章节 1：NVIDIA 的偏执与收购 Groq 的战略意图

📝 本节摘要：

本节作为访谈开篇，主持人 Matt Turck 介绍了嘉宾 Dylan Patel（SemiAnalysis）及其在半导体硬件分析领域的权威地位。核心讨论围绕 NVIDIA 为何在看似稳固的霸主地位下依然保持极度的“偏执”——Jensen Huang（黄仁勋）深知如果仅依赖通用 GPU，可能会在成本和性能上被针对特定工作负载（如极速推理）的竞争对手击败。Dylan 详细分析了 AI 模型架构的不确定性（如自回归、思维链等），指出 NVIDIA 即使拥有强大的通用算力，也必须通过收购（文中提及收购 Groq）和推出专用芯片（如针对 KV Cache 或 Decode 优化的 CPX）来覆盖所有潜在的技术路径，以维持其高昂的利润率和市场统治力。

[原文] [Speaker A]: this is the biggest change in human history maybe ever what's about to happen with AI this is the biggest revolution bigger than industrial revolution jensen is very paranoid about losing if he just kept making his mainline chip people crush him on cost and performance acquiring Grock is how you get those resources to make more solutions for different parts of the market to stay king at the end of the day this is an economic war if the US and the West win in AI China will not rise to be the global hedgeimony but without AI China definitely will rise they're just going to outrun America

[译文] [Speaker A]: 这可能是人类历史上最大的变革，AI 即将发生的一切是比工业革命更巨大的革命。Jensen（黄仁勋）对于失败非常偏执（paranoid）。如果他只坚持制造主流芯片，人们会在成本和性能上击垮他。收购 Groq（Grock）就是获取这些资源的方式，以此为市场的不同部分制造更多解决方案，从而保持王座。归根结底，这是一场经济战。如果美国和西方在 AI 领域获胜，中国将不会崛起成为全球霸主（global hegemony）；但如果没有 AI，中国绝对会崛起，他们会直接超越美国。

[原文] [Speaker A]: hi I'm Matt Turk welcome back to the Matt podcast today I'm joined by the one person Wall Street and Silicon Valley turn to when they need to cut through the hardware hype Dylan Patel of Semi analysis we dove into many of the most important topics of today nvidia's massive move to acquire Grock the truth about the capex bubble whether the US power grid can actually handle the AI boom and the geopolitical chess match playing out between the US and China but I have to warn you this conversation went off the rails in the best possible way and we ended up going into all sorts of fun tangents like the strange phenomenon of Chinese romance dramas set inside semiconductor factories and what's really like when three AI famous roommates live together in SF please enjoy this fantastic conversation with Dylan hey Dylan welcome

[译文] [Speaker A]: 嗨，我是 Matt Turk，欢迎回到 Matt 播客。今天加入我的是华尔街和硅谷在需要看穿硬件炒作时都会求助的那个人——SemiAnalysis 的 Dylan Patel。我们深入探讨了当今许多最重要的话题：NVIDIA 收购 Groq（Grock）的重磅举动、资本支出（Capex）泡沫的真相、美国电网是否真能承受 AI 爆发，以及正在美中之间上演的地缘政治棋局。但我必须警告你，这场对话在最好的意义上“脱轨”了，我们最后聊到了各种有趣的离题话题，比如以半导体工厂为背景的中国浪漫剧这一奇怪现象，以及三位 AI 界名人室友在旧金山同住的真实生活。请享受这场与 Dylan 的精彩对话。嘿 Dylan，欢迎。

[原文] [Speaker B]: hello how are you

[译文] [Speaker B]: 你好，你好吗？

[原文] [Speaker A]: i'm great i'd love to start with Grock and Nvidia since it's still fresh so not so long ago Nvidia was saying that uh one GPU could do it all and now they're doing this acquisition non-exclusive deal with Grock what does that mean from your perspective

[译文] [Speaker A]: 我很好。既然这事还很新鲜，我想从 Groq（Grock）和 NVIDIA 开始聊。不久前 NVIDIA 还说一个 GPU 就能搞定一切，现在他们却在做这个收购以及与 Groq 的非独家交易。从你的角度看这意味着什么？

[原文] [Speaker B]: it's very clear we're not sure where AI models are headed in terms of you know over the next few years what happens to the architecture but you know the thing that I think everyone is sort of like agreed on is models are pretty auto reggressive right next token generation is like the thing but beyond that right attention mechanisms changed the how how it works everything changes right could could change and so what's interesting is the reason Nvidia one is because they just took like the widest surface area bet and then people kept developing models on that and that kind of shape worked

[译文] [Speaker B]: 很明显，我们不确定 AI 模型在未来几年会朝哪个方向发展，架构会发生什么变化。但我认为大家某种程度上达成共识的一点是，模型在很大程度上是自回归的（auto-regressive），对吧？“下一个 token 生成”（next token generation）是关键所在。但除此之外，注意力机制（attention mechanisms）改变了它的工作原理，一切都在变，对吧？可能会变。所以有趣的是，NVIDIA 赢的原因是他们押注了最宽的表面积（widest surface area），然后人们继续在这个基础上开发模型，这种形态奏效了。

[原文] [Speaker B]: but now the workload is so large that there is room for specialization that will give you 10x increases in certain domains right in a general purpose workload grock doesn't work right you know it can't train it can't you know it can't inference really really large models um cost efficiently right you can't serve many many many users but what it can do is it can go bl screamingly fast right same with the cerebrous open AI deal but that's like one workload right uh very decode focused right gener doing auto reggressive tokens in a in a single stream super fast

[译文] [Speaker B]: 但现在工作负载太大了，以至于有了专业化（specialization）的空间，这可以在某些领域给你带来 10 倍的提升，对吧？在通用工作负载中，Groq 是行不通的，对吧？它不能训练，它不能——你知道——成本高效地推理非常非常大的模型，对吧？你无法服务海量的用户。但它能做的是它可以快得惊人（screamingly fast），对吧？Cerebras 和 OpenAI 的交易也是如此。但这就像是一种工作负载，对吧？非常专注于解码（decode focused），对吧？在一个单一流中极快地生成自回归 token。

[原文] [Speaker B]: another direction AI models could head right we don't know are models going to think in one token stream or is it actually they're constantly context switching right and they're going from they have this humongous humongous context and they're generating in multiple parallel streams right and so Google and openi have both released mechanisms of this with their pro models where the model actually doesn't just have one single chain of thought for reasoning it has multiple right and then I don't exactly like you know and and and how they choose which one and what the final answer to you delivers is is an area of research um but there there is room for that kind of chip right something that works on very parallel a lot lot of streams of chain of thought and maybe the latency requirements are not as crazy right

[译文] [Speaker B]: AI 模型可能朝向的另一个方向——我们不知道模型是会在一个 token 流中思考，还是实际上它们在不断地进行上下文切换（context switching），对吧？它们可能从拥有巨大的上下文，转变为在多个并行流中生成，对吧？Google 和 OpenAI 都在其 Pro 模型中发布了这种机制，模型实际上不仅仅只有一条单一的思维链（chain of thought）进行推理，它有多条，对吧？然后我不太——你知道——关于它们如何选择哪一条以及最终给你提供什么答案，这是一个研究领域。但是，确实有这种芯片的空间，对吧？那种在非常并行、许多思维链流上工作的芯片，也许延迟要求没那么疯狂，对吧？

[原文] [Speaker B]: maybe you don't want to go blindingly fast right maybe you're okay with it being you know because I can spin up 100 parallel you know streams of thought or agents or whatever you want to call them maybe I I care a lot about cost there and because it's 100 in parallel instead of one going super super fast it's not as deep right the tree search or the depth of the inference is not as deep but it is much wider you know there's other parts of inference hey process do creating the KV cache so Nvidia has a chip for that right that's the CPX so they they've made the CPX they bought Grock for decode and then they still have their general purpose GPU

[译文] [Speaker B]: 也许你不需要快得令人眩晕，对吧？也许你可以接受它——你知道，因为我可以启动 100 个并行的思维流或智能体（agents），或者不管你想叫它们什么。也许我在那里更在乎成本。而且因为它是 100 个并行而不是一个超级超级快的，它没那么深，对吧？树搜索（tree search）或推理的深度没那么深，但它要宽得多。你知道推理还有其他部分，比如处理创建 KV Cache（键值缓存）。所以 NVIDIA 有一款针对这个的芯片，对吧？那就是 CPX。所以他们制造了 CPX，他们买了 Groq 用于解码（decode），然后他们仍然拥有他们的通用 GPU。

[原文] [Speaker B]: so they've they're kind of trying to cover their bases because unlike the first wave of AI chip companies where they sort of just made chips and then tried to figure out where it would work right they had a thesis Grock and Cerebrus both as well as Samanova right which was put a lot of memory on the chip and not necessarily in the case of Cerebrus and Grock no memory off chip and in the case of Samanova less memory offchip or slower memory offchip with higher capacity you know they they sort of all made similar bets in that direction and it didn't work for a while until it kind of did right um because there's a workload that now necessitates it

[译文] [Speaker B]: 所以他们正试图覆盖他们的基础（cover their bases），因为不像第一波 AI 芯片公司，那时他们有点像是制造了芯片然后试图弄清楚它能在哪里工作，对吧？他们有一个论点——Groq 和 Cerebras 都是，还有 Sambanova，对吧？就是把大量内存放在芯片上，而不一定——在 Cerebras 和 Groq 的案例中是芯片外无内存，在 Sambanova 的案例中是芯片外内存较少或更慢但容量更高。你知道，他们都在那个方向上做了类似的押注。这在很长一段时间内都没奏效，直到它某种程度上奏效了，对吧？因为现在有一种工作负载需要它。

[原文] [Speaker B]: nvidia recognizes they're they're the leader they're at the tent pole hey in one respect they can just run faster than everyone but it's kind of hard to be 2x better than Google or or OpenAI or whoever else's internal chip right to justify their you know 75% plus margins right and then they have to be 2x to 4x better to justify 4x better to justify their margins because that's what they're charging above COGS you know the question is what what architecture will deliver that well yes keep the programmability of their GPUs is great for training and for a lot of workloads

[译文] [Speaker B]: NVIDIA 意识到他们是领导者，他们处于顶梁柱的位置。嘿，在一个方面他们确实可以比任何人都跑得快，但要比 Google 或 OpenAI 或其他任何人的内部芯片好 2 倍是有点难的，对吧？为了证明他们——你知道——75% 以上的利润率是合理的，对吧？然后他们必须好 2 倍到 4 倍，好 4 倍才能证明他们的利润率是合理的，因为那是他们在销货成本（COGS）之上收取的费用。你知道问题是，什么架构能实现这一点？是的，保留 GPU 的可编程性对于训练和许多工作负载来说很棒。

[原文] [Speaker B]: but you know guess what i think I think a lot of people will just be downloading an open source model downloading an inference framework and pressing go right a little bit more complicated than that but that's that's going to be the consumption method for a lot of enterprises a lot of uh startups a lot of tech companies is they're just going to do that or they're going to rent the G GPUs or or rent the chips and then download an open source framework and model and go right

[译文] [Speaker B]: 但你知道吗，猜猜看，我认为很多人只会下载一个开源模型，下载一个推理框架，然后按下“开始”，对吧？虽然比这稍微复杂一点，但这将是许多企业、许多初创公司、许多科技公司的消费方式。他们只会那样做，或者他们会租用 GPU，或者租用芯片，然后下载开源框架和模型并开始运行，对吧？

[原文] [Speaker B]: and Nvidia recognizes this and hey there is room for products that aren't general purpose right the general purpose GPU will still probably be the main line for training and for a lot of inference and for costefficient inference but maybe blindingly fast or workloads that have a ton of prefill i.e creating the the KV cache maybe that those workloads could be different chips right and the CPX chip they announced right they say it's for the context processing creating the KV cache it's also really useful for video models because video models don't care about memory bandwidth and so you know why pay for the expensive memory that the general purpose chip has or why do what Grock is doing which is tying hundreds or thousands of chips together and not having memory but keeping the entire model on chip

[译文] [Speaker B]: NVIDIA 意识到了这一点，嘿，非通用产品的空间是存在的，对吧？通用 GPU 可能仍将是训练和许多推理以及成本高效推理的主线，但对于极速（blindingly fast）需求，或者有大量预填充（prefill）即创建 KV Cache 的工作负载，也许那些工作负载可以是不同的芯片，对吧？他们发布的 CPX 芯片，对吧？他们说它是用于上下文处理、创建 KV Cache 的。它对视频模型也非常有用，因为视频模型不在乎内存带宽。所以你知道，为什么要为通用芯片拥有的昂贵内存付费？或者为什么要像 Groq 那样做，即把数百或数千个芯片绑在一起，没有（片外）内存，而是把整个模型保持在芯片上？

[原文] [Speaker B]: the trade-off for that of course is you need thousands of chips and you have less compute per chip and so like Nvidia's trying to capture the whole surface area because again you don't know where models are headed

[译文] [Speaker B]: 这样做的代价当然是你需要成千上万个芯片，而且每个芯片的计算量较少。所以就像 NVIDIA 试图捕捉整个表面积（market surface area），因为再说一次，你不知道模型会朝哪里发展。

章节 2：监管博弈与群雄逐鹿：反垄断阴影下的芯片战国时代

📝 本节摘要：

接续上文，对话转向了 NVIDIA 收购 Groq 背后的监管隐忧。Dylan 指出，尽管从反垄断角度看并不利于市场竞争，但对于初创公司而言，漫长的监管审批期（Limbo）往往是致命的。随后，Dylan 深入剖析了 NVIDIA 的核心驱动力——Jensen Huang（黄仁勋）继承了安迪·格鲁夫“唯偏执狂生存”的危机感。为了避免被针对特定领域的低成本“单点解决方案”（Point Solutions）击穿高利润防线，NVIDIA 必须通过收购来获取稀缺的芯片人才。最后，Dylan 盘点了当前的围剿势力：从 Etched 等新锐初创公司，到 Google、AMD、Amazon 等科技巨头，NVIDIA 正面临前所未有的全方位竞争。

[原文] [Speaker A]: and it's hard to say where the research is headed and do you think it's a good thing for the market yet another one of those deals that's structured as a as a license but really an acquisition

[译文] [Speaker A]: 很难说研究方向会去向何方。你认为这针对市场是一件好事吗？又是一个那种结构上设计为“许可交易”（license）但实际上是“收购”的交易？

[原文] [Speaker B]: i certainly think it's not good from an anti-competitive sense right i don't think people should just be able to buy companies without like any antitrust like process at all now in the case of like a large company buying a startup I'm completely fine with it the flip side is like hey we know the deal is happening right uh this happened for a company I was an adviser for Nvidia acquired in fabrica just maybe a few months before they did Grock and similar style of deal right if someone wanted to strike it down that's the biggest limbo right we've seen this happen in venture and you probably know more stories of this but like a company trying to get acquired they get stuck in limbo for like a year and then it falls apart stories

[译文] [Speaker B]: 我当然认为从反竞争（anti-competitive）的角度来看这并不好，对吧？我不认为人们应该能够在完全没有任何反垄断程序的情况下直接买下公司。现在的具体情况是，像大公司收购初创公司这种，我完全没问题。但反过来看，嘿，我们知道交易正在发生，对吧？这发生在我担任顾问的一家公司身上——NVIDIA 在收购 Groq 之前的几个月刚刚收购了 InfaBrica，也是类似风格的交易，对吧？如果有人（监管机构）想要否决它，那就会造成最大的“停滞期”（limbo），对吧？我们在风投圈见过这种情况，你可能知道更多这类故事：一家试图被收购的公司，陷入这种停滞状态长达一年，然后交易崩了。

[原文] [Speaker A]: yeah it falls apart the deal did because some regulatory BS and now the company was and the founders were focused on getting the deal done instead of like making the product better for a year and now they're like behind or you know they they they weren't focused on growth as much right you know you only have so much time as a founder so in that sense I like the license deals right

[译文] [Speaker A]: 是的，交易崩了，因为一些监管方面的扯皮（BS）。而现在这家公司和创始人把一年的时间都花在了搞定交易上，而不是把产品做得更好。现在他们落后了，或者你知道，他们没有那么专注于增长了，对吧？你知道作为创始人你的时间是有限的。所以从这个意义上说，我喜欢这种许可交易，对吧？

[原文] [Speaker A]: so now is uh Nvidia also dominating the the inference market is there any world where Nvidia is no longer the king or they seem to be getting stronger

[译文] [Speaker A]: 那么现在 NVIDIA 也要主宰推理市场了吗？是否存在这样一个世界，NVIDIA 不再是王，还是说他们看起来变得更强了？

[原文] [Speaker B]: I think the thing about Nvidia is they take the Andy Grove mentality like more serious than anyone else right like okay fine Google like implemented OKRs because Intel did it but that's like you know management stuff right only the paranoid survive right this is like core to the Bay Area um core to Nvidia um Jensen is very paranoid about losing right these specializations if he just kept making his mainline chip would mean people could you know point point solutions for specific parts of the market would crush him on cost and performance then he can't justify his margin that's a threat to Nvidia's business model as a whole

[译文] [Speaker B]: 我认为关于 NVIDIA 的一点是，他们比任何人都更认真地贯彻安迪·格鲁夫（Andy Grove）的心态，对吧？好吧，Google 实施了 OKR 是因为 Intel 做了，但这只是管理层面的东西，对吧？“唯偏执狂生存”（Only the paranoid survive），对吧？这是湾区的核心，也是 NVIDIA 的核心。Jensen 对于失败非常偏执（paranoid），对吧？如果他只坚持制造他的主流芯片，那些专门化技术——人们可以用针对市场特定部分的“单点解决方案”（point solutions）在成本和性能上击垮他。那样他就无法证明他的利润率是合理的。这是对 NVIDIA 整体商业模式的威胁。

[原文] [Speaker B]: especially if the best model only changes every 3 months or the model you want to roll out okay well then you're going to have three months to figure out how to make a model work on one chip architecture for that point solution and you know it's fine software software advantage of Nvidia is not that important then Jensen's super paranoid about losing and frankly it's really hard to hire enough talented chip people when you look across the market there is only a few companies who have successfully created a chip architecture software to run the models accurately run the run the models accurately right like cuz you can look at random APIs of say an Alibaba Quen model and different people are doing all sorts of tricks like quantizing it but also many other tricks which then end up like making the model quality lower

[译文] [Speaker B]: 尤其是如果最好的模型每 3 个月才变一次，或者你想推出的模型——好吧，那你就有 3 个月的时间去弄清楚如何让一个模型在那个单点解决方案的芯片架构上运行。你知道这没问题，那时候 NVIDIA 的软件优势就没那么重要了。Jensen 超级偏执于失败。坦率地说，雇佣足够多的有天赋的芯片人才真的很难。当你放眼市场，只有少数几家公司成功创造了芯片架构和软件来准确地运行模型——准确地运行模型，对吧？因为你可以看看比如阿里巴巴 Qwen 模型的随机 API，不同的人在做各种各样的把戏，比如量化（quantizing），还有许多其他的把戏，结果导致模型质量下降。

[原文] [Speaker B]: you know building a rack scale solution networking thousands of chips together and then deploying an API and Grock did the whole thing with frankly not that many people so now it's like okay well I'm Nvidia I want to make four different chip architectures and actually four different point solutions maybe the general purpose and then one here one here one here and in addition my general purpose thing is actually not just like a GPU chip it's like GPU chips CPU chips networking chips NV switch nicks like you know there's many many chips and each of those chips has many chiplets you don't have enough engineering resources right and so like acquiring gro is like how you get those resources to make more solutions for different parts of the market

[译文] [Speaker B]: 你知道，构建一个机架级（rack scale）解决方案，将成千上万个芯片联网在一起，然后部署一个 API，Groq 坦率地说用不了多少人就搞定了整件事。所以现在就像是，好吧，我是 NVIDIA，我想制造四种不同的芯片架构，实际上是四种不同的单点解决方案——也许一个是通用的，然后一个针对这里，一个针对那里，一个针对那里。而且，我的通用产品实际上不仅仅是一个 GPU 芯片，它是 GPU 芯片、CPU 芯片、网络芯片、NV Switch、网卡（NICs）等等，你知道有非常非常多的芯片，而每一个芯片又有许多小芯片（chiplets）。你没有足够的工程资源，对吧？所以收购 Groq 就像是你获取这些资源的方式，以便为市场的不同部分制造更多解决方案。

[原文] [Speaker B]: as far as like are they threatened like I think I think like obviously There's some cool startups out there right that are raising a lot right currently or have raised such as Etched Maddx uh positron these new age of AI companies there's also the prior age of like Cerebrris is is out there still right you know Tenstor etc and there's so there's a lot of AI chip companies on the startup side but then there's also you know Google's TPU AMD GPUs uh Amazon Tranium uh who are all really credible competitors and then you know Meta's MTIA is somewhat credible and then you know Microsoft Somaya is not credible but like you know maybe it will be one day right so you sort of have like a lot of competition they've got to hold the gates back

[译文] [Speaker B]: 至于他们是否受到威胁，我认为显然外面有一些很酷的初创公司，对吧？它们正在筹集大量资金或者已经筹集了资金，比如 Etched、Maddx、Positron 这些“新时代”的 AI 公司。还有“前一个时代”的公司，比如 Cerebras 仍然在场，对吧？你知道还有 Tenstor 等等。所以在初创公司方面有很多 AI 芯片公司。但然后你也知道还有 Google 的 TPU、AMD 的 GPU、Amazon 的 Trainium，这些都是真正可信的竞争对手。然后你知道 Meta 的 MTIA 在某种程度上是可信的，而 Microsoft 的 Maia 目前还不可信，但你知道也许有一天会变的，对吧？所以你确实面临很多竞争，他们必须守住大门。

[原文] [Speaker B]: and so I think is there a risk to them being I mean like there's there's risk from all of those companies that I mentioned and and you know effectively California/ Seattle right only two two places there's there's also chips from other parts of the world right obviously China has a number of different AI chip companies that are doing cool things anyone would have told you Grock was you know their business revenue their revenue was not like stellar right in fact they missed revenue last year significantly and yet they got bought right because the value of the IP was there and the value of the team anyone else would have been like well why the heck would I buy this right uh makes no sense there's definitely a credible threat

[译文] [Speaker B]: 所以我认为他们是否有风险？我是说，我提到的所有这些公司都构成了风险。而且你知道，实际上主要就是加州和西雅图这两个地方。世界上其他地方也有芯片，对吧？显然中国有许多不同的 AI 芯片公司正在做很酷的事情。任何人都会告诉你 Groq——你知道他们的业务收入——他们的收入并不亮眼，对吧？事实上他们去年的收入严重未达标。然而他们被收购了，对吧？因为 IP（知识产权）的价值在那里，团队的价值在那里。其他人可能会说：“好吧，我到底为什么要买这个？”对吧？这没道理。但（对 NVIDIA 来说）绝对存在可信的威胁。

章节 3：CUDA 护城河的演变与开源推理引擎的崛起

📝 本节摘要：

本节深入探讨了 NVIDIA 传统护城河（CUDA）在开源推理引擎（如 vLLM、SGLang）崛起背景下的演变。Dylan 指出，未来的 AI 芯片消费模式并非编写底层代码，而是直接调用开源引擎，这削弱了 CUDA 语言本身的壁垒。然而，NVIDIA 正在构建“新护城河”——即针对复杂工作负载（如编程 Agent 的频繁上下文切换）的系统级优化，例如 KV Cache 管理器。此外，本节还分析了 AMD 在硬件性能上对 NVIDIA 的追赶态势，以及其在软件生态上的差距。

[原文] [Speaker A]: yeah and do you think uh CUDA is going to remain that mode I guess a combination of CUDA and whatever came out of the Melanox acquisition like do do those persist as long-lasting advantages

[译文] [Speaker A]: 是的，那你认为 CUDA 会继续保持这种护城河（moat）吗？我猜是 CUDA 和收购 Mellanox 后产生的技术的结合，这些会作为长久的优势持续存在吗？

[原文] [Speaker B]: I think they do I think networking is super important I think uh the CUDA software mode is very important but it's also like changing rapidly right It's an incredible amount of the software that Nvidia GPUs run on is not from Nvidia it's it's the developer ecosystem that's open sourcing it when you look at for example VLM and SGLANG right these support AMD GPUs almost as first class citizens now and VLM is getting significant support for TPUs for tranium and there will be other chips coming out from startups that also support VLMs SGLang

[译文] [Speaker B]: 我认为会，我认为网络非常重要。我认为 CUDA 软件护城河非常重要，但它也正在迅速变化，对吧？NVIDIA GPU 上运行的软件有惊人的数量并非来自 NVIDIA，而是来自正在将其开源的开发者生态系统。当你看到例如 vLLM 和 SGLang 时，对吧？这些现在几乎将 AMD GPU 支持为一等公民，而且 vLLM 正获得对 TPU 和 Trainium 的重要支持，未来还会有初创公司推出的其他芯片也将支持 vLLM 和 SGLang。

[原文] [Speaker B]: now like how difficult is it you know the the the reason why CUDA is so important is like okay I can do whatever I need to do right programming a GPU i think most AI chips will not be consumed by people programming anything for it they will download an open source inference engine and they will download an open source model and then they will put it on the and it's really simple to download VLM and like make it work like it's not that hard to set up uh you know a server

[译文] [Speaker B]: 现在有多难呢？你知道，CUDA 如此重要的原因是，好吧，我可以做任何我需要做的事，比如给 GPU 编程。但我认为大多数 AI 芯片的消费方式并不是人们去为它编写任何程序，他们会下载一个开源推理引擎，下载一个开源模型，然后把它放上去。下载 vLLM 并让它运行起来真的很简单，搭建一个服务器并不难。

[原文] [Speaker B]: and Nvidia's putting out a lot of open source software like Triton inference server and and uh Dynamo and all these things to to make it easy because that is the consumption model ultimately for the majority of AI right is and it might be like oh it's my own inference engine but most servers will not run code besides the inference engine and the model it's like not like people are actually like researchers are like writing code for GPUs to see ideas if they'll work and train models and all these things or just mess around with them to figure out you know infra performance or whatever it is but most of it won't be there

[译文] [Speaker B]: NVIDIA 正在发布大量的开源软件，比如 Triton 推理服务器、Dynamo 以及所有这些东西，以使其变得简单。因为这最终是大多数 AI 的消费模式，对吧？也许你会说“哦，这是我自己的推理引擎”，但大多数服务器除了推理引擎和模型之外不会运行其他代码。这不像研究人员那样，为了验证想法是否可行、训练模型或搞清楚基础设施性能而真的去为 GPU 写代码，大部分情况并非如此。

[原文] [Speaker B]: and so CUDA as a mode CUDA language is like you know like it's like fine right like you know no one actually writes CUDA right like most people write PyTorch and then like torch compile and then they just run it on the GPU they don't write CUDA but a lot of this CUDA mode is like how does PyTorch translate into high performance GPUs and that surface area from when people were just writing like hardcore when people are hardcore writing CUDA kernels to like hey they're writing PyTorch and then it's compiling down to GPUs versus oh I'm just downloading VLM is it is a it is a curve of like not a ton of people that can do CUDA kernels a whole lot more people can do PyTorch right random you know PhDs and random people it's very simple right a crapload of people can do VLM download it run it on a server well if it now supports other chips what is the CUDA mode's recognized this and they've been building software that is not necessarily the CUDA remote and I I can give some examples

[译文] [Speaker B]: 所以 CUDA 作为一种护城河，或者说 CUDA 语言，就像——你知道——它还好，对吧？因为实际上没人真的写 CUDA，对吧？大多数人写 PyTorch，然后用 Torch Compile，接着就在 GPU 上运行了。他们不写 CUDA。但这其中很多所谓的“CUDA 护城河”是指 PyTorch 如何转化为 GPU 上的高性能表现。这个表面积（surface area）——从人们硬核地编写 CUDA 内核（kernels），到人们编写 PyTorch 然后编译到 GPU，再到“哦，我只是下载 vLLM”——这是一条曲线。能写 CUDA 内核的人不多；能写 PyTorch 的人要多得多，比如随便哪个博士或者普通人，这很简单；而能做 vLLM 下载并在服务器上运行的人更是一抓一大把。如果 vLLM 现在支持其他芯片了，那 CUDA 护城河是什么？NVIDIA 意识到了这一点，他们一直在构建不一定是传统 CUDA 护城河的软件，我可以举一些例子。

[原文] [Speaker B]: All right so the name of the game is fast tokens and lowest cost tokens right and lowest cost tokens happens by your chip being fast but there's also tricks right one example right like I mentioned with you know the CPX versus Grock right is processing your prefill contacts right super cheap CPX right if I'm if I'm care a lot about speed then Grock these are optimizations on the hardware side there's optimizations on the software side as well

[译文] [Speaker B]: 好的，所以游戏的核心是快速生成 token 和最低成本的 token，对吧？最低成本的 token 取决于你的芯片速度，但也有技巧，对吧？一个例子，就像我提到的 CPX 对比 Groq，对吧？处理你的预填充（prefill）上下文，CPX 超级便宜，对吧？如果我非常在乎速度，那就选 Groq。这些是硬件方面的优化，但在软件方面也有优化。

[原文] [Speaker B]: right and so one example is when I'm doing for example if I look at a cloud code or a cursor type application right the workload is like it takes your repo takes the relevant parts of your repo puts it in the context of the LLM it prompts it generates right and if it's an agent mode it it it circulates the context a couple times it'll collapse put things off to the side access different contexts

[译文] [Speaker B]: 一个例子是，比如我看 Claude Code 或者 Cursor 这类应用，对吧？它的工作负载是：获取你的代码仓库（repo），提取相关部分，放入大语言模型（LLM）的上下文中，提示它，然后生成，对吧？如果是代理模式（agent mode），它会循环处理上下文几次，它会折叠内容，把东西放一边，访问不同的上下文。

[原文] [Speaker B]: but what's you know especially when you think about an agent for software and you can see this in codeex you know Codex Codex actually not as good as cloud code but it can do work on time horizons of like 9 10 hours um and do like a big refactor better than cloud code can even though most of the times cloud code is better and and what's interesting about Codeex does is it'll like take your repo it'll identify parts if you're asking it to refactor it identify parts write stuff you know make like these notes for itself everywhere collapse the context switch from this part of the repo to that part of the repo to this part of the repo

[译文] [Speaker B]: 但是你知道，尤其是当你考虑一个用于软件的代理时——你可以在 Codex 中看到这一点。你知道 Codex 实际上不如 Claude Code 好，但它可以在 9 到 10 小时的时间跨度内工作，并且能比 Claude Code 更好地完成大型重构，即使大多数时候 Claude Code 更好。Codex 有趣的地方在于，它会获取你的代码库，识别部分内容——如果你要求它重构——识别部分，写东西，你知道，到处给自己做笔记，折叠上下文，从代码库的这一部分切换到那一部分，再切换到这一部分。

[原文] [Speaker B]: but when you think about it it's like oh if this thing is just generating tokens all the time plus it's switching what my context is constantly that's really expensive right if you think about like what's the cost of inference um I want to say it's like it's it's $10 per million tokens of output and or and $3 for decode or 10 for decode and three for prefill uh and so if you think about oh it just worked for nine hours on one task one refactor huge value but if it changed context a ton of times and your context is like 30k usually or 50k or you know heading to hundreds of thousands you know how long your how big your repository is and how much context switch now you're spending all this money on on prefill right not the decode tokens but actually why am I like regenerating the KV cache I can actually just like store the KV cache elsewhere and then when I need it again I can pull it and and plop it into CPU memory or into GPU memory

[译文] [Speaker B]: 但当你仔细想一想，如果这东西一直在生成 token，加上它不断地切换我的上下文，那真的很贵，对吧？如果你考虑推理的成本——我想说是每百万输出 token 10 美元，解码（decode）是 3 美元或 10 美元，预填充（prefill）是 3 美元。所以如果你想，哦，它在一个任务上工作了 9 个小时，一个重构，价值巨大。但如果它改变了无数次上下文，而你的上下文通常是 30k 或 50k，或者你知道，正朝着几十万发展——这取决于你的代码库有多大以及有多少上下文切换——现在你把所有的钱都花在预填充上了，对吧？不是解码 token。但我为什么要重新生成 KV Cache（键值缓存）呢？我实际上可以把 KV Cache 存储在别处，当我再次需要它时，我可以把它拉出来，直接扔进 CPU 内存或 GPU 内存里。

[原文] [Speaker B]: and so Nvidia's got this like KV cache manager and they've been working really hard on like making it so they can interface SSDs and stick the KV cache on there and pull it out whenever they want so for this kind of workload and then if you do this and you look at like coding as an application and you like look at these coding companies and how much they're paying for prefill versus decode actually majority of their cost is pre-fill tokens not decode tokens because their context is just so large and it's switching all the time even in agent modes you know if you can now not have to do the pre-fill your costs go down dramatically but that's a very complicated thing to do from a software perspective

[译文] [Speaker B]: 所以 NVIDIA 有了这个 KV Cache 管理器，他们一直非常努力地工作，使其能够连接 SSD，把 KV Cache 放在那里，并在需要时随时取出来。所以对于这种工作负载，如果你这样做，看看像编程这样的应用，看看这些编程公司在预填充与解码上花了多少钱，实际上他们的大部分成本是预填充 token，而不是解码 token，因为他们的上下文太大了，而且一直在切换。即使在代理模式下，你知道，如果你现在不需要做预填充，你的成本会大幅下降。但从软件角度来看，这是一件非常复杂的事情。

[原文] [Speaker B]: you know companies like Enthropic Google OpenAI have already done it but what about the wide world right and so Nvidia is trying to make the open source software for this and that's like CUDA mode but it's like actually no none of this is CUDA right like it's like memory management and like you know storage management and when do you call what and how do you transfer it and how do you like spread the KV cache across a bunch of different storage nodes and what happens when you read it and the network congestion just like all these things yeah it's like Nvidia's wheelhouse but it's not CUDA and I think like the easy way to say it is it is the CUDA mode right and so things like this KV cache manager and many other things they're trying to do to reduce the cost of inference like is how they build the new CUDA mode

[译文] [Speaker B]: 你知道像 Anthropic、Google、OpenAI 这样的公司已经做到了这一点，但广大的世界呢？所以 NVIDIA 正试图为此制作开源软件。这就像是 CUDA 护城河，但实际上不，这都不是 CUDA，对吧？这像是内存管理、存储管理，以及什么时候调用什么、如何传输它、如何将 KV Cache 分布在一堆不同的存储节点上、当你读取它时会发生什么、网络拥塞等等所有这些事情。是的，这像是 NVIDIA 的专长（wheelhouse），但这不仅仅是 CUDA。我认为简单的说法是，这就是 CUDA 护城河。所以像这个 KV Cache 管理器以及他们试图做的许多其他降低推理成本的事情，就是他们构建新 CUDA 护城河的方式。

[原文] [Speaker B]: because again today it's it's you know it is quite I mean AMD is like not fully there yet and TPU is being added right now and tranium is being added soon as well to VLM but all of them will have a very good UX for download model run model on VLM by the middle of the year I think right certainly AMD is already there by the end of this quarter we have something that like tests this right it's called inferencemaxa it's open source all the code is and the results are uh but we run across I think $60 million of GPUs which are donated to us by companies like Nvidia AMD openai Microsoft Amazon on Crusoe Core Weave Together AI

[译文] [Speaker B]: 因为再次强调，今天——我的意思是 AMD 还没完全到位，TPU 正在被加入，Trainium 也很快会被加入 vLLM。但到今年年中，我认为它们都将在 vLLM 上拥有非常好的“下载模型即运行”的用户体验（UX），对吧？AMD 肯定在这个季度末就已经到位了。我们有一个东西在测试这个，叫 InferenceMaxa，它是开源的，所有代码和结果都是。我们在大约价值 6000 万美元的 GPU 上运行测试，这些 GPU 是由 NVIDIA、AMD、OpenAI、Microsoft、Amazon、Crusoe、CoreWeave、Together AI 等公司捐赠给我们的。

[原文] [Speaker B]: uh all these companies are sponsoring GPUs for us to run this we're running VLM and SDG Lang every night on you know nine different kinds of GPUs on a variety of different models and different work uh context lens and all these things right to see the performance and you can see the performance moving every day or pretty often because the software changes all the time and so like the fact that this exists is the cuda boat right it's not that like AMD you can do this on their chips Nvidia can do this on their chips it's oh when the new model comes out how fast does it get to peak performance because you know it's it's a moving target or hey can I implement this KV cache management thing how hard is it how many engineers do I need oh just one great like or 10 great if I need a hundred people to develop it like Google and you know so on and so forth did then that's much harder

[译文] [Speaker B]: 所有这些公司都赞助 GPU 让我们运行这个。我们每晚在九种不同的 GPU 上运行 vLLM 和 SGLang，测试各种不同的模型、不同的上下文长度和所有这些东西，以查看性能。你可以看到性能每天或经常在变化，因为软件一直在变。所以这个东西的存在本身就是 CUDA 护城河，对吧？不是说 AMD 的芯片能不能做这个，或者 NVIDIA 的芯片能不能做这个，而是——噢，当新模型出来时，它多快能达到峰值性能？因为你知道这是一个移动的目标。或者，嘿，我能实现这个 KV Cache 管理功能吗？有多难？我需要多少工程师？噢，只要一个？太好了。或者十个？也行。如果我需要像 Google 那样的一百人去开发它，那就难多了。

[原文] [Speaker A]: do you think AMD can uh catch up

[译文] [Speaker A]: 你认为 AMD 能追上吗？

[原文] [Speaker B]: I think AMD will be caught up at times and very behind at other times like currently they're super far behind right because Blackwell is just way better than MI355 um and then you know Rubin comes out and they'll be way way behind but then AMD's new chip comes out and AMD will be caught up or evenlight ly ahead on a hardware perspective software's behind right and you have this like leaprogging and and AMD is a very credible second competitor i don't think they'll go beyond like I think they'll stay in single digits market share single digit percentage market share single digit percentage market share is still pretty good yeah i mean Nvidia's revenue this year is going to be like it's a lot the three gajillion dollars i think it's actually four gajillion

[译文] [Speaker B]: 我认为 AMD 有时会追上，有时会非常落后。比如目前他们超级落后，对吧？因为 Blackwell 比 MI355 强太多了。然后你知道 Rubin 出来后，他们会落后得更远。但接着 AMD 的新芯片出来，AMD 就会在硬件视角上追平甚至略微领先，但软件是落后的，对吧？你会看到这种交替领先（leapfrogging）。AMD 是一个非常可信的第二名竞争者。我不认为他们会超越——我认为他们会保持在个位数的市场份额。个位数百分比的市场份额仍然相当不错。是的，我的意思是 NVIDIA 今年的收入将会是——很多，大概“三无数亿”（three gajillion）美元，我想实际上是“四无数亿”。

章节 4：初创公司的突围：在巨头阴影下寻找生存空间

📝 本节摘要：

本节聚焦于 AI 芯片初创公司（如 Etched、Cerebras 等）如何在 NVIDIA 的统治下寻找生存之道。Dylan 指出，正面硬刚 NVIDIA 的通用性能是死路一条，因为 NVIDIA 掌控着供应链和最新技术。初创公司唯有通过“极致专业化”（Specialization）才有一线生机。然而，这种赌注风险极高，因为 AI 模型架构变化极快（如稀疏化、注意力机制变革），一旦赌错便满盘皆输。尽管 Dylan 认为初创公司的成功率可能低于 1%，但在视频生成（如 Midjourney）等特定高计算密度、低内存带宽需求的场景中，专用芯片仍有机会胜出，未来可能形成一个“多芯片并存”的世界。

[原文] [Speaker A]: what about all the startups you mentioned a few so there's a cerebrus on the one end of the spectrum and then newer ones edged and and others if if AMD has a you know uphill battle in front of them like do you think those guys can take significant market share

[译文] [Speaker A]: 那些初创公司呢？你提到了一些，比如光谱一端的 Cerebras，还有更新的 Etched 等等。如果连 AMD 面临的都是一场艰难的战斗（uphill battle），你认为这些家伙能夺取显著的市场份额吗？

[原文] [Speaker B]: you sort of the whole specialization game right you you have to specialize because you're never going to beat Nvidia at their own game right they're going to have the supply chain unlock they're going to get to the newest memory technology or process technology or whatever packaging technology whatever it is sooner than you and they're just going to crush you right if you play their game you have to AMD is trying to play Nvidia's game but AMD is like extremely good at engineering silicon right everyone else has to has to has to try something weird or different right

[译文] [Speaker B]: 你这就涉及到了整个“专业化游戏”（specialization game），对吧？你必须专业化，因为你永远无法在 NVIDIA 自己的游戏中击败它，对吧？他们拥有供应链的掌控力（supply chain unlock），他们会比你更早获得最新的内存技术、制程技术或任何封装技术，如果你玩他们的游戏，他们会直接碾压你。AMD 试图玩 NVIDIA 的游戏，但 AMD 在硅片工程方面极其出色，对吧？其他人都必须——必须——尝试一些奇怪或不同的东西，对吧？

[原文] [Speaker B]: and so when you look at Etched or Maddx or Posatron or Cerebrris or Tenstor you go to look at all these companies right there are unique things about what they're doing and it's not clear if AI models will still be within that realm when that comes out right uh does oh now people use like engrams and other sparse attention techniques is that like is does that change like some of the specializations people are doing or hey people are now doing like you know models are now sparse instead of being dense models does that change things there's so many optimizations and changes on the model side and you can't predict what's going to happen with the ML research easily at least you can't

[译文] [Speaker B]: 所以当你看看 Etched、Maddx、Positron、Cerebras 或 Tenstor，你去看所有这些公司，对吧？他们所做的事情都有独特之处。但不清楚的是，当他们的芯片问世时，AI 模型是否仍处于那个领域，对吧？噢，现在人们是不是开始使用 N-grams 和其他稀疏注意力（sparse attention）技术了？这是否改变了人们正在做的一些专业化设计？或者嘿，人们现在做的模型是稀疏的（sparse）而不是稠密的（dense），这会改变事情吗？模型端有太多的优化和变化，你无法轻易预测机器学习（ML）研究会发生什么，至少你做不到。

[原文] [Speaker B]: the thing you're optimizing for today has to be a vision of where AI will be in 2 years and Nvidia's fully accepted they don't know where that's going to be that's why they have a portfolio of chips now not just one GPU line right it's not just Hopper Blackwell Reuben now it's going to be you know it's not Ampure Hopper you know you know it's not that line it's like there's a variety of chips to serve the different markets um and different possible scenarios they think each of them has this vision today but oh it might turn out the general purpose one sucks and and actually AI models have developed in a way where CPX or Grock style chips are the best right well okay now we have a solution for that market

[译文] [Speaker B]: 你今天优化的目标必须是关于 AI 在 2 年后会是什么样子的愿景。而 NVIDIA 完全接受了这样一个事实：他们不知道未来会在哪里。这就是为什么他们现在拥有一个芯片组合（portfolio），而不仅仅是一条 GPU 产品线，对吧？不仅仅是 Hopper、Blackwell、Rubin。现在不仅是 Ampere、Hopper 那条线，而是有各种各样的芯片来服务不同的市场和不同的可能场景。他们认为每一个都有今天的愿景，但噢，也许结果证明通用芯片很烂，实际上 AI 模型发展成了 CPX 或 Groq 风格的芯片才是最好的，对吧？好吧，那我们现在也有针对那个市场的解决方案了。

[原文] [Speaker B]: and so I think that's the challenge with the startups with that said I think they're all taking very interesting bets i think it's I think it's much more exciting than the first wave of AI hardware uh bets graph course rebringing the memory on the chip they sort of just made a bet and they optimized for a certain kind of model all similar kinds of model and it didn't end up working out for a long time right they had to pivot and they had to work on a lot of things and it took a long time I think these companies have like a really clear vision of what they think models will look like right like Etch does Maddx does Posatron does and that's what's really cool about it between the three of them uh these new age so I mean I'm I'm excited for them i'm very very skeptical i don't know what uh what a venture capitalist views as likely chances of succeeding but I think all of them are less than 1% right

[译文] [Speaker B]: 所以我认为这就是初创公司的挑战。话虽如此，我认为他们都在进行非常有趣的押注。我认为这比第一波 AI 硬件押注要令人兴奋得多——比如 Graphcore 把内存带回芯片上，他们只是做了一个押注，针对某种特定类型的模型进行优化，所有类似的模型。这在很长一段时间内都没有奏效，对吧？他们不得不转型，不得不做很多事情，花了很多时间。我认为现在的这些公司对自己认为模型会是什么样子有着非常清晰的愿景，对吧？比如 Etched、Maddx、Positron 都有。这也是这三家“新时代”公司真正酷的地方。所以我是说，我为他们感到兴奋，但我非常非常怀疑。我不知道风险投资家认为的成功几率是多少，但我认为他们所有的成功率都低于 1%，对吧？

[原文] [Speaker B]: but you know that's that's that's a but the world where they win is a multi-silicon kind of world where any given customer uses a range of different GPUs it could it could or it could be any given customer has like one workload they care a lot about anthropic clearly does not give a crap about videogen image gen right they just don't care um on the flip side company like midjourney cares a lot about image and videogen right image and videogen is very very like like I mentioned like it's a very like it's not very memory bandwidth heavy it loves loves loves compute right whereas inference of large language models in the style of like you know this these you know say for example coding agents cares a lot about decoding for long streams of time um and that's very memory bandwidth heavy right

[译文] [Speaker B]: 但你知道，那是——但是，如果他们赢了，那将是一个“多芯片共存”（multi-silicon）的世界，任何给定的客户都会使用一系列不同的 GPU。或者可能是任何给定的客户有一个他们非常关心的工作负载。Anthropic 显然完全不在乎视频生成（VideoGen）或图像生成（ImageGen），对吧？他们就是不在乎。但在反面，像 Midjourney 这样的公司非常在乎图像和视频生成，对吧？图像和视频生成非常非常——就像我提到的——它对内存带宽（memory bandwidth）的要求不是很高，但它超级超级超级喜欢算力（compute），对吧？相比之下，大语言模型的推理，比如那些编程智能体（coding agents），非常在乎长时间流的解码，而那是极其依赖内存带宽的，对吧？

[原文] [Speaker B]: and so there's like that's like a simple example but there's a lot more nuance there in terms of like even like the size of like the matrix multiply you know the tensor cores that you you know the systolic arrays that you use or the ratios of networking and memory memory and like what's that memory hierarchy look like and you know what are you doing for different kinds of attention and like oh like all these sorts of things like there's a lot of specialization here and so some people are betting big on on different types of specialization

[译文] [Speaker B]: 这是一个简单的例子，但其中还有更多细微差别，比如矩阵乘法的大小、你使用的张量核心（Tensor Cores）、脉动阵列（systolic arrays），或者网络与内存的比例、内存层级结构是什么样的，以及你针对不同类型的注意力机制在做什么等等。这里有大量的专业化空间，所以有些人正在不同的专业化方向上押下重注。

[原文] [Speaker B]: and I I think like you could clearly see a world where companies do care about different stuff right like like if for example a chip optimized for video and image generation existed today and it was better than Nvidia or Nvidia made it i think Midjourney would absolutely only use that for inference i think for training they'd still use the general purpose thing and as would like Meta and Google would like they should do that right

[译文] [Speaker B]: 我认为你可以清楚地看到一个世界，公司确实在乎不同的东西，对吧？比如，如果今天存在一个针对视频和图像生成优化的芯片，并且它比 NVIDIA 的好，或者就是 NVIDIA 制造的，我认为 Midjourney 绝对只会用那个芯片进行推理。我认为训练方面他们仍然会使用通用芯片，Meta 和 Google 也是如此，他们应该那样做，对吧？

[原文] [Speaker B]: and hey Meta actually has two lines of AI chips there mtia there's a line that's focused on recommendation systems and then there's a line that's focused on Gen AI the GI one is a new line but that recommendation systems ch line is still continuing right it's not sexy no one cares because there's no and bite dance also has a recommendation system line of chips and it's not really focused on Jedi which is fine because you know this is a $200 billion business or something which is just deciding what ad to serve me right and what order to put my friends stories and you know things like this so so I think like it's perfectly fine for there to be specialized AI chips given the target market is big enough and you have to have vision to know what that target market is unless you're hyperscaler then you can like just like you can just use general purpose until you've like it's clearly there and then you can make your asich right

[译文] [Speaker B]: 嘿，Meta 实际上有两条 AI 芯片线，即 MTIA。一条专注于推荐系统，另一条专注于生成式 AI（Gen AI）。生成式 AI 那条是新的，但推荐系统那条芯片线仍在继续，对吧？它不性感，没人关心，因为没有——而且 ByteDance（字节跳动）也有一条推荐系统芯片线，它并没有真正专注于 Gen AI，这没问题，因为你知道这是一个 2000 亿美元或者类似规模的业务，它只是决定给我推送什么广告，对吧？或者把朋友的故事按什么顺序排列等等。所以我认为，只要目标市场足够大，存在专用 AI 芯片是完全没问题的。你必须有眼光去知道那个目标市场是什么。除非你是超大规模云厂商（hyperscaler），那你就可以先用通用芯片，直到市场明显存在了，然后你再制造你的 ASIC（专用集成电路），对吧？

章节 5：地缘政治与中国的“半导体狂热” (Semiconductor Pilled)

📝 本节摘要：

对话转向宏观的地缘政治话题。Dylan 分析了 NVIDIA 在中国市场份额的下降以及华为的崛起，指出虽然中国未能在高端芯片上完全达标，但在微控制器等中低端领域已实现高度国产化。他揭示了 ByteDance（字节跳动）如何通过在马来西亚租赁 Oracle 的算力来规避美国制裁。最引人入胜的是，Dylan 描述了中国社会这种独特的“半导体狂热”（Semiconductor Pilled）文化——半导体工程师成为了流行文化和浪漫偶像剧的主角，甚至地方政府之间为了争夺产业链也展开了残酷的“优胜劣汰”竞争，形成了极其细分的产业集群（如专门生产吉他或灯罩的城市）。

[原文] [Speaker A]: fascinating turning to the geopolitical aspect of of uh all of this which is always fun huawei and Nvidia in China last year that was like 10 or 12% of their overall revenue and this year they they were saying that their market share but has basically dropped to not very much is that Huawei chips is that restrictions is that tariffs uh what's happening

[译文] [Speaker A]: 真迷人，转向这一切的地缘政治方面，这总是很有趣。华为和 NVIDIA 在中国的情况——去年（中国市场）大概占他们（NVIDIA）总收入的 10% 或 12%，而今年他们说他们的市场份额基本上跌到了微不足道的程度。这是因为华为芯片吗？是限制措施吗？是关税吗？发生了什么？

[原文] [Speaker B]: it's a variety of things actually in in some in some quarters last year uh it was even north of 20 I think but I don't remember exactly but anyways you know if you look at 2022 China was almost the size of the US in terms of buying server hardware right almost not quite but getting there um and it looked like they were going to be the same size as America in like a year or two after that right and if you look at like global data center capacity global cloud capacity etc etc etc it's American companies and Chinese companies right that dominate the world American companies obviously doing a lot better here but both of those dominate the world

[译文] [Speaker B]: 实际上是多种因素。在去年的某些季度，我认为这个比例甚至超过了 20%，但我记不太清了。无论如何，如果你看 2022 年，中国在购买服务器硬件方面的规模几乎与美国相当，对吧？几乎，虽然还没完全赶上，但也快了。看起来在那之后的一两年内，他们就要达到与美国相同的规模了，对吧？如果你看全球数据中心容量、全球云容量等等，实际上是美国公司和中国公司主导着世界，对吧？显然美国公司在这方面做得更好，但这双方主导了世界。

[原文] [Speaker B]: and if you look at like every industry right you know it's It's it's very clear that like China wants to insource stuff right so in 2015 they made these 5-year plans for two 2020 and 2020 uh five where they set the percentage of semiconductors they wanted uh domestically produced and they've missed the goal both times which is fine right they set really aggressive goals and even you know shoot for the uh moon even if you miss you hit the stars right and that's sort of what's happened right like look China is not caught up on you know leading edge semiconductors but microcontrollers from China are almost as good as the microcontrollers are as good and cheaper than the ones from Texas Instruments or ST Micro or you know etc right or like this power random power chip is better than or the same as the one from like another company right

[译文] [Speaker B]: 如果你看每一个行业，很明显中国想要实现自给自足（insource），对吧？所以在 2015 年，他们制定了针对 2020 年和 2025 年的五年计划，设定了他们希望国产半导体的百分比。他们两次都错过了目标，但这没关系，对吧？他们设定了非常激进的目标，就像你知道的，“瞄准月亮，即使没打中，你也能落在星星之间”，对吧？这某种程度上就是发生的事情。看，中国在尖端半导体上还没有追上，但来自中国的微控制器几乎和德州仪器（Texas Instruments）或意法半导体（ST Micro）等公司的一样好，甚至更便宜，对吧？或者像这种随机的电源芯片，比其他公司的更好或者一样好，对吧？

[原文] [Speaker B]: and so they've really built up a semiconductor industry and started insourcing a lot more i don't see why China wouldn't be buying you know 30 40% of the world's AI chips and the US like 50 60% and then the rest of the world like you know and when I say US I mean US origin companies that seems like a more natural state for the world but there are restrictions and and hey this is the biggest change in human history maybe ever knowledge work and you know everything that's going to happen there and and then eventually like robotics and all these things like you know obviously there's there's a lot of geopolitical stuff and so there are restrictions Nvidia's been handcapped hand handicapped from selling their best chips to China and so that's obviously impacted the sales a lot because like why would you do that

[译文] [Speaker B]: 所以他们真的建立了一个半导体产业，并开始更多地进行内包。我不明白为什么中国不会购买全球 30% 到 40% 的 AI 芯片，而美国购买 50% 到 60%，剩下的世界购买其余的——我说美国是指源自美国的公司。这似乎是世界更自然的状态。但是存在限制，嘿，这是人类历史上最大的变革，也许是有史以来最大的——知识工作，以及那里将发生的一切，最终还有机器人技术和所有这些东西。显然这里有很多地缘政治因素，所以有限制。NVIDIA 被束缚住了手脚，无法向中国出售他们最好的芯片，所以这显然极大地影响了销售，因为（如果不能买最好的）你为什么要买呢？

[原文] [Speaker B]: and so when you look at who rents the most GPUs in the world it's three companies right so one of them is obviously OpenAI second one actually they were bigger than OpenAI they are bigger than OpenAI today or no they were bigger than OpenAI than OpenI eclipsed them recently is Bite Dance bite Dance runs rents tons of chips from Oracle and Google and and you know many other cloud companies because they couldn't get the chips they need in in China they're mostly just serving Tik Tok right okay well they they're not allowed to buy them and that sucks but you know they're they're allowed to rent them and so okay if I'm not allowed to get the best ones I'm going to rent externally and if Bite Dance is the second biggest renter of GPUs in the world that's substituting demand that would have been built in China in many cases it's instead being built in Malaysia and Oracle has over a gigawatt of capacity in Malaysia that Bite Dance is going to take right so things like this are you know you know hundreds of thousands if not millions of chips tens of billions of dollars of cap capacity that would go to China but it's not that it's going to Malaysia instead as an example

[译文] [Speaker B]: 所以当你看看谁在世界上租用了最多的 GPU，是三家公司，对吧？其中之一显然是 OpenAI。第二家——实际上他们曾经比 OpenAI 大，今天可能也比 OpenAI 大，或者不，他们曾经比 OpenAI 大，后来 OpenAI 最近超越了他们——是 ByteDance（字节跳动）。ByteDance 从 Oracle 和 Google 以及许多其他云公司租用了大量的芯片，因为他们在中国无法获得他们需要的芯片。他们主要只是为了服务 TikTok，对吧？好吧，他们不被允许购买这些芯片，这很糟糕，但你知道，他们被允许租赁。所以，好吧，如果不允许我买最好的，我就在外部租赁。如果 ByteDance 是世界上第二大 GPU 租户，那就是在替代本应在中国建设的需求，而在很多情况下，这些需求反而在马来西亚建设了。Oracle 在马来西亚有超过 1 吉瓦（gigawatt）的容量将被 ByteDance 拿下，对吧？所以像这样的事情——你知道是成千上万甚至上百万个芯片，数百亿美元的资本容量——本该去中国的，但没有，它去了马来西亚，这只是个例子。

[原文] [Speaker B]: another sort of point around this is China's like you know they've had these 5-year plans so and and you know the way these initiatives work from China is there is like some top down ordering but then they just kind of whip the whole like everyone just kind of gets into it and it's really cool like I don't think it's as top down as many people think like I think the entire country is like semiconductor pill right there are dramas where people fall in love in the fab or dramas where people fall in love and they're photovoltaic like solar cell researchers and engineers and it's like it's like this is just the backdrop and it's like actually this is it's like super cool for your like significant other to be that semiconductor engineer or to be that photovoltaic you know uh solar panel researcher as opposed to an influencer as opposed to an influencer right like I'm sorry love Island is I I I watched like for 10 minutes cuz I was forced to i was like this is freaking terrible um but you know like um we are so cooked no you know seriously we're cooked we're cooked

[译文] [Speaker B]: 另一点是，中国——你知道他们有这些五年计划。你知道中国的这些举措运作的方式是有一些自上而下的命令，但随后他们有点像是动员了整体——就像每个人都参与进来了，这真的很酷。我不认为这像许多人想的那样完全是自上而下的。我认为整个国家都像是“磕了半导体的药”（semiconductor pilled），对吧？那里有电视剧演人们在晶圆厂（fab）里坠入爱河，或者演光伏——比如太阳能电池研究员和工程师坠入爱河。这就像是背景设定，实际上如果你的另一半是那个半导体工程师，或者是那个光伏、你知道、太阳能电池板研究员，那是超级酷的事情。而不是什么网红（influencer），对吧？我很抱歉，《Love Island》（恋爱岛）——我看过 10 分钟，因为被迫看的——我就觉得这也太糟糕了。但你知道——我们真的完蛋了（we are so cooked），不，说真的，我们要完蛋了，我们要完蛋了。

[原文] [Speaker B]: and I think I think like when you think about like this happens it's like it's diffused into drama even people like like there's multiple dramas like taking place about semiconductor industry and and they're like romance comedy like the entire spectrum right drama like it's like it's like what the heck is going on anyways you have all these provinces you have all these local cities studying out ordinances and giving out subsidies and all sorts of stuff right it's truly like crazy like there's some national level stuff like "Oh no taxes on uh this oh we're going to ban a few things." But as far as I understand the national government has not banned Nvidia's H20 or H200 but the local ones have right a lot of local ones have said "No you know you must use China manufactured chips." And it's like who told you that you know you're here to uphold this it's like does it matter right i mean like it's it's it's cool because then you have this like survival of the fittest all these all these provinces and cities are trying to attract different companies with different types of subsidies and grants and industrial parks and like all these different things and then like the ones who succeed actually develop an industry and they take over this how one thinks of of China right

[译文] [Speaker B]: 我认为当你思考这一现象时，它已经渗透到了戏剧中，甚至人们——就像有多部关于半导体行业的电视剧正在上演，而且它们是浪漫喜剧，就像涵盖了整个光谱，对吧？这种戏剧让你觉得“这到底是怎么回事？”无论如何，你有所有这些省份，所有这些当地城市研究法规，发放补贴和各种各样的东西，对吧？这真的很疯狂。有一些国家层面的东西，比如“噢，这个免税，噢，我们要禁止一些东西。”但据我了解，国家政府并没有禁止 NVIDIA 的 H20 或 H200，但地方政府却禁止了，对吧？很多地方政府说：“不，你知道，你必须使用中国制造的芯片。”这就像是谁告诉你的？你知道你是来维护这个的——这重要吗？我的意思是，这很酷，因为你会看到这种“优胜劣汰”（survival of the fittest）。所有这些省份和城市都试图通过不同类型的补贴、赠款和工业园区以及所有这些不同的东西来吸引不同的公司。然后那些成功的确实发展出了一个产业，并接管了人们对中国的认知，对吧？

[原文] [Speaker B]: it almost sounds like more like the US or like with the federal government and states where the provinces have authority over their purchasing it's It's actually like uh great there's this one um Tik Tok or not Tik Tok Tik Tok and Instagram like uh person and they're like they they like sing it they're like if you want to if you want to buy things in China make sure you go to the right place and then they just say the most random shit and name the city and then you look into it and you're like wow this city has the entire supply chain for this um and it's like lampshades and then it names the city it's like what the fuck there's a city that specializes in lampshades like it's like and it's like microphone arms like microphones it's like it's like literally there's a city in China that specializes in guitars as well right this one one city that became the guitar capital of the world it's literally everything literally everything there's a city and it's not like hey specifically for uh camera arms for example there's ball bearings in this and the ball bearings are like there's ball bearings there's multiple manufacturers of ball bearings for camera arms and then like most of the camera arms in the world come from that one city it's like what the hell is going on

[译文] [Speaker B]: 这听起来几乎更像美国，或者像联邦政府和州，省份拥有其采购的权力。实际上这——呃——很棒。有一个 TikTok 或不是 TikTok——TikTok 和 Instagram 上的人，他们像是把它唱出来，他们说：“如果你想——如果你想在中国买东西，确保你去对了地方。”然后他们就说一些最随机的东西并报出城市名。然后你去调查一下，你会发现：“哇，这个城市拥有这个东西的完整供应链。”比如灯罩，然后它报出那个城市名。就像是：“搞什么？有一个城市专门生产灯罩？”就像是——还有麦克风支架，比如麦克风。就像是——简直了，中国还有一个城市专门生产吉他，对吧？这一个城市成为了世界吉他之都。简直是所有东西，字面意义上的所有东西都有一个城市。而且不是说，嘿，专门针对——比如相机支架，这里面有滚珠轴承，而这些滚珠轴承有多个制造商专门为相机支架生产，然后世界上大多数的相机支架都来自那一个城市。就像是：“这到底是怎么回事？”

章节 6：供应链的极致复杂性与华为的垂直整合

📝 本节摘要：

本节深入解析了半导体供应链的极度复杂性。Dylan 引用经典经济学随笔《我，铅笔》，指出芯片制造的链条远比铅笔复杂，哪怕是奥地利这样的小国，也能因垄断特定细分领域（如特定化学品或工具）而具备“关闭”全球芯片产业的能力。

对话随后转向中国在被封锁状态下的“垂直整合”能力。Dylan 认为，虽然中国在尖端光刻机（Lithography）等领域落后，但在成熟制程（10-20 年前的技术）上，中国拥有全球最完整的独立产业链，这是美国无法独自做到的。

最后，Dylan 特别强调了华为的“恐怖”之处：作为全球最垂直整合的公司，华为不仅在手机和通信领域曾超越苹果和诺基亚，更通过空壳公司和秘密渠道（如走私 HBM）突破制裁，成为 NVIDIA 极为忌惮的对手。

[原文] [Speaker B]: and you look across the semiconductor industry there's a famous economic essay called I pencil or something like that or talking about how the pencil like a simple pencil comes from like oh the rubber comes from like Indonesia for the eraser and the graphite comes from this mine here and and the wood comes from these aspen trees in Canada and like you actually can't make a pencil without aggregating this entire supply chain semiconductor industry is like way crazier because like I would say there's like 15 or 20 countries that could shut down the entire semiconductor industry right even like Austria could right and and it's like what and it's like well yeah there's two different companies there who have like 90% share in like some random niche stuff and it's like okay cool i guess Austria can and oh yeah those two companies only like have less than a billion of revenue but they just happen to have lynchpin critical things and there's lynch pin critical things everywhere because the process is so complicated

[译文] [Speaker B]: 当你审视半导体行业时，有一篇著名的经济学随笔叫《我，铅笔》（I, Pencil）之类的，它讲的是铅笔——像一支简单的铅笔，橡胶（橡皮擦）来自印度尼西亚，石墨来自这边的矿山，木材来自加拿大的白杨树。如果不聚合这整个供应链，你实际上造不出一支铅笔。半导体行业比这要疯狂得多，因为我想说大概有 15 或 20 个国家可以关闭整个半导体行业，对吧？甚至像奥地利都可以，对吧？你会觉得“什么？”。但这确实如此，那里有两家不同的公司在某些随机的细分领域拥有 90% 的市场份额。这就像是，好吧，酷，我猜奥地利确实可以。噢对了，那两家公司的收入可能不到 10 亿美元，但它们恰好掌握着关键枢纽（lynchpin critical things）。这种关键枢纽到处都是，因为这个过程太复杂了。

[原文] [Speaker B]: and so China's been trying to replicate this um is there one thing they're missing that they don't have yet i think there's a lot of things i think if you were to close your eyes and say or if you were to cut off every country and say there's no more globalism China has the most vertical stack in semiconductors today and they're the best at semiconductors in the world because their fabs could still run somewhat on a lot of things because they have built some of these chemical supply chains right like TSMC for certain kinds of chemicals 100% share from Japan right or Intel same thing right or you know for certain kinds of tools 100% share from Netherlands or 100% share from you know this American company or that you know Austrian company or this or that right like there's just all these like you know this Swiss company like there's just all these different places have 100% share it might be one company might be three companies but geographically or in the same area

[译文] [Speaker B]: 所以中国一直在试图复制这一切。你要问是否有一件东西是他们缺失的、还没有的？我认为有很多东西。但我认为，如果你闭上眼睛说，或者如果你切断每个国家的联系，说“不再有全球主义了”，那么中国拥有当今最垂直的半导体产业链（vertical stack），而且在这种（封闭）情况下他们是世界上最好的半导体国家。因为他们的晶圆厂在很多方面仍然可以某种程度上运行，因为他们已经建立了一些化学供应链。像台积电（TSMC），某些种类的化学品 100% 来自日本，对吧？或者 Intel 也是一样。或者你知道，某些工具 100% 来自荷兰，或者 100% 来自这家美国公司，或者你知道那家奥地利公司，或者这个那个。就像你知道的，还有这家瑞士公司。所有这些不同的地方都拥有 100% 的份额，可能是一家公司，也可能是三家公司，但在地理上或在同一个区域。

[原文] [Speaker B]: and China's built that up right because they've created this made in China initiatives which just plowed money into it and they've got this culture of like the diffused like you know these provinces like yeah I just decided I'm going to fucking focus on or might not even be might not even be the Right it may be the like you know someone brought it there and decided and then people were like "Oh wow you're doing that?" Me too

[译文] [Speaker B]: 而中国已经把这个建立起来了，对吧？因为他们创立了“中国制造”计划，把钱砸进去。而且他们有这种分散的文化，就像你知道的，这些省份会说：“是的，我决定我就要他妈的专注于这个。”或者甚至可能不是——可能不是最正确的决定，可能只是某人把它带到那里并决定做了，然后人们就说：“噢，哇，你在做那个？我也做。”

[原文] [Speaker B]: china's missing a lot of things right i would say like if you say minus 10 years tech China's complete and no one else is complete right taiwan is not complete their the fabs would shut down without foreign supply you know and you go down or you go across the stack uh but if you go to 10ear tech maybe maybe more like 20-year tech you could get a fully vertical supply chain in China which I do not think any country could do like America could not build a fully vertical fab without stuff from elsewhere even if it's 20-y old tech um probably not even 40-y old tech

[译文] [Speaker B]: 中国确实缺很多东西，对吧？我想说，如果你说是“落后 10 年的技术”，中国是完整的，而其他任何国家都不是完整的，对吧？台湾不是完整的，如果没有外国供应，他们的晶圆厂就会关闭。你知道，不管你是往下看还是横向看产业链。但是如果你看 10 年前的技术，也许——也许更像是 20 年前的技术，你可以在中国获得一个完全垂直的供应链。而在这一点上，我不认为其他任何国家能做到。比如美国如果不从别处获取东西，就无法建立一个完全垂直的晶圆厂，即使是 20 年前的技术，甚至可能连 40 年前的技术都做不到。

[原文] [Speaker B]: and so so that's interesting but then when the flip side is like well like you kind of do need specialization that's how that chemical gets the purest best you know most engineered you know or that that slurry of chemicals or that you know that gas or like that tool because every smart person or a lot of them in that country grew up around that culture and like every the supply chain is there and like everyone kind of knows and like it's like a a driveaway and like sort of like this is what makes supply chains work is that there is this specialization and the best of the best only comes when you have that hyper specialization so China doesn't have lithography their lithography is like 10 years behind and I think it'll be 5 years behind in a couple years right they're catching up fast

[译文] [Speaker B]: 所以这很有趣。但反过来说，你确实需要专业化（specialization）。这就是为什么那种化学品能做到最纯、最好、你知道——最精密的工程设计。无论是那种化学研磨液（slurry），还是那种气体，或者那种工具。因为那个国家的每个聪明人，或者很多人，都是在这种文化中长大的，供应链就在那里，每个人都懂行，就像开车就能到一样。这就是供应链运作的原因——存在这种专业化。只有当你拥有那种“超专业化”（hyper-specialization）时，才能诞生最好的产品。所以中国没有光刻技术（lithography），他们的光刻技术大概落后 10 年。我认为几年后会变成落后 5 年，对吧？他们追赶得很快。

[原文] [Speaker B]: i don't think they'll be as good as ASML for a long time you know maybe I don't know maybe they will be you know China you shouldn't ever underestimate China but like and Chinese engineers or you know but like for a while right or like you know I don't think they'll be able to make leading edge chemicals like many chi uh Japanese companies or many American companies and their tools and like you just go across the supply chain they're not hey forefront on really anything in the manufacturing supply chain on the design supply chain there's some things that they're starting to be similar par but like cheaper or like a year or two behind but cheaper and that's like fine for a lot of stuff

[译文] [Speaker B]: 我认为他们在很长一段时间内不会像 ASML 那么好。你知道，也许——我不知道，也许他们会。你知道，永远不要低估中国，或者中国工程师。但在一段时间内，对吧？或者你知道，我不认为他们能制造出像许多日本公司或美国公司那样的尖端化学品和工具。当你遍历供应链时，他们在制造供应链的任何方面都不是真正的前沿。在设计供应链上，有些东西他们开始达到类似的水平，但更便宜，或者落后一两年但更便宜。这对于很多东西来说已经够好了。

[原文] [Speaker B]: an example of that is Huawei right huawei in mobile phones was on par with Apple like entirely yeah and they had become Apple TSMC's biggest customer and they were designing the best thing and they are number one in telecom and their tech is just literally better and so when you think what happens you know is is is China missing anything it's like they're they don't they don't they don't have the best of much you know today in the AI supply chain they have a complete package and a couple years behind and they'll figure out how to make it cheaper slash do more slashcatch up and and create a robust industry but there's a reason like I don't think that like Jensen is scared of AMD really he's paranoid i mentioned he's paranoid i'm sure he's a little bit scared of them right like I think some of the things that they've done are reactions and competitive dynamics with AMD or Google's TPUs or whatever right

[译文] [Speaker B]: 一个例子就是华为，对吧？华为在手机领域曾完全与 Apple 并驾齐驱。是的，他们曾成为 Apple 也就是 TSMC 的最大客户，他们设计了最好的东西。他们在电信领域是第一名，他们的技术字面上就是更好。所以当你思考发生了什么——你知道，中国缺什么吗？这就像是，他们没有太多“最好的”东西。你知道，今天在 AI 供应链中，他们有一个完整的方案，虽然落后几年，但他们会弄清楚如何让它更便宜、做得更多、追赶上来，并建立一个强大的产业。但我认为有一个原因——我不认为 Jensen（黄仁勋）真的害怕 AMD。他很偏执（paranoid），我提到过他很偏执。我肯定他有点怕他们，对吧？我认为他们做的一些事情是对 AMD 或 Google TPU 等的反应和竞争动态，对吧？

[原文] [Speaker B]: there was a Core Weave deal today and I think that's directly the result of what Google's been doing yeah the two billion pipe that Nvidia announced into Nvidia invested two billion in core but what's more important is that that's like sort of just like the sticker what's really relevant is Nvidia is going to work with core reef to uh acquire um and and back stop and all these things the the land the power the energy the transmission that help build the data center all this capital side stuff that because Nvidia has so much money they can backs stop corore weave doing it because corore reweave then can be the one who generates demand anyways there's like because Google was doing And they did that with like a couple companies such as Fluid Stack and Terowolf and Cipher these are some public deals that have been announced and so Google is doing that with TPUs and Nvidia reacted right um and so in the same way I think Nvidia's reacted to AMD and in the same way I think the thing is Nvidia is like deathly terrified of Huawei

[译文] [Speaker B]: 今天有一个 CoreWeave 的交易，我认为这直接是 Google 所作所为的结果。是的，NVIDIA 宣布了对 CoreWeave 投资 20 亿美元。但这更像是贴牌价格（sticker），真正相关的是 NVIDIA 将与 CoreWeave 合作去获取、担保所有这些东西——土地、电力、能源、传输，帮助建设数据中心。所有这些资本层面的东西，因为 NVIDIA 很有钱，他们可以为 CoreWeave 做担保，因为 CoreWeave 可以成为产生需求的一方。无论如何，因为 Google 在做这些——他们和几家公司如 Fluid Stack、Terawulf 和 Cipher 做了这些（公开宣布的交易），Google 正用 TPU 做这些，所以 NVIDIA 反应了，对吧？同样的道理，我认为 NVIDIA 对 AMD 做出了反应；同样的道理，我认为 NVIDIA 对华为是感到“死亡般的恐惧”（deathly terrified）。

[原文] [Speaker B]: because Huawei has caught up to Apple and actually surpassed them as TSMC's biggest customer before they got banned right they did just crush Nokia Sony Sony Ericson etc right like the entire telecom supply chain they just like completely destroyed them and there's so many other areas like they straight up made a folding phone right you know I have a Samsung folding phone they have a folding phone that's better than Samsung's folding phone and it's like bro what like you know you know Huawei's really really cracked and so of course they're terrified of uh and and Huawei is the most vertical company in the world no company is more verticalized than Huawei which then leads to huge innovations it's something that we don't fully appreciate in the US

[译文] [Speaker B]: 因为华为追上了 Apple，实际上在被禁之前，他们已经超越 Apple 成为 TSMC 的最大客户，对吧？他们确实粉碎了诺基亚、索尼、索尼爱立信等等，对吧？就像整个电信供应链，他们完全摧毁了它们。还有很多其他领域，比如他们直接做出了折叠屏手机，对吧？你知道我有三星的折叠屏手机，而他们的折叠屏手机比三星的更好。这就像是，兄弟，什么鬼？你知道华为真的很强（cracked）。所以当然，他们（NVIDIA）对华为感到恐惧。华为是世界上最垂直整合的公司，没有哪家公司比华为更垂直化，这进而带来了巨大的创新。这是我们在美国没有充分意识到的事情。

[原文] [Speaker B]: but when you travel in Europe you see everybody who's like honors honor phones and it's like the the footprint of Huawei is huge in in phones in a way that people not just phones um you know security cameras actually they think they have like you know a lot of training on the that a captive group of testers exactly exactly um I think I think Huawei is terrifying right and and so like yes their chips are not as good today and is that is is that already happening i mean obviously the US and China are the two biggest markets but like for other markets I don't know UAE Middle East Europe are Nvidia and Huawei already uh headto-head in well they shipped a little bit but like mostly just like sticker capacity like there's nothing like no no like I would say like a little bit as in like a few servers not like a billion dollars worth of stuff right the thing is China's supply chain has to ramp up right

[译文] [Speaker B]: 但当你在欧洲旅行时，你会看到每个人都用 Honor（荣耀）手机。华为在手机领域的足迹是巨大的，不仅仅是手机，还有安防摄像头。实际上他们认为他们在那个领域有大量的训练数据——一群被俘获的测试者（captive group of testers），没错，没错。我认为华为是可怕的，对吧？所以是的，他们的芯片今天还没那么好。这已经在发生了吗？我是说，显然美国和中国是两个最大的市场，但对于其他市场，我不知道，阿联酋、中东、欧洲，NVIDIA 和华为已经在正面交锋了吗？嗯，他们出货了一点点，但主要只是名义上的产能（sticker capacity），没有什么——我会说只是一点点，比如几台服务器，而不是几十亿美元的东西，对吧？问题是中国供应链必须提升产能，对吧？

[原文] [Speaker B]: how was Huawei actually building chips well actually they were uh using shell companies to get chips from TSMC and using different methods of like sneaking HBM which is memory from you know Korea through Taiwan to China right like all sorts of crazy stuff we've reported on and and people it's like a whack-a-ole right they shut it down or like tools that get shipped to China and they shouldn't be for you know making leading edge chips but they actually are um and all these sorts of things are happening because they can't make everything and if they want to make the leading edge stuff they do need to rely on the foreign supply chain quite a bit

[译文] [Speaker B]: 华为究竟是如何制造芯片的？实际上，他们利用空壳公司（shell companies）从 TSMC 获取芯片，并使用不同的方法，比如把 HBM（高带宽内存）从韩国经由台湾偷运到中国，对吧？各种疯狂的事情，我们都报道过。这就像打地鼠游戏（whack-a-mole），对吧？他们查封一个，或者像那些运往中国的工具，本来不该用于制造尖端芯片，但实际上却被用上了。所有这类事情都在发生，因为他们无法制造所有东西。如果他们想制造尖端的东西，他们确实需要相当依赖外国供应链。

章节 7：经济战视角：芯片法案 (Chips Act) 与全球化制造

📝 本节摘要：

本节重点评估了美国芯片法案（Chips Act）的实际成效与局限性。Dylan 指出，虽然 500 亿美元的补贴是积极的，但在半导体行业庞大的资本规模面前（仅台湾就投入了逾 5000 亿美元），这只是杯水车薪。他揭露了法案通过的“幕后真相”：并非纯粹为了 AI 或国防，而是因为汽车制造商在疫情期间糟糕的“准时制”（Just-in-Time）库存管理导致芯片短缺，进而引发车价飙升，才迫使国会采取行动。尽管如此，Dylan 对美国本土制造（如 TSMC 亚利桑那工厂）仍持乐观态度，但他同时强调，追求完全的自给自足是不现实且不必要的，“全球主义”依然是半导体产业运作的最佳模式。

[原文] [Speaker A]: where do the uh US onshoring efforts fall in that category what do you make of them from the chipsack to like all the thing that is being built everything looks like it's massively delayed by the way which perhaps is not surprising

[译文] [Speaker A]: 美国的“回岸外包”（onshoring）努力属于哪个范畴？你怎么看这些举措？从芯片法案（Chips Act）到正在建设的所有东西，顺便说一句，所有东西看起来都严重延误了，这也许并不令人惊讶。

[原文] [Speaker B]: I think TSM MC's manufacturing wafers and they're like building real wafers and there's real fabs and like you know there's some other fabs that have been announced and like they're doing well and there's like a bunch of like different kinds of plants like a Korean company making a random gas plant in Texas for you know their chips right like uh for chips and all these like sort of things are happening um I think the chips act did really well with its $50 billion it's just I don't think people understand the scale of the semiconductor industry is the most complicated supply chain in the world right it's much bigger than you know say manufacturing airplanes it's much bigger than like you know really anything else right

[译文] [Speaker B]: 我认为台积电（TSMC）正在制造晶圆，他们在制造真正的晶圆，有真正的晶圆厂（fabs）。而且你知道还有其他一些已经宣布的晶圆厂，它们进展顺利。还有一堆不同类型的工厂，比如一家韩国公司在得克萨斯州建了一个随机的气体工厂，你知道是为了他们的芯片，对吧？所有这些事情都在发生。我认为芯片法案那 500 亿美元做得很好，只是我不认为人们理解半导体行业的规模——它是世界上最复杂的供应链，对吧？它比制造飞机大得多，比你知道的任何其他东西都要大得多，对吧？

[原文] [Speaker B]: if you look at the top 10 companies like of the world I think eight of them designed semiconductors right now obviously like Google designed semiconductors but it's like oh wait no but their cost of search would be like 10x higher if they didn't have TPUs and TPUs were super optimized for search right or like you know you you you go down the list right like Meta serves recommendation systems with their chips right like you go down the list it's everyone is making their own chips apple devices would be materially worse if they didn't have their own chips right um and you just go down the list it's like it's the most complicated supply chain and they they're spending something on the order of like $150 billion roughly in subsidies a year to the chip industry we are doing 50 over like a decade

[译文] [Speaker B]: 如果你看世界上市值前 10 的公司，我想其中有 8 家都在设计半导体。现在显然像 Google 这样的公司在设计半导体，这就像是——噢等等，如果没有 TPU，他们的搜索成本会高出 10 倍，而 TPU 是针对搜索超级优化的，对吧？或者你知道，你往下看列表，比如 Meta 用他们的芯片服务推荐系统，对吧？你往下看，每个人都在制造自己的芯片。如果 Apple 设备没有自己的芯片，性能会差很多，对吧？你继续往下看，这是最复杂的供应链。他们（全球其他国家）每年在芯片行业的补贴大概是 1500 亿美元左右，而我们在十年内才投了 500 亿。

[原文] [Speaker B]: there's a difference in scale here right the collective total amount of like capex that has been spent in Taiwan is like 500 billion plus right across the industry across all the companies that are making semiconductors in Taiwan and Taiwan doesn't have a domestic industry how is $50 billion of subsidies going to change America's needle right it does move it a little bit right i I want to be clear like the chips act is awesome i don't understand why like EVs or like solar was given this massive massive like trillion dollar package semiconductors were only given 50 like semiconductors need a lot bigger package to actually incentivize onoring i think what's happened so far has proven that it's working well tsmc is literally making chips for Nvidia and Apple and AMD and others in Arizona today right and I think that's really great

[译文] [Speaker B]: 这里的规模是有差异的，对吧？台湾在整个行业、所有制造半导体的公司中累积花费的资本支出（Capex）总额超过 5000 亿美元，对吧？而台湾甚至没有一个（完整的）国内产业。500 亿美元的补贴怎么能改变美国的局面呢？它确实能推动一点点，对吧？我想明确一点，芯片法案很棒。我不明白为什么像电动汽车（EV）或太阳能得到了这种巨大的、万亿美元级的包裹，而半导体只给了 500 亿。半导体需要更大的包裹才能真正激励回岸。我认为目前发生的事情证明它运作良好，TSMC 今天确实在亚利桑那州为 NVIDIA、Apple 和 AMD 等公司制造芯片，对吧？我认为这真的很棒。

[原文] [Speaker A]: is is your sense that the broad American government is just uh aware of of all of this that it's uh

[译文] [Speaker A]: 你感觉美国政府作为一个整体意识到这一切了吗？

[原文] [Speaker B]: I wouldn't say only passed because the automotive like prices went up because car manufacturers are like the worst because they do just in time inventory right or not worse but like this is just like a thing right just in time inventory systems covid happens sales plummet fabs that were making you know random power IC's or random microcontrollers for engines got repurposed to the boom from COVID which is which was data centers and PCs and smartphones so that stuff was booming

[译文] [Speaker B]: 我不会说它是仅仅因为汽车价格上涨才通过的，因为汽车制造商简直是最糟糕的——因为他们搞“准时制”（Just-in-Time）库存，对吧？或者不是最糟糕，但这就像是一种常态，对吧？准时制库存系统。COVID 发生了，销量暴跌。原本制造随机电源 IC 或引擎用微控制器的晶圆厂被重新分配去满足 COVID 带来的繁荣，也就是数据中心、PC 和智能手机，这些东西当时正在蓬勃发展。

[原文] [Speaker B]: and then when people were like "Oh wait actually like you know I have some money i stayed at home i didn't go out i didn't drink i have a lot of I have some cash right let me buy a car." They went out and bought cars and cars started skyrocketing in prices oh let's restart and let's let's Oh yeah can I can you sell me that microcontroller for the engine again it's like "No I I'm making a slightly different microcontroller that works for you know uh let's say a keyboard or a mouse right or whatever." And it's like and and and they actually didn't just leave me flatfooted and they were like a partner through co right you know versus you just left me screw you Ford or whoever Toyota um or automotive OEM you know you up that supply chain and so Chips Act did not get passed only got passed because that happened and people are like "Oh my god the semiconductors are why cars can't be made."

[译文] [Speaker B]: 然后当人们说：“噢等等，实际上你知道我有钱，我待在家里，没出门，没喝酒，我有很多——我有现金，让我买辆车吧。”他们出去买车，车价开始飙升。“噢，让我们重启，噢是的，你能再卖给我那个引擎用的微控制器吗？”对方就像是：“不，我在做一个稍微不同的微控制器，用于——你知道——比如键盘或鼠标或其他什么。”而且这就像是——他们并没有让我措手不及，他们在疫情期间是合作伙伴，对吧？而你（车企）却直接抛弃了我。去你的福特（Ford），或者不管是谁，丰田（Toyota），或者汽车 OEM，你知道你搞砸了那个供应链。所以芯片法案之所以通过，仅仅是因为这件事发生了，人们才惊呼：“天哪，半导体是造不出车的原因。”

[原文] [Speaker B]: If that didn't happen we wouldn't even have the chips act it's like it's like silly so like I don't know like I think you know whereas like and and even though that's what was pitched to all the senators like I know people who were running around Capitol Hill just pushing that narrative and story and that's why it finally got passed in reality it was all for advanced leading edge chips right nothing that goes in a car right and so it's like this like funny thing

[译文] [Speaker B]: 如果那没发生，我们甚至不会有芯片法案。这就像是——这很愚蠢。所以我不知道，我认为你知道，尽管这就是向所有参议员推销的内容——我认识那些在国会山跑来跑去推销这个叙事和故事的人，这就是为什么它最终通过了——但在现实中，这（法案的钱）全是为了先进的尖端芯片，对吧？根本没有用于汽车芯片的东西，对吧？所以这就像是一件滑稽的事情。

[原文] [Speaker A]: so in other words do you think my words my words not yours but is it is it hopeless that the US is going to

[译文] [Speaker A]: 所以换句话说，你是否认为——这是我的话，不是你的——美国要做到这一点是无望的吗？

[原文] [Speaker B]: I'm very optimistic okay i mean do you think there's a world where the US just decides to invest in semiconductor at the scale that you know I thought we just needed a bigger chips act but look Trump's kind of gotten TSMC to promise to invest a fuckload more and they're moving on it right they're like actually like just building it it's like I'm going to tariff the shit out of you unless you build a fab but it's like we'll build a fab and they're building it right now

[译文] [Speaker B]: 我非常乐观。好吧，我的意思是，你认为会有一个世界，美国决定以那种规模投资半导体吗？我认为我们只是需要一个更大的芯片法案。但你看，特朗普某种程度上让 TSMC 承诺投资更多钱，而且他们正在行动，对吧？他们实际上就在建设。就像是“我要对你征收巨额关税，除非你建个晶圆厂”，然后对方说“我们会建晶圆厂”，而且他们现在正在建。

[原文] [Speaker B]: the timelines for fabs just takes forever cuz again it's the most complicated thing in the world the cleanest space in the place in the world is not like a hospital or a biotech lab or whatever it's a semiconductor fab and the most expensive tools in the world are not you know any of these medical tools or whatever it's it's semiconductor tools or it's not a rocket it's a semiconductor tool right like everything you know I describe it as um I remember when I was a kid I was like I want to be a rocket scientist and then I was like oh I want to be a surgeon and I'm like wait chips are like rocket surgery but even cooler right

[译文] [Speaker B]: 晶圆厂的时间线确实很长，因为再次强调，这是世界上最复杂的东西。世界上最洁净的空间不是医院或生物技术实验室，而是半导体晶圆厂。世界上最昂贵的工具不是那些医疗工具，而是半导体工具；它不是火箭，它是半导体工具，对吧？就像所有东西——你知道我把它描述为——我记得小时候我想当火箭科学家，然后我想当外科医生，然后我想：“等等，芯片就像是‘火箭外科手术’（rocket surgery），但甚至更酷”，对吧？

[原文] [Speaker B]: like I think anyways like sort of like there there are fabs being built in America they won't take America to self-sufficiency i don't think that's a relevant i don't think that's a goal relevant like that's relevant right like globalism is generally just good hot take like in terms of economics we'll turn this into a short a YouTube short globalism globalism is good dude you're gonna get me like cancelled i tweeted about ice and it was a complete joke but so many people got mad at me because I can't be you know I'm too I'm too much of a joker you know these are serious things yeah yeah no I know the I know the feeling yes

[译文] [Speaker B]: 无论如何，美国正在建设晶圆厂。它们不会让美国实现自给自足，我不认为这是一个相关的目标。就像——全球主义总体上就是好的（globalism is generally just good）。这可是个暴论（hot take），从经济学角度来说。我们要把这个做成 YouTube Short 短视频：“全球主义是好的”。老兄，你会害我被“取消”（cancelled）的。我在推特上发过关于 ICE（移民局/或指内燃机，语境不明但指敏感话题）的玩笑，结果好多人生我的气，因为我不能——你知道我太爱开玩笑了。你知道这些是严肃的事情。是的，是的，不，我知道那种感觉。

章节 8：资本支出泡沫质疑、电力危机与“汉堡包”水资源迷思

📝 本节摘要：

本节内容极其丰富，涵盖了从社会舆论到硬核基础设施的多个维度。首先，Dylan 触及了公众日益增长的“反 AI”情绪（从新泽西电费上涨到网络垃圾内容）。随后，两人深入探讨了备受关注的“资本支出（Capex）泡沫”问题——Dylan 认为，只要 AI 模型能力持续提升，当前看似激进的投入（如 1000 亿美元收入对应 2500 亿美元基础设施折旧）就是合理的。

对话接着转向物理基础设施瓶颈：美国电网的陈旧与公用事业公司的迟缓迫使数据中心转向天然气发电（如 Elon Musk 的策略）。最后，Dylan 用一个生动的“汉堡包”比喻反驳了“AI 耗尽水资源”的观点，指出一个汉堡的耗水量远超 AI 查询，真正的用水大户是农业而非数据中心。

[原文] [Speaker B]: anyways um I think I think you know I think we are building fabs and I think it's like going to move and now even Elon's talking about building fabs now because he sees the shortages in the world right uh there's a lot of semiconductor related shortages for building out AI and and so I don't think it's hopeless i think I'm like very optimistic that we're going to do more and more and more and maybe this administration threatens tariffs and they get the deals and the next administration comes back with the carrot if it is the Democrats whatever happens I don't know like I was at a comedy club on Sunday night and like he's like "Oh I'm I use Chad GPT." And then like there were a couple people who booed and he's like "Yeah I'm one of those guys i know." And like it's like "Wow people hate AI."

[译文] [Speaker B]: 无论如何，我认为——我认为你知道——我们正在建设晶圆厂，我认为这将会推进。现在甚至 Elon（马斯克）也在谈论建设晶圆厂，因为他看到了世界上的短缺，对吧？建设 AI 存在许多与半导体相关的短缺。所以我不认为这是无望的，我非常乐观，我们会做得越来越多。也许这届政府用关税作为威胁达成了交易，而下届政府——如果是民主党的话——会带着“胡萝卜”（奖励政策）回来。不管发生什么，我不知道。就像周日晚上我在一家喜剧俱乐部，那个人说：“噢，我用 ChatGPT。”然后有几个人发出嘘声。他说：“是啊，我就是那种人，我知道。”这就像是：“哇，人们讨厌 AI。”

[原文] [Speaker B]: And that has has not even started right like the actual impact of AI or like New Jersey power prices are up right uh is it because of a data center new Jersey the governor's election like I think literally fl like there's like an election that changed recently in New Jersey because power prices were up and people blamed a Microsoft Nebius data center in New Jersey for that reason but in reality that data center has nothing to do with power prices going up it's super storm standy like five years ago knocking or whatever how many years ago knocking down the state's electrical infrastructure and then the then improving all these improvements and then those improvements have to be paid by someone and it turns out the consumer has to pay for them with higher power prices right and so like you know like there's like there's a lot like going on in that regard right um that kind of is uh sad um and and people hate AI and they're blaming AI on it and artists hate AI and like you know you see all this deep fake stuff

[译文] [Speaker B]: 而这甚至还没开始，对吧？AI 的实际影响。或者像新泽西州的电价上涨了，对吧？是因为数据中心吗？新泽西州长的选举——我认为真的——最近新泽西的一次选举发生了变化，因为电价上涨，人们因此指责微软（Microsoft）的 Nebius 数据中心。但实际上，那个数据中心与电价上涨毫无关系。那是五年前——或者不管多少年前——的“超级风暴桑迪”（Superstorm Sandy）摧毁了该州的电力基础设施，然后进行了所有这些改进，而这些改进必须由某人买单，结果就是消费者必须通过更高的电价来买单，对吧？所以你知道，在这方面发生了很多事情，这有点——呃——令人悲哀。人们讨厌 AI，并把这归咎于 AI。艺术家讨厌 AI，你也看到了所有这些深度伪造（deep fake）的东西。

[原文] [Speaker B]: and like I think I think it'll be the hottest button issue especially as like we're really getting into like I think last year Google spent $3 billion on Whimo and we're waiting for their guide for this year $3 billion on Whimo taxis but their t their Whimos went from like 300k to like 100k or 90k the new Whimo car and they're going to spend more than three because they've just launched in like four cities now right or five cities and and they're testing it a lot and the same a robo taxi like people are going to hate AI for that reason people are going to hate AI because the slop on the internet people are going to hate AI because you know the perceived job replacement people are going to hate AI for all these reasons and so yeah it's going to be a hot button political issue

[译文] [Speaker B]: 我认为这将成为最热门的话题，尤其是当我们真正进入——我想去年 Google 在 Waymo 上花费了 30 亿美元，我们在等他们今年的指引。30 亿美元用于 Waymo 出租车，但他们的 Waymo 成本从大概 30 万美元降到了 10 万或 9 万美元，也就是新款 Waymo 车。他们将会花费超过 30 亿，因为他们刚刚在四个城市——或者五个城市——推出了服务，并且正在进行大量测试。同样的，Robotaxi（自动驾驶出租车）——人们会因为这个原因讨厌 AI。人们会因为互联网上的垃圾内容（slop）讨厌 AI。人们会因为——你知道——感知到的工作被替代而讨厌 AI。人们会因为所有这些原因讨厌 AI。所以是的，这将成为一个热门的政治话题。

[原文] [Speaker A]: don't you think yeah talking about that so um capex is there a capex bubble are we u investing too much or actually are we investing not enough given what you were saying earlier about uh the the the rate of revenue increase and and therefore implied demand that you expect for this year

[译文] [Speaker A]: 你不这么认为吗？是的，说到这个，那么——资本支出（Capex），是否存在资本支出泡沫？考虑到你之前所说的收入增长率以及因此隐含的今年预期需求，我们是投资过多了，还是实际上投资得不够？

[原文] [Speaker B]: i'm obviously a maxi i think we're going to need a lot of infra and I think I'm literally paid to like analyze the supply chain and do consulting like that's what my company does so like obviously I'm very biased i think I think we're pretty good at calling when when things go down though right before like a part of the supply chain reb anyways you know again going back to the economics of it it's north of hundred billion dollars of revenue exiting this year for AI from a base of you know sub1 billion gen AI from a base because ads and stuff is like already a multiundred billion dollar AI industry right you know go back to 2023 it was like less than a billion right and 2024 I don't know exactly what number maybe let's call it 10 and 25 was maybe like 30 40 it'll be north of 100 easily

[译文] [Speaker B]: 我显然是个“最大化主义者”（Maxi），我认为我们需要大量的基础设施。而且我想我是真的拿钱来分析供应链和做咨询的，这就是我公司做的事，所以显然我有很大的偏见。不过我认为我们在预测行情下跌方面也做得不错，比如在供应链某部分反弹之前。无论如何，回到经济学上，今年 AI 领域的退出收入（revenue exiting this year）将超过 1000 亿美元。这是从一个——你知道——生成式 AI（Gen AI）不到 10 亿美元的基础上发展起来的。因为广告之类的本身已经是一个数千亿美元的 AI 产业了，对吧？回到 2023 年，它可能不到 10 亿。2024 年我不知道确切数字，也许叫它 100 亿。2025 年也许是 300、400 亿。它将很容易超过 1000 亿。

[原文] [Speaker B]: if you're talking about hundred billion of revenue let's say at a 50% gross margin so that's $50 billion of gross profit um and $50 billion of COGS that $50 billion of COGS needs to run on infra which cost roughly if a five if you're talking about fiveyear depreciation call it $250 billion right of infra for hundred billion of revenue mhm okay what is what is the actual spend on AI infra this year it's going to be like it's I mean it depends on what layer if you're talking about energy those are longer lived assets and all these other things right um data centers are longer lived assets the chips are not as much people are putting capex down um and the hyperscalers capex is going to be like $500 billion this year or something like this

[译文] [Speaker B]: 如果你说的是 1000 亿美元的收入，假设 50% 的毛利率，那就是 500 亿美元的毛利，以及 500 亿美元的销货成本（COGS）。这 500 亿美元的 COGS 需要在基础设施上运行。如果按 5 年折旧计算，大概需要 2500 亿美元的基础设施来支撑 1000 亿美元的收入。嗯，好的。那么今年实际的 AI 基础设施支出是多少？这取决于你在哪一层。如果你说的是能源，那是长寿命资产；数据中心也是长寿命资产；芯片则不完全是。人们正在投入资本支出。超大规模云厂商（Hyperscalers）今年的资本支出大概会是 5000 亿美元左右。

[原文] [Speaker B]: and then besides them there's also a lot more hyp uh capex elsewhere um and so you know is it a bubble i mean theoretically like you know it's twice as much as it should be but it's also like well no there's an R&D component to this and the excess spent that wasn't revenue generating last year is what led to models being so good this year um and led to like everyone who can using cloud code and like that changing their life this is like it's not a bubble right i don't think it's a bubble yet um I think if AI model progress stops and that's the main thing right the moment model progress stops all the spending is for not but so far we've had consistent improvement as you put in more compute you get more performance and better models

[译文] [Speaker B]: 除了他们之外，其他地方还有更多的资本支出。所以你知道，这是泡沫吗？我是说理论上，它比“应该”有的水平高出两倍。但也像是否定的，因为这里有一个研发（R&D）的成分。去年那些没有产生收入的超额支出，正是导致今年模型变得如此出色的原因，也导致了像所有能用 Claude Code 的人都在用它，并因此改变了生活。这不像是一个泡沫，对吧？我不认为它目前是一个泡沫。我认为如果 AI 模型进步停止了——那是关键所在，对吧？模型进步停止的那一刻，所有的支出就都白费了。但到目前为止，我们看到了持续的改进：你投入更多的算力，你就获得了更高的性能和更好的模型。

[原文] [Speaker A]: yeah model performance being the lagging indicator of hardware progress or data center yeah of of capex right yeah ultimately the capex that Microsoft spent in 2024 for OpenAI is what results in in 2025 for OpenAI Cory or whoever is what results in their models being so good this year same with Enthropic and Amazon Google and their models now being so good now is is that capex and actually they still haven't paid for those chips yet because those chips are still have a useful life for another few years right I think model progress is very clear um the moment that stops happening right if we hit a wall there's no new research directions um then then it's cooked yeah right

[译文] [Speaker A]: 是的，模型性能是硬件进步或数据中心——是的，资本支出的滞后指标，对吧？最终，Microsoft 在 2024 年为 OpenAI 花费的资本支出，才是导致 2025 年 OpenAI——或者不管是谁——模型变得如此出色的原因。Anthropic 和 Amazon、Google 也是一样，他们的模型现在这么好，是因为那笔资本支出。实际上他们还没付清那些芯片的钱，因为那些芯片还有几年的使用寿命，对吧？我认为模型进步是非常清晰的。一旦那停止发生，对吧？如果我们撞墙了，没有新的研究方向了，那就完蛋了（cooked）。

[原文] [Speaker A]: but oh that $50 billion capex was spent in year one what about energy in the in the data center world you had this fun post about the gas replacement for for energy so is uh is AI basically uh uh destroying the grid

[译文] [Speaker A]: 但是，噢，那 500 亿美元的资本支出是在第一年花的。那么能源呢？在数据中心领域，你有一篇关于用天然气替代能源的有趣文章。所以，AI 基本上是在摧毁电网吗？

[原文] [Speaker B]: it would if the utilities were willing to let it but I think the utilities are so slow and dumb that they don't want to not destroy but like expanding the grid yeah um I think the US could have a way better grid but we just don't want to like no one's made the effort or initiative you know there's not enough power america's not built power for 50 years really right it's like converted from coal to gas and like things like this but like really just have not built wholesale new power on a large scale

[译文] [Speaker B]: 如果公用事业公司（utilities）愿意让它摧毁的话，它会的。但我认为公用事业公司太慢太蠢了，他们不想——不是不想摧毁，而是不想扩张电网。是的，我认为美国本可以拥有一个好得多的电网，但我们就是不想，好像没人做出努力或倡议。你知道电力不足，美国实际上已经 50 年没有建设电力了，对吧？就像是从煤炭转换为天然气之类的，但真的没有大规模地建设全新的电力。

[原文] [Speaker B]: and there have been a lot of times where the industry blew up right independent power producers IPs have blown up multiple times in the 2010s when uh Korean and Japanese investors like flooded the market with because they saw such a good return there or before in the early 2000s power was growing a little bit for a little bit and so people overbuilt on power so power industry has been burned a couple times but no one really built power and then you've got data centers now all of a sudden coming online and going from 2% to 10% of the US grid in just a handful of years and so you've got this humongous humongous change in the industry

[译文] [Speaker B]: 行业曾多次崩溃，对吧？独立发电商（IPPs）在 2010 年代多次崩溃，当时韩国和日本投资者因为看到了良好的回报而涌入市场。或者更早在 2000 年代初，电力增长了一点点，人们就过度建设了电力。所以电力行业被烧伤了几次，但没人真正建设电力。然后现在数据中心突然上线，在短短几年内从占美国电网的 2% 增长到 10%。所以你在行业中面临着巨大的、巨大的变化。

[原文] [Speaker B]: we don't have the labor right i think ultimately that's the biggest problem is the equipment and the labor and equipment is basically you know again labor and time takes time to build a factory so you can build the things i think the equipment side of things will be solved like more reasonably and one one example was like gas right people initially thought oh you can only use like the two vendors right uh Seammens or G Vernova for gas turbines but they have the they have the best ones the most efficient ones it's like okay well like okay also Mitsubishi exists and they're ramping up production fast oh Ducson and Korea exist and they're ramping up production fast oh actually I can just take Cumins engines right like you know if you've ever like ridden a pickup truck or like you know like diesel trucks like everyone loves Cumins right you know you see the Ram on the street and has the Cumins like badge it's like it's like a that's like an aura symbol for a certain kind of redneck from South Georgia which I have a little bit of anyways I I don't have a I don't have a truck i have though um

[译文] [Speaker B]: 我们没有劳动力，对吧？我认为最终最大的问题是设备和劳动力。而设备基本上——你知道，又是劳动力和时间，建工厂需要时间才能造出东西。我认为设备方面的问题会得到更合理的解决。一个例子就是天然气，对吧？人们最初认为，噢，你只能用那两家供应商的燃气轮机，西门子（Siemens）或 GE Vernova，因为他们有最好的、最高效的。但这就像是，好吧，三菱（Mitsubishi）也存在，而且他们正在快速提升产量。噢，韩国的斗山（Doosan）也存在，他们也在快速提升产量。噢，实际上我也可以直接用康明斯（Cummins）的引擎，对吧？你知道如果你开过皮卡或柴油卡车，大家都爱康明斯，对吧？你在街上看到道奇公羊（Ram），上面有康明斯的标志。那就像是——那就像是某种来自乔治亚州南部的红脖子（redneck）的光环象征——我自己也有一点这种特质。无论如何，我没有卡车。

[原文] [Speaker B]: but anyways like the there's like all these engines like people are figuring out how to make the equipment you know solar sucks it's too intermittent wind sucks it's too intermittent nuclear sucks it takes forever to build coal sucks it's way too dirty how do you make power for data centers besides gas and like okay the grid's not willing to put the gas on your site right that's what Elon did now everyone's doing it right this other cool post just uh last week or two weeks ago that was about water consumption uh did you want to talk to that yeah yeah so there's this annoying thing where everyone's like "Oh AI is using all the water oh wow AI and data centers are going to like use up all the water and now we don't have any water." And it's like that's so silly

[译文] [Speaker B]: 无论如何，就像所有这些引擎，人们正在想办法制造设备。你知道太阳能很烂，太间歇性了；风能很烂，太间歇性了；核能很烂，建设时间太长；煤炭很烂，太脏了。除了天然气，你还能怎么为数据中心发电？好吧，电网不愿意把天然气送到你的站点，对吧？这就是 Elon（马斯克）所做的，现在每个人都在这样做，对吧？还有那个很酷的帖子，就在上周或两周前，关于水资源消耗的。你想谈谈那个吗？是的，是的。有一个很烦人的说法，每个人都在说：“噢，AI 正在用光所有的水，噢哇，AI 和数据中心要耗尽所有的水了，现在我们没有水了。”这就像是——这太愚蠢了。

[原文] [Speaker B]: uh water is a distribution problem not a like we don't have enough problem right like you look at California so California has shitloads of water but people decide to make oat milk which consumes like 1,000x the water of like anything else like regular milk even and and cows obviously eat a you know consume a lot of water um but anyways like you know data centers consume very little water actually right so the US grid will get to like 10% of power by like 28 27 is data centers for water consumption it's not even going to crack 1% by the end of the decade and what was the metric um and so so the the comparison we made is because like you know it was a bit of a shit post but it was like serious research yeah basically like we were doing serious research because we keep getting this like question and debunking it and we would do it seriously but then I was like no no no this is like too like complicated like let's make it very simple so I was like "Guys why don't we just compare it to like hamburgers right

[译文] [Speaker B]: 水是一个分配问题，而不是我们不够用的问题，对吧？你看加利福尼亚，加州有大量的水，但人们决定生产燕麦奶，它的耗水量是其他任何东西的 1000 倍，甚至比普通牛奶还多。显然牛也消耗大量的水。无论如何，数据中心实际上消耗的水非常少，对吧？美国电网中数据中心的用电量到 2027、2028 年可能会达到 10%，但在水资源消耗方面，到本十年末甚至不会突破 1%。那个指标是什么来着？所以我们做的比较是——你知道那有点像是个恶搞帖子（shit post），但也是严肃的研究。是的，基本上我们在做严肃的研究，因为我们要不断反驳这个问题。本来我们想严肃地做，但我后来觉得，不不不，这太复杂了，让我们把它变简单点。所以我说：“伙计们，我们干脆把它和汉堡包做比较吧？”

[原文] [Speaker B]: cuz cuz you know I've heard that argument from some like vegetarian people before or some Hindus or like I'm Hindu myself although you know and I I I do eat beef sometimes but you know like I I'm Hindu but like you know so so we made this comparison to hamburgers right hamburgers require a shitload of water cuz cows you know when to for them they require a ton of water and when a cow's taking a lot of water it's not the cow itself it's all the feed you're feeding them right because no one grass feeds their cows you know and just lets the rain take care of the grass they like either rain the the grass or most likely they do mass industrial farming of corn soybean alalfa etc which uses shitloads of water right like you know or like almond milk like uses tons and tons of water like produce is like the main user of water i think the uh metric was the entirety of Elon Musk's Colossus data center right uses as much water as two and a half in-n-outs um because that's you know you do the calculation on how many how many b what's the average revenue per in-n-out and how many hamburgers does that translate to right

[译文] [Speaker B]: 因为你知道，我之前从一些素食主义者或者印度教徒那里听过这个论点。我自己也是印度教徒，虽然你知道我有时也吃牛肉，但我确实是印度教徒。所以我们做了这个与汉堡包的比较，对吧？汉堡包需要大量的水，因为牛——你知道养它们需要吨级的水。当一头牛消耗大量水时，不是牛本身喝掉的，而是你喂给它们的饲料，对吧？因为没人是用草喂牛然后指望雨水来浇灌草的。他们要么给草浇水，或者更可能是进行玉米、大豆、紫花苜蓿等的大规模工业化种植，这使用了大量的水，对吧？或者像杏仁奶，使用了成吨成吨的水。农产品才是水的主要使用者。我想那个指标是：Elon Musk 的整个 Colossus 数据中心，其用水量仅相当于 2.5 家 In-N-Out 汉堡店。因为你知道，你计算一下 In-N-Out 的平均收入是多少，那转化成多少个汉堡，对吧？

[原文] [Speaker B]: if everyone's ordering like a combo right okay let's ignore the drink let's ignore the fries let's just talk about the hamburger let's ignore the bread which does use have grain let's just do the meat and the cheese and all of a sudden all this water is there's so much water right like a single query like all of your AI usage from chat GBT of the average user is like a hamburger right like it's like okay this is nothing right you know because these things are the data centers actually are like they're mostly closed loops and like sure they evaporate some water for like cooling reasons but like by doing evaporative cooling they're using less power right and that's actually better for the environment than uh than not using evaporative cool there's all all these reasons why this myth or hoax of AI of AI using all the water is just nonsense right

[译文] [Speaker B]: 如果每个人都点一个套餐，好吧，我们忽略饮料，忽略薯条，只谈汉堡。我们忽略面包——虽然它确实含有谷物——我们只算肉和奶酪，突然间所有的水——有那么多的水，对吧？比如一个单一查询，或者一个普通用户在 ChatGPT 上的所有 AI 使用量，就像一个汉堡包，对吧？这就像是，好吧，这根本不算什么，对吧？因为这些东西——数据中心实际上大多是闭环的。当然，为了冷却原因它们会蒸发一些水，但通过蒸发冷却，它们使用的电力更少，对吧？实际上这比不使用蒸发冷却对环境更好。有所有这些原因说明“AI 正在用光所有的水”这个神话或骗局纯属胡扯（nonsense）。

[原文] [Speaker B]: like Meta's data center in Louisiana is getting protested because the water it's it's going to be the largest data center in the world it's going to be like four or five gigawatts at least announced so far we're tracking some other ones that are that may be as big or bigger uh but Meta is getting protested because the local population around that area is like "Oh the water's dirty it's because of this meta data center." And like there's these trucks on these big trucks on these back roads that used to be empty completely they're just like mad and annoyed about that right but at the end of the day what actually made the water dirty is that that's an area where you go fracking like fracking is absurdly worse and almost all of that gas is being shipped to an LG terminal and being shipped to Asia like you know you know like Japan or Taiwan or China or Korea and some Europe as well right like like actually all of this water is dirty because of regulation fracking like I support fracking by the way but you know that's that's an insane take too maybe um but like water usage is is is like not a relevant argument

[译文] [Speaker B]: 比如 Meta 在路易斯安那州的数据中心正在遭到抗议，因为水。那将是世界上最大的数据中心，至少目前宣布的是 4 或 5 吉瓦（gigawatts），我们正在追踪其他一些可能一样大或更大的。但 Meta 遭到抗议是因为那里的当地居民说：“噢，水脏了，是因为这个 Meta 数据中心。”而且那些偏僻的道路上现在有了大卡车，以前那是完全空的。他们只是对此感到生气和恼火，对吧？但在一天结束时，真正让水变脏的是那个地区在进行水力压裂（fracking）。水力压裂要糟糕得荒谬，而且几乎所有的天然气都被运往 LNG 终端，运往亚洲，比如日本、台湾、中国或韩国，也有一些运往欧洲。实际上所有的水变脏是因为监管和水力压裂——顺便说一句，我支持水力压裂——但这可能也是个疯狂的观点。但是，水资源使用并不是一个相关的论据。

章节 9：资本支出 (Capex) 泡沫与循环融资争议

📝 本节摘要：

针对市场关于“循环融资”（Circular Financing）的质疑——即 NVIDIA 投资 CoreWeave 或 OpenAI，而这些公司反过来购买 NVIDIA 芯片——Dylan 进行了详细辩护。他认为这种安排在当前阶段是合理的：大型科技公司（如 Microsoft）急需算力，但建设者（如 CoreWeave）缺乏足够的信用记录来获得巨额贷款，因此巨头们通过签署长期合同或提供担保（Backstop）来协助融资。同理，OpenAI 用股权融资换取现金来租赁算力，本质上是预支未来收入以换取当下的基础设施建设，这在快速增长的科技行业中虽显激进（funky），但并非不可接受的泡沫。

[原文] [Speaker A]: yeah I wanted to come back quickly to uh that um Nvidian core wave deal that you mentioned as we sort of close the discussion on uh on capex and a and a bubble it seems like there is circular deals but also a lot of debt kind of like flushing around so I don't know the specifics of of that deal but like I did hear variations of this where effectively you have a large player guaranteeing the debt being the last recourse uh for a lot of infrastructure build is sort of uh this plus the whole like oracle commitment there there is a fragility into this whole thing that can be a little unnerving what do you make of it

[译文] [Speaker A]: 是的，我想快速回到你提到的 NVIDIA 和 CoreWeave 的交易，作为我们结束关于资本支出（Capex）和泡沫讨论的一部分。这看起来像是存在“循环交易”（circular deals），而且还有大量的债务在流动。我不知道那笔交易的具体细节，但我确实听到过各种版本，大意是一个大型玩家在为债务提供担保，作为许多基础设施建设的最后追索权（last recourse）。这加上整个 Oracle 的承诺，这整件事里有一种脆弱性，让人有点不安。你怎么看？

[原文] [Speaker B]: i think it's like completely fine and I think like people are like freaking out and making narratives where there really is shouldn't be one it's like well okay Google doesn't have enough data center capacity they need people to build data centers but no one can build a data center because they don't have the capital like don't have you know many cases capital is not the you know they don't have capital right or like no one will give them a loan because they don't trust some random fucking company

[译文] [Speaker B]: 我认为这完全没问题。我觉得人们在歇斯底里，编造一些本来就不该存在的叙事。这就像是，好吧，Google 没有足够的数据中心容量，他们需要别人来建数据中心。但没人能建，因为他们没有资本——比如你知道，在很多情况下资本不是——他们就是没钱，对吧？或者没人愿意给他们贷款，因为银行不信任某些随机的不知名公司。

[原文] [Speaker B]: and it's like but then Google's like well no we've due diligence to them we think they can build it here we'll like even guarantee we'll buy the thing or start using it once they build it you know just having a customer alone spoken for it was enough right um in the case of Cororewave they were actually able to no backs stop right right they were able to just say "Hey hey look here's our Microsoft contract for this many GPUs i want to put in that data center that data center that data center here's the contract for renting those GPUs i want to hire these people i want to do this." No one will like they don't have any money but then they were able to like have it work out because they were able to get people to lend to them

[译文] [Speaker B]: 然后 Google 会说：“不，我们对他们做了尽职调查，我们认为他们能建成。甚至我们可以担保，一旦他们建成，我们就会买下或者开始使用它。”你知道，仅仅有一个客户承诺使用就已经足够了，对吧？在 CoreWeave 的案例中，他们甚至不需要（某些类型的）担保，对吧？他们可以直接说：“嘿，看，这是我们的 Microsoft 合同，我们要在这个数据中心、那个数据中心放这么多 GPU。这是租赁那些 GPU 的合同，我要雇这些人，我要做这个。”没人会——虽然他们没钱，但他们最终能搞定，因为他们能让别人借钱给他们。

[原文] [Speaker B]: i think like Cororeweave did that and there was no circular financing but that was when there was like the scale of investment was like single digit billions or less than a billion right now the scale of investment is hundreds of billions um and so the question is like oh well if I want data center capacity how do I how do I get data center capacity i just go to everyone who's going to build it looks smart is smart enough to do it but can't afford to do it and tell them I'll I'll take it and in fact I won't just take it i'll go to your debtor and be like I'll guarantee you yeah because you know obviously you're a new company i've vetted you but the debtor hasn't and so you know like you know you know they don't want me to just be able to walk away because like in the Microsoft Corwave deals Microsoft could have walked away if Corwe fucked it up right there's no I mean yeah there's there's always like uh sort of like cancellation or whatever possibilities and so this is just a further form of guarantee

[译文] [Speaker B]: 我认为 CoreWeave 做到了这一点，那时还没有所谓的循环融资，但那时的投资规模只是几十亿或者不到十亿。现在的投资规模是数千亿。所以问题变成了：噢，如果我想要数据中心容量，我该怎么获得？我就去找每一个看起来聪明、有能力做这件事但没钱做的人，告诉他们：“我会包下它。”而且事实上，我不只是包下它，我还会去找你的债权人（debtor），对他们说：“我给你担保。”是的，因为显然你是一家新公司，我审查过你，但债权人没有。所以你知道，他们不想让我能轻易一走了之。因为在 Microsoft 和 CoreWeave 的交易中，如果 CoreWeave 搞砸了，Microsoft 本来是可以一走了之的，对吧？我是说，是的，总是存在取消或其他可能性。所以这（现在的做法）只是一种更进一步的担保形式。

[原文] [Speaker B]: um as far as on like a lot of these back stops as far as on like Oracle getting the money and then OpenAI getting money and Nvidia you know paying and it's a whole circular it's kind of nonsense because it's like Nvidia's getting equity in OpenAI they're basically saying "Hey every gigawatt you buy we'll also buy some equity." Yeah right okay well cool now Nvidia owns an asset which they think is valuable openai right open AAI is turning around and is like trying to rent those uh use the equity they buy what do they what was their use of equity people's cash pay isn't that great right it's mostly just 99 plus% of their spend at the company is probably just compute

[译文] [Speaker B]: 至于许多这类担保，至于 Oracle 拿到钱，然后 OpenAI 拿到钱，NVIDIA 付钱，说这是整个“循环”的——这其实有点胡扯（nonsense）。因为这就像是 NVIDIA 获得了 OpenAI 的股权，他们基本上是在说：“嘿，你每买一吉瓦的算力，我们也会买一些股权。”是的，好吧，酷，现在 NVIDIA 拥有了一项他们认为有价值的资产——OpenAI，对吧？OpenAI 转过身来，试图租赁那些——使用他们通过股权融资拿到的钱。他们的资金用途是什么？人员的现金薪酬并没有那么高，对吧？他们在公司的支出可能有 99% 以上都只是算力。

[原文] [Speaker B]: uh so so sort of like it's like okay well then I I raise this money i'm going to do the the whole thing I explained earlier right year one and two I lose money year three four five I hope to make money on it right um and open has been doing that right so I'm going to Okay I'm going to go out there i've raised $50 billion i've raised $10 billion i'm going to raise it i'm going to rent a cluster for five years for $65 billion and I've rented that contract and now I only have enough to pay for the first year to be clear but I think you know you trust me Oracle you think I'm going to grow and you think I'll be able to pay for it oracle's like "Yeah or if you're not I think I'll be able to sell it to someone else." So like okay cool i'm going to spend $50 billion this year yep to build that data center and and and this these this is like for a gigawatt um and so is it like circular that OpenAI is every amount of GPUs they consume and gives an investment that investment is turned around to pay for the first year of the rent to the cluster um or second year then first two years go you know it's sort of like it's fine yeah yeah like it's like it's like it is a little bit funky but like I don't think it's a big deal

[译文] [Speaker B]: 所以这就像是，好吧，那我筹集了这笔钱，我要做我之前解释过的整件事，对吧？第一年和第二年我亏钱，第三、四、五年我希望从中赚钱，对吧？OpenAI 一直在这样做。所以我要去——好吧，我要去筹集 500 亿美元，或者 100 亿美元。我要去筹钱，我要租一个集群 5 年，价值 650 亿美元。我签了租赁合同，但我现在只够付第一年的钱——说清楚点。但我想你知道，你信任我，Oracle。你认为我会增长，你认为我能付得起。Oracle 会说：“是的，或者即使你付不起，我想我也能把它卖给别人。”所以，好吧，酷，我今年要花 500 亿美元来建那个数据中心。这就像是为了那一吉瓦。所以，OpenAI 消耗多少 GPU，NVIDIA 就给多少投资，然后这笔投资转头被用来支付集群第一年的租金，或者第二年——这是循环的吗？你知道，这还好。是的，是的，这确实有点激进（funky），但我认为这并不是什么大事。

章节 10：模型的未来：AI 编程 (Claude) 与技术文化

📝 本节摘要：

作为访谈的最终章，话题转向了 AI 模型的未来应用与硅谷独特的硬核文化。Dylan 极其看好 Anthropic 的 Claude（文中口误为 Cloud Code）等模型对编程工作的重塑，指出即便是没有任何技术背景的分析师现在也能通过 AI 完成复杂的代码任务，这可能导致中级（L4）工程师的需求大幅下降。他还预测了 OpenAI 和 Anthropic 在下一代模型（如 Opus 4.5）上的竞争态势。最后，对话以轻松的轶事结束：Dylan 分享了与 AI 界名人室友（如 Sholto Douglas 和 Dwarkesh Patel）同住的趣事，展现了旧金山科技圈“生活即工作，工作即娱乐”的独特氛围。

[原文] [Speaker A]: yeah love it contrary intake maybe let's finish with the models and the software side of things we talked extensively about hardware and supply chain and all the things i get a sense that you are super super bullish on uh what's happening next in in AI your roommate Schulto I assume was the roommate that you were talking about earlier on this pod effectively making the point that we're just starting to scratch the surface and there was so much low hanging fruit around you know RL and all the things you were in Silicon Valley circles is that is that your sense as well and what are you tracking on the model side one thing is like you know simple stuff like uh GitHub commits other things are like what's the amount of usage how much are people using like all these sorts of things

[译文] [Speaker A]: 是的，我喜欢这种反直觉的观点（contrarian take）。也许让我们以模型和软件方面的内容来结束。我们广泛讨论了硬件和供应链以及所有这些事情。我感觉你对 AI 接下来发生的事情超级超级看好。你的室友 Sholto——我猜就是你之前在播客里提到的那位室友——实际上提出了一个观点，即我们才刚刚开始触及表面，在强化学习（RL）和所有这些方面还有很多唾手可得的果实（low hanging fruit）。你身处硅谷圈子，你也是这种感觉吗？你在模型方面追踪什么？一方面是像 GitHub 提交量这种简单的东西，另一方面是比如使用量是多少，人们用了多少，像所有这类事情。

[原文] [Speaker B]: i think there's so many different alternative data sources for tracking AI model progress area tokconomics uh token economics tokconomics and so that's like an entire practice for us are you rebranding the term from crypto i yeah I don't believe in crypto people like I've always hated them um so now you're taking the term yeah yeah and Jensen's used it now so I've like I've convinced him to use the word he's used it as sovereigns and so I think I think we've won that's awesome congratulations i've said it to him we've written it in articles it's an entire practice of consulting that I just I started in like 23 2023 uh was token economics and we've been trying to build out these like you know

[译文] [Speaker B]: 我认为有太多不同的替代数据源来追踪 AI 模型的进展了。“Area Tokenomics”（面积代币经济学）——呃，代币经济学（Token Economics），Tokenomics。这对我们来说是一个完整的业务板块。你是在从加密货币（Crypto）那里重塑这个术语吗？是的，我不相信加密货币圈的人，我一直很讨厌他们。所以你现在把这个词夺过来了？是的，是的，Jensen（黄仁勋）现在也用了，所以我好像说服了他使用这个词，他把它用在“主权”（sovereigns）这一概念上，所以我认为我们赢了。太棒了，恭喜。我对他说过，我们在文章里也写过。这是一个我在 2023 年才开始的完整咨询业务板块，就是代币经济学，我们一直在试图构建这些——你知道。

[原文] [Speaker B]: but basically I think the main things are like people who don't code can use cloud code now right i think people don't understand that like even if you don't code you've never had any training in software development you've never take had a job as a software developer you can code let's take an an example of what one of the one of the analysts at my company did right comes from a engineering background but on like semiconductor systems right uh like worked on mechanical systems worked on these sorts of things and they coded this thing which was they wanted to do an analysis of area of clean rooms right clean rooms are the building that you the fab has all the tools in the most complicated kind of building in the world has every all sorts of chemical systems and all this area of that a company who builds systems builds these systems and revenue of that company right

[译文] [Speaker B]: 但基本上，我认为主要的事情是那些不会写代码的人现在可以使用 Claude Code（注：原文听录为 cloud code，结合上下文实指 Anthropic 的编程工具 Claude）了，对吧？我认为人们不明白，即使你不写代码，你从未受过任何软件开发培训，从未做过软件开发的工作，你也可以写代码。举个例子，我公司的一位分析师做了什么。他有工程背景，但是是在半导体系统方面，比如从事机械系统之类的工作。他写了一个程序，目的是分析“洁净室”（clean rooms）的面积。洁净室是晶圆厂放置所有工具的建筑，是世界上最复杂的建筑，拥有各种化学系统。他想分析一家建造这些系统的公司的洁净室面积与该公司收入的关系，对吧？

[原文] [Speaker B]: and so it was like okay uh we have this fab data set pointed it at it was like hey here's this fab data set what's the square footage of all of them and we have this like thing that we built which uh just pulls with cloud code separately which for data centers and and and fabs and everything else just calculates the area of something from a from a satellite image right very simple so we have the square footage of all these things points at that here's the company name okay go find the filings so it dig dug through all these filings it it pulled the data right okay great now told it to um compare these two make a chart great oh wait there's this like weird inflection oh that's because they bought a company five years ago can you do a proform of this analysis without those financials of that of that company they acquired okay great and then like we were able to like like figure out an investment case for our clients as well as like you know some other interesting details from someone who's never really coded just using clawed code and it like doing this all and this is like not even their and it wrote the note and they just like they didn't even like work on this full-time for like 3 hours right they just told the model and would go work on other things and told the model and worked on other things they just did this

[译文] [Speaker B]: 于是就像是，好吧，我们要把这个晶圆厂数据集指给它看，就像是：“嘿，这是晶圆厂数据集，它们的平方英尺面积是多少？”我们有一个用 Claude 独立构建的东西，可以针对数据中心和晶圆厂等从卫星图像计算面积，非常简单。所以我们有了所有这些东西的平方英尺面积。指向那个，这是公司名称，好吧，去找财报文件（filings）。所以它挖掘了所有的财报，提取了数据。好的，太棒了。现在告诉它：“比较这两个，做一个图表。”太棒了。“噢等等，这里有个奇怪的拐点。”“噢，那是因为他们五年前收购了一家公司。你能做一个剔除那家被收购公司财务数据的预估（pro-forma）分析吗？”“好的，太棒了。”然后我们就能为客户弄清楚一个投资案例，以及其他有趣的细节。这一切都来自一个从未真正写过代码的人，只是使用 Claude。它不仅做了所有这些，甚至还写了注释。而他们甚至没有全职做这个，大概只花了 3 个小时，对吧？他们只是告诉模型，然后去忙别的事，再回来告诉模型，再去忙别的事。他们就这样搞定了。

[原文] [Speaker B]: people don't understand that like the skill sets that like I think like if you go talk to an analyst right a very junior analyst at any right whether it's venture or especially growth venture or public markets or private equity their their job is like finding data cleaning it making charts it's like this is cloud code now you don't need junior analysts just like a lot of companies have stopped hiring L4 engineers because it's useless why would I hire an L4 engineer i just tell Claude to do it you you sort of like have this has happened and this is a really big like shift I guess like is that like low-level knowledge work just doesn't matter right why would I why would I use Excel when I can just tell Claude to manipulate CSVs why would I use Word when Claude will just generate the markdown and I can copy and paste the markdown directly into our WordPress and then you know and that WordPress is fully formatted now and it's like oh my god like what's the point of Word right um and what's the point of doing all sorts of stuff

[译文] [Speaker B]: 人们不理解技能组合的变化。我认为如果你去和任何一个分析师谈——无论是风险投资（VC），特别是成长型风投，还是公开市场或私募股权（PE）的非常初级的分析师——他们的工作就是找数据、清洗数据、制作图表。这现在就是 Claude 的工作了。你不需要初级分析师了，就像很多公司已经停止招聘 L4（中级）工程师一样，因为没用了。我为什么要雇一个 L4 工程师？我只要告诉 Claude 去做就行了。这种事情已经发生了，这是一个非常巨大的转变。我猜低级别的知识工作变得不再重要了，对吧？既然我可以告诉 Claude 去处理 CSV 文件，我为什么要用 Excel？既然 Claude 可以直接生成 Markdown，我可以把 Markdown 直接复制粘贴到我们的 WordPress 里，而且格式完全正确，我为什么要用 Word？这就像是，天哪，Word 还有什么意义？做各种杂事还有什么意义？

[原文] [Speaker B]: i think when we look at model progress that's just for Opus 4.5 open's new model I think will be better than Opus 4.5 and it's coming like somewhat soon in Marchish um time frame i maybe February Marchish but yeah um because OpenAI has a better RL stack than Enthropic today it's just their pre-trained models suck compared to Enthropic's pre-training right and so like if they catch up a lot on pre-training and keep their better RL stack they would actually have a model that's much better right flip side Google has a better pre-trained model than Anthropic or OpenAI but their RL stack sucks so if they catch up on RL like these models are going to get ridiculously and then Anthropic is obviously advancing as well right and so and then and then you look across the ecosystem everyone's advancing really fast progress these moments are happening right you know chat GPT was a moment gibbly was a moment those were more consumer those were less like I mean there chat GBT is everyone using it for work too but like I think cloud code is like a new moment right 4.5 on cloud code is a new moment where the way you work has forever changed

[译文] [Speaker B]: 我认为当我们看模型进展时，这还只是 Opus 4.5（注：可能是指 Claude 3.5 或未来的 4.5 预测）。我认为 OpenAI 的新模型会比 Opus 4.5 更好，它可能会在三月左右——也许二月或三月——推出。因为 OpenAI 目前拥有比 Anthropic 更好的强化学习（RL）技术栈，只是他们的预训练模型相比 Anthropic 的预训练模型很烂，对吧？所以如果他们在预训练上大幅追赶，并保持更好的 RL 技术栈，他们实际上会拥有一个好得多的模型。反过来说，Google 拥有比 Anthropic 或 OpenAI 更好的预训练模型，但他们的 RL 技术栈很烂。所以如果他们在 RL 上追上来，这些模型会变得强得离谱。然后 Anthropic 显然也在进步。所以你看整个生态系统，每个人都在飞速进步。这些时刻正在发生，对吧？你知道 ChatGPT 是一个时刻，Ghibli（可能指某种图像模型或口误）是一个时刻，那些更多是消费者层面的。我是说 ChatGPT 大家也用于工作，但我认为 Claude Code 是一个新的时刻，对吧？基于 Claude Code 的 4.5 是一个新的时刻，你工作的方式已经永远改变了。

[原文] [Speaker B]: and so now we're trying to force everyone in my company there's 54 people here i think like half of them have coded the other half we're trying to force them to use like cloud code and it could be like oh well actually you come from a consult a semiconductor consulting background oh you come from like a semiconductor like engineering of like package oh you worked in a fab right like these kind of people they're using cloud code now right and and their productivity is being boosted and it's like you know workspace cloud workspace is new it sucks compared to cloud code but it'll get there right he he he said he coded it entirely in cloud code you know that right or that was on your pod right yeah yeah so like um you I've heard that and I think maybe that might have been from your pod uh original uh disclosure my pod was before that but yes oh okay okay it was I had as the guy on my pod subsequently said that okay i think it's like a brand new age and and like there's so much low hanging fruit as Shto said on the episode when he was here there's so much low hanging fruit

[译文] [Speaker B]: 所以现在我们试图强迫我公司的每个人——这里有 54 个人，我想一半人会写代码，另一半我们正试图强迫他们使用 Claude。不管你是来自半导体咨询背景，还是来自半导体封装工程背景，或者你在晶圆厂工作过，这些人都开始用 Claude 了，对吧？他们的生产力得到了提升。你知道 Claude Workspace 是新的，比起 Claude Code 还很烂，但它会好起来的。他说他完全是用 Claude 写的代码，你知道吧？或者那是在你的播客上说的？是的，是的，所以我听说过那个。我想那可能来自你的播客——原谅我披露一下，我的播客在那之前，但是是的。噢，好的，好的，那就是我在我的播客上采访的那个人后来那么说的。好的。我认为这是一个全新的时代，就像 Sholto 在他来这期节目时说的那样，有太多唾手可得的果实了。

[原文] [Speaker A]: and speaking of Schultoe we both agreed that he was a a perfect specimen dude I' I've been I'm straight but I've been accused of being uh homosexual which is perfectly fine for for how much I like praise this man because like think about it right he's like 6'4 he's like really good-looking he's like Australian accent sounds amazing like you've heard his I I have like a annoying voice probably his voice sounds amazing he's absurdly good at coding he was an Olympian level fencer like like he picks up any sport he's really good at it right because he's athletic it's like "Holy crap you're a specimen." Yeah yeah this clip and sent him for sure yeah it must be uh you know I guess uh may maybe some people don't follow the playbyplay on on Twitter and like don't haven't haven't heard of like the fact that all of you guys are roommates or you roommate with Scholto and then with Dwarish and Darkish is like the podcasters podcaster so it must be absolutely What's a podcasters podcaster mean uh the podcaster that other podcasters uh aspire to to to become or learn from yeah yeah

[译文] [Speaker A]: 说到 Sholto，我们都同意他是个完美的“样本”（specimen）。

[Speaker B]: 兄弟，我是直男，但我因为太赞美这个男人而被指责是同性恋——这完全没问题。因为你想想看，对吧？他身高 6 尺 4（约1.93米），长得真的很帅，有澳洲口音，听起来很迷人。你知道，我的声音可能很烦人，但他的声音听起来很棒。他写代码强得离谱，还是奥运级别的击剑手。就像他拿起任何运动项目都能玩得很好，因为他很有运动天赋。就像是：“天哪，你简直是个完美的样本。”

[Speaker A]: 是的，是的，这段剪辑肯定要发给他。是的，我想可能有些人没有在 Twitter 上实时追踪，没听说过你们所有人都是室友的事实——或者是你和 Sholto 以及 Dwarkesh（注：Dwarkesh Patel）是室友。而 Dwarkesh 是那种“播客主的播客主”（podcaster's podcaster）。所以那肯定绝对是——“播客主的播客主”是什么意思？

[Speaker A]: 呃，就是其他播客主渴望成为或者想要学习的那种播客主。是的，是的。

[原文] [Speaker B]: his his when he's preparing you know it's like he's he's he's he's so locked in and he prepares so hard for interviews it's great no he's he's just uh incredible and then and then he might only say like a hundred words on the episode but he's prepared so hard and then like I think people just realized oh wow he's not just like you know it's like oh he just has good guests no no no like he's preparing really hard but you can't tell if you're not like realizing that and then once he started writing more and he started writing more people like oh wow he's actually really really smart it's like yeah cuz he's studying like crazy like it's like "Oh I'm interviewing an AI researcher who worked on this i'm gonna try and train a freaking model." Yeah right it's like that's the level of like commitment he goes to when he records this stuff

[译文] [Speaker B]: 他的——当他准备的时候，你知道就像是他完全进入状态（locked in），他为采访准备得如此努力，这太棒了。不，他真的很不可思议。虽然他在那一集里可能只说了一百个字，但他准备得非常充分。我认为人们才刚刚意识到，噢哇，他不仅仅是——你知道，不仅仅是“噢，他只是请到了好嘉宾”。不不不，他准备得非常努力，但如果你没意识到那一点你是看不出来的。一旦他开始写更多东西，人们就会觉得：“噢哇，他实际上真的非常非常聪明。”那是当然，因为他学习起来像疯了一样。就像是：“噢，我要采访一位研究这个的 AI 研究员，我要试着自己训练一个该死的模型。”是的，对吧？这就是他在录制这些内容时投入的程度。

[原文] [Speaker A]: what do you guys talk about when you bump into each other is that is that AI non-stop or you talk about everything but AI

[译文] [Speaker A]: 你们互相碰面时聊些什么？是不停地聊 AI，还是聊除了 AI 以外的所有事情？

[原文] [Speaker B]: with Shoto it's like the Age of Empires game you know because we we got super into it for a bit we talked only about that in his RTS that he made uh with with with Dwarash it's I mean it's all sorts it's like normal roommate stuff it's like "How's your dating life?" "Oh okay you went on a date it wasn't well it didn't go well." "Okay well okay." Yeah you know like oh you know like that's me that's me you know my days don't go well no I'm just kidding um or like it's like oh you want to like have dinner we can invite a few friends like yeah great or like you know it's like all sorts of like normal stuff too um al obviously we also do talk about a lot about tech right like we are like this is our lives um and tech is the most fun thing

[译文] [Speaker B]: 和 Sholto 在一起就是聊《帝国时代》（Age of Empires）游戏，你知道，因为我们有一阵子超级迷这个，我们只聊那个以及他做的那个 RTS 游戏。和 Dwarkesh 在一起，我是说各种各样的事，就像正常的室友那样。比如：“你约会生活怎么样？”“噢，好吧，你去约会了，不太顺利？”“好吧，好吧。”是的，你知道就像——噢，那是说我，那是说我，你知道我的约会不太顺利，不，我开玩笑的。或者就像是：“噢，你想吃晚饭吗？我们可以叫几个朋友。”“好啊，太棒了。”或者你知道，也是各种各样的正常琐事。呃，显然我们也确实聊很多关于科技的事，对吧？因为这就是我们的生活，而且科技是最有趣的事情。

[原文] [Speaker A]: awesome well great great San Francisco lore uh Dylan thank you so much uh that was absolutely fabulous really enjoyed it learned a lot so really appreciate uh your coming on the pub thank you so much

[译文] [Speaker A]: 太棒了，这就是精彩的旧金山传说（SF lore）。Dylan，非常感谢你，这绝对是太精彩了，真的很享受，学到了很多，所以真的非常感谢你来上这个播客。非常感谢。