Gavin Baker - AI Market Jitters - [Invest Like the Best, EP.485]

频道: Invest Like the Best with Patrick O'Shaughnessy
视频: https://traffic.megaphone.fm/CLS6565465184.mp3
原文语言: en
统计: 共 74 轮 · Patrick O'Shaughnessy 6 · Gavin Baker 64

[0:00]

Ramp is the only platform built to make your finance team leaner, faster, and better,saving businesses 5% annually on average so you can stay focused on growth.Ramp customers grew revenue 3.2 times faster than the average American business.Visa, Vercel, Cursor, Stripe, Notion, 11Lab, Shopify, and 70,000 other businesses all runon Ramp.Mine does too, and so should yours.Learn more at ramp.com slash invest.OpenAI, Cursor, Anthropic, Perplexity, and Vercel all have something in common.They all use WorkOS.To achieve enterprise adoption at scale, you have to deliver on core capabilities like SSO,SCIM, RBAC, and Audit Logs.Instead of spending months building these mission-critical capabilities yourself, you can just use WorkOSAPIs to gain all of them on day zero.That's why so many of the top AI teams you hear about already run on WorkOS.WorkOS is the fastest way to become enterprise-ready and stay focused on what matters most,your product.Visit WorkOS.com to get started.Felix by Rogo is a personal finance agent that turns a single prompt into finished,client-ready work using your firm's own templates, context, and standards.Send Felix an email like,take these comments and turn them for me,

Ramp 是唯一一个专为让你的财务团队更精简、更快、更好而打造的平台，平均每年为企业节省 5% 的开支，让你能专注于增长。Ramp 的客户收入增速是美国企业平均水平的 3.2 倍。Visa、Vercel、Cursor、Stripe、Notion、ElevenLabs、Shopify 以及另外 70,000 家企业都跑在 Ramp 上。我自己的公司也是，你的也该是。到 ramp.com/invest 了解更多。

OpenAI、Cursor、Anthropic、Perplexity 和 Vercel 有一个共同点：它们都用 WorkOS。要在规模上拿下企业客户，你必须交付几项核心能力——单点登录（SSO）、账号自动开通与同步（SCIM）、基于角色的权限控制（RBAC）和审计日志（Audit Logs）。与其花几个月自己造这些关键能力，不如直接用 WorkOS 的 API，第 0 天就全部拥有。这就是为什么你听说过的那些顶级 AI 团队大多已经跑在 WorkOS 上。WorkOS 是让你「企业级就绪」、并把精力持续放在最重要的东西——你的产品——上的最快路径。访问 WorkOS.com 开始使用。

Rogo 出品的 Felix 是一个个人金融智能体，它能把一句提示词变成可以直接交付给客户的成品，用的是你所在机构自己的模板、上下文和标准。你给 Felix 发一封邮件，比如「把这些意见帮我改进去」，

[1:11]

or update my tracker with the context of these emails.And Felix sends back finished PowerPoint decks, Excel models, and sourced research.Felix works the way your team already does,delivering work quickly and accurately around the clock.Learn more at rogo.ai slash Felix.Hello and welcome, everyone.

或者「用这些邮件里的上下文更新我的追踪表」。Felix 就会给你回一份做好的 PowerPoint 演示稿、Excel 模型和带出处的研究材料。Felix 按你团队本来的工作方式干活，全天候、快速而准确地交付。到 rogo.ai/felix 了解更多。

大家好，欢迎收听。

[1:30] Patrick O'Shaughnessy

I'm Patrick O'Shaughnessy, and this is Invest Like the Best.This show is an open-ended exploration of markets, ideas, stories, and strategiesthat will help you better invest both your time and your money.If you enjoy these conversations and want to go deeper,check out Colossus, our quarterly publication with in-depth profilesof the people shaping business and investing.You can find Colossus along with all of our podcasts at Colossus.com.Patrick O'Shaughnessy is the CEO of Positive Sum.All opinions expressed by Patrick and podcast guests are solely their own opinionsand do not reflect the opinion of Positive Sum.This podcast is for informational purposes onlyand should not be relied upon as a basis for investment decisions.Clients of Positive Sum may maintain positions in the securities discussed in this podcast.To learn more, visit PSUM.VC.Gavin, it's only been two months.Like the model release cycles, the gap between our podcast episodes are shortening.We're basically, you and I are basically on a model release cadence at this point.Well, I was sensitive to criticism that I think somebody pointed outthat our podcasts were coincident with like local market peaks.And nobody can say that after this.

我是 Patrick O'Shaughnessy，这里是《Invest Like the Best》。本节目是一场开放式的探索——关于市场、想法、故事和策略，帮你更好地投资你的时间和金钱。如果你喜欢这些对话、想更深入，可以看看 Colossus，我们的季刊，里面有对那些塑造商业与投资的人物的深度特写。Colossus 和我们所有播客都能在 Colossus.com 找到。

Patrick O'Shaughnessy 是 Positive Sum 的 CEO。Patrick 和播客嘉宾表达的所有观点仅代表他们个人，不代表 Positive Sum 的观点。本播客仅供参考，不应作为投资决策的依据。Positive Sum 的客户可能持有本播客中讨论的证券。了解更多请访问 PSUM.VC。

Gavin，才过去两个月。就像模型发布周期一样，我们两期播客之间的间隔也在缩短。我俩现在基本上是按模型发布的节奏在录了。

是啊，之前有人指出我们的播客总是正好撞在市场的局部高点上，我对这个批评还挺敏感的。这回没人能这么说了。

[2:44] Gavin Baker

What's on your mind?It's been a crazy...Yeah, I would describe July as 2022 in a month.Yeah.There are some fundamental negatives which we should talk.But on the whole, the ballots of fundamentals, I think, is improving significantly.Loads of AI names are down 50, 60% from their highs.We'll call it 40 to 60% in a month in a straight line.And I asked you before we started, you've been out here for the summer.Have you heard a single negative quantitative metric about AI?

你最近在想什么？

最近真是疯了……是啊，我会把七月形容成「一个月里过完了 2022 年」。

是啊。

有一些基本面上的负面因素，我们应该聊聊。但整体上，基本面的天平我认为在显著改善。一大批 AI 标的从高点跌了 50%、60%——就说一个月内直线下跌 40% 到 60% 吧。

开录之前我问过你，你整个夏天都待在这边（硅谷）。你有没有听到过任何一个关于 AI 的负面量化指标？

[3:22] Gavin Baker

A single instance of deceleration?Nothing.Nothing.In fact, every metric is accelerating.And to your point, not just blind optimism from people excited about AI.Yeah.But like, here's some data that they can show you from their different vantage points.Absolutely.I mean, however you cut it, whether you cut GPU availability, whether you cut GPU rental pricing,whether you cut like the spot price of DRAM this month, token growth, everything is actually accelerated.And I do think a big part of the problem is, one, the market does not have visibility intoanthropic open AI.And then I would say these open source inference clouds that monetize inference here in America,Fireworks, Base 10, Modal together.And the picture looks very different when you see that.Because open source has accelerated massively because of GLM 5.2, KBK3.And then Nematron continues to kind of chug a log.We had a great, very small American open source model release.OpenAI has accelerated.An anthropic continues to grow really strongly and is almost certainly pumping out significantamounts of free cash flow.And I just think if, you know, there's this chart that everybody looks at of semiconductorcash flow going like this and hyperscale free cash flow going like that, and you're missing

哪怕一次减速的迹象？

没有，一个都没有。事实上每一个指标都在加速。而且按你说的，这不是那些对 AI 兴奋的人的盲目乐观。

是啊。

而是「这是我从我的观察点能拿给你看的数据」。

完全正确。我是说，你怎么切都一样——不管你切 GPU 可得性、切 GPU 租赁价格、切 DRAM 这个月的现货价、还是切 token 增长，所有东西实际上都在加速。

我确实认为，问题很大一部分在于：第一，市场看不见 Anthropic 和 OpenAI 的情况。其次，还有这些在美国把推理变现的开源推理云——Fireworks、Baseten、Modal、Together。你把这些看进去，画面就非常不一样了。

因为开源已经大规模加速了，靠的是 GLM 5.2 和 Kimi K3。然后 Nemotron 也在稳步往前拱。我们还有一个很棒的、非常小的美国开源模型发布。OpenAI 加速了。Anthropic 继续强劲增长，而且几乎可以肯定正在产出大量自由现金流。

我就觉得，大家都在看的那张图——半导体现金流这样往上走、超大规模云厂商（hyperscaler，指微软/谷歌/亚马逊/Meta 这类自建巨型数据中心的公司）的自由现金流那样往下走——你漏掉了

[4:49] Gavin Baker

these private companies.But I also think that that chart misses something very important, which is just that you haveeveryone in 24 and 25, even if you were really bullish, you thought that GPU prices, if you'rereally bullish, you thought they would decline slowly.If you're bearish, you thought it would decline precipitously.I don't think anyone in 24 or 25 thought that the prices of old GPUs would be going vertical.Everybody thought, hey, we're going to be smart.We're going to sign these long-term contracts.And to some degree, like a lot of the neoclouds had to do that because they needed an offtakeagreement to finance the GPUs.And so essentially, you have the contracted base of installed compute trading at a massivediscount to the current spot market.And as those contracts roll off and compute gets repriced higher,spot can decline and compute will still get repriced higher.I think you're going to see a lot of acceleration that's going to answer these ROI questions.You've started to see that this quarter.If we look at operating cash flow, not free cash flow, operating cash flow from Microsoft,Meta, and Amazon has reported accelerated from 28 to 32.There are some actually pretty big unusual items now, like these hyperscalers.

这些私营公司。

但我也认为那张图漏掉了非常重要的一点：在 24 年和 25 年，哪怕你非常看多，你也只是认为 GPU 价格会缓慢下跌；如果你看空，你认为它会断崖式下跌。我不认为 24 年或 25 年有任何人想到过，老一代 GPU 的价格会直线向上。

所有人当时都想「嘿，我们要聪明点，我们要签这些长期合同」。而且某种程度上，很多新云厂商（neocloud，专做 GPU 出租的新兴云）不得不这么做，因为他们需要一份承购协议（offtake agreement，即客户承诺长期包用的合同）才能给 GPU 融到资。

所以本质上，已装机算力的合同价基数，相对当前现货市场是在大幅折价交易的。随着这些合同到期、算力按更高的价格重新定价，现货价可以下跌，而算力仍然会被重新定价到更高。

我认为你会看到大量加速，而这会回答那些关于投资回报率（ROI）的质疑。这个季度你已经开始看到了。

如果我们看经营性现金流——不是自由现金流，是经营性现金流——微软、Meta 和亚马逊报出来的（同比增速）已经从 28% 加速到 32%。现在这些超大规模云厂商确实有一些相当大的非经常性项目。

[6:13] Gavin Baker

They always seem to have billions of dollars of legal expenses that are unusual, mostlyfines to the EU.But there is an unusual amount of one-timers this quarter.And if you adjust for that, we went from 28 to 35.And that's a material acceleration at this scale.And that's really before they start to light up the Rubens, which will come at a meaningfulpremium, before these contracts reprice.It's been a challenging month.Is it helpful to kind of like walk through the month, how we got here?

他们似乎总有几十亿美元的「非经常性」法务开支，主要是欧盟的罚款。但这个季度的一次性项目多得不寻常。如果你把这些调整掉，就是从 28% 加速到 35%。在这个体量上，这是实质性的加速。

而且这还是在他们开始点亮 Rubin 之前——Rubin 会带来相当可观的溢价——也是在这些合同重新定价之前。

这个月挺难熬的。要不要把这个月是怎么走到今天的捋一遍？

[6:45] Gavin Baker

So first, Meta is going to rent out compute.And this is seen as like very bearish.They have excess capacity.They're going to cut CapEx.This is a disaster.This is not at all what it was.They just reported they didn't cut CapEx.What it was is they saw SpaceX have a big installed base of compute and sell some big trading optimizedclusters into the market at a truly massive premium to these contracted rates.And at least analysts like that, they saw an opportunity.There's a lot of speculation they're going to raise capital.So maybe what they're thinking is like, hey, we will show on a small chunk of capacity thatwe can generate really strong IRRs.Then we're going to raise equity capital and we'll be off to the races and probably raise CapEx.That doesn't look like that's what they're doing.But nonetheless, the market sold off because it interpreted this very negatively.And I was really sure it wasn't negative.A lot of telemetry into Meta's CapEx plans.None of that telemetry had shifted at all.If anything, they're continuing to get more aggressive.And then shortly after that, they released their best model in a long time, Muse 1.1,which is actually a very good model.I mean, it was overshadowed by Grok 4.5, but it was a good model.

首先，Meta 要把算力租出去。这被看成极度看空的信号：他们产能过剩，他们要砍资本开支（capex），这是灾难。

完全不是那么回事。他们刚发的财报显示他们并没有砍资本开支。真实情况是，他们看到 SpaceX 有一大批已装机算力，并把一些大的训练优化集群按远高于合同价的巨大溢价卖进了市场。至少分析师喜欢这个。他们看到了机会。

市场上有很多猜测说他们要融资。所以也许他们的想法是「嘿，我们先用一小块产能证明我们能做出非常强的内部收益率（IRR），然后再融股权资本，接着就一路狂奔，很可能还会上调资本开支」。看起来他们做的并不是这个。

但不管怎样，市场因为把这件事解读得非常负面而抛售了。而我非常确定这不是负面的。我对 Meta 的资本开支计划有很多信号来源，所有这些信号一点都没变。要说有变化，那就是他们还在变得更激进。

然后没过多久，他们发布了很久以来最好的模型 Muse 1.1，这确实是个非常好的模型。当然它被 Grok 4.5 盖过了，但它是个好模型。

[8:06] Gavin Baker

Way better than anything in two years.So just no chance they're taking their foot off the gas.Then Kimmy comes out.And then there's this huge freak out about open source.And at the same time, this silicon data token index kind of dips and flattens.And the two are connected.What the silicon data token index captures is mix.And they don't see all the tokens.But because of GLM 5.2 and then Kimmy, although it took a while to layer in, there's kind of a mix shift in this data from more expensive frontier tokens,which probably have an inference margin.We can debate whether it's 80, 90, or 95.But super high towards open source tokens.And for whatever reason, the market thought this was negative.But the reality is a token is a token.And you need the exact same amount of compute to make a token all else equal.It takes the same amount of flops, the same amount of memory, the same amount of watts.Tokens are not equal, but broadly speaking, all open source taking share does is take margin dollars out of the frontier model layer.There is elasticity, thereby driving token demand.You need more demand for compute.And the margins, you know, Anthropic and open source, they all run on the same underlying cloud providers who charge the same amount of compute.

比过去两年的任何东西都好得多。所以他们绝无可能松油门。

然后 Kimi 出来了。接着就是关于开源的巨大恐慌。与此同时，SiliconData 的 token 指数有点下探然后走平了。这两件事是连着的。

SiliconData token 指数捕捉到的其实是「结构（mix）」。而且他们也看不到所有 token。但因为 GLM 5.2、然后是 Kimi——虽然叠加进来花了一段时间——这个数据里出现了一种结构性转移：从更贵的前沿模型 token（它们的推理毛利率是 80%、90% 还是 95% 可以争论，但反正超高）转向开源 token。

不知为何，市场认为这是负面的。但现实是：token 就是 token。在其他条件相同的情况下，生产一个 token 需要的算力是一模一样的——同样多的浮点运算（flops）、同样多的内存、同样多的瓦特。

token 并不完全等价，但大体上说，开源抢份额唯一做的事，就是把利润从前沿模型这一层里拿走。而且存在需求弹性，反而因此推高了 token 需求。你需要更多的算力需求。

至于利润率，Anthropic 和开源都跑在同样的底层云上，这些云对算力收同样的钱。

[9:31] Gavin Baker

So you're literally just taking margin from frontier models and essentially driving more margin dollars into the AI infrastructure layer.That was the catalyst.This combination of things.Well, yeah.Jensen is the world's largest supporter of open source.He is like a super idealistic guy.He's a patriotic American.I think he always does what's right.But does it really stand to reason that Jensen would be the world's biggest supporter of open source if it was bad for his business?

所以你实际上只是把利润从前沿模型那里拿走，本质上把更多利润美元推进了 AI 基础设施这一层。

这就是导火索，是这几件事叠在一起。

嗯，是啊。黄仁勋（Jensen）是全世界最大的开源支持者。他是个超级理想主义的人，是个爱国的美国人。我认为他总是做正确的事。

但是，如果开源对他的生意不利，黄仁勋会成为全世界最大的开源支持者，这真的说得通吗？

[10:01] Gavin Baker

You know, he'd still support if it was the right thing for the world.And by the way, I think open source is really important to worlds where there's just one or two dominant frontier models that charge like 90% margins.It's not good for humans.It might not be good for society.And I think we want a lot of models, as we've discussed before.So then it's like, okay, the market digests that and comes to work with it.Then China has a DUV machine.You know, everybody's in these baskets.This causes a huge sell-off in semi-cap equipment.And then we get to what I think is, in a lot of ways, the real concern, which is real yields have gone up, which makes sense.You know, we're investing a lot to fund this investment.And for sure, credit is an increasing part of it, even if the majority is still funded out of operating cash flows.So real yields go up and spreads widened.Meta priced a bond last week, and it did not price where you would think a Meta bond would price.And this just shows that the credit market...NVIDIA CDS was blowing out.All of these CDS...CDS for everybody is blowing out.And, you know, very smart private capital people just like that, hey, this is just exactly what you'd expect.

你知道，如果这对世界是对的事，他还是会支持。

顺便说，我认为开源在「只有一两个主导性前沿模型、还收 90% 毛利」的世界里非常重要。那对人类不好，对社会可能也不好。我认为我们想要很多模型，这个我们之前聊过。

所以接下来是：好，市场消化了这件事，慢慢接受了。然后中国有了 DUV（深紫外）光刻机。大家都在这些「篮子」里，这引发了半导体设备（semi-cap）板块的大幅抛售。

然后我们就来到了我认为在很多意义上才是真正令人担心的事：实际收益率上去了。这说得通——我们在投入大量资金来给这轮投资融资。而且可以肯定，信贷在其中的占比在上升，即使大部分资金仍然出自经营性现金流。

所以实际收益率上去了，信用利差走阔了。Meta 上周发了一笔债，定价并不在你以为 Meta 的债该定的位置。这就说明信用市场……

NVIDIA 的信用违约互换（CDS，本质是给债券买的违约保险，价格越高说明市场越担心违约）在大幅走阔。所有人的 CDS 都在走阔。

而且，很聪明的私募资本的人就会说，嘿，这完全就是你该预期到的。

[11:11] Gavin Baker

These are just banks hedging their commitments.But nonetheless, it doesn't look good.And these are undeniable facts.CDS is up.Spreads widened.Real yields are up.That would be really, really scary if we needed debt to finance this buildout.And that's where I think it's this differential between spot and contract pricing for the installed base of compute is so important.It's so important to understand what the financing will be like for the next six months or something.The degree to which this buildout is going to require credit.Right.Which would be the classic capital cycle.Absolutely.Overextend ourselves with debt, and that's where things get scared.100%.And then debt-fueled buildouts, they demand immediate repayment.So if supply and demand get a little bit out of whack, things can unwind very, very quickly.That's what happened to the internet.If one believes, as I do, rightly or wrongly, after this month, I'm super open.You know, I'm looking like I've been pressure testing all of these.And, like, I really went deep on credit because, hey, this is real.It's undeniable.And if we need credit to fund this buildout, this is a significant negative.And if you model it out, if you look at the amount of gigawatts that are supposed to come on and consensus estimates for hyperscalers, they're effectively modeled.

这些只不过是银行在对冲自己的承诺敞口。

但不管怎样，这看着不好看。而且这些是无可否认的事实：CDS 上去了，利差走阔了，实际收益率上去了。

如果我们需要债务来给这轮建设融资，那这些会非常非常吓人。这就是为什么我说，已装机算力的现货价与合同价之间的差如此重要。理解未来六个月左右的融资环境会是什么样，这非常重要。

这轮建设会在多大程度上需要信贷。

对。这就会是典型的资本周期。

完全正确。用债务把自己撑过头，那才是事情变吓人的地方。

百分之百。而且债务驱动的建设是要求立刻还钱的。所以只要供需稍微失衡一点，事情就可能非常非常快地崩掉。互联网泡沫就是这么破的。

如果一个人像我一样相信——不管对错——经过这个月，我是超级开放的。你知道我一直在压力测试所有这些。我在信贷上真的挖得很深，因为，嘿，这是真的，这无可否认。如果我们需要信贷来给这轮建设融资，这就是重大利空。

而如果你把模型建出来，如果你看看预计要上线的吉瓦数和超大规模云厂商的一致预期，它们实际上是被这样建模的——

[12:33] Gavin Baker

And these are gigawatts of Blackwell and Rubin.Rubin being NVIDIA's next chip.Blackwell being the current chip.They are essentially modeled to monetize roughly at the rate of Ampere, which is two generations behind.Not at Hopper, but Ampere.So there's $1.3 to $1.4 trillion in hyperscale operating cash flow.If you just assume, I think it's very unlikely they monetize at the rate of Ampere.We could go into why.Some of it comes from just seeing what is happening on the ground with demand here for real quantitative metrics.But like, let's just say they monetize at a discount to current Blackwells.Then it's more like $2 trillion of operating cash flow.And that kind of takes $700 billion of credit demand out, ironically, as that improves all the credit ratios.As these installed bases of compute reprice, we're going to continue accelerating.Consensus is modeling at a deceleration, which I think is unlikely.Then the credit metrics look better.And then all of a sudden, it gets easier to finance with credit.Now, whether they choose to do that or not, we'll see.This is all a little bit, you know, I think we spoke.Two months ago.No, but the time before that about kind of the risks of a Blackwell air pocket.

而这些是 Blackwell 和 Rubin 的吉瓦。Rubin 是 NVIDIA 的下一代芯片，Blackwell 是当前这代。

它们在模型里的变现速率大致被假设成和 Ampere 一样——Ampere 是往前数两代。不是 Hopper，是 Ampere。

所以（在这个假设下）超大规模云厂商的经营性现金流是 1.3 万亿到 1.4 万亿美元。如果你只是假设——我认为他们按 Ampere 的速率变现是非常不可能的，我们可以展开说为什么，其中一部分理由就来自看到这边真实的需求量化指标——但我们就说，他们按低于当前 Blackwell 的速率变现好了，那经营性现金流就更像是 2 万亿美元。

这就一下子把 7000 亿美元的信贷需求给拿掉了。讽刺的是，这反而改善了所有的信用比率。

随着这些已装机算力重新定价，我们会继续加速。一致预期建的是减速，我认为这不太可能。那样的话信用指标看起来会更好。然后突然之间，用信贷融资反而变容易了。

至于他们会不会选择这么做，我们走着瞧。

这些多少有点……我想我们聊过。

两个月前。

不，是再上一次，我们聊过 Blackwell「气穴（air pocket）」的风险。

[13:50] Gavin Baker

Oh, yes.Where you're spending hundreds of billions of dollars on Blackwells.They're mostly being used for trading initially.Trading does not generate a return.This could be a risk.You actually really saw that in the first quarter.I think one reason to the podcast two months ago, I got comfortable with that risk was just that you were seeing such incredible things out of Anthropic.And then it's like, okay, well, the market's kind of going to look past this.And it did look past it in April, in May, in June.And then in July, because of this kind of confluence of things, stopped looking past it.Just as the operating cash flow started to really accelerate.And this is just a fact.It is accelerating at big scale.You know, like Microsoft, they brought on a huge slug of capacity in the month of June.That didn't even show up in the second quarter.So essentially, what this all comes down to is, do you believe that the quantitative demand signals seeing on the ground here in Silicon Valley from private companies are going to continue,such that the installed base of compute reprices higher as contracts roll off, operating cash flows go up, and you could fund most of this out of operating cash flows.

哦，对。

就是你花几千亿美元买 Blackwell，而它们一开始主要用于训练。训练不产生回报。这可能是个风险。

你在第一季度确实非常明显地看到了这一点。

我想两个月前那期播客里，我对这个风险变得安心的一个原因，就是你当时看到 Anthropic 那边出来的东西太惊人了。然后就是，好吧，市场大概会看穿这一层。它在 4 月、5 月、6 月确实看穿了。

然后到了 7 月，因为这一堆事情凑到一起，市场不再看穿了——而恰恰就在这个时候，经营性现金流开始真正加速。这就是事实，它正在大体量地加速。比如微软，他们在 6 月这一个月里上线了一大批产能，这在第二季度里都还没体现出来。

所以本质上，这一切归结为一个问题：你相不相信，我们在硅谷从私营公司身上看到的这些量化需求信号会持续下去，使得随着合同到期、已装机算力被重新定价到更高，经营性现金流上升，而你可以用经营性现金流覆盖这里面的大部分。

[15:07] Gavin Baker

Maybe all of it.Like if it reprices at current rates, you could probably fund all of it for the next several years.It has been a very unusual episode in the market.We should talk about what the fundamentals are that are getting better that I'm talking about.Technicians would say it's actually in 22.Okay, the market is worried about a recession, rates going up, inflation.That's what the market was worried about in 22.You knew exactly what it was.Okay, deep seek, you know what it's worried about.Liberation day, you know what it's worried about.There's something very clear.And in a weird way, that's comforting.Pretty sure.And here, you know, we talked about a lot of specific things, but it just feels all those specific things, with the exception of credit, are just kind of ridiculous.And so the fact that it is still going down, a technician would say, hey, that's a little scary.It's definitely the bullet you don't see that gets you.You know, I think we've talked before about how, like, I think the three most important words in investing aren't margin of safety, but I don't know.But just, you've been out here for two months.I've been out here.I literally spoke to a company this morning who rented a cluster of several, and this is one of the sexiest startups that people want to be in business with.

甚至可能是全部。如果它按当前价格重新定价，未来好几年你大概能全部用经营性现金流覆盖。

这是市场里非常不寻常的一段。

技术派会说这其实就是 2022 年。好吧，2022 年市场担心的是衰退、利率上行、通胀——那时候市场担心的就是这些，你清清楚楚知道是什么。DeepSeek 那次，你知道它在担心什么。「解放日」（Liberation Day，指特朗普宣布大规模关税那天）那次，你知道它在担心什么。总有一个非常清楚的东西。

而且以一种奇怪的方式，那反而让人安心。

非常确定。

而这一次，我们聊了很多具体的事，但感觉那些具体的事——除了信贷——都有点荒谬。

所以，市场居然还在跌，技术派会说，嘿，这有点吓人。真正打中你的永远是你没看见的那颗子弹。

我想我们以前聊过，我认为投资里最重要的三个词不是「安全边际」，而是「我不知道」。

但就说，你在这边待了两个月，我也在这边。我今天早上刚跟一家公司聊过，他们租了一个几千卡的集群——这还是那种人人都想跟他们做生意的、最性感的创业公司之一。

[16:25] Gavin Baker

And they had rented a cluster of several thousand black wells, and we'll just call it somewhere in the mid $2 per GPU hour.They're renting the exact same size cluster, essentially identical in every way, B200s, no differences.And they're hoping, seven months later, to pay just under $4 today.Like, that's pretty crazy, because again, you would expect a really gentle decline in prices would be bullish.Instead, we're up, depending on the starting point, 50% to 60% in six or seven months.There have been so many anecdotes like that.Like, I think one of the inference clouds, I think it was based in, I'm not sure, they went on a podcast and they essentially said, we are planning to pay 100% more for black wells when our contract expires.And that just means that essentially all the hyperscalers are under-earning.My main kind of mission out here this week...Is like pressure test?

他们当时租了一个几千张 Blackwell 的集群，价格就算它每 GPU 小时 2 美元出头吧。他们现在要租一个完全同样规模的集群，B200，各方面基本一模一样，没有任何差别。

而他们希望——七个月之后——能付到接近 4 美元。这挺疯狂的，因为你本来会预期价格温和下跌才是利多；结果我们是涨的，取决于起点，六七个月里涨了 50% 到 60%。

这样的实例太多了。我想有一家推理云——我不太确定是不是 Baseten——他们上过一期播客，基本上说：我们合同到期时，准备为 Blackwell 多付 100% 的价钱。

这就意味着，本质上所有超大规模云厂商现在都在「少赚」。

我这周来这边的主要任务……

就是压力测试？

[17:26] Gavin Baker

Yeah.Yeah.Tell me something negative.Like, you know, the question I asked you, is there one negative quantitative metric you've heard?That's been what I've been asking everyone.The main thing people are saying is the third-party data suggests that the entropic curve started to go off of its trajectory a little bit.That's like the only thing that I...I think that may very well be true.But then you have OpenAI and open source massively accelerating.Yeah, the complexes.And if you look at the sum, it is net accelerating.Like, I think open source is a little bit of a...You know, they talk about dark matter in the universe.Like, open source is kind of dark matter to the public markets.It's hard for public markets to measure it.But like, if you just track what these inference clouds are saying, people saying things on podcasts or people saying things in meetings, they're not audited financials.And demand is clearly accelerating, which makes sense because you had this huge capability leap with GLM 5.2 and Kimi K3, which I think we're going to see continue.I think you're going to see NVIDIA bring Nemotron steadily closer to the frontier.But man, it has been a humbling, challenging month.But just, it's also like, wow, I've kind of pressure tested every assumption.

对，对。给我讲点负面的。就像我问你的那个问题：你有没有听到过哪怕一个负面的量化指标？我一路都在问所有人这个。

大家主要说的是，第三方数据显示 Anthropic 的曲线开始稍微偏离它原来的轨迹了。这基本上是我听到的唯一一条……

我觉得这很可能是真的。但同时你有 OpenAI 和开源在大规模加速。

是啊，整个复合体。

如果你看总量，它是净加速的。

我认为开源有点像……你知道人们说宇宙里的暗物质。开源对公开市场来说有点像暗物质，公开市场很难测量它。

但如果你就去追踪这些推理云在说什么、人们在播客上说什么、人们在会议里说什么——那些不是审计过的财报——需求显然在加速。这说得通，因为 GLM 5.2 和 Kimi K3 带来了巨大的能力跃升，我认为这个还会继续。我认为你会看到 NVIDIA 把 Nemotron 稳步推近前沿。

但天哪，这真是让人谦卑而难熬的一个月。不过也是那种「哇，我把每一个假设都压力测试了一遍」的感觉。

[18:39] Gavin Baker

The underlying fundamentals are improving.NVIDIA is actually, as we record this, at its lowest forward PE of the last 10 years.Crazy.The only time the SIMIs have been cheaper were Liberation Day and DeepSeek.And those were kind of B bottoms.And that means to you just that the market thinks they're significantly over-earning?

底层基本面在改善。

在我们录这期节目的当下，NVIDIA 的前瞻市盈率处于过去 10 年的最低点。

太疯狂了。

半导体股比现在更便宜的时候，只有「解放日」和 DeepSeek 那两次。而那两次基本上都是 V 型底。

那对你来说是不是就意味着，市场认为它们严重「多赚」了？

[19:02] Gavin Baker

Yeah, the market 100% thinks they're significantly over-earning.And we need to be humble.Maybe they are.Maybe they are.But my kind of mission out here this week was to look for negative data points as hard as I could.And normally, you come to Silicon Valley, and there's a mixture of here's something negative,here's something positive, da-da-da.Unbalance, it's positive.Tech, it creates value over time.But I haven't been able to find one that is like a quantitative metric.But that anthropic third-party data, I would say that seems to be hotly contested by the anthropic shareholders who are chopping at the bit to tell you what they know.We're also very scared they're not going to get an IPO allocation.If it gets back to the company that they're the ones who said, actually, things are great.You know, you can just see anthropic shareholders.Like, they want to be like, it's not true.You know?

是的，市场 100% 认为它们严重「多赚」了。我们需要保持谦卑。也许它们真的是。也许真的是。

但我这周来这边的任务，就是拼命去找负面的数据点。

通常你来硅谷，会听到一堆混合的东西：这有个负面的，这有个正面的，等等等等。整体上是正面的——科技会随时间创造价值。

但我没能找到任何一个负面的量化指标。

不过关于 Anthropic 的那份第三方数据，我会说，它似乎被 Anthropic 的股东们激烈反驳，那些人急得要命想告诉你他们知道的情况。

他们同时又很害怕拿不到 IPO 的配售——万一被公司知道是他们出来说「其实情况好得很」。你能想象 Anthropic 股东的样子，他们特别想说「这不是真的」。你懂吧？

[20:00] Gavin Baker

It's hard for me to believe that open source and open AI have accelerated to the extent they did.But yeah, anthropic is clearly in the pole position.And oh, by the way, Grok and Cursor have also, you can see from third-party data, like,July was a pretty transformational month with Grok 4.5, Grok builds coming out.So it has been a tricky month.And I have a friend, I have a friend of fidelity, who just says the way to have navigated the last three years is just do the dumbest, most superficial thing as quickly as possible and just cycle between them.What is that now?

我很难相信开源和 OpenAI 加速到了那种程度。但是啊，Anthropic 显然还在领跑位置。

哦顺便说，从第三方数据你也能看到，Grok 和 Cursor 也是——7 月因为 Grok 4.5、Grok Build 的推出，是相当有转折意味的一个月。

所以这是个棘手的月份。

我有个在富达（Fidelity）的朋友，他说过去三年正确的打法就是：用最快的速度做最蠢、最表面的那件事，然后在这些事之间来回切换。

那现在这件事是什么？

[20:38] Gavin Baker

Yeah.Well, that's just, that has been to cut risk all month in response to these narratives that factually, except for credit, are not true.And the work we've done makes me think that credit just isn't going to matter how's the street prices.Let's just say you do need credit to, like, build the flops we need.Well, if credit's not there, it just means the flops that are there are going to be even more valuable.And then eventually that will improve the metrics.And then it's like credit is there.So as long as we're in a compute shortage, which I'm just, like, desperately trying to find a single sign that we're not in one and that it's not actually getting worse almost by the day, it's almost like the problem becomes the solution.And then this company, Black Forest Labs, I think that's their name.I hope I got it right because there is an interesting essay that got sent to me.You know, I think we've talked before about Mike Momison's theory that breakdown in diversity is kind of what leads bubbles and crashes.And essentially everyone I know in the public equity investment business, whether retail or institutional, every piece of news gets fed into Claude.And Claude, Claude Code, sometimes a Claude agent, and it's probabilistic.

是啊。那就是整个月都在降风险，为的是应对那些——除了信贷之外——事实上并不成立的叙事。

而我们做的工作让我认为，信贷不会成为问题，不管街上怎么定价。

就算你真的需要信贷来建我们需要的那些算力好了。那如果信贷不到位，就只意味着已经存在的算力会更值钱。然后这最终会改善那些指标。然后就变成：信贷到位了。

所以只要我们还处在算力短缺里——而我正在拼命找哪怕一个迹象说明我们没在短缺、说明它没有几乎一天比一天更严重——这几乎就变成了「问题本身就是解药」。

然后还有一家公司，Black Forest Labs，我想是叫这个名字吧。我希望我没记错，因为有人给我发过一篇很有意思的文章。

我们以前聊过迈克尔·莫布森（Michael Mauboussin）的理论：多样性的崩塌才是导致泡沫和崩盘的原因。

本质上，我认识的公开市场投资圈里的每一个人，不管是散户还是机构，每一条新闻都会被喂进 Claude。而 Claude、Claude Code，有时候是一个 Claude agent，它是概率性的。

[21:55] Gavin Baker

There's probably not that much variation in the way it's interpreting this news.And so it's almost like we're back to, in stock market terms, there's never really been this way in the stock market before, but people talk about the fragmentation of media and how it used to be like Walter Cronkite, only voice of truth.And now we don't have that anymore.It's like Claude, it's kind of Walter Cronkite for the stock market.And everybody just believes whatever it says.By the way, it's really smart, but it's not always right.Its interpretation isn't always correct.And with the stock market, you are fundamentally dealing about, you know, a probabilistic Bayesian interpretation of the future.It feels like in the market, here's this piece of news.It gets fed through Claude.Claude interpreted it this way.It's a huge chunk of people trade on Claude's view.And so you've seen stuff.There's this guy, TBU.He's like part of the anonymous semiconductor mafia on X.Yeah.Actually a very smart guy.I know him in real life.But he posted this amazing chart of Japanese capacitor stocks.And he said, we've had an entire capacitor cycle in six weeks.And it's true, you know, the stocks like, whether they double, triple, quadruple, I don't know.

它解读这条新闻的方式，大概不会有太大差异。

所以这几乎就像是我们回到了……用股市的话说，股市里以前从来没有这样过，但人们会谈媒体的碎片化，说以前是沃尔特·克朗凯特（Walter Cronkite，美国新闻主播，冷战年代被称为「全美最可信的人」），唯一的真相之声，而现在我们没有那个了。

现在就像是 Claude——它有点成了股市的沃尔特·克朗凯特。所有人都相信它说的任何话。

顺便说，它非常聪明，但它不总是对的。它的解读不总是正确的。而在股市里，你本质上打交道的是对未来的、概率性的贝叶斯解读。

现在市场上的感觉就是：这里有条新闻，它被喂进 Claude，Claude 这样解读它，然后一大批人按 Claude 的观点去交易。

所以你会看到一些现象。有个人叫 TBU，他是 X 上匿名半导体帮的一员。

是啊。

其实是个非常聪明的人，我现实里认识他。他发了一张关于日本电容股的神图，他说：我们在六周里过完了一整个电容周期。

这是真的。那些股票——翻倍、翻三倍、翻四倍，我不知道具体多少——

[23:12] Gavin Baker

But like vertical and then whoosh.Like the actual fundamentals haven't even hit.And yet you've already had what probably would have normally been a three-year cycle in like six weeks.Vanta automates security and compliance for over 16,000 fast-moving companies like Ramp, Cursor, and Harvey, keeping them audit-ready around the clock.

但就是直上，然后「咻」地下来。

而实际的基本面甚至还没落地。可你已经把通常需要三年的一个周期，在大概六周里跑完了。

Vanta 为超过 16,000 家快速成长的公司（比如 Ramp、Cursor 和 Harvey）自动化安全与合规，让它们全天候处于「随时可审计」的状态。

[23:35]

It's the number one agentic trust platform.And it now helps companies like yours watch for the risks that show up between audits across your vendors, your AI tools, and your whole environment.Every new tool your team signs up for, every vendor that turns on AI features, is an opportunity for something to go wrong.And most security programs weren't built for AI's pace of growth.The Vanta agent works like a 24-7 GRC engineer in the background, finding issues, drafting fixes for you, and cutting vendor assessment time by up to 50%.Whether you're a fast-growing startup or a global enterprise, Vanta helps you earn and prove trust.Invest like the best listeners get a special offer for $1,000 off at vanta.com slash invest.Ridgeline is the first end-to-end system of record with embedded AI for investment management firms, running portfolio accounting, reconciliation, reporting, trading, and compliance on one unified platform.Firms are moving off legacy technology and onto Ridgeline because of how far ahead Ridgeline's AI features are compared to anything else in investment management software.Which is why I believe that firms that come out ahead in the AI era will be the ones running on Ridgeline's unified platform.

它是排名第一的智能体信任平台。现在它还能帮你这样的公司盯住审计间隙期里冒出来的风险——覆盖你的供应商、你的 AI 工具和你的整个环境。

你团队新开通的每一个工具、每一个打开 AI 功能的供应商，都是一次出事的机会。而大多数安全体系并不是为 AI 这种增长速度设计的。Vanta 的智能体就像一位 7×24 小时在后台工作的治理·风险·合规（GRC）工程师，发现问题、替你起草修复方案，把供应商评估时间最多缩短 50%。

不管你是高速成长的初创还是全球性企业，Vanta 都能帮你赢得并证明信任。《Invest Like the Best》的听众可以在 vanta.com/invest 拿到 1,000 美元优惠的专属报价。

Ridgeline 是第一个面向投资管理机构的端到端记录系统，内嵌 AI，在一个统一平台上运行组合会计、对账、报告、交易和合规。各家机构正在从遗留技术迁移到 Ridgeline，因为 Ridgeline 的 AI 功能相比投资管理软件里的任何东西都遥遥领先。这也是为什么我相信，在 AI 时代跑赢的机构，会是那些跑在 Ridgeline 统一平台上的机构。

[24:43] Patrick O'Shaughnessy

If you're serious about your firm's AI strategy, Ridgeline should be part of that conversation.You can request a demo at ridgeline.ai.What's your sense of being out here especially, it makes me especially curious about this, the innovation that is going on here to improve the efficiency and every aspect of serving inference of training models, etc.And how that will affect like public markets over time.Like have you learned anything interesting about the long lead time innovation type stuff that has you especially excited or curious?

如果你认真对待你公司的 AI 战略，Ridgeline 应该出现在这场对话里。你可以在 ridgeline.ai 申请演示。

你在这边（硅谷）的感受是什么？这让我特别好奇——就是这边正在发生的、为了提升训练和推理服务各个环节效率的创新，以及它长期会怎么影响公开市场。

你有没有学到什么关于「长周期创新」的有意思的东西，让你特别兴奋或者特别好奇的？

[25:18] Gavin Baker

Yeah, I am very curious.A lot of people seem to feel like they are very close to solving continual learning and sample efficient learning, which we've talked about before.And it is possible that if those are solved, that could that be like a temporary discontinuity in demand?

是的，我非常好奇。很多人似乎觉得，他们已经非常接近解决持续学习（continual learning，模型上线后还能继续从新经验里学）和样本高效学习（sample efficient learning，用少得多的数据就学会）了，这个我们之前聊过。

有没有可能，如果这些被解决了，会造成需求上的一次暂时性断层？

[25:39] Gavin Baker

And if instead of, I was trained on effectively 20 billion tokens, and then it's like these models are trained on 300 trillion tokens.And if, you know, you can trade something on 10 trillion tokens and then let it out into the world and learn sample efficiently, that doesn't sound good for training demand.But like training has a percentage of semiconductor demand to compute is going to asymptote to something not approaching zero, but very small.But I would say that is the most interesting.And, you know, who knows if it's long horizon or short horizon.You know, SSI says that they're going to come out with their model in August.There's this whole generation of new labs that are focused on this.And this would be good for the world.This would be amazing for the world.This would be awesome for the world.We all want this.Yeah, we want this.It would be amazing for the world.And it's just, it's hard for me to believe that that would actually be negative for AI infrastructure demand.But again, trying to be really, really open-minded.I would say that was probably like the biggest scientific or technical takeaway.You know, it's also...We just don't know.Well, yeah.And also like NVIDIA is heavily involved with all of these startups.

如果不是像现在这样——我这个人大概是在 200 亿 token 上「训练」出来的，而这些模型是在 300 万亿 token 上训练的——如果你能用 10 万亿 token 训练出一个东西，然后把它放到世界里去、让它以样本高效的方式学习，那对训练需求听起来就不太妙。

但训练作为半导体算力需求的占比，本来就会渐近趋向某个不接近零、但非常小的数字。

不过我会说，那是最有意思的一件事。而且谁知道它是长周期还是短周期呢。你知道 SSI（Safe Superintelligence）说他们 8 月要发模型。有一整代新实验室都聚焦在这上面。

这对世界会是好事。这对世界会是极好的事。这对世界会是超棒的事。我们都想要这个。

是啊，我们想要这个。

这会非常美妙。而且我很难相信这会对 AI 基础设施需求是负面的。但话说回来，我在努力努力再努力地保持开放心态。

我会说那大概是最大的科学或技术层面的收获。你知道，还有……

我们就是不知道。

是啊。而且 NVIDIA 跟所有这些创业公司都深度绑定。

[26:52] Patrick O'Shaughnessy

If you were just forced to come up with the set of circumstances that would really switch you around and get you really scared, would it just be that this operating cash flow thing doesn't play out and therefore we just need to finance this?

如果非要你想出一组会真正让你掉头、让你真的害怕的情形，那是不是就是：经营性现金流这件事没有兑现，因此我们就必须靠融资？

[27:04] Gavin Baker

Yeah, if the operating cash flow does not continue to accelerate, that would be negative.And that to some degree is going to be a function of how Anthropic, OpenAI, GrokCursor, and open source do.If there was a pretty dramatic contraction in GPU prices that was kind of sustained, the market would react to that instantly.That would be worrisome.If it started to get to be really easy to get GPUs.I mean, have you heard anyone say they have too many GPUs?

是的。如果经营性现金流不再继续加速，那就是负面的。而这在某种程度上取决于 Anthropic、OpenAI、Grok、Cursor 和开源做得怎么样。

如果 GPU 价格出现相当剧烈且持续的收缩，市场会立刻反应。那会让人担心。

如果 GPU 开始变得非常容易搞到。

我是说，你有听到过任何人说他们 GPU 太多了吗？

[27:36] Gavin Baker

Like not a single person.No, in fact, it's the opposite.It sounds like a drug market or something.Yeah, it really does.It's just wild.But yeah, I mean, I think there's a long list of pretty obvious things.If the sum of these labs plateaus or starts to decline, that's really negative.Unless it's just because open source tokens are net growing the pie and taking share.And I do really think the future is multi-model.Particularly for the AI natives, they're going to want to take an open source model.It's got all these inference clouds.They've gotten really good at supervised fine tuning and reinforcement learning.So you can take your data, customize an open source model, and then get something that you can put behind a router.And the router routes it to often first your model and then Claude, Frontier model, whatever Claude, Grock checks it.And you can, in a lot of cases, get slightly better outcomes at half the cost.But again, that half the cost, I think a lot of people hear that.They're like, that's bad for AI demand.It's actually not at all because the cost the user pays is just a function of the margin on the tokens.And you're literally just shifting tokens from really expensive tokens with like 90% gross margins to tokens with maybe, let's call it a 30% gross margin.

一个人都没有。

没有，事实上正相反。

这听起来像个毒品市场之类的。

是啊，真挺像的。太野了。

但是啊，我是说，有一长串相当明显的（风险）。如果这些实验室加起来见顶或者开始下滑，那非常负面——除非那只是因为开源 token 在净做大蛋糕、在抢份额。

我确实非常认为未来是多模型的。特别是对 AI 原生公司，他们会想拿一个开源模型。现在有这些推理云，他们在监督微调（SFT）和强化学习（RL）上已经做得很好了。

所以你可以拿自己的数据去定制一个开源模型，得到一个可以放在路由器后面的东西。路由器把请求分发过去——通常先给你自己的模型，然后 Claude、前沿模型，随便什么 Claude、Grok 再检查一遍。

很多情况下，你能用一半的成本拿到略好一点的结果。

但话说回来，「一半的成本」——我觉得很多人一听到这个就想「这对 AI 需求不好」。其实完全不是，因为用户付的成本只是 token 上那层利润率的函数。你只不过是把 token 从毛利率 90% 的超贵 token，转移到了毛利率大概 30% 的 token 上。

[29:01] Gavin Baker

And that's where the savings are coming from.But the tokens cost the same amount of compute to produce.And then also all these things are kind of happening on different cycle times.All these big public companies are like, oh my God, my AI spend is 20x.I've burned my budget in three months.So they set up a router.And that actually cuts their AI spend.But it doesn't really impact.It may actually increase the amount of tokens that they are generating just by shifting them to these cheaper open source tokens.And that's just more compute.So a company getting smarter about which model to use for which task, that may lead to a stabilization of their spend or even a decline.But it actually has nothing to do with the amount of GPU compute hours they are effectively consuming behind these model layers of this router.The GPU compute hours probably are going up as you shift to these cheaper tokens you can use more of.That's happening to like a cutting edge of public companies.And then you have this whole wave of AI natives.They're leaning into this so hard.And they're not hiring humans.They're just putting it mostly into tokens.They're not slowing down.And then you have companies on the east coast of America who have like barely adopted AI.

省下来的钱就是从这里来的。但生产这些 token 消耗的算力是一样的。

另外，所有这些事情是在不同的周期时间上发生的。

所有这些大型上市公司都在说「我的天，我的 AI 支出翻了 20 倍，我三个月就把预算烧完了」。所以他们搭了个路由器。这确实削减了他们的 AI 支出。

但这并不真的产生影响。它甚至可能反而增加了他们生成的 token 量——只是把这些 token 转移到了更便宜的开源 token 上。而那就是更多算力。

所以一家公司变得更聪明地选择「哪个任务用哪个模型」，可能会让它的支出稳住甚至下降。但这跟它在路由器背后的这些模型层里实际消耗的 GPU 算力小时数毫无关系。随着你转向这些更便宜、因而能用得更多的 token，GPU 算力小时数很可能是在上升的。

这是发生在最前沿的一批上市公司身上的事。

然后你还有一整波 AI 原生公司。他们疯狂地押注这个。他们不招人，基本上把钱都投进了 token。他们没有放慢。

然后你还有一批美国东海岸的公司，他们几乎还没开始用 AI。

[30:23] Gavin Baker

Companies, broadly speaking, you know, not on the coast who maybe are as cutting.And then Europe who's just trying to figure out how to regulate AI.Like, you know.Before using it.Yeah.So just like there's kind of these differential waves of adoption all happening at the same time.But the thought I can't get out of my mind is like I think I said it maybe last time.But just Yalak Sassadra, like 500,000 people in the world, 250,000 maybe are using agentic AI.And we're in an acute compute shortage.There's 7 or 8 billion people on the planet.What happens when we go from 500,000 to 100 million, you know, to 500 million?

还有那些不在沿海地区、可能也就那样的公司。然后是欧洲，他们还在琢磨怎么监管 AI。

在用它之前就先想着管。

是啊。

所以就像是有好几波不同的采用浪潮同时在发生。

但我脑子里挥之不去的一个念头是——我想上次可能说过——就是伊利亚·苏茨克维（Ilya Sutskever，此处音频不清、人名待核）说的那个数：全世界大概 50 万人，也许 25 万人在用智能体 AI（agentic AI）。而我们已经处在急性算力短缺里了。

地球上有 70 亿、80 亿人。当我们从 50 万人变成 1 亿人、变成 5 亿人的时候，会发生什么？

[31:04] Gavin Baker

It is interesting.You know, a lot of people.I do think it's like helpful to post on X to see the pushback.And a lot of people are saying we accept your argument that hyperscalers are under earning.It is compute reprices.Their operating cash flow is going to accelerate.And maybe we could fund this.But like where is that operating cash flow going to come from?

这挺有意思的。你知道，很多人……

我确实觉得发到 X 上看看别人的反驳挺有帮助的。很多人说：我们接受你的论点，超大规模云厂商现在是「少赚」的，随着算力重新定价，他们的经营性现金流会加速，也许我们能给这个融到资。

但那笔经营性现金流要从哪来？

[31:23] Gavin Baker

Where is the customer?And kind of definitionally, it has to either come from faster economic growth through productivity.Kind of Satya's comments like either we're going to start growing 10% or we're not.Or labor substitution.And for sure, I think in a lot of these AI natives, you're seeing labor substitution, but not because they're firing people.They're just not hiring nearly as many humans.The gross profit dollars per FTE and A16Z, Icotic, a bunch of companies that have done this work.They're vertical, particularly relative to past generations of startups.And that it is interesting.Are you doing any surveys of your companies and their token spend relative to labor spend?

客户在哪里？

从定义上讲，它要么来自通过生产率提升带来的更快的经济增长——有点像萨提亚（微软 CEO 纳德拉）说的，要么我们开始以 10% 的速度增长，要么就不是——要么来自对劳动力的替代。

可以肯定，在很多这些 AI 原生公司里，你看到的是劳动力替代，但不是因为他们裁员，而是他们招的人少得多。

a16z、ICONIQ 和一堆做过这方面研究的公司发现，人均（FTE，全职员工）毛利美元是垂直向上的，尤其是相对于过去几代创业公司。

这挺有意思的。

你有没有对你投的公司做过调研，看他们的 token 支出相对于人力支出是多少？

[32:07] Gavin Baker

Oh, yeah.I mean, it's tokens as a percent of total comp spend or something like this.And what are the ranges you've seen?I mean, like in the really pilled companies, like it gets really high, 20%, 25%.Our friend Dylan Patel at his company, he's an ASI maxi, but he's at 30%.That's probably the highest one I've heard.I've actually heard of 50.And there's $25 trillion in knowledge work.Let's take your 20% number.That's $5 trillion.And that either comes out of labor substitution or faster economic growth.And we really, really, really want as humans to come from faster economic growth.One interesting thing I heard this morning from one of the great leading technology CEOsthat's founded several companies.If you look at the founder-led and controlled companies and adjust for some of the like COVIDera overhiring, nobody's really laying people off.These are the people that would probably be most quick to adopt AI to become more efficientor whatever.Like they're not really doing, jack aside, like huge scale layoffs, which probably tellsyou something about where they think there will be lots of opportunity to still have peopleplus.100%.Well, the bull case, you've seen charts from Cognition, RAMP, and Stripe, that the companies

哦，有。我是说，就是 token 占总薪酬支出的百分比之类的。

你看到的区间是多少？

我是说，在真正「吃透了」的公司里，这个数会很高，20%、25%。我们的朋友 Dylan Patel 在他自己的公司——他是个 ASI（超级智能）极端信徒——他是 30%。那大概是我听过最高的。

我其实听说过 50% 的。

而知识工作是一个 25 万亿美元的盘子。就取你说的 20%，那就是 5 万亿美元。而这要么来自劳动力替代，要么来自更快的经济增长。作为人类，我们非常非常非常希望它来自更快的经济增长。

今天早上我从一位创办过好几家公司的顶尖科技 CEO 那里听到一件有意思的事：如果你看那些创始人领导并控制的公司，并且把疫情期间的超额招聘调整掉，其实没人在真的裁员。

这些人本该是最快采用 AI 来提效的那批人。而他们并没有真的搞大规模裁员——这大概说明了一些事情，说明他们认为还会有大量机会是需要人的。

百分之百。

而且看多的理由是，你看过 Cognition、Ramp 和 Stripe 出的那些图，那些

[33:19] Gavin Baker

that are spending the most on AI are growing meaningfully faster.Yeah, I love that Cognition Index.Yeah, the Cognition Index is wild.All the skeptics will point out rightfully.It's not really controlling for industry.But then if like you dig down into it, I think one of them gave an example of, I forget ifit was a plumber or an HVAC contractor, but like everybody who's a blue collar worker isdoing great because of AI.By the way, something that I think we should touch on, and we can do it now or later, isjust everybody is signing these LTAs.Everything is at a shortage.If there's weakness, it's just because we can't energize the gigawatts fast enough.The gigawatts are going to get energized, like regulatory policies moving in a good way.The turbine manufacturers, the diesel jet manufacturers, you're ripping turbines offold airplanes and reconditioning them and then repurposing them.There's crazy things happening.Capitalism is very, very good at this.But I do think one of the most important questions in the market and like a transition of themarket that I got wrong is we are shifting, particularly for memory more than anythingelse, from crushing numbers in the short term to their trading short-term upside for these,

在 AI 上花钱最多的公司，增长明显更快。

是啊，我很喜欢那个 Cognition 指数。

Cognition 指数太野了。

所有怀疑论者都会正确地指出，它并没有真正控制行业变量。但如果你往下挖，我记得其中一个例子讲的是——我忘了是水管工还是暖通空调（HVAC）承包商——反正所有蓝领工人都因为 AI 过得很好。

顺便说，有件事我觉得我们该聊聊，现在聊或者待会儿聊都行：所有人都在签这些长期协议。所有东西都短缺。如果说有什么疲软，那也只是因为我们没法足够快地把这些吉瓦通上电。

这些吉瓦是会通上电的，监管政策正在往好的方向走。涡轮机制造商、柴油机/航空发动机制造商——人们在从旧飞机上拆涡轮、翻新，然后改作他用。有很多疯狂的事情在发生。资本主义在这方面非常非常在行。

但我确实认为，市场上最重要的问题之一——也是我判断错了的一个市场转变——是：我们正在从「短期把数字做爆」转向……尤其是内存（memory）比什么都明显……他们在用短期的上行空间去换这些——

[34:32] Gavin Baker

what are they called supply chain agreements, long-term agreements, LTAs.There's many flavors, but the customer prepays, there's a floor and a ceiling.And this comes back to the point about labor because a lot of people after firing too manypeople during COVID were really reluctant to lay people off.They talked about labor hoarding, if you remember a few years ago.You remember this?

——那些叫什么来着，供应链协议、长期协议（LTA）。有很多种形式，但基本上是客户预付，设一个价格下限和一个上限。

这又回到了关于劳动力的那一点，因为很多人在疫情期间裁了太多人之后，非常不情愿再裁员。几年前大家谈「囤劳动力（labor hoarding）」，你还记得吗？

[34:54] Gavin Baker

Let's just think about the game theory of breaking an LTA.So there's four companies that matter at scale.There's Amazon with their tradiums.There's Google with their TPUs.There's AMD.And then there's NVIDIA, who's like much bigger than everybody else combined.Let's just say it's 2027.And it's very important to realize memory is the more memory you put with flop for a givenunit of compute, the more tokens you get out.It's the single most important thing you could do to increase token output per unit of compute.And then that obviously, definitionally, actually lowers costs, which is why the demand hasn'tresponded at all negatively.There's been no elasticity just because it's the axis that is dominating all others.And this is at some level like a giant Game of Thrones or IMPERS between these companies.Okay, it's 2027 or 28.You're vaguely tempted to break one of these LTAs and try and get a lower price.But to a large degree, market shares, I think for the next several years, are going to bedetermined by supply chain allocations and kind of what you have pre-purchased.So if you break the LTA, this is assuming we're not in a severe oversupply situation.The game theory even holds in a severe oversupply situation.

我们来想想「撕毁 LTA」的博弈论。

在规模上真正重要的公司有四家：有亚马逊和它的 Trainium，有谷歌和它的 TPU，有 AMD，然后是 NVIDIA——NVIDIA 比其他所有人加起来还大得多。

我们就假设是 2027 年。

而且非常重要的一点是要意识到：在给定单位算力的情况下，你给每单位浮点算力配的内存越多，能吐出来的 token 就越多。这是提升「单位算力 token 产出」你能做的最重要的一件事。而这显然从定义上就降低了成本，这也是为什么需求完全没有出现负面反应——完全没有弹性，就因为这条轴压过了其他所有轴。

这在某种层面上就像这些公司之间一场巨大的《权力的游戏》或者帝国博弈。

好，现在是 2027 年或 2028 年。你隐隐有点想毁掉其中一份 LTA，去拿一个更低的价格。

但很大程度上，我认为未来几年的市场份额，将由供应链配额、由你预购了什么来决定。

所以如果你撕毁 LTA——这是假设我们没有处在严重的供给过剩里。其实即使在严重供给过剩的情况下，这个博弈论也成立。

[36:17] Gavin Baker

If you break your LTA and then in the next two or three years, for any reason, leverageshifts back to the memory, guys, you're out of business.It's over.Let's just say Google breaks an LTA.There's an oversupply.I'm making this up at 28, 29.They break their LTAs.Well, if they're breaking their LTAs, it probably means your oversupply prices are coming downand then capacity naturally contracts.Well, what do you think is going to happen to Google's allocations?

如果你撕毁你的 LTA，然后在接下来两三年里，因为任何原因，筹码又回到内存厂手里，那你就完了，出局了。

我们就说谷歌撕毁了一份 LTA。假设 28、29 年出现供给过剩——我瞎编的。他们撕毁了 LTA。

那么，如果他们在撕毁 LTA，那大概意味着供给过剩、价格在往下走，然后产能自然收缩。

那你觉得谷歌的配额会怎么样？

[36:47] Gavin Baker

This is a cyclical industry and oversupply is followed by undersupply.What do you think they think is going to happen to their allocations next time?So I just think given that this is the axis around which kind of everything is revolving,you might blow up your entire business and your franchise by breaking an LTA.And that was never the case before.Apple, who cares?

这是个周期性行业，供给过剩之后跟着的就是供给不足。

你觉得他们（内存厂）认为下一次谷歌的配额该怎么给？

所以我就觉得，鉴于这是万物围着转的那条轴，你可能会因为撕毁一份 LTA 而把整个生意、整个特许经营权炸掉。

而这在以前从来不是这样的。

苹果？谁在乎。

[37:11] Gavin Baker

They don't have a competitor.They're overwhelmingly the largest purchaser.This is, you know, going back three, four, five years.They know they can do whatever they want with no consequences because their volume is so bigthat even if they like super screw Hydex, Micron will of course take them.This is just different.You know, you have at least four players.Did you have all the startups?

他们没有竞争对手。他们是压倒性的最大买家。这是回看三年、四年、五年前的情况。

他们知道自己可以为所欲为而没有后果，因为他们的量太大了，就算他们把海力士（Hynix）坑得再惨，美光（Micron）当然还是会接他们。

现在完全不同了。你至少有四个玩家。

那那些创业公司呢？

[37:32] Gavin Baker

You're an investor in Etched.If you break an LTA, they just say, okay, fine, great.You broke the price agreement.We're going to break the volume agreement.And screw you.We're going to give the volume to your competitor.You just lost share, you know?

你是 Etched 的投资人。

如果你撕毁一份 LTA，他们就会说：好啊，行，很好。你毁了价格协议，那我们就毁掉量的协议。去你的。我们把量给你的竞争对手。

你就这么丢了份额，懂吧？

[37:48] Gavin Baker

NVIDIA's dominance, the current environment they stood to which it favors NVIDIA.It is a little hard for me to understand why it's trading at such a low multiple.In other words, if you need to be able to finance the chips, and you do,nothing's more financeable than an NVIDIA GPU.Nothing.If you need to get land and power, well, they're doing a very good job of playing that chess gameand matchmaking.And then they've rolled out this really clever new business model, which I would describeas kind of like a credit wrapper with a revenue share if GPU prices are above the floor.Yeah, yeah, yeah.Yeah.And this could lead to them having a really giant cloud business effectively through royaltiesreally quickly.And it is another way of kind of alleviating this cash flow mismatch.Like, hey, we're making all the cash.This isn't really vendor financing because they're not loading them the money.Somebody else is loading the GPU buyer the money.They're still making equity investments, but it's not like you're just putting money intosomeone that some of that money was used to buy your chips, even though NVIDIA said thatthey write into all their equity investments that the money can't be used to buy NVIDIA

NVIDIA 的统治地位，以及当前这个对 NVIDIA 有利的环境……我有点难理解它为什么会以这么低的倍数交易。

换句话说，如果你需要能给芯片融到资——你确实需要——那没有什么比一块 NVIDIA GPU 更好融资的了，没有。

如果你需要拿到土地和电力，他们在下这盘棋、在做撮合方面也做得非常好。

然后他们还推出了一个非常聪明的新商业模式，我会把它描述成一种「信用包装（credit wrapper）」加上一个收入分成——只要 GPU 价格高于某个下限就分。

对对对。

是啊。而且这可能让他们很快就通过分成实质性地拥有一个巨大的云业务。这也是缓解现金流错配的另一种方式——就像是，嘿，钱都是我们在赚。

这并不真的是「厂商融资（vendor financing）」，因为不是他们把钱借给买家，是别人借钱给 GPU 买家。

他们仍然在做股权投资，但这不是那种「你把钱投给某人、而那笔钱里有一部分被用来买你的芯片」——尽管 NVIDIA 说他们在所有股权投资里都写明这笔钱不能用来买 NVIDIA

[38:59] Gavin Baker

chips, but obviously money is fungible.And money thing makes no sense.Yeah.But, you know, I think at some level it probably makes everybody feel better.What would you do if you were the member?Like if you were the CEO of Hynix?

的芯片，但显然钱是可替代的，那条规定说不通。

是啊。

不过我觉得在某种层面上，这大概能让所有人心里舒服点。

如果你是内存厂你会怎么做？比如如果你是海力士（Hynix）的 CEO？

[39:11] Gavin Baker

I'd do the exact same thing NVIDIA is doing right now.Which is?I would be going to the buyers of GPUs, tradiums, and whoever and say, I'll participate in theNVIDIA credit wrapper.Now, their business is just inherently less stable and predictable, but in some way, andmaybe they just put up some cash up front.So it's like they're not on the hook.I'm just making this up.But like, do something like you can, because you have money now and credit markets arerevolting.I'm sure our friends at Blackstone and Apollo are suggesting some variant of this to thememory companies.But hey, we will put up some amount of money from our cash flow today.And then it's gone.It's surety that makes the person who's extending the debt feel better.But we want some sort of a cut of the ongoing revenues as well.That is 100% what I would do.And it's almost like a logical extension of the LTAs where they're trading upside for durability.Here, you can effectively get a royalty on recurring revenues.And that is what NVIDIA is doing.And I do think that is very misunderstood.And I think it would serve NVIDIA well to really explain this.One, they're really bullish on AI.Essentially, every time they haven't taken an equity stake in something, it's been a mistake.

我会做跟 NVIDIA 现在做的一模一样的事。

也就是？

我会去找 GPU、Trainium 之类的买家，说：我愿意参与 NVIDIA 那个信用包装。

当然，他们的生意本身就更不稳定、更难预测，但某种程度上——也许他们就先拿出一些现金放在前面，这样就不用一直担着风险。我这是瞎编的。

但就是做点什么，因为你现在有钱，而信用市场在反抗。我敢肯定我们在黑石（Blackstone）和阿波罗（Apollo）的朋友正在向内存公司提议某种变体：嘿，我们今天从现金流里拿出一笔钱来。然后这笔钱就出去了，它是一种确定性，让放债的人感觉更好。但我们也想要在未来的持续收入里分一杯羹。

那 100% 就是我会做的事。

这几乎就是 LTA 的一个逻辑延伸——LTA 是用上行空间换持久性；而在这里，你实际上能拿到经常性收入的分成。

这就是 NVIDIA 正在做的事。我确实认为这被严重误解了。而且我认为 NVIDIA 把这件事好好解释清楚会对他们很有好处。

第一，他们真的非常看多 AI。本质上，每一次他们没有拿股权的时候，事后都被证明是个错误。

[40:38] Gavin Baker

They've taken equity stake in everything, essentially, except the memory companies thatfor a long while, Anthropic, that they took an equity stake in Anthropic.But like, why not if you have cash flow and you're bullish on AI and Jensen because hesees every lab.He knows all the advances, you know, like all these continual learning labs, you know, safesuperintelligence is now working with them.He sees everything and like what he sees makes him bullish.So one, have some equity upside.And then two, have a revenue share.And you're generating hundreds of billions of dollars of free cash flow and helping tokind of bridge what is clearly kind of a gap, at least given everybody's got free cash flownegative until the operating cash flow accelerates enough that you can internally fund this.It's very opportunistic in a good way.And it significantly increases their revenue per gigawatt.And then it also strengthens their competitive position.You and I, we both have startups, but okay, that's great.Use that startup's chip.What prices are they paying at Taiwan Semi?

他们基本上什么都拿了股权，除了内存公司，以及很长一段时间里的 Anthropic——后来他们也在 Anthropic 拿了股权。

但是，如果你有现金流、又看多 AI，而且黄仁勋因为见过每一家实验室、知道所有的进展——所有这些做持续学习的实验室，Safe Superintelligence 现在也在跟他们合作——他什么都看得到，而他看到的东西让他看多。

那为什么不呢？第一，拿点股权上行。第二，拿一份收入分成。

而且你正在产出几千亿美元的自由现金流，帮着去弥合一个明显存在的缺口——至少在所有人的自由现金流都还是负的、直到经营性现金流加速到足以内部自筹之前。

这是非常机会主义的，而且是好的那种机会主义。它显著提高了他们的「每吉瓦收入」。

然后它还强化了他们的竞争地位。你和我，我们俩都投了（芯片）创业公司，但好吧，很好，就用那家创业公司的芯片。

那他们在台积电付的价格是多少？

[41:39] Gavin Baker

Higher than NVIDIA and all these guys.What prices are they paying for HBM DRAM?Higher.Can you finance those chips easily at the same rate as NVIDIA?No.And so it's always like there's a real burden.Particularly if you use HBM DRAM, you're in the crosshairs of this.Unless like actually they made really different architectural choices.Everything that's happening is actually pretty good for him.By the way, going back to game theory, Anthropic, if they had been as aggressive on compute asOpenAI had been, they would have run away with it.Now OpenAI is back in the game.I think Grok is in the game.Those are the companies on the Pareto frontier.And they have the compute.Do you think after watching that, anyone is going to let off the gas?

比 NVIDIA 和这些大厂都高。

那他们买 HBM DRAM 的价格是多少？

更高。

你能像 NVIDIA 那样、用同样的利率轻松给那些芯片融资吗？

不能。

所以总是有一个真实的负担。特别是如果你用 HBM DRAM，你就正处在这件事的准星里——除非他们真的做了很不一样的架构选择。

正在发生的每一件事其实都对他（黄仁勋）挺好的。

顺便说回博弈论：Anthropic，如果他们在算力上像 OpenAI 那样激进，本来是可以一骑绝尘的。

现在 OpenAI 又回到了牌桌上。我认为 Grok 也在牌桌上。这些就是处在帕累托前沿上的公司。而且它们有算力。

你觉得看完这一切之后，还有谁会松油门吗？

[42:28] Gavin Baker

Right.It was, I think, four months ago that Dario was talking about how it was a really thoughtfulcommentary, but he's like, it's really, really hard because if you buy too much compute,you could go bankrupt at the scale of these things.But if you don't buy enough, you could lose.Well, OpenAI just got back into the game.And now SpaceX is in the game in a big way with Glock 4.5 and Cursor.After watching that from a game theory perspective, is anybody going to back off anytime soon,especially if it can be funded out of operating cash flow?

对。我想大概是四个月前，Dario（Anthropic CEO 达里奥）谈到过——那是段很有思考的评论——他说这真的真的很难：因为如果你买太多算力，在这种体量下你可能会破产；但如果你买得不够，你可能会输。

而 OpenAI 刚刚重新回到了牌桌上。现在 SpaceX 也带着 Grok 4.5 和 Cursor 大举入局。

从博弈论的角度看完这一切，还有谁会在短期内退让吗？特别是如果这能用经营性现金流来覆盖的话。

[43:02] Patrick O'Shaughnessy

Have you met anyone in your travels out here that you would say is like way more bullishthan you?And if so, what do they believe that you don't?I mean, essentially everyone out here is more bullish than me, man.I read this thing that Dorkesh wrote and I was like-The 3x compute price thing or whatever?

你在这边跑这一趟，有没有遇到过哪个人你会说比你还看多得多？如果有，他们信什么是你不信的？

我是说，这边基本上每个人都比我看多，老兄。

我读了 Dwarkesh（Dwarkesh Patel）写的那个东西，我当时就……

算力价格 3 倍那个？

[43:18] Gavin Baker

Yeah.Well, he was, I forget what it was.No, no.It was like 15x or something.Yeah.No, but just basically that renting an H100 for a year would cost $250,000.And that's 15x the current spot or something.Exactly.Like, wow.You know, that was just like-That wasn't in my book.That wasn't in my, forget my like Bayesian probability space of expected outcomes.That wasn't even in my considered but dismissed his totally unlikely outcomes.Dorkesh, he's a very smart guy.He's very plugged in.Then he pointed out that margins on compute are going up.The amount of compute is going up and inference margins going up.And if you multiply those three, that's how you're getting this crazy accelerationin the sum of the labs plus open source, although the margins on open source are not really going up.I look at what's happening in the stock market and I feel like a foolish optimist.And then when I talk to people, whether it's people at the labs, anyone in this ecosystem,I'm like bearish relative to essentially everyone.Just a strange state of affairs.What do you make of the DUV news out of China where I've seen reactions really along a spectrum of like,this is the equivalent of like what ASML had in 2001 or something.

对。呃，他是……我忘了具体是什么了。

不不。是 15 倍还是多少。

对。不是，基本上就是说，租一块 H100 一年要花 25 万美元。那是当前现货价的 15 倍左右。

正是。就像，哇。那个……

那不在我的剧本里。那不在我贝叶斯概率空间里的预期结果里。那甚至都不在我「考虑过但当作极不可能而排除掉」的结果里。

Dwarkesh 是个非常聪明的人，消息非常灵通。

然后他指出，算力的利润率在上升，算力的量在上升，推理的利润率也在上升。你把这三个乘起来，这就是这些实验室加上开源的总和为什么会有这种疯狂加速的原因——虽然开源的利润率并没有真的在上升。

我看股市正在发生的事，感觉自己像个愚蠢的乐观主义者。而当我跟人聊——不管是实验室的人还是这个生态里的任何人——我又觉得自己相对所有人都算看空的。真是个奇怪的处境。

你怎么看中国在 DUV 上的消息？我看到的反应跨度非常大：有人说这相当于 ASML 在 2001 年的水平之类的，

[44:35] Gavin Baker

Or like, no, this is actually the first bit of news in a new story for how we should think aboutthe global supply of cutting edge compute.I think both could be true.Make an analogy.Like, let's just say a DUV machine was a jet turbine.And now an EUV machine is like a warp drive.DUV machines like a propeller plane.EUV is like a jet turbine.They didn't have it before.And now they allegedly do.And that is like a phase transition.You've gone from like liquid to solid.Now that solid, that jet engine prop plane, whatever, is 25 years behind.But still, it's important.And I don't think should be dismissed.But I also, you know, it's kind of funny.You just see this in the stock market.The stock market massively overreacts.And then if this ever hits ASML's orders, maybe it hits it in five years.And like the market has forgotten about it, gotten worried about it, forgotten about it,gotten worried about it, forgotten about it multiple times along the way.So I do think that was probably an overreaction.But we shouldn't dismiss that either.And if you're China, like this is really important to you.You know, there are some reports that like an EV machine had been smuggled into China.And I mean, what a feat of espionage because those things are like giant and delicate.

也有人说，不，这其实是一个新故事的第一条消息，关乎我们该怎么思考全球尖端算力的供给。

我觉得两者都可能是真的。

打个比方。我们就说 DUV 机器是喷气涡轮，而 EUV 机器是曲速引擎；或者说 DUV 像螺旋桨飞机，EUV 是喷气涡轮。

他们以前没有，现在据称有了。这是一次相变，你从液态变成了固态。

但那个固态——那个喷气发动机也好、螺旋桨飞机也好——落后 25 年。不过它仍然重要，我不认为该被轻视。

但同时，这也挺好笑的，你在股市上就是会看到这个：股市会大幅反应过度。然后如果这件事真的哪天打到 ASML 的订单上，可能是五年后，那时市场早就把它忘了、又担心过、又忘了、又担心过，来回好几轮。

所以我确实认为那大概是反应过度了。但我们也不该轻视它。

如果你是中国，这对你真的非常重要。有报道说有一台 EUV 机器被走私进了中国。我是说，那得是什么样的间谍壮举——因为那些东西又巨大又娇贵。

[45:57] Gavin Baker

Yeah, they're huge.I don't know if that's true.You know, there's some noise about it.But, you know, China, they're really, really good.They're really, really smart.They work brutally hard.And they see this as super important for them as a country.But are they going to go from the year 2001 to 2026 or even 2030?

是啊，它们非常大。

我不知道那是不是真的，有一些风声。

但你知道，中国真的真的很强。他们真的真的很聪明。他们工作起来极其拼命。而且他们把这件事看作对国家极其重要。

但他们能从 2001 年跳到 2026 年，甚至 2030 年吗？

[46:18] Gavin Baker

It's a learning by doing.And you can't accelerate the doing.You can't teleport into the future.You actually have to go through those learning cycles.Is it significant?Yes.Did the market overreact?Probably.Honestly, it's very hard as an American to really understand what is happening in Chinaand like have total conviction and clarity, you know, like for better or worse, we are decoupling.That is a process that has been set in motion.And at this point, it almost feels like self-reinforcing on each side.That's unfortunate.We are where we are.They're not going to stop.Neither are we.Any commentary on like every other company in America?

这是「干中学」。而你没法加速那个「干」。你没法瞬移到未来，你必须真的走完那些学习循环。

这重要吗？重要。

市场反应过度了吗？大概是的。

老实说，作为一个美国人，真的很难理解中国正在发生什么、很难有完全的确信和清晰度。不管好坏，我们正在脱钩。这个过程已经启动了。到这个时候，它几乎感觉在两边都在自我强化。这很不幸。但我们就在这儿了。他们不会停，我们也不会。

对美国其他所有公司有什么评论吗？

[47:00] Gavin Baker

Like I feel like right now it is 10 companies, couple private.Not last month.Everything but AI was vertical.And I do think open source getting closer to the frontier and companies like Fireworks making it really easy to customize a model such that you can get, in some cases, better than Frontier performance for a meaningfully lower cost.That is a godsend for the software industry.And it's also a godsend for all these AI natives.You know, it's like our friend Vishria, I think he said two years ago.I've never seen more companies go from being founded to like $50 million a year in revenue and generating cash flow in like whatever it is, nine months.And it's hard to know if any of them are durable because a lot of people would dismiss them as chat GPT wrappers.Well, now with open source, you've generated some data that's unique to your use case, whatever your vertical you're going after as a wrapper is.Fireworks did come out with a really cool product called Nexus.And if you're using Cloud Code, OpenAI Codex, Grok Build, it is literally three lines of code, like 20 words.And Fireworks ingest your data.They can RL a model.And then there's a router that sends the query.And they've had amazing results.

我感觉现在就是 10 家公司，外加几家私营的。

上个月不是。除了 AI 之外的所有东西都在垂直向上。

而且我确实认为，开源越来越接近前沿，加上 Fireworks 这样的公司让定制模型变得非常容易——某些情况下你能用显著更低的成本拿到比前沿更好的表现——这对软件行业是天赐之物，对所有这些 AI 原生公司也是天赐之物。

就像我们的朋友 Vishria（音，人名待核）两年前说过的：我从没见过这么多公司，从成立到年收入 5000 万美元并且开始产生现金流，只用了九个月还是多久。

而且很难判断它们有没有持久性，因为很多人会把它们贬为 ChatGPT 套壳。

但现在有了开源，你就产生了一些对你的用例独有的数据——不管你这个「套壳」瞄准的是哪个垂直行业。

Fireworks 确实推出了一个很酷的产品，叫 Nexus。如果你在用 Claude Code、OpenAI Codex、Grok Build，它字面上就是三行代码、大概 20 个词。然后 Fireworks 把你的数据吃进去，用强化学习（RL）训一个模型，再用一个路由器来分发请求。他们的效果非常惊人。

[48:20] Gavin Baker

And this is kind of the solution for every AI native.And that's why you saw Harvey before it was acquired.Cursor leans so heavily into this.Harvey, Lagora, all of them.Because if you can go from just using one, two, or three frontier models to using those frontier models for whatever it is, 30% to 60% of your token consumption, and then use your own RL model, all of a sudden you're not a wrapper.You're way more defensible.I was so interested by that cursor thing that came out.I think it was cursor, where it's sort of like AI is speed running, like what we've learned amongst humans, which is you could use the frontier model to plan and then farm out tasks to the dumber models.And it's 15 times more efficient or whatever the metric was.And it may be, and this is like super ironic, lower margin open source tokens that are just a little bit behind the frontier.We have friends who believe that once a frontier model hits RSI, it will actually have a dramatically lower cost to serve at every level of intelligence by kind of distilling this.And then there's no place for open source.And I would say that's like a anthropic, open AI, Grok, maximalist view.We shouldn't dismiss anything.Anything is possible.

这基本上就是每一家 AI 原生公司的解法。

这也是为什么你看到 Harvey 在被收购之前、还有 Cursor 都如此重仓这条路。Harvey、Legora，全都是。

因为如果你能从「只用一个、两个或三个前沿模型」，变成「用那些前沿模型处理 30% 到 60% 的 token 消耗，其余用你自己的 RL 模型」，你一下子就不是套壳了，防御性强得多。

那个 Cursor 出来的东西我特别感兴趣。我想是 Cursor 吧——就是 AI 在速通我们人类之间早就学到的那套东西：你可以用前沿模型来做规划，然后把任务外包给更笨的模型。效率高 15 倍还是什么指标。

而且这些「更笨的模型」可能——这就非常讽刺了——就是那些利润率更低、只比前沿差一点点的开源 token。

我们有些朋友相信，一旦某个前沿模型达到递归自我改进（RSI），它在每一个智能水平上的服务成本都会通过蒸馏大幅下降，那样就没有开源的位置了。

我会说那是一种 Anthropic / OpenAI / Grok 的极端主义观点。

我们不该轻视任何可能性。任何事都有可能。

[49:41] Gavin Baker

We want to be very humble.I particularly want to be humble after the month I've had.But that doesn't seem that likely to be.One, because there are so many of these AI natives that have actually generated a decent amount of domain-specific proprietary data.And before open source had this moment and these inference clouds and these routers really developed, you kind of didn't have a choice.Like whatever the terms of service were, you accepted them.But if you can now get off that treadmill, that gives you a degree of independence, maybe durability, safety.But going back to your point, it may be that these cheaper tokens massively inflate the value of the most cutting-edge frontier tokens.Because if today, if you have, you know, I'm going to make this up.120 IQ open source models, and they're really cheap to run.Doesn't that make a 160 IQ model that can orchestrate them more valuable?

我们要非常谦卑。经历了这样一个月之后我尤其想谦卑。

但那件事（前沿吃掉一切）看起来没那么可能。

第一，因为有这么多 AI 原生公司实际上已经积累了相当可观的、领域特定的专有数据。

而在开源迎来这个时刻、这些推理云和这些路由器真正成熟起来之前，你其实是没得选的——服务条款是什么样，你就接受什么样。但如果你现在能从那个跑步机上下来，那就给了你一定程度的独立性，也许还有持久性和安全感。

不过回到你说的那点：也可能这些更便宜的 token 会大幅抬高最尖端的前沿 token 的价值。

因为如果今天你有——我瞎编个数——智商 120 的开源模型，而且跑起来非常便宜，那这不就让一个能编排它们的、智商 160 的模型更值钱了吗？

[50:46] Gavin Baker

We talked last time about how I've been really surprised that so much of the economic returns have accrued to the frontier.That is changing with what we're seeing with these inference clouds.Together, modal, base 10, they're all working in a very cash-efficient way.What's shocking about those business models is they're growing almost as fast as the frontier labs in the early days, but burning very little cash.Right.It's pretty extraordinary.To go back to silly SaaS metrics, like, you know, the rule of 40 perspective, like, these are crazy numbers.Do you think there's a lot of instruction in just, like, the distribution of pay inside of an organization?

我们上次聊过，我一直很惊讶经济回报里有这么大一部分归到了前沿。

而随着我们在这些推理云身上看到的东西，这个正在改变。Together、Modal、Baseten，他们都在以非常省现金的方式运作。

这些商业模式让人震惊的地方在于，他们的增长几乎跟前沿实验室早期一样快，但烧的现金非常少。

对。

挺不可思议的。

回到那些没啥意义的 SaaS 指标，比如「40 法则」的视角，这些都是疯狂的数字。

你觉得从一个组织内部的薪酬分布里，能得到很多启示吗？

[51:25] Gavin Baker

Like, the CEO makes X times more than the median person at a company, and maybe that's frontier tokens versus, you know, open source tokens.Yeah.Something simple.Yeah, it may be that what we discussed last time, where, you know, frontier tokens, like, the pie is growing really, really fast.They may continue to capture the overwhelming majority of economic value, but not all of it the way they have been.And open source tokens might be the majority of tokens processed.Again, going back, that's great for infrastructure demand because a token is a token, and it takes the same amount of flops, watts, space, cooling to make.What's the worst thing that could happen in AI?

比如 CEO 赚的是公司中位数员工的 X 倍，也许那就对应前沿 token 和开源 token 的关系。

是啊。

一个简单的类比。

是的，可能就是我们上次讨论的：前沿 token，蛋糕在非常非常快地变大。它们可能会继续拿走绝大部分的经济价值，但不再像过去那样拿走全部。

而开源 token 可能会占到被处理的 token 里的大多数。

再说一遍，这对基础设施需求是好事，因为 token 就是 token，生产它需要同样多的浮点算力、瓦特、空间和冷却。

AI 里可能发生的最糟的事是什么？

[52:05] Gavin Baker

Is it regulatory?I think regulatory has to be the biggest risk.It's the most obvious risk.And so that was kind of one reason I was excited to be here this week.I want to be scared, you know?I, like, I don't want to feel like a lunatic watching these stocks get cheaper, thinking the expected forward returns are going up,while it feels like the on-the-ground fundamentals have pretty materially improved in July relative to even June.But I still come away thinking, like, regulation, it just has to be the biggest risk.Like, you just can't ignore New York making a data center moratorium.We're living in this weird, post-factual, post-logical, political world.And, I mean, I think the AI industry, it has done a terrible job of PR.And I do think that...I think it at least realizes that now.Yeah.Maybe if not fixed it, it realizes it.Yeah, but, like, the political narrative, I think amongst a lot of ordinary Americans, is, like, data centers,they're going to raise your electricity prices, they're going to take all your water,and then they're going to take your job.The reality is, given the deals that are being cut now, when a data center goes in,electricity prices actually generally go down for everyone around there because of behind-the-meter deals.

是监管吗？

我认为监管必然是最大的风险。它是最明显的风险。

所以这也是我这周很兴奋来这边的原因之一：我想被吓到，你懂吧？我不想觉得自己像个疯子——看着这些股票越来越便宜、觉得预期的远期回报在上升，而与此同时，地面上的基本面在 7 月相比 6 月还实质性地改善了。

但我最后还是觉得，监管必然是最大的风险。你就是没法忽视纽约州搞数据中心暂停令。

我们活在一个奇怪的、后事实、后逻辑的政治世界里。

而且我认为 AI 行业在公关上做得非常糟糕。我确实认为……

我觉得它至少现在意识到了。

是啊。就算没修好，它至少意识到了。

是啊，但我认为在很多普通美国人当中，政治叙事是：数据中心会推高你的电费，会把你的水都用光，然后还会抢走你的工作。

而现实是，以现在正在谈成的这些条件，当一个数据中心落地时，周围所有人的电价通常实际上会下降，因为有「表后（behind-the-meter，指发电直接接到用电方、不走公共电网）」交易。

[53:21] Gavin Baker

This is that, like, data center pledge that Trump asked people to sign.Generally, the data center developer used to be they just had to get the police department and the fire departments,like, new trucks and new cars and new body armor or whatever.Now it's like, we're going to build you a hospital, a school, a new police station, and a fire station,and we're going to lower your power bills.How does that sound?

这就是特朗普让大家签的那个「数据中心承诺」。

以前数据中心开发商一般只需要给警察局和消防局配新卡车、新车、新防弹衣之类的。现在变成了：我们要给你们盖一家医院、一所学校、一个新警察局和一个消防站，还要把你们的电费降下来。你觉得怎么样？

[53:43] Gavin Baker

And, by the way, the jobs are ongoing because it turns out that you kind of need these plumbers,electricians, HVAC contractors, data centers.In a lot of ways, the best thing to happen for blue-collar wages in my lifetime.And yet you have the Democrats, who ostensibly represent the blue-collar workers, taking those jobs away.It's just kind of wild how, what is the phrase, like, a lie could go around the world.Faster than truth gets out of bed, yeah.Yeah, faster than truth gets out of bed.But an author made a mistake in a book.It overestimated the amount of water usage in data centers by 10,000x.Not a little bit.Like, not one order of magnitude.Not two orders of magnitude.Not three.She's admitted that mistake many times.I was completely wrong.It's, like, been super debunked.It's like the Popeye effect.You ever hear that example?

而且顺便说，这些工作是持续性的，因为事实证明数据中心是需要这些水管工、电工、暖通空调（HVAC）承包商的。

在很多意义上，这是我这辈子见过对蓝领工资最好的一件事。

可你却看到民主党人——名义上代表蓝领工人的那些人——在把这些工作拿走。

这真挺离谱的。那句话怎么说来着，谎言能绕地球一圈……

在真相起床之前，是啊。

对，在真相起床之前。

但有位作者在一本书里犯了个错，把数据中心的用水量高估了 10,000 倍。不是高估了一点点，不是一个数量级，不是两个，也不是三个。

她已经多次承认了那个错误：我完全错了。这个已经被彻底证伪了。

这就像「大力水手效应」。你听过那个例子吗？

[54:36] Gavin Baker

No.You know, Popeye eats spinach.The reason was same deal.In an academic book, they placed the decimal two things wrong.So spinach does not have more iron than everything else.It was just this one source, and then that propagated.Did people still say it has more iron?

没有。

你知道大力水手吃菠菜。原因是一样的：在一本学术著作里，他们把小数点点错了两位。所以菠菜的含铁量并不比其他所有东西都高，那只是一个来源出错，然后就那样传播开了。

现在人们还在说菠菜含铁高吗？

[54:50] Gavin Baker

I literally had, I thought it had more iron.I mean, that's wild.Just like 80 years ago.That's wild.I literally thought spinach had more iron.That's amazing.It'd be crazy.Yeah, you learn something new every day.Yeah, same thing, though.Yeah, it's the same thing.And it's just, so somebody just needs to tell the truth.I feel like the industry, geez, maybe if nobody else is going to do it, like, I'll do it.There needs to be some sort of foundation.Maybe it's a pack.That runs ads during the Final Four, during NFL games, during college football games.Here's the virtues.World Series.Here's what a data center does.Your power, a data center that signed this pledge in your community.Your power prices are going to go down.They're almost certainly going to contribute to the community in a material way.You're going to see a massive influx of super high-playing blue-collar jobs that are going to persist.And I think a lot of people thought that they were one time, and they're just not.Like, there's for sure a spike, and then that moves to the next data center.But there is an ongoing need for RMA and then upgrades at these data centers, and technology is changing.So you're going to have more jobs.

我真的一直以为它含铁高。

我是说，这太离谱了。就是 80 年前的事。

太离谱了。我真的一直以为菠菜含铁高。

太神奇了。这也太疯狂了。

是啊，每天都能学到新东西。

是啊，不过是一回事。

对，是一回事。

所以就是，得有人来说真话。

我感觉这个行业……天哪，也许如果没别人愿意做，那我来做吧。

需要有某种基金会，也许是一个政治行动委员会（PAC），在 NCAA「四强赛」、NFL 比赛、大学橄榄球比赛期间投广告。

把好处讲出来。世界大赛（World Series）也行。

讲清楚数据中心是干什么的：一个在你们社区签了这份承诺书的数据中心，你的电价会下降，它几乎肯定会实质性地回馈社区，你会看到大量高薪蓝领工作涌入，而且这些工作会持续下去。

我认为很多人以为这些是一次性的，其实不是。当然会有一个高峰，然后那波人转去下一个数据中心。但这些数据中心有持续的返修换件（RMA）和升级需求，而且技术在变。所以你会有更多工作。

[55:52] Gavin Baker

You're going to have cheaper power.You're going to have a wealthier community.There's going to be no impact on water, no impact on the environment.But it's easy to build the data center 10 miles out of town, you know?

你会有更便宜的电。你会有一个更富裕的社区。对水没有影响，对环境没有影响。

而且把数据中心建在城外 10 英里就行了，对吧？

[56:04] Gavin Baker

So that story needs to be told, along with, we heard a story, I think we talked about it last time, about how AI is increasingly really saving lives, curing rare diseases.I think it was at ASCO this year, the vibe was, like, this is the most scientific breakthroughs we've ever seen at a single conference.And for sure, some of that is due to AI.And so we need to tell those stories, like, you know, if you have a sick child, sick parent, a sick loved one, like, AI meaningfully increases the odds of them recovering.Everybody needs to tell this.And I think people out here, all of this is so blindingly obvious to them.They can't process that this is a true but wildly divergent view from most Americans.The industry really needs to tell its story better.New York, it just feels like, is the first of many.And even in some of these deep red states that are super pro-growth, they're just like, hey, you guys are not doing a good job telling your story.We can't tell your story.If you tell your story, though, we can retell it.But, like, you're the experts.If you do not speak your own truth, no one else will.What have we missed?

所以这个故事需要被讲出来。还有——我们听过一个故事，我想上次聊过——就是 AI 正越来越多地在真正救命、治愈罕见病。

我想是今年的 ASCO（美国临床肿瘤学会年会），那种氛围是：这是我们在单场会议上见过的最多的科学突破。而这里面肯定有一部分要归功于 AI。

所以我们需要讲这些故事。如果你有个生病的孩子、生病的父母、生病的亲人，AI 会实实在在地提高他们康复的几率。每个人都需要讲这件事。

我觉得这边（硅谷）的人，这一切对他们来说太显而易见了，以至于他们没法理解：这是一个真实的、但与大多数美国人极度背离的看法。

这个行业真的需要把自己的故事讲得更好。纽约感觉只是众多之中的第一个。

甚至在一些超级支持增长的深红州，他们也在说：嘿，你们这个故事讲得不行。我们没法替你们讲。但如果你们讲了，我们可以帮着复述。可你们才是专家。如果你们不说出自己的真相，就没有别人会说。

我们还漏了什么？

[57:21] Gavin Baker

I do think something that is missing from all of this conversation about compute is what is going to happen when you put these SRAM-based accelerators that are not constrained by HBMDRAM and are often made on older nodes that are not competing with, like, the latest and greatest GPUs.When you disaggregate inference, people talk about pre-fill and decode.But decode is two parts, attention and feedforward network.And, like, the ultimate holy grail is if you could do pre-fill on one chip that probably doesn't have HBMDRAM.Do the attention on a super high-powered chip with HBMDRAM and then do the feedforward network on one of these SRAM chips.But, like, the ROI on adding these SRAM accelerators to the existing installed base of compute and new compute.But, like, what we're seeing is you do better.You just can't beat SRAM in particular for that feedforward network.And no matter how much you try and get the ratio of compute to HBMDRAM to SRAM on the chip correct, the workloads are always changing and there's different workloads.Being able to disaggregate into these three parts, this is going to be really, really positive for the ROI on AI.For some reason, I just thought of a funny question, which I love the framing of Game of Thrones versus all these people.

我确实觉得，关于算力的这整场讨论里漏掉了一件事：当你把这些基于 SRAM 的加速器放进来时会发生什么？它们不受 HBM DRAM 的约束，而且往往用更老的制程制造，不跟最新最强的 GPU 抢产能。

当你把推理拆解开——人们会说预填充（pre-fill）和解码（decode）。但解码其实分两部分：注意力（attention）和前馈网络（feedforward network）。

终极的圣杯是：你能不能在一块大概不带 HBM DRAM 的芯片上做预填充，在一块带 HBM DRAM 的超强芯片上做注意力，然后在一块 SRAM 芯片上做前馈网络。

把这些 SRAM 加速器加到现有的已装机算力和新算力上，那个投资回报率……我们看到的是，你的表现更好。在前馈网络这一块，你就是干不过 SRAM。

而且不管你多努力去把芯片上「算力 : HBM DRAM : SRAM」的比例调对，工作负载总是在变，而且有各种不同的工作负载。

能够拆解成这三部分，这对 AI 的投资回报率会是非常非常正面的。

不知为什么，我突然想到一个好玩的问题——我很喜欢你把这帮人比作《权力的游戏》这个框架。

[58:42] Patrick O'Shaughnessy

Can you imagine a player that is not currently on everyone's mind becoming relevant at, like, the major Game of Thrones scale?Like, that could be, like, Micron all of a sudden.Someone that becomes as important as Anthropic, OpenAI, Microsoft, Amazon.So, like, a dark horse.Like, Leipu is probably a dark horse.Lit at Fireworks.She is an absolute killer.I think our friend Scott Wu.Cognition is kind of...You're here to that one.Yes.I think those are the most obvious names.What about SpaceX?

你能想象出一个现在还不在所有人视野里的玩家，突然变得像《权力的游戏》主角级别那么重要吗？比如说，突然是美光？某个变得跟 Anthropic、OpenAI、微软、亚马逊一样重要的公司？

就是说，一匹黑马。

比如说 Leipu（音，公司名待核）大概算一匹黑马。还有 Fireworks 的 Lin（林俏），她是个绝对的狠角色。

我想还有我们的朋友 Scott Wu，Cognition 也算……

这个你可太熟了。

是的。

我觉得这些是最明显的名字。

那 SpaceX 呢？

[59:19] Gavin Baker

What's it been like watching that be digested by public markets, at least initially?Do you think the market understands it as a company, the most important new company to be public?It doesn't really feel like it does.The fundamentals have gotten better since an IPO.Like, ROC 4.5, the cursor acquisition.Cursor has clearly accelerated meaningfully.They've shown over the last three years they can bring on more compute faster than anyone at lower prices.And now we know that they can even adjusting for the spot first contract gap.Their big advantage was they came into the market, hit those spot highs.And in a strange way, like one of the more bullish things for compute is they put a vast amount of compute into the market overnight.And it wasn't even really a blip.It was like the market just utterly absorbed it.The freight trade didn't slow down at all.Well, a sub-stack writer will fund AI.They think that SpaceX is going to try and bring on 8 gigawatts of compute over the next 18 months.So 8 gigawatts over the next 18 months.I will never bet against Elon.But I mean, that would be a truly incredible feat.Rates have gone up since they signed those last contracts.Not down.And they're monetizing at something like $50 billion a gig.

看着它（SpaceX）被公开市场消化，至少是最初这个阶段，是什么感觉？你觉得市场理解它作为一家公司——最重要的新上市公司——吗？

感觉并不太理解。

自 IPO 以来基本面是变好了。比如 Grok 4.5、收购 Cursor。Cursor 明显在显著加速。

他们过去三年已经证明，他们能比任何人更快、以更低的价格把算力上起来。而现在我们知道，即使调整了现货与合同的价差，这也成立。他们的一大优势是，他们进场的时候正踩在现货高点上。

而且用一种奇怪的方式说，对算力来说最看多的事情之一，就是他们一夜之间把海量算力投进了市场——结果连个水花都算不上，市场就把它完全吸收了，行情一点没放缓。

有位 Substack 作者，Funder AI（音，名字待核），他们认为 SpaceX 会试图在未来 18 个月里上 8 吉瓦的算力。

18 个月 8 吉瓦。

我永远不会跟埃隆对赌。但我是说，那会是一个真正难以置信的壮举。

自从他们签下上一批合同以来，价格是涨了，不是跌了。而他们的变现大约是每吉瓦 500 亿美元。

[1:00:39] Gavin Baker

And consensus estimates for next year are $73 billion.So forget Starlink V3.Forget Starlink Direct to Sell.Grok 4.5 and Cursor.I think that the sum of that probably hits a $10 billion ARR pretty quickly.Forget all of that.You know, forget the core base Starlink business.If they bring on anywhere near that, the consensus estimate is $73 billion.And that's 8 gigs at $50 billion a gig.And obviously, that would not all be lit up at the beginning of 27.And it seems very implausible to me.Like, I almost don't believe the funder report.But to this day, the only companies that have brought on more than 500 megawatts of power in a yearare the hyperscalers, Corweave, Crusoe, and SpaceX.And SpaceX has kind of brought on the most, the fastest, at the lowest cost.And then people do actually really like their clusters.But again, it's kind of like the market is going to need to see that.That would not be the market's interpretation of SpaceX today.No, no.And it does feel like, you know, there's this big New York hedge fund short case on it.And I think they think the spot price for compute is going to go down 90%.You're going to bring on all this compute.It's not going to generate, you know, nearly as much revenue as you think.

而明年的一致预期是 730 亿美元。

所以先别管星链 V3，先别管星链直连手机（Starlink Direct to Cell）。Grok 4.5 和 Cursor，我觉得这俩加起来很快就能到 100 亿美元的年度经常性收入（ARR）。这些都先不管。

连星链核心业务也先不管。如果他们能上到接近那个数——一致预期是 730 亿美元——而 8 吉瓦按每吉瓦 500 亿美元算……

显然那不会在 27 年年初就全部点亮。而且这在我看来非常不合情理，我几乎不相信那份 Funder 的报告。

但直到今天，一年内上过 500 兆瓦以上电力的公司，只有那几家超大规模云厂商、CoreWeave、Crusoe 和 SpaceX。

而 SpaceX 大概是上得最多、最快、成本最低的。而且人们确实非常喜欢他们的集群。

但话说回来，市场需要亲眼看到这些。

这不会是今天市场对 SpaceX 的解读。

不不。

而且确实感觉，纽约有家大对冲基金在做空它。我想他们认为算力的现货价会跌 90%——你会上这么多算力，而它产生的收入远没有你想的那么多。

[1:01:58] Gavin Baker

Maybe.But I also want to be really clear.Like, I've seen Elon's companies do really impressive things.The funder AI report of 8 gigawatts in 18 months.I mean, I'm just quoting that because it's public.It's available to everyone.I think one of Elon's phrases is, we specialize in making the impossible late.I've never heard that.That's great.Yeah.Yeah.You know, there's like kind of a lot of truth to that.Yeah, yeah.But I just think very little is built in from my perspective to that stock for the amountof compute that they might be able to bring on.And again, I don't think it's anywhere near eight.And it's going to be really hard.And energizing these GPUs is really hard.But they've been good at it.And it doesn't feel like that's in estimates or really in people's thinking.I'm thinking about that funny meme that says SpaceX, the data center company.Oh, 100%.Yes, absolutely.And then I would also just say, like, I did spend a lot of time at Starbase.And orbital compute feels more real every day.Pretty cool to see that Starship landing the other day.Pretty cool to see the Starship landing.And it's, you know, it is funny.Our friends at Benchmark, they funded StarCloud.And I don't know, last time StarCloud is an orbital compute company that like SpaceX is

也许吧。

但我也想说得很清楚：我见过埃隆的公司做成过非常了不起的事。

Funder AI 那份「18 个月 8 吉瓦」的报告，我引用它只是因为它是公开的，所有人都能看到。

我想埃隆有句话是：我们专精于把不可能的事做晚。

我没听过这句。太棒了。

是啊。你知道，这话还挺有几分道理的。

是啊是啊。

但我就是觉得，从我的角度看，这只股票几乎没有把他们可能上起来的算力量计入价格。

再说一遍，我不认为会接近 8 吉瓦，而且会非常难，给这些 GPU 通电非常难。但他们一直做得很好。而且感觉这既不在预期里，也不太在人们的思考里。

我在想那个好玩的梗：SpaceX，一家数据中心公司。

哦，百分之百。是的，绝对的。

然后我还想说，我在星舰基地（Starbase）待了很长时间，轨道算力（orbital compute）每天都感觉更真实了。

前几天看到星舰着陆挺酷的。

看到星舰着陆挺酷的。

而且你知道，挺好玩的，我们在 Benchmark 的朋友投了 StarCloud。我不知道，上次……StarCloud 是一家轨道算力公司，SpaceX

[1:03:12] Gavin Baker

kind of partnering with.They're going to, I think, let them use the Starlink laser technology, which is reallyimportant for orbital compute.But I do think that's like kind of a good sanity check.Last time I checked, the Benchmark guys were pretty smart.And they're not coming from the Elon ecosystem at all.And they chose to fund an orbital compute company at a decent valuation without the internallaunch costs that SpaceX gets.To me, that's a good like, hey, am I crazy?

算是在跟他们合作。我想他们会让 StarCloud 用星链的激光技术，那对轨道算力非常重要。

但我确实觉得这算是一个不错的理智检验。上次我看的时候，Benchmark 那帮人还是挺聪明的。而且他们完全不是埃隆生态里的人。

他们选择用一个还不错的估值去投一家轨道算力公司，而且没有 SpaceX 那种内部发射成本的优势。

对我来说，这是一个很好的「嘿，我是不是疯了？」的检验。

[1:03:42] Gavin Baker

Am I crazy?And it's like, well, maybe I'm crazy.And maybe Elon's crazy.And maybe Benchmark is also crazy.And maybe the SpaceX engineers are also crazy.But man, that just doesn't seem that probable to me.Should we say whose offices we're in?

我是不是疯了？

然后就是，好吧，也许我疯了。也许埃隆也疯了。也许 Benchmark 也疯了。也许 SpaceX 的工程师们也疯了。

但老兄，那在我看来实在不太可能。

我们要不要说一下我们在谁的办公室里？

[1:03:59] Patrick O'Shaughnessy

Yeah, we're sitting in the Benchmark offices.Yes, this is their famous table for their famous dinners.So thank you, Benchmark.Thank you, Benchmark, for this episode.Thanks, Richard.Yes, thanks, Eric.And Sheila, we should thank them all.Eric coordinated for me, so he gets a special shout out.Thank you, Eric.Thank you, all of the partners.Thank you, Eric.But I mean, we will see where all of these stocks are in a year.And the great thing is, time will tell.People are going to be right or wrong.The future is probabilistic.But it's an exciting moment.Well, if we keep doing this on the model release cycle, I'll see you in a couple weeks.Yeah, it's crazy.Maybe you're going to benchmark.It's always a blast to do with you.If you enjoyed this episode, visit Colossus.com.You'll find every episode of this podcast complete with hand-edited transcripts.You can also subscribe to Colossus, our quarterly print, digital, and private audio publicationfeaturing in-depth profiles of the founders, investors, and companies that we admire most.Learn more at Colossus.com slash subscribe.You know how small advantages compound over time?

好啊，我们正坐在 Benchmark 的办公室里。

是的，这就是他们办那些著名晚宴用的那张著名桌子。

所以谢谢你，Benchmark。

谢谢 Benchmark 赞助这一期。

谢谢 Richard。

是的，谢谢 Eric。还有 Sheila，我们该谢谢他们所有人。Eric 帮我协调的，所以他要特别点名感谢。谢谢你，Eric。谢谢所有合伙人。谢谢 Eric。

不过我是说，我们一年后看看这些股票都在什么位置。而最棒的是，时间会给出答案。人们会被证明对或错。未来是概率性的。但这是个激动人心的时刻。

好，如果我们继续按模型发布周期来录，那我们过几周见。

是啊，太疯狂了。

也许到时候还来 Benchmark。

跟你做这个总是很过瘾。

如果你喜欢这一期，请访问 Colossus.com，你能找到本播客的每一期，都配有人工编辑的文字稿。你也可以订阅 Colossus——我们的季度印刷版、数字版和私人音频出版物，深度特写我们最欣赏的创始人、投资人和公司。到 Colossus.com/subscribe 了解更多。

你知道小小的优势会随时间复利吗？

[1:05:19]

That's true in investing and just as true in how you run your company.Your spending system is your capital allocation strategy.Ramp makes it smarter by default.Better data, better decisions, better economics over time.See how at ramp.com slash invest.As your business grows, Vanta scales with you.Automating compliance and giving you a single source of truth for security and risk.Learn more at vanta.com slash invest.The best AI and software companies from OpenAI to Cursor to Perplexityuse WorkOS to become enterprise-ready overnight, not in months.Visit workos.com to skip the unglamorous infrastructure work and focus on your product.Ridgeline is redefining asset management technology as a true partner, not just a software vendor.They've helped firms 5x in scale, enabling faster growth, smarter operations, and a competitive edge.Visit ridgelineapps.com to see what they can unlock for your firm.Every investment firm is unique and generic AI doesn't understand your process.Rogo does.It's an AI platform built specifically for Wall Street, connected to your data,understanding your process, and producing real outputs.Check them out at rogo.ai slash invest.W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W ca W

在投资里是这样，在你怎么经营公司这件事上同样如此。你的支出体系就是你的资本配置策略。Ramp 让它默认更聪明：更好的数据、更好的决策、随时间更好的经济效益。到 ramp.com/invest 看看。

随着你的业务成长，Vanta 与你一同扩展，自动化合规，为安全与风险提供唯一的事实来源。到 vanta.com/invest 了解更多。

从 OpenAI 到 Cursor 再到 Perplexity，最好的 AI 与软件公司都用 WorkOS 一夜之间做到企业级就绪，而不是花上几个月。访问 workos.com，跳过那些不光鲜的基础设施活儿，专注于你的产品。

Ridgeline 正在把资产管理技术重新定义为一个真正的伙伴，而不只是一个软件供应商。他们已帮助多家机构把规模做到 5 倍，实现更快的增长、更聪明的运营和竞争优势。访问 ridgelineapps.com，看看他们能为你的机构释放什么。

每一家投资机构都是独特的，而通用 AI 不理解你的流程。Rogo 理解。它是一个专为华尔街打造的 AI 平台，连接你的数据、理解你的流程、产出真实的成果。到 rogo.ai/invest 看看他们。

（此处为音频尾部的转写噪声，无实际内容。）