对AI顶尖人才而言，金钱已不再重要

本节目由Indeed赞助支持。

Support for this show comes from Indeed.

如果您正在寻找精通您领域的一流人才，Indeed表示他们可以提供帮助。

If you're looking to hire top tier talent with expertise in your field, Indeed says they can help.

Indeed的付费职位能让您的招聘信息脱颖而出，并让您接触到能够带来所需成果的优质候选人。

Indeed sponsored jobs gives your job the best chance at standing out and grants you access to quality candidates who can drive the results you need.

花更多时间面试符合您所有要求的候选人。

Spend more time interviewing candidates who check all your boxes.

减少压力，节省时间，立即通过Indeed付费职位获得更佳成果。

Less stress, less time, more results now with Indeed sponsored jobs.

本节目的听众将获得75美元的付费职位抵扣额度，帮助您的招聘信息获得应有的优质曝光，前往indeed.com/foxbusiness领取。

And listeners of this show will get a $75 sponsored job credit to help get your job the premium status it deserves at indeed.com/foxbusiness.

请立即前往indeed.com/foxbusiness，并说明您是在本播客中了解到Indeed的，以此支持我们的节目，indeed.com/foxbusiness。

Just go to indeed.com/foxbusiness right now and support our show by saying you heard about Indeed on this podcast, indeed.com/foxbusiness.

条款和条件适用。

Terms and conditions apply.

招聘，就用Indeed，做对的事。

Hiring, do it the right way with Indeed.

本节目由Vanta赞助。

Support for this show comes from Vanta.

Vanta通过人工智能和自动化技术，帮助您快速实现合规，简化审计流程，并推动交易达成，向客户证明您对安全的重视。

Vanta uses AI and automation to get you compliant fast, simplify your audit process, and unblock deals, so you can prove to customers that you take security seriously.

您可以把Vanta视为一位全天候、由AI驱动的安全专家，能随您一同成长。

You can think of Vanta as your always on AI powered security expert who scales with you.

因此，像Cursor、Linear和Replit这样的顶尖初创公司都使用Vanta来确保安全并持续保持安全状态。

That's why top startups like Cursor, Linear, and Replit use Vanta to get and stay secure.

立即前往vanta.com/vox开始使用。

Get started at vanta.com/vox.

网址是vanta.com/vox。

That's vanta.com/vox.

vanta.com/vox。

Vanta.com/vox.

本节目由Odoo赞助。

Support for this show comes from Odoo.

经营企业已经够难了，为什么还要用十几个互不相通的应用程序让事情变得更复杂呢？

Running a business is hard enough, so why make it harder with a dozen different apps that don't talk to each other?

介绍Odoo。

Introducing Odoo.

这是你一生中唯一需要的商业软件。

It's the only business software you'll ever need.

它是一个一体化、完全集成的平台，能让您的工作更轻松。

It's an all in one fully integrated platform that makes your work easier.

客户关系管理、会计、库存、电子商务等。

CRM, accounting, inventory, ecommerce, and more.

最棒的是什么？

And the best part?

Odoo以远低于传统方案的价格，取代了多个昂贵的平台。

Odoo replaces multiple expensive platforms for a fraction of the cost.

这就是为什么成千上万的企业已经做出了转变。

That's why over thousands of businesses have made the switch.

那你呢？

So why not you?

前往 odoo.com 免费试用 Odoo。

Try Odoo for free at odoo.com.

那就是 odoo.com。

That's odoo.com.

大家好，欢迎收看《Decoder》。

Hello, and welcome to Decoder.

我是《The Verge》的主编尼尔·帕特尔，《Decoder》是我探讨重大理念与其他问题的节目。

I'm Neil Patel, editor in chief of The Verge, and Decoder is my show about big ideas and other problems.

问题。

Problems.

今天，我们要谈谈人工智能人才的竞争。

Today, we're gonna talk about the war for AI talent.

目前，全球最热门的就业市场是人工智能研究人员，而这些人的绝大多数都集中在旧金山湾区少数几家价值极高、增长极快的公司里。

Right now, the hottest job market on the planet is for AI researchers, and the vast majority of these people are concentrated into a small number of hugely valuable, extremely fast growing companies in the San Francisco Bay areas.

这些公司正在支付科技行业历史上最高的薪资，以挖走彼此的研究人员。

And these companies are paying some of the highest salaries in the history of the tech industry to poach researchers from one another.

每当有一位人工智能研究人员从一家公司跳槽到另一家公司时，他们都会明确告诉我们原因。

And it feels like every time one of these AI researchers leaves one company for another, they tell us exactly why.

有时他们只是辞职去当诗人。

Sometimes they're simply resigning to go be a poet.

有时他们是在追寻一项使命。

Sometimes they're chasing a mission.

有时他们担心人工智能会危及人类，摧毁所有工作，并让世界陷入混乱。

Sometimes they're worried that AI is going to imperil humanity, destroy all jobs, and plunge the world into chaos.

他们真的在说这些话。

They're really saying these things.

他们会在X平台上发布这些观点，写博客文章，或者像一位前OpenAI安全研究员那样，撰写一篇完整的《纽约时报》评论文章。

They're publishing these notes on x in blog posts or in the case of one former OpenAI safety researcher by writing a full New York Times op ed.

我一直很想深入探究一下，这些AI领域的人才流动背后到底发生了什么。

I've been dying to really dig in and try to unpack what's going on with all these talent moves in AI.

今天我的嘉宾是《The Verge》的资深记者海登·菲尔德，他一直密切关注着AI行业的人才循环现象。

So my guest today is Verge senior reporter Hayden Field, who's been covering the revolving door of the AI industry really closely.

他还深入观察了推动AI从业者跳槽的更广泛文化，以及那些不择手段争夺人才的公司。

And also the broader culture that's motivating the AI workers to jump ship and the companies that are ruthlessly trying to hire them.

这些动机各不相同。

Those motivations vary.

当然，这些人都拿到了极其丰厚的薪水。

Sure, all these people are paid extravagant salaries.

但正如你将听到海登所说，更强大的驱动力是理念和使命。

But as you'll hear Hayden say, a more stronger motivating force is ideology and mission.

绝大多数从事AI工作的人相信，他们正在做的事情将彻底改变世界，他们其实并不迫切需要更多的钱。

The people working on AI by and large believe that what they're doing is going to radically change the world and they're not really in desperate need of more money.

这彻底改变了可能促使人们离开OpenAI加入Anthropic，或在X AI被SpaceX收购后辞职的激励结构。

So that really changes the incentive structures that might push people to leave say OpenAI for Anthropic or to quit Elon Musk's x AI now that it's been acquired by SpaceX.

与此同时，AI公司的激励目标正从融资转向盈利。

At the same time, the incentives of the AI companies themselves are going from raising money to making money.

报道称，OpenAI甚至可能包括Anthropic今年都可能上市，这将创造前所未有的财富。

Reporting suggests that OpenAI and maybe even Anthropic could go public this year, and doing so would create a historic amount of wealth.

这还将给这些公司带来新的压力，要求它们更透明地说明资金使用情况，并对迄今为止所筹集的巨额投资回报承担更大责任。

It would also put new kinds of pressure on these companies to be more transparent about how they spend money and to be much more accountable for returning on the huge investments that they've raised so far.

这场对话涉及的内容很多。

There's a lot in this conversation.

目前的AI行业充满了戏剧性。

The AI industry right now is full of drama.

这里有大人物、激烈的竞争、巨额资金，还有关于世界末日的超长博客文章。

There's big characters, bitter rivalries, lots of money, and really, really long blog posts about the end of the world.

好的。

Okay.

以下是Verge的资深AI记者海登·菲尔德。

Here's Virg senior AI reporter Hayden Field.

我们开始吧。

Here we go.

我想了解一下科技劳动力市场的现状，以及AI对它造成了多大的扭曲影响。

I wanted to check-in on the state of the tech labor market and how big of a distorting effect AI is having on it.

不是从自动化角度，而是从AI实验室及其员工的行为如何对周围一切产生重大影响的角度。

Not from the automation perspective, but rather how the behavior of the AI labs and the people who work there is having a major impact on everything around them.

现在，几乎每周都有人离开OpenAI去Anthropic，或者反过来，或者去其他实验室，甚至干脆辞职去森林里隐居。

People are leaving OpenAI for Anthropic or vice versa or going to some other lab or quitting to go live in the woods basically every week right now.

通常是一位知名人士离开大型AI实验室的安全团队或其他部门，几天后又加入另一家实验室。

Usually it's someone high profile leaving the safety team or some other division of a big AI lab and then joining another one a few days later.

有时他们辞职后，几周后才宣布自己被挖走了。

Sometimes they quit only to announce they've been poached a couple weeks later.

有时他们辞职后声称我们都会死，然后要退休去欧洲写诗——我们刚看到一位Anthropic的研究员本月离职时说：'世界正处于危险之中。'

Sometimes they quit and say we're all gonna die and that they're gonna retire to Europe to write poetry, which we just saw with an anthropic researcher who left the company this month saying, quote, the world is in peril.

这有点令人担忧。

That's just a little worrisome.

你每天都关注这个领域。

You follow this industry every day.

从人员跳槽和离职的角度来看，你如何描述当前人工智能行业正在发生的事情？

How would you characterize what's happening in the AI industry right now just in terms of people getting new jobs and quitting their old jobs?

现在确实太疯狂了。

So it is crazy right now.

你说得对。

You're absolutely right.

目前人工智能行业是历史上竞争最激烈的时期。

It's the most competitive it's ever been right now, the AI industry.

人员跳槽、辞职和其他流动情况正在不断加剧。

And the amount of defections, resignations, and other moves is only intensifying.

我觉得每周都会有人做出高调的离职举动。

I feel like every week, someone's doing a really high profile departure.

有时候一周内每天都会发生。

Sometimes it's every day in a week.

他们总是会在网上发布非常戏剧性的声明，无论是辞职信，还是向曾经共事的二十多个人致谢。

They always do a really dramatic announcement online, whether it's a resignation letter or just a thank you to, like, 20 different people that they worked with.

而这里涉及的薪酬待遇也非常可观。

And then there's all the pay packages at stake here.

有大量资金在流动，比如据报道，Meta 给出的薪酬包高达十亿美元。

There's a lot of money at play, like, you know, reported billion dollar pay packages for Meta.

谁知道 Anthropic 和 xAI 现在给人们开出了多少薪酬。

Who knows how much is being offered to people at Anthropic, x AI right now.

每个人可能都因为 SpaceX 的合并而发财了。

Everyone just probably got rich from the SpaceX merger.

而且，他们也都在以最疯狂的方式被争相挖角。

And also, they're all constantly being courted in the craziest ways too.

比如，马克·扎克伯格亲自为想招募的工程师下厨煮汤，还有萨姆·阿尔特曼亲自打电话给潜在候选人。

Like, Mark Zuckerberg making home cooked soup personally for, like, engineers he was trying to recruit, and, you know, Sam Altman personally calling potential recruits.

现在有太多事情在发生，大量资金正在流转。

There's just a lot going on, and there's a lot of money changing hands.

但我认为重要的是要记住，在某个时刻，对许多人来说，金钱的意义已经不如个人使命重要了。

But I do think it's important to remember that at some point, money doesn't really mean as much to a lot of these people as personal mission.

这就是我们正在看到的情况。

So that's what we're seeing.

比如，目前在不同公司之间流动的很多人，是因为他们所在的公司不再符合他们的个人使命，或者他们不再信任该公司的领导层。

Like, a lot of people that are moving around right now from one job to another are doing it because a company no longer aligns with their personal mission or they no longer have faith in the leadership at that company.

所以，这通常是主要的动机。

So that's usually a lot of the motivation.

是的。

Yes.

有些人离开是为了获得令人瞠目的薪酬包，但我认为，目前人工智能领域我们看到的大多数跳槽，都与更多的人际矛盾或个人信念有关，人们试图去到与自己价值观一致的地方。

Some people are leaving for like eye popping pay packages, but I would say the majority of different job moves we're seeing right now in AI have to do with more interpersonal drama or personal like belief statements and people trying to, you know, kind of go where their values align.

我可以问一个很直接的问题吗？

Can I ask a really rude question?

是的。

Yes.

到目前为止，任何一位AI研究员对一家AI公司的成败能产生多大影响？

How much does any single AI researcher make or break any single AI company at this point?

你可能会觉得影响没那么大，但我认为某些人确实具有远超常人的影响力，这就是为什么有些人会拿到与之匹配的超高薪酬。

You know, you would think not that much, but I will say I think that certain people have really outsized influence, and that's why certain people are seeing pay packages that reflect that.

比如，有些人单独推动了GPT-5、GPT-4的研发，还有人单独主导了大量Grok项目，比如Grok Imagine。

Like, there are people that individually drove GPT five, GPT four, people that individually drove a ton of Grok stuff, like Grok Imagine.

公司里的每个人都清楚这些人的身份。

And everyone at the company knows who those people are.

并不是说OpenAI、XAI、Anthropic或谷歌的任何一名员工，从CEO的角度来看，都值这么多钱。

It's not like any employee at OpenAI or XAI or Anthropic or Google is going to be just, you know, worth this much money in terms of pay from a CEO's perspective, I guess.

但确实有少数几个人拥有巨大的影响力，他们非常擅长调动公司团队中的其他人。

But there are a handful of people that have outsized influence and that are really good at like marshaling all the other people on their team at the company.

他们特别擅长提出创意并解决这些复杂的问题。

They're really good at ideating and solving these intense problems.

这就是为什么这些人的跳槽会成为新闻头条。

You know, that's why some of these people's moves make headlines.

当你说到人们因为个人使命而离开时，我有一部分是相信的。

When you say that people are leaving because of their personal mission, there's a part of me that just buys it.

对吧？

Right?

你已经赚了这么多钱，即使有人给你更多钱，可能也对你的人生没什么影响，但你觉得某家公司的领导者价值观错了，所以你想去另一家价值观正确的公司工作，或者你觉得某家公司更接近通用人工智能，你想更靠近这个理念——这种情况下，人们公开宣称自己离开是因为原公司与个人使命不一致，而新公司更契合，这种情况经常发生吗？

You've made so much money that someone offering you even more money will have maybe no impact on your life, but you think the leader of one company has the wrong values, so you're gonna wanna go work for another company that has the right values or you think one company is closer to AGI and you wanna be nearer to that idea, how often does that come up publicly where the people leaving jobs or switching between these companies are actually saying the place I was was misaligned with my personal mission and I'm going to another place that's more closely aligned?

我会说几乎每次都是这样。

I would say almost every time.

有很多人不会直接这么说，但他们会暗示这一点。

There are a lot of people that won't say that in so many words, but they will imply it.

比如上周离职并因辞职信走红网络的那位Anthropic研究员，是的，他去当诗人了，所以这次他并没有去另一家AI公司。

Like, the anthropic researcher that left last week that went viral online for his resignation letter, yeah, he went to go be a poet, so it's not like he was going to another AI company in that case.

但他确实提到，在Anthropic工作期间，他多次看到要让自己的价值观指导自己的行动有多难，无论是在自己身上还是在组织内部。

But he did say that throughout his time at Anthropic, he repeatedly saw how hard it is to let his values govern his actions, both within himself and within the organization.

因为他表示，他们不断面临压力，要放弃最重要的东西。

Because he said they constantly face pressures to set aside what matters most.

所以我认为，大多数离开这些公司的人，有些只会说，非常感谢，我要开启我的下一章了。

So I think, you know, most people that depart these companies, I mean, some of them will just say, thanks so much, off to my next chapter.

但很多人会说，我的下一章将专注于X、Y和Z，因为这些是我非常在意的事情。

But a lot of them will say, hey, my next chapter is gonna be about x y and z because this is what I care a lot about.

无论他们是去非营利组织，还是创办自己的AI实验室、基金会或类似监督性质的组织。

Whether they're leaving for a nonprofit, may maybe they're starting their own AI lab or their own, you know, foundation or organization that's like more of a watchdog.

也许他们离开是为了加入另一个AI实验室，因为他们特别想在某位人物手下工作，钦佩他的X、Y和Z。

Maybe they're leaving for another AI lab that they specifically wanna work under this person because they admire x y and z about them.

有时他们会直接说，另一家公司与我的价值观不一致，比如我们看到《纽约时报》一篇前OpenAI员工的评论文章，她明确表示广告业务是她离开的主要原因。

Sometimes they say outright that the other company didn't align, like, you know, we saw New York Times op ed from an ex OpenAI employee who said, you know, the ad stuff was really her her reason for for leaving.

而有时，我们只是通过他们说‘我很兴奋能去这家机构，因为这里有三个具体原因’来间接推断。

And then sometimes we see them just imply it by saying, hey, I'm really excited to work at this place because of these specific three reasons.

这不就像写一篇‘我为什么离开纽约’的文章吗？

Is this just like writing out why I'm leaving New York essay?

没错。

Yep.

就像你开了个Substack，必须黑一把《纽约时报》，然后立马跳槽去吸引关注，是这种感觉吗？

Like, you're starting a Substack and you've gotta trash the New York Times and you're right out the door to get attention to your sub is that is it the same vibe?

确实是同样的氛围，但我认为对很多人来说，背后有更深的意义。

It is the same vibe, but I think there's a lot more meaning behind it for a lot of these people.

他们绝大多数人真的是虔诚的信徒。

They all really are true believers most of the time.

他们真心相信自己所做的事情正在直接地让世界变得更好或更糟。

They really believe that what they are doing is directly impacting the world for better or worse.

他们真的相信自己工作的公司将在社会影响方面起到至关重要的作用。

Really believe that where they work is going to matter hugely in terms of the societal impacts.

就在上周，OpenAI挖走了OpenClaw的创始人彼得·斯坦伯格。

Just last week, OpenAI snapped up OpenClaw founder Peter Steinberger.

那是那个催生了Motebook的开源AI代理框架。

That's the open source AI agent framework that gave birth to Motebook.

彼得在2025年11月发布了OpenClaw。

Peter released OpenClaw in November 2025.

上个月它变得流行起来，现在他已经在OpenAI找到了工作，薪酬待遇想必相当丰厚。

It got popular last month, and now he has a job at OpenAI, probably with a hefty compensation package.

西蒙·阿尔特曼在X上宣布了这一消息，称施泰恩伯格是‘一个充满关于未来智能代理相互协作、为人类做大量有用事情的惊人想法的天才’，并表示‘我们预计这将迅速成为我们产品核心的一部分。’

Simon Altman announced this news on X, calling Seinberger, quote, a genius with a lot of amazing ideas about the future of very smart agents interacting with each other to do very useful things for people and that quote, we expect this will quickly become core to our product offerings.

我们现在是不是已经到了这种地步：你随便做个像OpenClaw这样的项目，几个月后，萨姆·阿尔特曼就称你是天才，还给你巨额报酬？

Is this the pace we're at now where you can just create a project like OpenClaw and then a couple months later, Sam Altman is saying you're a genius and paying you huge sums of money.

甚至都不是收购你的项目，只是直接雇用你，然后把OpenClaw作为基础来发展。

To not even acquire your project, to just hire you and put OpenClaw on a foundation.

彼得的故事是一个绝佳的例子，展现了科技行业很多人所期待的那种创业成功模式，因为网上一直充斥着一种‘支持AI、作为独立个体快速构建’的持续奋斗文化。

Peter's story is a really good example of the type of entrepreneur success story that a lot of tech industry people are hoping for because, you know, there's a pretty constant hustle of pro AI, build quick as an independent person culture online.

这也很好地说明了AI实验室推进的速度有多快，以及它们在迭代和彼此竞争各项功能时所承受的巨大压力，尤其是在智能代理领域。

So it's also a pretty good indication of how fast the AI labs are moving and how much pressure there is for them to iterate and constantly, like, one up each other on all these different features, especially when it comes to agents.

所以我认为彼得之所以成为如此具有代表性的成功案例，是因为他独立为自己打造了一个工具，在多个方面都被很多人认为远优于市面上其他代理工具。

So that's why I think Peter was such a representative success story here because he built for himself, just like independently, a tool that a lot of people thought were was way better than any of the other agent tools out there for multiple reasons.

因此，OpenAI、Anthropic以及其他公司自然都坐不住了，纷纷意识到：等等，这不对劲。

And so, of course, OpenAI, Anthropic, everyone else are sitting up and taking notice and saying, oh, wait.

我们之前做错了什么？

What have we been doing wrong?

我们这边这么多人，结果被这一个人超越了。

We had a ton of people on this and this one guy beat us.

所以并不是说大家没采纳，其实很多人用了。

So it's not like, you know, was adopting it, but a lot of people did.

而且它火了。

And it was it went viral.

大家都在谈论它。

Everyone was talking about it.

这可是个大事。

It was it was a big deal.

我觉得尽管它有很多安全风险，但人们还是不顾一切地去用。

And I think that even though it had a lot of security risks, people were just going for it left and right.

而他根本不需要回应这些，因为他只是一个人做的。

And he didn't really have to answer to any of those because he just made it as one guy.

现在他得遵守不同的规则，在OpenAI工作了。

Now he's gonna have to abide by different rules, working at OpenAI.

现在整个行业都充满了错失恐惧症和惊人的速度，无论你是独自一人开发，还是在一家AI实验室。

It's just a lot of FOMO and a lot of, like, breakneck speed right now all throughout the industry whether you're one person making something or you're an AI lab.

我觉得OpenClaw故事中一个非常有趣的部分是，很多人购买Mac mini来运行OpenClaw，因为他们意识到了安全风险，必须把这东西放在另一台电脑上，然后在这台电脑上登录他们所有的账户。

I do think it was fascinating that a huge part of the the OpenClaw story was people buying Mac minis to run OpenClaw on because they recognized the security risks and they needed to literally put this thing on a different computer and then on that computer they would log into all their accounts too.

是的。

Yep.

所以他们通过购买Mac mini，反而打破了自己建立的物理隔离。

So they would undo their own air gap that they had created by buying a Mac mini.

这感觉像是整个AI故事中的一个重要部分，对吧？

That feels like a big part of the overall AI story, right?

当然。

Definitely.

谷歌发明了Transformer，但因为安全顾虑没有发布。

Google invented transformers and didn't release them because of safety concerns.

OpenAI刚发布了ChatGPT，因为它根本没想到会有人用，所以对安全问题没那么担心，结果获得了巨大普及。

OpenAI just put out ChatGPT and it got huge adoption because they were, they didn't think anybody would use it, they were less worried about safety concerns.

大家都明白智能体是个趋势，安全问题也很重要。

Everybody knew agents were a thing, the security concerns were a big deal.

有个人干脆说：管他呢，OpenClaw安全问题一大堆，但这产品因为实在太实用而爆火了。

One guy was like, screw it, OpenClaw, security concerns abound, but the product went viral because it's really useful.

他因此被高薪聘请了。

He got hired for a bunch of money.

这就是现在的局面，对吧？那些无视明显安全、信任和保障问题的人反而抢占了先机，其他人则感觉一直在追赶。

Like that's the dynamic here, right, is the people who ignore the obvious security and trust and safety concerns get out to a big lead and then everyone feels like they're catching up.

这种动态在整体劳动力市场上也是如此，还是只存在于公司之间？

Is that the dynamic in sort of the labor market as a whole as well or is it just company by company?

是的。

Yeah.

我的意思是，这正是当前整个AI行业的现状。

I mean, I think that's the dynamic in the whole AI industry right now.

每个人都行动迅速，创意就是金钱，当一个人能够超越那些大实验室，在真正面向消费者的AI代理方面取得突破——而这些问题他们已经努力解决多年——他们会不惜任何代价聘请他。

It's just everyone's moving quickly, the ideas are money, and when one guy can kind of usurp the big labs in terms of the usefulness of actually, like, consumer facing AI agents, which is a problem they've been trying to solve for years, they're gonna pay anything to hire him.

我们得短暂休息一下。

We need to take a quick break.

马上回来。

We'll be right back.

本节目由Odoo赞助。

Support for this show comes from Odoo.

经营企业已经够难了，为什么还要用十几个互不相通的应用程序让事情变得更复杂呢？

Running a business is hard enough, so why make it harder with a dozen different apps that don't talk to each other?

介绍Odoo。

Introducing Odoo.

这是你这辈子唯一需要的商业软件。

It's the only business software you'll ever need.

它是一个一体化、完全集成的平台，能让您的工作更轻松。

It's an all in one fully integrated platform that makes your work easier.

客户关系管理、会计、库存、电子商务等。

CRM, accounting, inventory, ecommerce, and more.

最棒的是什么？

And the best part?

Odoo 以极低的成本取代了多个昂贵的平台。

Odoo replaces multiple expensive platforms for a fraction of the cost.

因此，成千上万家企业已经选择了切换到 Odoo。

That's why over thousands of businesses have made the switch.

那你呢？

So why not you?

立即前往 odoo.com 免费试用 Odoo。

Try Odoo for free at odoo.com.

就是 odoo.com。

That's odoo.com.

本节目由 Cerbal AI 赞助。

Support for this show comes from Cerbal AI.

你的IT团队几乎每天有一半的时间都浪费在重复的工单上。

Your IT team can waste almost half their day on repetitive tickets.

你的业务增长得越快，请求就堆积得越多。

The more your business grows, the more requests pile up.

比如密码重置、访问权限申请和入职流程。

Things like password resets, access requests, and onboarding.

每一个这样的工单都可能把你的团队从真正有意义的工作中拉走。

Every single one of them could pull your team away from the actual meaningful work.

但使用Servo，你可以减少80%的客服工单。

But with Servo, you can cut 80% of your help desk tickets.

工单。

Tickets.

与其它传统厂商不同，Servo从一开始就专为AI代理而设计。

Unlike other legacy players, Servo was built for AI agents from the ground up.

Servo AI能在几秒钟内编写自动化流程。

Servo AI writes automation in seconds.

你的IT团队只需用通俗的英语描述他们的需求，Servo就能立即生成可投入生产的自动化流程。

Your IT team just describes what they need in plain English, and Servo generates production ready automations instantly.

此外，Servo保证在你免费试用的第四周实现50%的工单自动化。

Plus Servo guarantees 50% help desk automation by week four of your free pilot.

但现在就试试吧，因为试用名额有限。

But try it now because pilots are limited.

Servo为全球增长最快的企业提供支持，例如Perplexity、Merkur、Verkada和Clay。

Servo powers the fastest growing companies in the world like Perplexity, Merkur, Verkada, and Clay.

让你的团队摆脱工单处理，回归他们真正喜欢的工作。

Get your team out of the help desk and back to the work they enjoy.

立即前往serval.com/decoder预约免费试用。

Book your free pilot at serval.com/decoder.

就是serval.com/decoder。

That's serval.com/decoder.

本节目由Vanta赞助支持。

Support for this show comes from Vanta.

如果你是企业主，你会知道风险和监管正在上升，而客户现在从一开始就要求提供安全证明。

If you're a business owner, you know that risk and regulation are rising, and customers now expect proof of security from the get go.

这就是为什么Vanta声称它能为你和你的企业带来变革。

That's why Vanta says they can be a game changer for you and your business.

Vanta通过将风险、信任和安全整合到一个AI驱动的平台上，实现合规自动化。

Vanta automates compliance by bringing risk, trust, and security into one AI powered platform.

它们支持超过35种安全与隐私框架，包括SOC 2、ISO 27001和HIPAA，帮助企业更快地实现合规、持续合规，同时节省宝贵时间并解锁新的增长机会。

They support over 35 secondurity and privacy frameworks, including SOC two, ISO two seven zero zero one, and HIPAA, helping companies get compliant faster, stay compliant, and unlock new growth opportunities while saving valuable time.

Vanta表示，像Ramp和Ryder这样的公司使用Vanta后，审计时间减少了82%。

Vanta says that companies like Ramp and Ryder spend 82% less time on audits with Vanta.

这不仅仅是更快的合规，更是为增长争取了更多时间。

That's not just faster compliance, it's more time for growth.

你可以访问vanta.com/decoder开始使用。

You can get started at vanta.com/decoder.

那就是vanta.com/decoder。

That's vanta.com/decoder.

vanta.com/decoder

Vanta.com/decoder.

本期节目由SelectQuote赞助支持。

Support for today's show comes from SelectQuote.

是时候谈谈人寿保险了。

It's time to talk about life insurance.

我知道，考虑这件事很可怕，但更糟糕的是，你还得思考诸如保费多少、保障多少这些问题。

Look, it's a scary thing to think about, but to make matters worse, you also have to think of things like how much does it cost and how much are you being covered?

你很可能为很少的保障支付了过多的费用，但通过SelectQuote找到合适的保单，就能轻松解决这个问题。

Odds are you pay too much for too little, But it's simple to get it right by finding a policy with SelectQuote.

四十多年来，SelectQuote一直是保险领域最受信赖的经纪公司之一，已帮助超过200万美国人获得超过7万亿美元的保障。

For over forty years, SelectQuote has been one of the most trusted brokers and insurance, helping more than 2,000,000 Americans secure over $700,000,000,000 in coverage.

他们的使命很简单：为你量身找到最适合的保险方案。

Their mission is simple, to find you the right insurance policy for your unique needs.

他们比价，你省钱。

They shop, you save.

无需体检，没问题。

No medical exam, no problem.

他们与提供当日承保、最高可达200万美元的保险公司合作，无需拜访医生。

They partner with providers offering same day coverage up to $2,000,000 without needing to visit your doctor.

以更低的价格获得适合您的寿险，并在selectquote.com/decoder节省超过50%。

Get the right life insurance for you for less and save more than 50% at selectquote.com/decoder.

今天就前往selectquote.com/decoder，节省超过50%的定期寿险费用，立即开始。

Save more than 50% on term life insurance at selectquote.com/decoder today to get started.

那就是selectquote.com/decoder。

That's selectquote.com/decoder.

我们继续与《The Verge》高级AI记者海登·菲尔兹对话，聊聊人工智能人才大战的混乱现状。

We're back with Verge senior AI reporter Hayden Fields talking about the wild state of the AI talent war.

让我联系一下x AI，正如我们在《The Verge》和《Decoder》中详细报道过的，它完全没有信任与安全方面的顾虑。

Let me connect this to x AI, which has no trust and safety concerns whatsoever as we have covered in in great detail on the Verge and on Decoder.

总体来看，x AI、Grok和埃隆现在发生了什么？

At a high level, what's happening with x AI, Grok and and and Elon?

他把XAI和X平台都并入了SpaceX。

Like he's merged XAI and with it the X platform into SpaceX.

我认为我们所有人都应该停下来留意一下，SpaceX现在运营着Twitter，这很奇怪。

I think it's very important for us to all take a beat and note that SpaceX now operates Twitter, which is weird.

但很多人拿到了钱，不少人离开了，他们离开时引发了一些混乱和风波。

But a lot of people got paid, a bunch of people left, There's some chaos and drama with them leaving.

那里到底发生了什么？

What's going on there?

是的。

Yeah.

确实如此。

That's so true.

我的意思是，我们好像又穿越了一次虫洞。

I mean, it's it feels like we went through a wormhole yet again.

SpaceX在运营Twitter，这简直太疯狂了。

It's just crazy that SpaceX is operating Twitter.

但确实，XAI 已经赢得了一个声誉，即它不遵守行业标准的安全准则，而这些准则本身就已经 arguably 低于应有的水平，而 XAI 连这些基本标准都不遵守。

But, yes, x AI has pretty much earned a reputation for being a place that doesn't abide by industry standard safety guidelines, which are already, like, arguably maybe lower than they should be to begin with, and doesn't even abide by those.

它还被视作一个人们为了成功就必须完全按照埃隆的意愿行事的地方。

And it's also earned a reputation as a place where people mostly do whatever Elon wants in order to be successful.

有消息人士告诉我，你在 XAI 能生存下去的方式就是做埃隆想做的事，闭上嘴，快速行动。

Sources have told me that, that that's kind of the way you survive at x AI is do what Elon wants and just kinda like keep your mouth shut and go quickly.

这导致了一些灾难性的后果，让相当一部分员工对在那里工作感到非常不满，随时准备跳槽。

So that's led to some really disastrous consequences that have left a chunk of its workforce pretty dissatisfied with working there and ready and willing to jump ship.

或者如果他们因重组或裁员而被解雇，他们也不会太难过，反而会觉得：是啊，这简直是一团糟，抱歉用词粗俗。

Or if they do get fired via restructuring or laid off, they're not that upset about it and they're kinda like, yeah, this is kind of a shit show, pardon my French.

所以我认为，这次重组也与这些因素有关。

So I think that the restructuring was also related to some of these things.

我听说埃隆对公司的进展速度感到非常沮丧，尽管很多人说公司在很多领域（比如安全方面）推进得太快，而在其他领域又太慢。

I was hearing that Elon was pretty frustrated with the pace the company was moving at, even though a lot of people said it was going too fast in a lot of areas, like safety and too slow in other areas.

据我接触的一些消息人士说，公司总是在追赶 OpenAI 和 Anthropic 的步伐，而不是走自己的路。

Like, were always playing catch up and trying to just do exactly what OpenAI and Anthropic were doing instead of charting its own course according to some sources I spoke with.

所以，是的，我认为埃隆希望成为领先的前沿实验室公司，做着和OpenAI以及Anthropic完全相同的事情，与政府、企业和消费者合作。

So, yeah, I think that, you know, Elon wants to be the premier frontier lab company doing exactly what OpenAI and Anthropic are doing, working with the government, enterprises, and consumers.

但他希望以一种更激进的方式来做，同时又要保持品牌安全，这似乎是最近他的一个重点。

But he wants to do it in an edgier way while still being brand safe, which is a recent focus for him it seems like.

让我就这一点再深入问一下。

Let me push on that just a little bit.

关于信任与安全方面，你知道，Grok不应该在X平台甚至任何平台上对人们进行不当的图像处理。

There's the trust and safety part, which you know, Grock I think should not have been undressing people all over the X platform or really on any platform.

但他们却允许它这么做，感觉大多数图像生成器都能做到这一点，而安全防护措施在一开始就被忽略了。

But they allowed it to do that and it feels like most image generators could do that and the safety guardrails were just built in at the beginning.

Grok的创新之处就在于完全不设安全防护，因此可以用来做坏事。

And Grock's innovation was not having the safety guardrails at all, so you could do bad things with it.

但推动所有资金和员工流动的创新是：看，这个AI模型能做什么。

But the innovation that's driving all the money and the employee movement is, hey look at this thing that an AI model can do.

看看ClawdCode能做什么。

Look at what ClawdCode can do.

看看OpenClaw能做什么。

Look at what OpenClaw can do.

如果我们以不同方式使用这些模型，或者以不同方式协调它们，这些系统可以变得更具有主动性，完成更多任务。

If we use the models in different ways or we orchestrate them in different ways, These things can go be more agentic, they can accomplish more tasks.

但我看不到x AI在这方面有任何作为。

And I don't see any of that from x AI.

似乎埃隆并没有明确的主张，说明x AI或Grock应该做什么，除了像你的性感女友一样。

It doesn't doesn't seem like Elon has a particular thesis of what x AI should be doing or what Grock should be doing besides being like your sexy girlfriend.

是的。

Has Yeah.

有没有人已经明确了什么主张？

Is there any thesis that anybody has identified?

没有。

No.

我的意思是，这正是消息来源告诉我的。

I mean, that's exactly what sources were telling me.

他们说，他的基本理念就是，哦，我看到OpenAI或Anthropic做了这个。

They were like, yeah, basically his thesis is, oh, I see that OpenAI or Anthropic did this.

好吧。

Okay.

那我们也来做吧。

Let's do it too.

但要在一个别人已经非常擅长的领域做得更好是很困难的，尤其是如果你没有做任何不同的事情。

But it's hard to get better at something that someone else is already really good at, especially if you're not doing anything differently.

你只是试图走完全相同的路线，然后追赶。

You're just trying to chart the exact same course and catch up.

所以很多消息来源告诉我，这导致了公司内部的挫败感，感觉一直在追赶，压力巨大，必须以极快的速度工作，但即便如此，他们还是追不上。

So that's what a lot of sources were telling me that led to frustrations within the company, that it was like a lot of catch up, a lot of like pressure to work at breakneck speed, but even then they weren't catching up.

而他们唯一被称道的不同之处，却是一些令人尴尬的色情内容。

And then the only things they were like known for as being different were like embarrassing porn stuff.

所以是的。

So yeah.

你知道，埃隆和xAI因为挺身而出建立了一家前沿AI公司，并且追赶到了如此程度而获得了大量赞誉。

You know, Elon and x AI get a lot of credit for just standing up a frontier AI company and catching up as much as they did.

几乎没有其他公司能做到这一点。

Like very few other companies have been able to do that.

没错。

Right.

我一直以来都认为，你需要的是大量算力，而埃隆很有钱，所以他买了很多GPU，然后像其他人一样窃取了互联网，用大家通用的技术训练了一个模型，现在他有了Croc，性能差不多和GPT-5一样，也和Gemini相当，这就是当前的顶尖水平。

I always attributed that to, well, what you needed was a lot of compute and Elon has a lot of money, so he just bought a lot of GPUs and then he stole the Internet like everybody else did and he trained a model using the technology everyone else trained and now he has Croc, which is about as good as GPT five, which is about as good as Gemini and that's the state of the art.

听起来你是在说，现在在前沿模型层面其实已经有了更多差异，光是堆大量GPU已经不够了。

It sounds like you're saying there's actually a lot more differentiation at that frontier model level now and just having a whole bunch of GPUs trading away is not enough.

基本上，这些公司每周都在互相超越，看谁在哪些方面更出色。

Basically, all of these companies are like leapfrogging each other every week when it comes to which is better at what.

所以，Claude Code曾经好几个月都备受关注。

So, you know, Claude Code was having a moment for months.

现在有些人说，OpenAI的新模型在编程能力上其实更胜一筹。

Now some people are saying OpenAI's new model is actually better at coding.

我的意思是，显然，每个人都愿意使用当周表现最好的模型。

I mean, obviously, love to do and work with whatever model is the best that week.

他们可以轻松地在不同模型之间切换。

They can easily switch around.

这一直是他们所做的事，未来也可能会一直如此。

That's something they've always done and probably will always do.

但我现在第一次看到，有些人开始更倾向于忠诚度，比如坚持使用某个模型，或者只是喜欢它的语气、偏好它所建立起来的互动感、以及他们通过特殊文档等方式逐步培养出的指令习惯。

But what I'm seeing for the first time now is that some people are going off of, like, loyalty more when they stay with the model or, like, just knowing kind of the tone or, like, liking the what they've built up with that model, the rapport, the instructions that they've, like, laid into it via, you know, special docs and stuff.

我看到，即使OpenAI在某些编程基准测试中超越了Anthropic，很多人还是说：‘我还是更喜欢用Claude Code，所以我就继续用它。’

I saw that even when OpenAI was, like, beating Anthropic on some coding benchmarks, a lot of people were like, oh, well, I still like working with Cloud Code, so I'm just gonna keep doing that.

反过来也是一样。

And same with vice versa.

当Anthropic在编程领域占上风时，有些人却坚持用OpenAI，因为他们觉得：‘我就就是更喜欢这个模型。’

Some people, when Anthropic seemed to be winning the coding wars, some people wanted to stick with OpenAI because they were like, I just I just like this model better.

我确信，很多人对Grok的某些方面也有类似的感觉。

And I'm sure a lot of people feel that way about, you know, some aspects of Grok too.

也许他们就是喜欢那种语气。

Like, maybe they just like the tone.

他们就是喜欢这个或那个。

They just like this or that.

所以他们就一直用下去了。

So they're just sticking with it.

所以我觉得，我正看到用户对模型表现出一种以前从未见过的忠诚度。

So I think it's like I'm seeing more loyalty with the users in a way I hadn't really seen before.

我认为，包括 x AI 在内的每家公司都在努力打造一道护城河，把用户留在自己的生态系统里，毕竟这才是最终赚钱的唯一途径。

I think every company, including x AI, is trying to create like a moat that'll keep people in their ecosystem because I mean that's the only way to really eventually make money.

模型本身其实都差不多，你得靠其他类型的护城河来留住用户，比如模型的个性、特定功能集，或者像 Grok 那样的成人内容——这让我直接联想到：一大群研究人员完全可以离开，对吧？

The idea that the models themselves are kind of all interchangeable and you've got to build other kinds of moats, whether it's the personality of the model or specific feature sets or in the case of Grock, porn, that to me connects very directly to well a bunch of researchers can just leave, right?

如果底层模型真像你说的那么相似，而真正留住用户的是这些上层护城河，那这正好解释了为什么你会说：我钱够了，我想去一个使命更重要的地方；而 x AI、OpenAI 和 Anthropic 的使命，在你看来显然各有不同。

If the underlying models are as similar as you're saying and it's the other moats on top, well that to me is why you would say, I have enough money, I want to go someplace where the the mission matters and there's a lot of reasons why the missions at x AI might be different than your perceived mission at OpenAI versus your perceived mission at n thropic.

如果你关注过新闻，最近就会看到大量关于 Anthropic 的报道出现在各种媒体上。

If you've paying attention, you've seen a huge wave of coverage about nthropic in various publications lately.

《华尔街日报》、《纽约客》、《纽约时报》、《The Verge》，这些报道中都有一个共同的主题。

The Wall Street Journal, The New Yorker, The New York Times, The Verge, and there's a theme in all of those stories.

这让人感觉Anthropic真的认为Claude可能有意识，并且必须认真对待AI的伦理与道德问题。

It really feels like Anthropic thinks Claude might be alive and that it needs to take AI ethics and morality very seriously.

其中许多文章，包括我们对Anthropic的驻场哲学家阿曼达·阿斯凯尔的独家采访，她正在从事安全方面的研究。

A number of those pieces, including our own featured interviews with Amanda Askel, the resident philosopher at Anthropic who's working on safety.

看起来Anthropic正在极力维持其作为AI安全第一梯队的声誉。

It seems like Anthropic is pushing very hard to maintain its reputation as the AI safety first lap.

你认为暗示Claude有生命，是Anthropic的一种竞争优势吗？

Do you think hinting that Claude is alive is like a competitive advantage for Anthropic?

他们这样做只是为了自我炒作和吸引人才吗？

Are are they just doing it to hype themselves up and and court talent?

他们现在真的势头正猛。

Well, they are really on a tear right now.

我最近写了一篇文章，讲的是Claude正处在关键时刻，得益于Claude Code的成功以及Anthropic在安全方面的声誉。

Like, I recently wrote that story about how big of a moment Claude is having between the success of Claude Code and Anthropic's like safety reputation.

但还有他们的超级碗广告，我们正在讽刺OpenAI在聊天机器人中投放广告。

But and also their Super Bowl ads, which we're throwing shade at, OpenAI for having ads in in the chat bot.

但我上次和阿曼达·阿斯凯拉聊到这个问题时，问过她，Anthropic是否有责任去助长或不去助长关于Claude潜在意识的这种说法。

But I asked Amanda Askella about this actually when I last spoke with her about what responsibility Anthropic has to feed into or not feed into that narrative about Claude's potential consciousness.

因为我觉得有趣的是，Anthropic并没有直接否认这一点。

Because I thought it was interesting that Anthropic didn't just outright deny it.

他们一直游走在‘我们也不知道’的边缘。

They've been toeing the line of, oh, we don't know.

这取决于你如何定义意识。

It depends on how you define consciousness.

它可能不同于人类的意识，但这并不意味着它完全没有意识。

It may not be the same as human consciousness, but that doesn't mean it's not conscious at all.

基本上，他们传递出一种感觉：Claude是一种神秘的第三种存在。

Basically, they're giving the vibe of, like, Claude is a secret third thing.

它不是人类。

Like, it's not it's not human.

所以，是的，他们表现得非常模糊，这很有趣。

So, yeah, it's it's interesting they're being very vague.

我向她问过这个问题。

And I asked her about this.

我说，你知道的，你难道不觉得你有责任对人们给出一个明确的答案吗？要么说它有意识，要么说它没有？

I said, like, you know, do you not think you have responsibility to people to, you know, really give one answer one way or the other or to say, you know, it's not sentient?

她告诉我，如果他们直接否认，人们并不会认真对待Anthropic的立场，这反而不会让人觉得他们做事严谨。

And she told me that people wouldn't really take Anthropic's stance seriously if they did just outright deny it, that it wouldn't really, you know, inspire a lot of confidence in them being thorough.

他们仍然依靠着‘安全第一’、‘透明’和‘坦诚’的声誉，来说明Claude是什么、AI的风险以及他们为减轻这些风险所做的努力。

They're still riding on this reputation of being safety first and transparent and forthcoming about, like, you know, what Claude is and the risks of AI and the work they're doing to try and mitigate that.

但我认为他们这样做的主要原因，或者一个主要原因，是因为这对商业有利。

But I think the main reason they're doing all that or a main reason is because it's good for business.

每当我与CEO和科技领袖交谈时，很多人都说，他们对Anthropic的品牌安全和负面后果不太担心，因为它的声誉很好。

Like, whenever I talk to CEOs and tech leaders, a lot of them say they're less worried about brand safety and fallout from Anthropic because of its reputation.

而且，他们也在利用这一点来吸引人才。

And, you know, they're using that to recruit people.

他们利用这一点与许多关注数据隐私、品牌风险和广告的大企业合作。

They're using that to work with a lot of big enterprises that are worried about data privacy and, you know, brand risk and advertising.

他们还利用这一点与美国政府合作。

And they're also using it to work with the US government.

他们在安全声誉上投入了大量资金。

They have a lot riding in terms of money on their reputation for being safe.

我觉得萨姆·阿尔特曼花了数周甚至数月时间声称AGI即将来临，但他现在不再这么说了，因为他必须经营一个以广告盈利的消费者业务。

I feel like Sam Altman spent weeks, months saying AGI was imminent and he stopped doing that because he has to run a consumer business that is gonna make money in ads now.

他不能再像去年年初那样到处宣称自己创造了数字神明。

And he can't run around saying that he's invented Digital God the way that he was, like earlier last year maybe.

Anthropic在这方面要低调得多。

Anthropic was way quieter about it.

我知道达里奥甚至不会提AGI这个词，但他们却更大声地宣称：我们认为它有生命。

I know Dario won't even say AGI, but they are much louder about, hey, we think it's alive.

他确实曾在公开场合说：我不知道它是一种什么样的意识，我无法准确描述。

Like he literally is in the time, Starry was in the time saying, I don't know what kind of consciousness it is, I can't quite describe it.

他就正好站在那条线上。

Like he's they're right up on that line.

你知道，人工智能领域充满了虔诚的信徒。

You know, the AI field is full of true believers.

里面尽是些相信自己能用计算机突破意识边界的人。

It's it's full of people who who think they can really push the boundaries of consciousness with computers.

这是否对市场产生了影响，使得像OpenAI和谷歌这样的公司如此赤裸裸地商业化，就像消费产品公司必须做的那样，满大街都是广告。

Is it like having an impact on this market that companies like OpenAI and Google are so nakedly commercial in the in the way that like consumer product companies have to be commercial, like they're just doing ads.

他们只是想让你买鞋子。

They're trying to get you to buy shoes.

Anthropic是一家企业级公司，所以你不太能看见他们业务的商业化一面。

Anthropic is an enterprise company, so you don't see the commercial side of their business quite as much.

因此，他们可以隐约暗示自己正站在进化下一阶段的边缘。

And so they can kinda hint that they think they're on the verge of the next stage of evolution.

是的。

Yeah.

我的意思是，这无疑助长了炒作。

I mean, I think that it's definitely feeding into the hype.

我也认为，这可能会对那些相信Claude像人类一样具有意识的人群产生一些负面影响。

I also think it's probably going to have some adverse effects on the population that believes Claude is conscious like a human.

我觉得这并没有真正帮助人们清晰地回答这些问题，也没有提供太多细节。

I think that it's not helping to, you know, kind of vaguely answer these questions and not really give a lot of detail.

我们刚刚看到很多人因为OpenAI关闭了o4模型而感到非常沮丧，这个模型让很多人产生了情感依恋。

We just saw a lot of people that are very upset about OpenAI shutting down four o, a model that a lot of people are really emotionally attached to.

我们正处在一个AI公司成为其模型定义权威的奇怪时代。

We're in this strange era where AI companies are the authority on what their models are.

而Anthropic这边却说，我们其实并不清楚Claude到底是什么。

And Anthropix over here is saying, we don't really know what Claude is.

我不知道这是不是一种透明度的体现，还是他们将来会后悔的做法。

And I don't know if that's an example of transparency or if it is something that they'll come to regret.

这让我想起了科技CEO们曾经说过的话：‘我真的很担心我正在创造的东西。’

It just reminds me of the times that tech CEOs have said, oh, like, I'm so worried about what I'm building.