being like, actually, we're going to hire a completely new team, we're going to start from scratch, this all again is pretty insane. Kratzheimer Zuckerberg in that regard, again, you could decide whether that's a good idea or not. And then SpaceX AI, data centers in space, the theory is there. Do they have to own their own model, though, to do that? They'd get better margins if they do. Then again, if we actually run out, whether through political opposition or power or whatever it might be, if we run out of data centers on Earth, they can run whatever model they want, as we're seeing with selling their capacity to Anthropic right now. They're all pretty interesting. Probably the case for SpaceX AI is probably the weakest thing. Because the data center and space play is so highly differentiated. If that plays out, I'm not sure to what extent they need to even have their own model. So why are you wasting billions and billions of dollars in the meantime? That's a fair question. From a tactical perspective, I love the cursor acquisition. That makes so much sense for both companies. And so I've been intrigued to see what they do. Meta is probably the most interesting. You've written a lot about this recently. to not be on the frontier. If you're a digital company. The counter to meta is actually Microsoft. Microsoft is not on the frontier. The reason why Microsoft has... $40 million of free cash for the last quarter. Microsoft paid a $10 billion dividend last quarter. There's some money, but their play is, oh, we're going to play all these off each other. We're going to provide middleware. We're going to provide the platform that enterprises will build on us, and we're going to sort of disintermediate the models. I think it's a rational play. It's the IBM play of the 90s. History echoes. Everyone talks about Google. Google, like, following Microsoft. Microsoft follows IBM. And you can see that to an extent. What did IBM do? What's the analogy? Well, so IBM had this dominant. We talked about it in the 70s. And then you fast forward to the 90s. And IBM is this very distressed asset. And the thought was IBM needed to break up into all these different pieces they had. So Lou Gerstner comes in and takes it over. Gerstner's real key insight to IBM is we're pretty mediocre at everything. It's kind of like what I talked about before. And that's the price of monopoly. capacity to be good. because you didn't need to compete anymore. And I think a lot of tech Incoming companies have this problem. It didn't matter what they did, they were gonna rake in money. And if you don't have the pressure, if you don't have the incentive, if you don't have the fear of death, or the fear of God, as we talk about these model companies, then you don't do your best work. And the problem is that once you lose that muscle, it's gone, you're just sort of fat and flabby. [55:00] But what Gerstner realized is actually the worst thing IBM could do would be to break it up into component pieces because all those component pieces are actually not very good. Our biggest asset is that we're big. Say what? No, what does it mean we're big? It's the 90s, this internet thing's coming along, there's all these companies that kind of know they have to figure out the internet and they don't know what to do, they need someone who can come in. understand their business and help them get online. That's basically what IBM did. So they built out, and this is an echo of what's happening now, huge consultant force. And they put all their time into building, basically it was middleware. where they would go in and they'd put this layer between a company's old school mainframe, which all these companies had, And then modern web services, on the other hand, they could have websites and e-commerce sites and all this sorts of things. And it gave IBM a 30-year lease on life. Yes, in theory, you could go get point solutions from all these hot Silicon Valley startups, but you understand that. You don't know how to do that. You know us. We'll come in. We'll create all this middleware, build this big consulting force to help you implement it, and you'll get online. And IBM basically brought all of corporate America online. That's Microsoft's playbook. Microsoft will help you figure out AI. It will help you figure out in a way where you're not giving away the crown jewels to these companies. We're going to build this platform, this harness, this sort of middle layer. We're dependable. We're stable. You know us. We have backwards compatibility to the 80s. You can build on us and then we'll manage all the changing models and what's updating and do all those sorts of things. And does that mean you'll get the absolute best experience? No, middleware saws off the sharp edges. You sort of get a lowest common denominator capacity. But if you value in this, the oldest enterprise sales motion, how did Oracle go to market? Oracle went to market in 1980s, Larry Allison, with another technology taken from IBM, or just IBM didn't want it, relational databases. And they're like, you don't want to be locked into IBM. Relational database you could run anywhere. Come with us. The reason this is a joke is because Oracle locks you in more than anyone, right? But all of enterprise sales is companies whose long-term goal is to lock you in, getting you on board by trying to make you scared of being locked into somebody else. All the cloud companies are like, oh, portability, whatever, you'd be whatever. They're like, oh, just use our service that only runs in our cloud. And now you're locked in. That's a Microsoft's playbook. It's a very rational playbook, and I think it makes sense. That's why they have extra money, 'cause they're not on the frontier. They are building massive data centers, but they're building data centers for inference. They're not building it for training. And their story about investing in time in response to customer demand, is more believable in that regard. They're not having to tell a fungibility story where we're building big data centers for training that will be used for inference down the road, maybe. Go back to this notion that it's reckless to not be in the front. The reason why that's concerning, though, is at the end of the day, why are we using Microsoft products again? Because we did before. Like, to what extent does it actually make sense to have... all these artifacts, all these documents, all these email inboxes. Can't AI just do that? There's a real threat here where to Microsoft software business, the whole systems of record thing is funny because. One reason why systems of records are so powerful is it's so hard to move them to somewhere else because it's a very tedious, repetitive job. Oh, AI is actually surprisingly good at that. I'm not sure how good the systems of record. Microsoft isn't so much systems of record. They do have some of the dynamics business. It's user interface. It's like where you actually interact with the computer. That's the part when you see Codex, Quad, Cowork, or whatever, it is aimed like an arrow to the heart of what Microsoft has. In the long run, all digital companies are, but Microsoft is very much. Their strategy is sound. It's also sound. desperate in a existential way and also in a they might pull it off because they're desperate sort of way. Meta is not threatened immediately. But this is where my bullish view of AI comes in. I think all digital companies are threatened. And Meta is a digital company. They have software. Now, one worry is AI takes up more and more time. Time ultimately is Meta's currency. We saw... I tried the Sora thing, didn't really take off. Social network is actually pretty hard. Also cost a lot of money. It's kind of really interesting. So this came up with the creator payment stuff. So YouTube very famously as paid creators kind of from the beginning. And that's a much bigger drag on the business than people appreciate because YouTube has marginal costs to their content. Now, unlike a Netflix, they don't have to pay that cost upfront. They'll pay it after the fact. So their sharing revenue is a better model than a Netflix model. Netflix has to pay upfront for content and then ideally make more money. YouTube pays along the way. But Facebook or Meta, they pay nothing. Instagram is this unbelievable product that generates all this money for which Facebook pays zero dollars for content. It's unbelievable. It's funny because you could see a world where for YouTube, AI generated content could theoretically be a positive because the inference cost of generating content could be. [60:00] last time what they're sharing with creators. for meta AI generated content to the extent they're the ones generating it is actually a worse margin profile than what they have today, because what they have today is free. So they have attention. There is a bullish world where meta... is actually very well placed because in a world where we interact with the AI all the time, the desire for a human connection becomes greater. And it's sort of like a meta going back to their roots meta. One of their biggest mistakes, actually, Meta was always a social network company. They killed Snapchat or stopped Snapchat's growth by realizing Snapchat is a great product. Let's layer it onto our network. They brought the network to bear to kill Snapchat. The reason why TikTok was a blind spot for them is TikTok is classified as a social network. And it's not a social network at all. TikTok is an entertainment product. It doesn't matter who you follow on TikTok. What you see on TikTok is a function of what you watched. And you're going to get more of the same. It's a user generated content network. And the insight from TikTok was the way to get the best content to limit it to your social network is an artificial constraint. We're going to give you the best content from across the whole network. And the vast majority of content is going to be crap. But this is like the absolute question before. You don't think about margins. You think about absolute numbers, the absolute amount of great content, even if the margin for great content is infinitesimal. If we have a ton of content, the absolute amount of great content is going to be very large. Then meta is like we're a social network. Meta is serving you content from your network of people you know, and TikTok serving you the best content from around the world. That's why they took a huge chunk out of them. Meta had to shift. That's what's happened with Instagram and with Reels is – It's not really a social network. It is a entertainment product that pulls from the entire network. And social networking is like the group chat. It's possible in AI. Actually, social network is important again, because like we actually want humans. We want to have some sort of connection to them. That'll be interesting to see how that plays out. But. The other thing with the models is they're so impactful on advertising. The biggest impact of the models, the biggest monetization right now is probably not anthropic opening. It's the incremental gain that is happening for Google and meta. Most of the stuff is pre LLM. But we're getting to LMS, whether it be generating advertising content, What do we want? We want verifiable domains. How do you verify if a generated image is good for an ad? Does the ad sell or not? they actually can validate their image creation and their text creation. In a way, no one else can. And their validation is the ad marketplace running a gazillion A-B tests on all these different things. See what works. See what doesn't. Most ads are a throwaway. It's fine. The vast majority of ads don't convert. They have this massive advantage, this huge liquid market that is a verification machine where the verifiers are humans deciding whether they click on that ad and make a purchase or not. But they're doing it at global scale that can actually have a feedback loop to make their products better. You're also going to get a world where ad matching is actually still fairly crude. Here's the qualities of the person. Here's the qualities of the ad. And it's like you create an embedding, like a vector calculation and see what numbers match. And then you sort of match an ad to the person. What do LMs do? LMs predict. We're going to move to this world where Meta is going to look at people and say, this person probably wants to see this next. And they're going to go find that thing and show it to them. They only need to increase... a few percentage points for the returns to be billions and billions of dollars. This alone is worth them investing in being on the leading edge in having these amazing models. I think a big problem that it has is they don't tell this story. It's weird, but Mark Zuckerberg has the same problem Sam Altman does. He doesn't love ads. They have the best ad business in the world. They have an ad business that I think is a societal positive. You and I have set up these little content businesses that make great money. Content is you get a ride on social media. I grew up on Twitter, people sharing my links. It was amazing. If you're selling some product that The beauty of the internet is there is a niche out there that wants that product. The question is, how do you find the niche? Facebook advertising. That's what it does. It helps products find the people who didn't even know they wanted that product, but when they get it, they're so happy they got it. That's a huge societal positive. You have new business from a new entrepreneur making a new product. You have customers who are happy they got something that they didn't know they would get otherwise. Those customers, by the way, got lots of free entertainment and they'd have to pay for it along the way. And Meta made a bunch of money for themselves and their shareholders, which is basically everywhere in the world. This is why advertising is great. And Meta's advertising in frustrated that Meta doesn't talk about that. Mark Sanford has never really talked about the societal benefits of advertising except in passing. In 20 years, he's handed it off to other people to take care of. And maybe there's a bit where him not paying attention is why there is a certain grit and grind that goes in building advertising business. People get frustrated or have questions about as far as data and all. [65:00] those sorts of things. And maybe there was a bit where he didn't want to be involved in it and wipe his hands of it. But you saw this when Apple passed ATT, app tracking transparency was one of the worst antitrust violations in the history of technology. Apple unilaterally passed. of obliterating all these business models while they're simultaneously building their own as far as advertising goes and doing all this tracking. Why? Trust us. And meanwhile, they're running these advertisements. I remember that advertisement of people on the bus, like overhearing everyone around them, what they're saying. That was such a dishonest representation of how advertising works on the Internet. You had Tim Cook in Congress talking about companies selling data. Facebook's not selling your data. That's value to them. Why would they sell the data? Meta was not prepared to respond. I think you got this with Sheryl Sandberg back in the day. She, in every call, would talk about advertising, how great it is, and have a bunch of case studies. people who are benefiting from advertising and these new entrepreneurs. And then she left and it's kind of like that hole never got filled. It feels like it's a company that's kind of like embarrassed. We make a lot of money from ads, but we got glasses and we're doing AI. It's like you have ads and ads are awesome. I think if they had communicated that more consistently, they would. would be in a better place generally from a PR perspective. They would have been a better place relative to Apple. And I think they would have an easier time right now convincing Wall Street that let us invest. The other problem is, They've spent... cumulative hundred some billion dollars on Oculus, which I dated all along. And so there's a bit where why should we let you spend money again? The one major player and company that we haven't talked about much is Jensen and NVIDIA. And I'm curious how you would tie this back to the notion of not understanding commodity markets in Silicon Valley, whether or not you think compute ultimately is a commodity. I'm curious whether or not you think intelligence will ultimately be a commodity. It's interesting that intelligence and compute, which seem to be by far the most interesting and important topics in tech, both might be commodities and less differentiated than products. The most interesting thing about the Internet is free distribution. Bandwidth is a commodity. The fact that I can pull out my phone right now and connect to any information source in the world for free, free on a marginal cost basis, is because it's a commodity. It changed the world. Commodities change the world. There is a aspect of differentiated products by definition have lower TAMs. Because there's an elasticity aspect to it. Not everyone can afford to pay for it. People's willingness to pay is going to differ. Your market is going to be constrained. Apple's never going to serve the whole world by selling a device. Whereas... Google can because it's free. That matters. You're paying for a commodity, but to the extent it is available to everyone is the extent it is impactful. The Internet is a commodity. It changed the world. So I don't think it'd be weird that intelligence ends up a commodity and changes the world. Commodities often are not thought of as as good of businesses as these differentiated harm reduction products. Curious for your thoughts on Jensen and NVIDIA specifically. NVIDIA's position is I think definitely unnatural. You look at NVIDIA, they've maintained all their margins. Isn't that amazing? It's 2026 and everyone's coming for them and they're still charging. however much money for a chip, but they're actually not maintaining the margins because This whole question of circular financing is people talk about Lucent and things like that. And, you know, this whole deal and NVIDIA is probably a 25% backstop. But if you actually ascribe a value to that, to NVIDIA's, taking equity in the new clouds or whatever. They guarantee they're going to buy all their compute to 2030. Why do they do that? So that the entity in question can get a lower cost of capital, so they can buy a stock market. et cetera, but implicit in that, why do they get a lower cost of capital? They get a lower cost of capital because NVIDIA assumed risk. This is my point before risk never disappears. It just appears somewhere else. Taking on risk has a price. There is a world where AI takes off, it never stops, and everything is fine. And NVIDIA captured all the upside of their risk. But there's also a world where Say that this is Neocloud they backed up. A ton of compute comes to market. The HyperScouters have plenty of compute. They don't have enough compute. NVIDIA is paying for a compute that no one wants. They just lost a bunch of money. If you think about it, there's an expected value of that investment. That expected value, it's not zero. It's not 100%. It's somewhere in the middle. But that is a diminution of NVIDIA's profitability. If you actually look at their business holistically, what that is is a price cut. Now, the price cut didn't show up in margins. It didn't show up in what they're offering. But a lot of what NVIDIA is doing is how can we maintain our margins effectively? Even if the wide view sort of discounted cash flow, expected value, holistic view of our company, people do discount cash flows. But are you actually considering all these pieces? The reality is, is that moving stuff off the balance sheet by and large works. But they're doing all these deals to maintain what feels somewhat unnatural. we have seen price cuts. They're just manifesting in these very bizarre sort of ways. Now in the long run, I think the challenge is, their ultimate competitors are the hyperscalers, particularly Google and Amazon. So Google and Amazon aren't just building [70:00] in their own chips. but they're also looking to sell those chips externally. Google already made a deal to sell like 20% of their TPUs, Genthropic. On the last earnings call, Andy Jassy practically confirmed that they'll be selling Trinium 3s, or maybe Trinium 4s, or they'll be selling Trinium chips sort of eventually externally, which makes sense. That gives them a long-term buy into these companies. There's a huge amount of R&D that goes into developing chips. They get more leverage on their spend. It all makes sense. And by the way, they're not selling their chips on differentiation. They're selling their chips as commodities. NVIDIA is the one selling differentiation. People aren't going... to Amazon to use Tranium. So they're not cannibalizing the attractiveness of their cloud by selling Trinium outside. So they're NVIDIA's biggest problem. Because what's the number one advantage that, the hyperscalers have. Scale. Lower cost of capital. It's a capital fight. They have a lower cost of capital than the neoclouds do. The neoclouds are... They'll buy Nvidia left, right and center. And by the way, it also makes total sense that why SpaceX, I like Elon's out there, we will always buy Nvidia because they're the best. No, you'll buy Nvidia because they're the most fungible. Nvidia is true. It is the most fungible. Kuda's moat is dramatically diminished. Because the models don't care what they run on. And that's what actually matters, what's built on top of the models. But it still matters. It's still something of a mode. If you want to play the game SpaceX is doing, where we're going to build a lot and rent it out, but reserve the right to pull it back, of course you're going to be on NVIDIA because the easiest way to rent it out is to be on NVIDIA. You saw this very early, by the way, you back to 2024, 2023, NVIDIA starts talking about all these sovereign clouds. They start talking about they try to call these Neotron models. They have this thing in 2024. I remember is the first one where it's like the rock star GTC at San Jose and like the huge Coliseum. Just know what comes out. It was a very boring GTC. The old ones used to be NVIDIA demonstrating like 50 gazillion things as they're throwing stuff at the wall. They knew they had something with GPUs and they're trying to like Find the use case. What's LM showed up? It's like, oh, we have the use case. But they were coming up with all these enterprise offerings. I can't remember what they were called. They're like these modules, basically, that, of course, they were free, but they only ran on NVIDIA. And you could see what they were doing is they were trying to lock people in. Intel is a good example here. AMD cleaned them out in hyperscaler sales because the hyperscalers were put in the effort to get stuff working on AMD versus Intel. There are still small differences, even though they're x86, because they're buying at such scale. The investment to do it is worth it to get a better chip or lower price or whatever it might be. The part of Intel's business that never floundered was selling to government and selling to enterprises. They don't have the resources of a hyperscaler. They're not buying at that scale. They're just going to keep buying what they had before. to sovereign clouds. That's why they talk about selling to enterprises, because they want to get in these markets where they're not going to be balancing this chip versus that chip. The hyperscalers have always been the threat to NVIDIA for that reason. They're actually bigger. So you have this issue where the hyperscalers are the threat. The Librascalers have a better cost of capital than the other companies if he wants to buy them. That's how you get this deal this week. I see this deal as a response. That's why it goes with the Google deal. Google can just issue equity. The shareholders don't love it, but their monetization capacity is much higher than NVIDIA or NVIDIA's customers are. I think what NVIDIA is hoping for, maybe they wouldn't say this in so many words, but if we get to a world where we actually run out of power, that's probably good for NVIDIA. Because in a world where we're totally constrained on power, everyone wants the best. We have to get the best efficiency, the best token efficiency. And I think NVIDIA is still the most token efficient. So that is a good role for them. Probably the, Biggest problem for NVIDIA over the last couple of years is I think the U.S. has actually brought a lot more power online than expected. It surprised me. Whether it be what Elon did sort of behind the meter, which has been replicated, or West Texas and natural gas, but even like restarting nuclear plants. We love how the U.S. responds to these things. That's awesome. It's actually one of the biggest... encouraging signals about the U.S. is I was writing early on, assume this is a bubble. You want there to be a long-term payout. The dot-com, we got fiber in the ground. And by the way, Google's played this game before. Google built its business by buying up dark fiber. They had the killer search engine, but so much of the power of what they do is because they bought up all this dark fiber that was basically free after the dot-com era, our core internet still runs on WorldCom Fiber. That was a lasting benefit. The railroads, BNSF is throwing off money that's going to Google from Northern Pacific and Jay Cook selling bonds to retail investors. You want a bubble that produces something that lasts. And very long, it's like, what's gonna last from AI? The GPUs don't last that long. Data centers, okay, fine. But what is it gonna be? Power, it has to be power. If we're in a world where this all blows up and we have way too much power, that is an amazing world to be. We've always been energy constrained. everything. What would it be like to live in a world of energy abundance? It's hard to even imagine because our minds are so constrained by the fact [75:00] actually always been an energy scarcity. I think we've done an unbelievable job. Power for sure is a constraint. It's going to be a constraint. But I think it has taken longer to become a constraint than anyone expected. And I wouldn't be surprised if that includes Jensen Huang. I think he thought insufficient power was going to be Nvidia's moat sooner than it happened. It turns out that the longer we have enough power, the the more time Amazon has to make Trinium better, the more time Google has to make TPUs competitive from an efficiency standpoint. And if we get in a world where it's a world where those margins seem very hard to sustain. I love hearing your takes on just everything going on. It's the most interesting time I've ever observed in this world that you love so much. So thank you so much for your time. Thank you very much. If you enjoyed this episode, visit Colossus.com. You'll find every episode of this podcast complete with hand edited transcripts. You can also subscribe to print, digital, and private audio publication featuring in-depth profiles of the founders, investors, and companies that we admire most. Learn more at Colossus.com slash subscribe. Outro Music. You know how small advantages compound over time? That's true in investing and just as true in how you run your company. Your spending system is your capital allocation strategy. Ramp makes it smarter by default. Better data, better decisions, better economics over time. See how at ramp.com slash invest. As your business grows, Vanta scales with you, automating compliance and giving you a single source of truth for security and risk. Learn more at vanta.com slash invest. not in months. Visit workos.com to skip the unglamorous infrastructure work and focus on your product. Ridgeline is redefining asset management technology as a true partner, not just a software vendor. They've helped firms 5x in scale, enabling faster growth, smarter operations, and a competitive edge. Visit ridgelineapps.com to see what they can unlock for your firm. Every investment firm is unique and generic AI doesn't understand your process. Rogo does. It's an AI platform built specifically for Wall Street, connected to your real outputs. Check them out at rogo.ai slash invest.