Anthropic 机密提交 IPO 申请，Nvidia 推出新款 PC 芯片，前 Meta CTO 分享其 AI 战略

欢迎各位收看Information的TITV节目。

Welcome everyone to the Information's TITV.

我是阿卡什·马斯里查。

My name is Akash Masricha.

今天是六月一日，星期一。

It is Monday, June 1.

今天开场就有一条突发新闻。

We have breaking news to start the show today.

Anthropic公司今天悄悄提交了IPO申请。

Anthropic today filed confidentially for an IPO.

就在上周，这家公司刚刚完成了最新一轮私募融资，融资前估值高达9000亿美元，超过了OpenAI最新一轮的估值。

It was just last week that the company closed its latest private funding round at a $900,000,000,000 valuation before the investment, and that figure is bigger than OpenAI based on its latest funding round.

此前有多家媒体报道称，OpenAI正试图加快IPO的申报进程，我们会密切关注他们何时提交申请。

Now, there have been a number of reports that OpenAI has been looking to expedite its IPO filing process, so we'll be closely watching to see if and when we see OpenAI's filing.

后续我们还会持续跟进这条新闻，敬请期待。

We're going to have more coverage for this to come, so stay tuned.

今天节目里，我们会详细聊聊英伟达最新发布的面向PC的芯片，这是它和微软一起公布的。

Today on the show, we will unpack Nvidia's latest PC focused chip that it unveiled alongside Microsoft.

我们还会提前预览一下微软的Build大会，这周就要开始了。

We'll also preview Microsoft's Build Conference, which is kicking off this week.

接下来我们会深入介绍丹尼斯·德雷瑟，她是OpenAI新上任的首席营收官，之前是Salesforce旗下Slack的CEO。

We're then going to unpack our profile of Denise Dresser, OpenAI's new chief revenue officer, who was previously the CEO of Slack at Salesforce.

我们还会聊聊前置部署工程，以及这类岗位在AI生态里的火爆趋势。

We'll also talk about forward deployed engineering and the boom across that category of roles across the AI ecosystem.

节目最后，我们请到了Meta前首席技术官Mike Schroepfer，聊聊他的基金Gigascale Capital最新一轮2.5亿美元的融资情况。

And to close out the show, we've got Meta's former CTO, Mike Schroepfer, coming on the show to talk about his fund, Gigascale Capital's big new $250,000,000 in fundraising.

今天的节目非常精彩，咱们直接开始吧。

It's going to be a fun show, so let's get right on into it.

微软和英伟达发布了一款全新的PC芯片，这意味着英伟达正式杀入了个人电脑市场。

Microsoft and Nvidia unveiled a new chip for PCs, putting Nvidia squarely in the game of personal computing.

下面有请微软记者Aaron Holmes，为大家详细解读这一消息。

I want to bring on our Microsoft reporter, Aaron Holmes, to break down that announcement.

Aaron，欢迎再次回到节目。

Aaron, welcome back to the show.

很高兴你今天能来。

It's great to have you here.

能来我很开心。

Happy to be here.

好的。

Okay.

那微软和英伟达今天发布了什么？

So what did Microsoft and Nvidia unveil today?

这次新闻里最重要的，是英伟达推出了一款新芯片，叫N1X。

So the biggest part of this news is Nvidia's new chip called the N1X.

一直以来，英伟达在数据中心芯片市场都占据主导地位。

And specifically, you know, Nvidia has for a while dominated the the market for chips in data centers.

但这次，他们真正推出了首款面向个人电脑的主处理器。

But with this chip, they're actually, you know, releasing their first core processor for PCs.

所以现在他们要进军PC市场了，这个市场长期以来一直由英特尔和高通等竞争对手主导。

So they are now entering into the PC market, which has, for a long time been dominated by Intel and Qualcomm and some other competitors.

除此之外，我们还看到微软以及其他一些PC厂商预览了将搭载这款芯片的新机型。

And on top of that, we saw Microsoft as well as some other PC makers preview new machines that will use this chip.

就微软而言，那就是他们最新款的Surface笔记本电脑。

So in Microsoft's case, that is the the newest version of their Surface laptop.

他们目前还没透露太多细节，比如具体售价是多少。

They haven't really said many details, including, you know, what it'll actually cost.

不过，这款笔记本在很多方面性能更强，而且能比之前的 Surface 版本在本地运行更多的 AI 功能。

But the idea is that this laptop will have in in a lot of ways, higher performance and be able to run more AI features locally than previous versions of the Surface.

那在此之前，英伟达在 PC 芯片方面做到什么程度了？

And to what extent did Nvidia have PC chips before this?

因为我知道英伟达确实是以游戏领域起家的，而且我知道，我的意思是，市面上有大量的电脑

Because I know Nvidia had certainly made its name in gaming, and I know, I mean, there are tons of PCs out there

那些

that

很适合玩游戏。

are good for gaming.

那他们之前就已经在这个市场里了吗？

So was it already in this market?

英伟达之前主要做的是图形处理器，也就是负责处理电脑图形的那些芯片，还有数据中心用的GPU。

So Nvidia had mostly been making graphics processing units, which are essentially what handle graphics on PCs as well as know, GPUs and data centers.

他们在中央处理器方面做得比较少，也就是电脑运行的核心处理器。

They hadn't done as much central processing units, which is, you know, the the core processors for what your PC runs on.

所以这对他们来说绝对是一次市场扩张。

So this is definitely an expansion for them.

而且，你也知道，现在CPU正处于供应短缺的状态。

And, you know, we also have seen that CPUs are in short supply right now.

所以我认为他们觉得自己已经做好了准备，能够抓住这次供应紧张带来的机会。

So I think that they think that they are well positioned to capitalize on, you know, that supply crunch.

那么，你觉得在这种情况下，Nvidia在CPU市场上对其他的巨头们构成威胁的程度有多大呢？

So to what extent do you think then Nvidia could actually then pose a threat in the CPU market to the other CPU giants?

我在想ARM，还有英特尔，微软这类公司过去一直依赖的芯片供应商。

I'm thinking about ARM, I'm thinking about Intel, the companies that Microsoft and the likes of them have traditionally relied on for these chips.

我的意思是，现在缺货这么严重，有货就不错了吗？

I mean, is it just a scenario where there's such a shortage right now that any supply is good?

有趣的是，这款芯片将内置在Nvidia所谓的RTX Spark里，其实就是把常规的处理器、GPU和内存整合在一个系统里。

Well, know, what's interesting is that this chip is going to sit inside of Nvidia's what it's called RTX Spark, which essentially combines the standard computer processor as well as, you know, a GPU and a memory in one system.

我觉得，如果Nvidia能提供这种整合方案，价格更实惠、性价比更高，那对电脑厂商，甚至对普通消费者都会很有吸引力。

And I think the the idea there is that, you know, if Nvidia can offer all of that for essentially what is a a more, you know, cheaper or, you know, more bang for your buck, then that might be appealing to, you know, PC makers, but also to the end customers.

不过，我们其实还不清楚这东西最终会卖多少钱。

Again, we don't really know how much this is actually going to cost.

所以很难判断它到底有多吸引人，是普通买家都会想要，还是只适合追求极致性能的人。

So it's hard to say how appealing that will be and whether this is going to be something that, you know, any PC buyer would want versus someone who wants extremely high performance.

但不管怎样，这确实给Nvidia提供了一个切入市场的绝佳机会。

But either way, it it definitely, you know, seems like an opportunity for Nvidia to seize on that market a bit.

那微软这边呢？

And what about for for Microsoft?

你觉得这对微软的业务和其PC产品线意味着什么？

What do you think this means for Microsoft's business and and its PC line?

你知道，微软一直想让Surface设备在竞争中脱颖而出。

You know, Microsoft has been trying for a long time to get their Surface devices to stand out from the competition.

近年来，Surface经历了不少重大改版。

And we've seen Surface undergo, you know, a number of overhauls in recent years.

他们曾经大力推广Copilot+ PC，也就是那些专为使用高通处理器本地运行AI功能而设计的电脑。

They, for a while, were marketing copilot plus PCs, which were essentially, you know, PCs that were designed to run more AI locally using Snapdragon processors.

目前还不清楚这种产品是否真的打动了消费者，但我觉得，有了这次与英伟达的合作，他们又有了一个全新的卖点来推广Surface。

It's not clear if that really caught off with consumers, But I think that, you know, with this Nvidia partnership, they have essentially a new angle to promote the surface.

而且在很多方面，他们似乎想更多地抢占MacBook的市场，尤其是那些希望用个人设备开发和运行更复杂应用的开发者。

And, you know, in a lot of ways, I think they're trying to encroach more on the MacBook market, especially for know, developers who wanna be able to use their personal device to develop and run more advanced applications.

至于这能不能成功，我们还得拭目以待。

So whether whether that pays off, we we, you know, have to wait and see.

我能不能问你个简单的问题？

Can I can I just ask you a quick question here?

你刚才提到高通处理器？

You said Snapdragon processor?

高通处理器是什么东西？

What is a Snapdragon processor?

那只是他们之前用过的一个处理器品牌。

That that's just a previous brand of of processor that they use.

说实话，这类产品的命名确实挺有创意的。

Well, I guess, I mean, the names, as we know, in this category are sometimes very creative.

至少比5.4或者4.7那些命名强多了，这点我可以肯定。

So, it's better than 5.4 or 4.7, whatever the other naming convention is, I'll tell you that much.

我想聊聊你今天发的那篇关于Anthropic的文章。

Okay, I want to talk about a story that you published today, Anthropic.

你深入分析了公司们是怎么使用Anthropic的Claude模型的。

You went deep on how companies are using Anthropic's mythos model.

再帮大家回忆一下，这个模型可是当时一出来就引爆市场的那个。

And just to refresh, folks, I mean, this was the killer model that came out.

大家都很害怕，因为它的能力实在太强了。

Everyone was scared because its capabilities were so strong.

Anthropic 表示，他们只会让少数几家公司使用这个模型。

Anthropic said, We're only going to give access to a couple of companies here to use the model.

你深入到了网络安全领域。

You went out into the world of cybersecurity.

你采访了几家正在使用它公司。

You talked to some of the companies using it.

你发现了什么？

What did you find?

是的。

Yeah.

我们本来就知道，Mythos 的成本会比 Anthropic 之前推出的模型更高，而那些模型本身就已经属于高价 AI 了。

So essentially, we already knew that Mythos was going to be more expensive than Anthropic's earlier models, which were already on the pricier end of AI models.

但我跟一些正在测试它的人聊过，了解了实际使用时意味着什么。

But I spoke to people who were testing it and got a sense of what that actually means in practice.

我了解到，很多测试人员发现，他们在短短几周内就花掉了价值一百万美元的令牌来测试这个模型。

And I learned, you know, a lot of testers saw that they ran through $1,000,000 worth of tokens testing the model in just a couple of weeks.

比如帕洛阿尔托网络公司，他们用它来测试内部源代码，仅仅三周就用掉了超过一百万美元的令牌。

You know, Palo Alto Networks has been using it to test their internal source code, and within three weeks, they had used over $1,000,000 worth of tokens.

但更有趣的是——

But what's also interesting is-

为了让你有个概念，这数额真的很大，对吧？

And that's just to put this in context, that's a lot, right?

要知道，这些可都是市值数千亿美元的公司。

Like, you know, because these are companies, hundreds of billions of dollars companies.

按他们平时的开销来看，这仍然是一笔巨款，对吧？

This is still a lot, right, based on what they usually spend?

是的。

Yeah.

才用几周就这么贵，确实不便宜。

For for just a couple of weeks of usage that is, you know, pretty expensive.

有意思的是，尽管花的钱不少，公司们还是没打算专门为此设预算，因为他们觉得这钱花得值——毕竟用这个模型发现了大量原本根本发现不了的代码漏洞，或者至少要花上几倍的时间才能找到。

And what's interesting is that companies still aren't planning to budget for this just because the, you know, amount that they're spending, they think is worth it because they've found so many vulnerabilities in their code that they wouldn't have found otherwise, or at least it would have taken them a lot longer without this model.

所以我再确认一下，比如CrowdStrike这家公司，是用这个模型来扫描自己开发的代码，而他们卖的产品本身就是网络安全软件，对吧？

So just to clarify here, so this is a company like CrowdStrike using the model to find vulnerabilities within its own code, which is itself the cybersecurity product that it offers to sell to customers, right?

没错。

Yes.

而且不光是网络安全公司这么做。

And it's not just cybersecurity companies doing this.

像微软这样的大公司，还有其他一些拥有广泛应用软件的企业，都在用这个模型扫描全部源代码，快速发现并修补漏洞。

A lot of companies like Microsoft and and others that have widely used software are using the mythos model to scan essentially all of their source code and just find vulnerabilities and quickly patch them.

他们的想法是，得赶在黑客拿到这种模型、开始用它来挖掘漏洞并趁还没修复前发动攻击之前，先把问题解决掉。

And the idea is that they want to do that before a model like this falls into the hands of hackers who could start doing the same thing and exploiting vulnerabilities before they're patched.

从预算的角度来看，他们发现实际花的钱比预想的要多。

On the budget side of all this, so they're finding it's more expensive than they may have thought.

高管们有没有跟你说，他们打算增加预算来支持这类模型？他们现在是怎么考虑的？

Are executives telling you that they are planning to expand their budget to make room for models like Methos, or where do they stand there?

对。

Yeah.

Mean 还联系了 Zscaler，这是一家也在测试该技术的公司。

Mean, also spoke to Zscaler, which is another company that is testing this.

他们告诉我，虽然这确实是个巨大的飞跃，成本也比以前的代码扫描工具高，但他们已经在为下一年的预算做安排了。

They told me that while it is essentially a step change, more expensive than their previous code scanning tools, they are just budgeting for it in the year ahead.

而且，你知道的，在某些情况下，这可能意味着需要跟财务总监进行一场艰难的对话，说：‘看，这笔账单看起来太离谱了，但相信我，这笔钱花得值。’

And, you know, in in some cases, that might have looked like a difficult conversation with the CFO saying, look, this bill looks crazy, but trust me, it's going to be worth it.

据我了解，我觉得CFO、CEO甚至董事会开始意识到这件事的严重性了，特别是围绕Methos和类似模型带来的那种担忧。

And from what I've heard, I think, you know, CFOs, CEOs and even boards are starting to understand the gravity of this, especially with just the amount of, I think, fear that's surrounding, you know, mythos and comparable models.

所以据我了解，公司愿意承担这笔费用。

So from what I've heard, companies are willing to put up that cost.

但与此同时，他们也在谨慎使用这些昂贵的模型，确保不会花冤枉钱。

But at the same time, they're trying to be very judicious about how they use these expensive models to make sure they're not spending more than they need to.

没错。

Right.

好的。

Okay.

所以网络安全公司都在说，我们会为这个项目腾出预算。

So the cybersecurity companies are saying, Hey, we're going to make room in our budget for this.

那这对网络安全客户的支出意味着什么？

What does this mean then for spending from cybersecurity customers?

我在想CrowdStrike和Zscaler的客户会怎么样。

I'm thinking about CrowdStrike and Zscaler's customers.

成本总得转嫁出去吧？

I mean, the costs are going have to get passed on, right?

所以价格会上涨，还是企业会在安全上花更多钱？

So are prices going to go up, or are we going to see enterprises spending a whole lot more on security?

你觉得最有可能的情况是什么？

What's our best guess there?

是的。

Yeah.

我也跟一些客户聊过，主要是首席信息安全官，他们普遍表示，预计自己也会增加投入，不管是不是直接使用Mythos，还是用基于Mythos的软件。

I spoke to customers as well, chief information security officers, and essentially what they are saying is they also expect to spend more, you know, whether that's actually using mythos or software powered by mythos.

有些公司干脆直接用了现成的模型，比如Anthropic的Claude和Opus，这些模型虽然不如Mythos强大，但在发现漏洞方面已经相当不错了。

Or in some cases, you know, companies have started to just use the models that are already available, like, you know, Anthropic, Claude, Opus models, which are also, you know, not as good as mythos, but pretty good at finding these vulnerabilities.

他们说，现在每年要多花几百万美元，就为了运行这些模型，持续扫描自家软件，找出潜在的安全漏洞。

And they say that they're budgeting more sometimes in the millions of dollars per year just to run those types of models to to constantly scan their software and and find potential vulnerabilities.

总的来说，我听到的都是一个共识：企业都意识到AI威胁即将到来，所以愿意提前加大投入，避免被曝光重大安全事件。

So what I'm hearing across the board is that, you know, companies see this potential wave of AI threats coming, and they're willing to spend ahead of it to make sure that they're not the victim of an embarrassing hack.

Mythos现在会向更多客户开放吗？

Is Methos going to be available at all to a wider slate of customers now?

刚推出的时候，他们说只提供给顶级银行和网络安全公司。

Initially, when it came out, they said, We're only giving it to top banks and cybersecurity companies.

现在看来，会不会扩大范围，向更多用户开放？

Is it looking like it's going to be wider or have a wider release at all?

是的。

Yeah.

所以 Anthropic 表示，在接下来的几周里，他们将开始发布与 Mythos 相当、面向更广泛用户的模型。

So Anthropic has said that in the coming weeks, they're going to start releasing models that are comparable to mythos more broadly.

但还不清楚这些模型是否真的能像 Mythos 那样执行网络攻击类任务，因为 Anthropic 也会加入安全限制，让模型明确表示不会去入侵你的系统。

But it's not clear if those models would actually be capable of doing the type of cyber offensive work that mythos does because Anthropic also is going to put in guardrails that essentially will make the model say, I'm not going to hack into your system.

其实也有办法做到这一点，理论上可以规定只在受控的安全环境中，且针对用户自有系统进行测试时才允许操作，但这实施起来会有些复杂。

There are ways for them to do that that would potentially theoretically say we're only going to hack into your system in a safe environment where you're doing it to your own systems, but that can be a little bit complicated.

而 Anthropic 对具体的发布计划一直比较低调，没有透露太多细节。

And Anthropic has been a little bit tight lipped about what exactly that rollout will look like.

所以目前还不太清楚这些能力什么时候能广泛普及。

So it's not exactly clear how much of these capabilities are going to become broadly available anytime soon.

明白了。

Got it.

好的。

Okay.

我想再回到微软的话题上。

I want to go back to Microsoft for just a second.

这周是Build大会。

This week is the Build Conference.

你这回会亲自去旧金山现场。

You're going to be there on the ground in San Francisco.

你打算在现场向与会者和高管们问哪些关键问题？

What are the big questions that you're planning to ask all the attendees and executives on-site?

是的。

Yeah.

微软这次Build大会，我从消息源那里听说，重点会放在推广自家研发的AI模型，吸引开发者。他们不仅会推出自家的代码生成模型，还有推理、语音转文字、图像生成这类模型。

So Microsoft, with this Build conference, one theme that I have been told by sources that's going to come up is Microsoft trying to essentially pitch its own homegrown AI models as appealing to developers, and specifically they're releasing their own in house coding models as well as models for reasoning and things like, you know, transcription and image generation.

我对开发者最想问的是，到底要怎么样，你们才会更广泛地使用微软的模型？

And, you know, my question for developers is just, you know, what would it take for you to start using Microsoft's models more broadly?

是价格问题吗？

Is it about cost?

还是性能问题？

Is it about performance?

是不是只是因为除了Anthropic和OpenAI这两家，大家想多点别的选择？

Is it just having, you know, different options besides the Anthropic and OpenAI duopoly on models.

所以我想了解一下，微软这个宣传对开发者来说到底有多吸引人。

So I'm going to try to get a sense of how appealing that pitch that Microsoft is making is with developer customers.

另外，我想接着我们上周的报道说一下，我很想知道客户们会怎么看待微软可能要发布的芯片。

Well, and also, was going to say, piggybacking off our reporting, I think it was last week, I am curious to see how customers might be thinking about the chips that Microsoft could be looking to unveil.

我们以前没觉得微软会在芯片领域认真竞争，但根据我们的调查，他们似乎打算推出一些更便宜、更专用的芯片，至少能产生一些影响。

And we hadn't thought that Microsoft was going to be a serious competitor in the chip category, but based on our reporting, I mean, they are going for maybe more cheaper, specialized chips that could at least make a dent in things.

所以现在去现场确实挺有意思的。

So it's certainly an interesting time to be there.

Aaron，谢谢你今天来参加。

Aaron, I want to thank you for coming on.

这位是Aaron Holmes，我们《The Information》的微软记者。

That is Aaron Holmes, our Microsoft reporter, here at The Information.

最近，前排部署工程师成了热门话题，我的同事Laura Bratton和Kevin McLaughlin在本周末的《周日洞察》专栏里深入探讨了这个新角色如何开始改变硅谷。

Forward deployed engineers have become all the rage as of late, and my colleagues Laura Bratton and Kevin McLaughlin went deep on how the new role is starting to shape Silicon Valley in this weekend's Sunday Insights column.

接下来请劳拉来聊聊她的发现。

I want to bring on Laura to talk about what she found.

劳拉，欢迎回到节目。

Laura, welcome back to the show.

很高兴你能来。

It's great to have you here.

嘿，阿卡什。

Hey, Akash.

好的。

Okay.

部署型工程师。

Forward deployed engineers.

提醒一下大家，这难道只是顾问的另一种说法吗？

Just remind us, is it just a fancy term for consultants?

现在这个岗位的具体职责到底是什么？

What exactly is the job description nowadays?

过去，顾问和软件工程师虽然会合作，但各自负责不同的工作。

Historically, consultants and software engineers would work together but do separate roles.

而前排工程师的职责是把这两部分都包下来。

The role of a forward deployed engineer is to do both.

他们既要像顾问一样，坐在客户身边，搞清楚客户对AI的需求——该用什么模型、需要开发哪些应用，这些原本都是顾问的工作。

They are both the consultant who will sit with the customer and figure out what their need for AI is, which models they should use, what kind of applications they might need, which is traditionally the role of a consultant.

然后他们还会亲自参与打造或部署这个产品。

And then they'll actually help build that product or implement that product.

明白了。

Okay.

那你这个周末为什么偏偏想写这个话题呢？

So what made you want to write specifically about it this weekend?

我们其实已经聊这个话题很久了。

Because we've been talking about this for quite a while now.

是有什么新的变化吗？

Did something change with the landscape?

对。

Yeah.

我觉得主要是因为越来越多的公司在财报电话会上提到，他们正在招聘部署工程师。

I think it's really just that more companies are talking about the fact that they're hiring for deployed engineers in their earnings calls.

Workday在最近一次财报电话会上也提到了这一点，他们今年开始招聘部署工程师，目前团队里已经有大约25人了。

Workday mentioned it in their latest earnings call, and they started hiring for deployed engineers this year and have about 25 on staff now.

然后更大的新闻是，Meta和Google都表示他们将引入前端部署工程师。

And then in bigger news, Meta and Google have said that they're gonna bring on forward deployed engineers.

这些显然都是我们报道中提到的那些大厂。

And those are obviously the bigger the bigger names that we cover.

所以我觉得，其实就是那些最大的科技公司现在都在模仿帕兰提尔（Palantir）当初的模式，也就是部署四名工程师。

So I would say it's really just that the biggest of big tech is now copying what was once just Palantir's model of having four deployed engineers.

这也是他们推动企业界大规模采用人工智能的关键举措之一。

And it's really a big part of their push to scale up enterprise adoption of AI.

不过，你文章里最有趣的一点是，这个岗位现在可是大家挤破头都想进这些公司的热门职位。

Well, one of the most interesting parts of your column though is, I mean, this the is role that people want to get at these companies.

这事儿看起来竞争特别激烈。

I mean, it seems like it's highly competitive.

薪资水平高得吓人，仅次于顶尖的AI研究员。

It seems like people are getting paid top dollar, second to maybe being a top AI researcher.

你文章里提到的薪资范围，有些情况确实很惊人。

I mean, the salary bands that you talked about in your story are pretty impressive in some cases.

所以，这个职位现在是不是就像当年想进Meta当产品经理一样，成了人人追捧的热门选择？

So what, this role has become the most popular thing since wanting to be a product manager at Meta or something like that?

是的。

Yeah.

我觉得现在还太早，不好下定论，因为几年前这种岗位的人还特别少，但增长速度却快得惊人。

I think it's still early days, so it's hard to say because there were so few forward deployed engineers just a couple of years ago, and the role was growing really rapidly.

LinkedIn告诉我，今年到目前为止，美国市场上新增了4004个部署工程师的职位。

So LinkedIn told me that their data shows 4,004 deployed engineers have been added in The US job market as role posting so far this year.

这些公司包括Cursor、OpenAI、Anthropic、Notion、软银等等。

That includes from companies like Cursor, OpenAI, Anthropic, Notion, Softbank.

现在这些岗位招聘量很大，你知道的，比去年新增的部署工程师职位数量多了大约3.5倍。

Companies are hiring these roles and that's, you know, much higher than about 3.5 times the amount of new job listings for forward deployed engineers we saw last year.

所以我觉得这个岗位确实越来越受欢迎，越来越多的公司开始设立这个职位。

So I think it's definitely more popular and more companies are taking on this role.

而且我觉得它的薪资也比一般的软件工程师岗位高，基本工资大概高出10%到15%。

And I think that, you know, it's also higher paid than a typical software engineering role, about 10% to 15% higher in terms of base salary.

我感觉不同公司对部署工程师的定义可能不太一样。

I get the sense that maybe the definition of a forward deployed engineer could vary across different companies.

这和你看到的情况一致吗？

Does that align with what you're seeing?

我聊过的几位顾问说，这个岗位在技术层面上其实是一样的。

Some consultants I spoke to said that the role is technically the same.

你真正需要的只是对AI有深入的技术理解。

You really just need a deep technical understanding of AI.

不过，如果你在Palantir工作，和在Salesforce、ServiceNow这些更传统的SaaS公司工作，这个岗位的具体内容肯定会有所不同。

But obviously the role is going to look different if you work at Palantir versus if you work at Salesforce or ServiceNow or one of these more traditional SaaS companies.

而且可以这么说，如果你在为 Anthropic 或 OpenAI 这样的模型提供商做前部署工程师，你对他们模型的技术理解可能会更深入，因为你能近距离看到他们的开发过程。

And the argument can be made that if you work as a forward deployed engineer for a model provider like Anthropic or OpenAI, you might have a better technical understanding of their models because you really have a front row seat to see how they're developing.

对。

Right.

我想转到你上周末和我们的 OpenAI 及 Anthropic 记者 Sri Mupiti 一起写的一篇人物专访。

I want to pivot to a profile that you wrote with our OpenAI and Anthropic reporter, Sri Mupiti, over the weekend.

你们俩都写了 OpenAI 首席营收官 Denise Dresser 的故事。

You both profiled Denise Dresser, who is the chief revenue officer at OpenAI.

这是我们周末的深度报道，你们俩深入探讨了她的背景、她带着什么样的资历进入这个角色，以及 OpenAI 究竟指望她来做什么。

It was our weekend big read, and you both went deep on who it is, what background she's coming into the role with, what it is OpenAI is relying on her for.

为什么在这个时间点，你们决定把焦点放在 Denise 身上，从商业角度进行报道？

Why did you decide to focus on Denise at this moment for business?

我觉得现在到了一个关键节点，Anthropic 的 AI 工具在企业客户中迅速普及。就像之前提到的，部署型工程师岗位的兴起，说明企业采用 AI 的方式比去年这个时候要严肃得多。

Well, I think we're really at a point where Anthropic AI tools have taken off with enterprise customers and enterprises, you know, like the rise of forward deployed engineers speaks to enterprises are really beginning to adopt AI in a much more serious way than we saw the same time last year.

这意味着，随着企业转向使用这些 AI 工具，模型提供商开始希望从企业的软件预算中分得更大的一块蛋糕。

And and what that means is that the model providers are starting to want to capture a larger share of their software budgets as enterprises shift towards using these AI tools.

Anthropic 在这方面确实做得非常成功。

And Anthropic has really been successful at doing that.

他们大约 80% 的收入都来自企业客户。

About 80% of their revenue comes from business customers.

OpenAI 也想做到这一点。

And OpenAI wants to be that too.

他们希望从一家深受消费者欢迎的科技公司，转型为专注企业市场的公司。

They want to change from being a super popular consumer focused tech company to a company for enterprise customers.

所以他们请来 Denise，就是为了帮他们推进这个战略。

So Denise was brought on to help them with that strategy.

她有整整十四年在 Salesforce 的工作经验。

She has, you know, fourteen years of experience at Salesforce.

她简直是这份工作的不二人选，因为你需要一个真正懂如何向企业销售 SaaS 的人。

She's pretty much the perfect person for this job, because you really want somebody who's been focused on selling SaaS to enterprises.

她不是 Slack 的 CEO 吗？

She the CEO of Slack, right?

我的意思是，那——

I mean, that's-

她并不是Slack的首席执行官。

She wasn't the CEO of Slack.

是的。

Yes.

没错。

Right.

她在Salesforce待了十多年，主要负责与大型科技公司建立和维护关系。

She spent, you know, more than a decade at Salesforce, basically managing a lot of the relationships with larger profile tech companies.

我接触过的很多人都提到，她帮助建立了与AWS的合作关系，并成功管理了这些合作。

She was credited by a lot of folks I spoke to for, for example, helping build a relationship with AWS and managing those relationships.

在她任期的最后几年，Salesforce收购了Slack之后，她转任Slack的首席执行官，并因成功推动Slack与Salesforce的整合而广受赞誉——毕竟两家公司的文化差异不小。

And then towards the last couple of years of her tenure, she transitioned after Salesforce acquired Slack to be the CEO of Slack and was really credited with helping successfully integrate Slack into Salesforce because there was sort of a culture clash among those two companies So

你能跟我们说说，她在Salesforce和Slack的经历中，有哪些经验可能会影响她未来在OpenAI的工作方式？

tell us, what did you learn about her time at Salesforce and at Slack that you think could inform the way that she might approach this role at OpenAI?

你刚才提到了她和客户，比如 AWS 这样的大型云服务商之间的关系。

You just talked about the relationships that she had with customers, big cloud providers like AWS, etcetera.

OpenAI 目前的情况正如你所说，挺有意思的是，根据我们的报道，他们的企业业务整体规模，虽然得看最新数据，但有一段时间，仅凭企业订阅量，它其实比 Anthropic 的企业业务还要大。

OpenAI is in this moment where, like you said, they It's funny because, you know, based on our reporting, actually, the enterprise business in aggregate, and we'd have to look at the latest numbers for this, but at one point, it was still larger than Anthropic's enterprise business just based on the enterprise subscriptions.

但我们显然知道，他们正试图在编程模型领域大举进军，让企业购买那套产品。

But we obviously know that the coding model is where they're trying to make a lot of inroads, getting companies to buy that suite of products.

所以，我很好奇，我们是否了解 Denise 的领导风格？她是如何处理这个角色的？她可能会如何调整公司架构，以便在 OpenAI 的企业市场竞争？

So, I I just wonder, we know anything about Denise's leadership style, how she approached the role, you know, how she might structure the company to compete in enterprises at OpenAI?

对。

Yeah.

我是说，Shree 对 OpenAI 那边的事更了解，而我写这篇报道的重点是深入挖掘她在 Salesforce 的经历，以及这段经历如何为她后来在 OpenAI 的工作打下基础。

I mean, can Shree knows more about the OpenAI side of things, and my real background for the story was diving into her history at Salesforce and how it set her up for her time at OpenAI.

我想说两点。

And what I would say two things.

她的同事们反复告诉我，她非常以客户为中心，对同事要求严格但也很公平。

Her colleagues told me over and over that she is super focused on customers and very tough, but fair with her colleagues.

所以我觉得有一个特别明显的例子。当时正好是财年末，她马上要签下一笔大单。

And so I think one really, stark example was that, you know, it was the end of the fiscal year, and she was about to close a deal.

结果因为某些原因，那笔交易没在午夜前敲定，她因此失去了这笔订单的佣金资格。

It didn't for some reason close by midnight, and she was no longer eligible for commission on the deal.

但她还是把同事们留下来，一直熬到凌晨三四点，直到把单子签下来。

But she still kept her colleagues there until, like, three or four in the morning to close the deal.

她就是这样的人，如果客户不满意，她会在周末、晚上十点甚至凌晨五点把团队叫来开会，或者发短信给同事。

She, she also, is the type of person to call a team into a meeting on a weekend or at 10PM or text her colleagues at five in the morning if a customer is unhappy.

所以她真的非常非常注重客户，我觉得这对像OpenAI这样的公司来说非常理想，因为过去这里更多的是AI研究人员，专注于构建更大、更好的模型。

So she's really, really focused on customers in a way that I think is ideal for a company like OpenAI, where historically you have more AI researchers that are focused on building bigger and better models.

现在你有一位领导者，她专注于客户，以及他们能否成功部署OpenAI的产品。

Now you have a leader who's focused on on customers and and how successfully they're able to implement OpenAI's products.

最后我还想提一点，我觉得这也是个很好的例子。

One last thing that I wanna mention is she also I thought this was a a really great example as well.

她提出了Slackbot这个概念，这在它正式推出至少一年前就出现了，而现在它已经成为Salesforce战略的核心部分。

She pitched Slackbot, which has now become central to Salesforce's strategy, you know, at least a year before it launched.

我听说，其他Salesforce高管当时更关注Agent Force，但这个产品刚推出时表现并不理想。

And other Salesforce executives, I'm told, were prioritizing agent force, which kind of struggled, when it was initially launched.

我觉得这充分体现了Denise的前瞻性眼光——她不仅能妥善处理客户关系，还能准确预判哪些产品未来会真正赢得客户青睐。

And I think that that that was a really big tell of Denise's forward looking abilities and and how she not only can manage relationships with customers, but also really see into the future about like what products might be successful for customers.

很好。

Great.

Laura，非常感谢你来做客，我也希望大家去读一读那篇专访。

Well, Laura, I want to thank you for coming on, and I encourage everyone to check out that profile.

文章就在我们的网站上。

It is on our website.

这是我们周末的重磅阅读推荐。

It is our weekend big read.

这位是Laura Bratton，我们TITV的AI记者，也是《应用AI》通讯的作者。

That is Laura Bratton, our AI reporter and author of our Applied AI newsletter here on TITV.

Meta的前首席技术官迈克·施罗普弗，如今创立了Gigascale Capital，专注于能源、材料与可持续发展交叉领域的投资。

Meta's former Chief Technology Officer, Mike Schroepfer, is now the founder of Gigascale Capital, a new fund focused on the intersection of energy, materials, and sustainability.

这个基金募集了2.5亿美元，我想请迈克来聊聊他的投资策略。

The fund raised $250,000,000 and I want to bring on Mike to talk more about his playbook for his fund.

迈克，欢迎来到节目。

Mike, welcome to the show.

很高兴你能来。

It's great to have you here.

谢谢。

Thank you.

很高兴能参加。

Glad to be here.

你是在两年前左右创立了Gigascale Capital，对吧？

So you founded Gigascale Capital, I think it was a little over two years ago now.

现在你们已经有了2.5亿美元的新资金。

You've got $250,000,000 in new funding.

能说说你当初设立这个基金的初衷吗？这笔资金将如何帮助你实现目标？

Tell us about what your vision for the fund originally was and how this capital is going to help you get there.

是的。

Yeah.

我们三年前刚起步时，投资理念其实挺反主流的——我们认为，实体经济、如何构建经济体系，才是打造风险投资规模公司的最有意思的领域。

When we started three years ago, our thesis was quite out of the norm, which was that the physical world, how we build our economy, was the most interesting place to build venture scale companies.

那时候这可不是主流看法。

And that was not the common wisdom.

现在最让人欣慰的是，我们的方向真的行得通了。

I would say the fun thing about where we are now is it's working.

现在根本不用再说服谁。

I don't have to convince anyone.

我们投的每一家公司，大家都不再问‘有没有人需要他们的产品’。

All of our companies are in a place where the question isn't do people want what they're building.

大家问的是：‘他们能多快把产品推出来？’

It's how fast can they get it there?

我们关注的是数据中心的清洁能源。

We're talking clean power for data centers.

我们谈的是任何你想制造的东西所用的清洁材料。

We're talking clean materials for anything you want to build.

我们谈的是更快速、更高效、更便宜的计算能力，无论是用于推理还是训练。

We're talking, you know, compute in faster, better, cheaper for inference and for training.

现在，对这些更好、更快、更便宜的技术的需求被提前拉到了眼前，而这些技术彻底颠覆了过去一百年我们做事的方式。

This is just like a tremendous amount of demand has been pulled forward to build all of these things that are better, faster, cheaper than the way we've been doing it for the last one hundred years.

在这么多领域和这么多具体技术中，因为能源市场实在太大了。

So of all these segments and of all these really specific technology, because energy is so big.

可持续性至关重要，整个领域里到处都是亟待解决的问题。

Sustainability is huge and there are problems all across the category that need to be solved.

告诉我，最近七天里，有没有哪一项创新让你特别着迷，让你觉得非得在这一个细分领域大举投资不可？

Tell us, what is the one innovation that just last seven days alone that you have just been so obsessed with that you could see yourself really making a ton of investments in this one specific category?

我给你举个具体的公司和产品吧，这样更清楚。

Well, I'll give you a specific company and a specific product, because I think that's even better.

有一家公司叫Heron Power，他们把一种整整一百年前就有的技术——变压器——重新做了一遍。

There's a company called Heron Power, and they've taken this literally 100 year old technology, the transformer.

我说的是电力变压器，不是AI模型里的那个变换器。

And I'm talking about a power electric transformer, not the transformer in an AI model.

这东西现在直接卡住了数据中心AI建设的脖子。

Have to be But it is literally the piece of equipment that is bottlenecking data center AI build out right now.

交货周期要等四五年呢。

That's, you know, four to five year back order.

它依赖一种特别定制的特种钢材。

It relies on this very special kind of bespoke steel.

其他所有部件你都得自己造。

You've got to make all the rest of it.

Heron Power的艾伦说，过去十年我们一直在研发这种电力电子新技术。

And Aaron Power drew back, Alino said, hey, we've been for the last ten years developing a new technology for this power electronics.

现在每辆电动车里用的都是这个技术。

It's in every electric vehicle that shipped.

我们已经实现了规模化应用。

So we've done this at scale.

咱们可以把电动车里的这项技术用到电网上去。

Let's like take that technology in the EV and put it on the grid.

我能做出一个体积只有三分之一大小的设备。

I can build a box that's a third the size.

它可以通过软件来编程。

It's software programmable.

以前我得在制造硬件时就定死所有参数，现在只要把它放上去，就能直接用于数据中心。

So instead of deciding when I build my piece of hardware, exactly what the specs have to be, I drop it on a pad and I can use it for a data center.

也可以用在太阳能电站。

I can use it for a solar farm.

还能用在电动车充电桩上。

I can use it for an EV car charger.

不管哪种场景，它都能正常工作。

It works in any of these conditions.

这就像是一个万能的乐高积木，哪儿都能用。

So it's sort of a single Lego block that works for everything.

所以这家公司明年就要开始量产商用产品了。

And so this is a company that, you know, will be shipping commercial products next year.

他们是去年才成立的。

They were founded last year.

另一个类似的情况是，他们的订单多到都忙不过来。

And they, you know, another one of these, they have more demand than they know how to deal with.

这正是新技术解决老问题的一个例子。

So it's an example of new technology solving an old problem.

你觉得，有什么反主流的观点在影响你对巨能资本的投资决策？

What would you say is a contrarian belief of yours that is informing the way you plan to invest at Gigascale Capital?

我这么问是因为，能源问题其实人人都看得见。

And I ask this because, energy, it is an obvious problem.

谁都明白这个问题必须解决。

Everyone knows it needs to be solved.

不管谁投钱进来，都是好事。

Any money going into the space is good.

那你个人的逆向思维观点是什么呢？

But what is the contrarian view that you hold?

我觉得一个例子是，我敢打赌到2030年，海底的算力会比太空里的还多。

Well, I think one is, I bet you we'll have more compute in the ocean than we do in space by 2030.

具体说说。

Say more.

为什么？

Why?

我们拥有巨大的、尚未开发的10太瓦波浪能资源。

We have this massive 10 terawatt untapped resource of wave power.

造一艘船的成本比发射一颗卫星还要低。

It is cheaper to build a ship than it is to launch a satellite.

所以从基本原理来看，我们在公海上应该进行更多的计算，而不是在太空中。

So on first principles, we should be doing a lot more compute in the open ocean than we should be in space.

你们也在海洋领域有投资吗？

And you have investments in the ocean?

对，有一家公司叫 Anthalasa。

Yeah, there's a company called Anthalasa.

还是那句话，算是早期押注。

Again, early bet.

那是很多年前的事了。

We made it many years ago.

属于逆向投资。

Contrarian bet.

不过现在彼得·蒂尔刚投了一大笔钱给他们。

But now Peter Thiel has just come in and gave them a bunch of money.

所以他们现在有足够的资源去推进这个项目。

So they now have the sort of resources to go ahead and build this.

他们在俄勒冈海岸外已经有设备在运行了，接下来就要进入全面部署和开发阶段。

They've had units in the ocean off the coast of Oregon working, and now they're going towards full scale deployment and development.

所以 hopefully 明年或者后年，他们就能开始运行一些负载了。

So they're going be running a few payloads hopefully next year or year after.

听起来你对太空数据中心并不是特别看好。

So it sounds like you're not that bullish then on data centers in space.

我觉得埃隆·马斯克会把这件事做得非常出色。

I think Elon Musk is gonna do a great job of it.

但我觉得这对其他人来说是个陷阱。

I think it's a trap for everyone else.

我的意思是，他在卫星成本和发射成本上拥有批发价优势。

I mean, he's got wholesale costs for satellite costs and for launch costs.

它将会

It's gonna

非常难以超越。

be really hard to beat that.

所以我确定，我是说，星链，你知道，V3版本大概是20千瓦。

So I'm sure I mean, and Starlink, you know, V3 is probably 20 kilowatts.

所以他已经在那儿了。

So he's of already there.

所以我觉得，没人能挡得住埃隆·马斯克做成这件事。

And so I wouldn't bet against Elon Musk doing it for him.

我觉得其他想追这个梦的人，迟早要吃大亏。

I think everyone else chasing that dream is in for a rough ride.

但话说回来，你不觉得SpaceX能为其他公司打开市场吗？

But I mean, you don't think that SpaceX could open up the market for other players?

我知道这公司垂直整合得特别厉害，但听起来你并不认为SpaceX会真正带动整个太空数据中心市场，除了他们自己搞之外。

I I know this is a very vertically integrated company here, but it sounds like you don't think that SpaceX is going to open up the market for space data centers more broadly, other than just them launching their own.

Starlink有为卫星互联网打开市场吗？

Has Starlink opened up the market for satellite internet?

没有。

No.

至少我没看到有。

Not to my recollection.

这生意太棒了。

Great business.

产品非常棒。

Awesome product.

我坐飞机只选那些有星链服务的航空公司，比如联合航空，谢谢。

I only fly, you know, airlines, United, thank you, that have Starlink.

在这方面，埃隆做得非常出色。

Elon is crushing it in this regard.

再说一次，我觉得他在建设SpaceX数据中心方面会做得很好。

Like, again, I think he'll do a great job building SpaceX data centers.

我觉得我投的其他项目里，有不少在海洋、陆地还有视觉融合方面都挺让人兴奋的。

I think there are many other things that I'm investing in that are are exciting in the ocean, on land, vision fusion.

我们在下一代太阳能电池方面有很多可以大展身手的地方。

We have a lot of places to build next generation solar batteries.

我们可以把电网建设得更快。

We can build the grid much faster.

所以我对陆地上的解决方案挺看好的。

So I'm sort of bullish on terrestrial solutions for this.

我觉得埃隆在太空领域也会做得非常出色。

And I think Elon will do a great job in space.

你如何看待中国在能源创新方面的表现？

How do you think about all the energy innovations in China?

他们在这一块已经做了很多工作。

They have done a lot of work in that area.

你有在那边的投资吗？

Are you making any investments in companies over there?

我们目前没有在中国投资。

We are not investing in China.

我们的投资完全集中在美國、歐洲和世界其他地區。

We're wholly in The US and Europe and the rest of the world.

你如何看待中国在能源领域的创新，以及这些创新如何助力他们在人工智能竞赛中取得优势？

How do you think about the energy race insofar as all the innovations that China has made in energy and how that could help them in the AI race altogether?

我觉得中国有很多值得借鉴的地方。

Well, I think there are lots of lessons from China.