AI 革命内幕：睡眠、压力与规模扩张

你睡多少觉？

How much are you sleeping?

你知道，我从来不是个睡得很好的人。只能说，我正在学习在异常压力下放松和入睡的艺术。

You know, I've never been someone who slept all that well. Let's just say I'm learning the art of finding ways to relax and sleep through moments of unusual pressure.

一切发展得太快了。内部感觉如何？

It is all moving so fast. How does it feel on the inside?

这是一种指数级的感觉。就像你乘坐飞船以相对论速度加速离开地球。根据狭义相对论，你睡一觉醒来，地球上已经过去了两天，所以你一天要处理两天的事；然后你再睡，因为继续加速，地球上又过去了三天，接下来一天过去四天——这大概就是那种感觉。

It's this feeling of like the exponential. Like suppose you were to accelerate away from earth on a spaceship at relativistic speed. The way special relativity works is you go to sleep and you wake up and two days have gone by on earth and so you have to deal with two days in one day and then you go to sleep and then because you've continued to accelerate three days have gone by on earth and then the next day and four days have gone by and that's a little bit what it feels like.

你睡觉时会不会一直担心醒来后会发生什么？

I mean do you go to bed constantly paranoid about what you'll wake up to?

我们有足够多清晰而紧迫的问题需要处理，我一直在处理这些问题，同时思考如何准备。但我不认为偏执或担心醒来后会发生什么是有效的。我观察过历史上那些应对高压局面的人。你需要学会理性应对，不要夸大或低估危险。这种在“我不担心”和“天哪，我们今天必须恐慌”之间摇摆——我认为这是不成熟决策的标志。真正成熟的决策是：你不能忽视这一点。我们不能自满。事实上，风险越来越大。但我们必须理性应对。就像外科医生做手术，或者军官执行军事行动，或者任何影响很多人的决策者，都必须理性决策，理解风险，同时保持基本的冷静。

There are enough clear and present issues that we have to deal with that I'm constantly dealing with those while thinking about how we can prepare. But I don't think paranoia or worrying about what you'll wake up to is productive. I've looked at people in history who've dealt with these very high pressure situations. And you need to learn to respond rationally and not put dangers out of proportion to each other. This yo-yoing between 'I'm not worried' and 'oh my god, we need to panic today' — I think that's a hallmark of immature decision-making. The actual mature decision-making is you can't ignore this. We can't be complacent. In fact, it's getting to be a bigger and bigger risk. But we have to respond rationally. Like a surgeon would deal with an operation, or like a military officer would deal with a military operation, or someone making decisions that affect a lot of people has to make those decisions rationally and they have to understand the risk. But they have to maintain a basic sense of calm.

我儿子昨天说：“我能用你的 Claude 协作账号吗？”我说绝对不行，我需要我的额度。

So my son yesterday was like 'can I use your Claude co-work account?' And I was like absolutely not. I need my tokens.

我们在消费领域也看到越来越多这样的需求。我们原本想更偏向企业级公司，但即使我们没有投入太多精力，消费端也开始快速增长。

We're seeing more and more of them even in the consumer space. We wanted to be more of an enterprise company but even consumer without us putting that much effort is starting to grow fast.

你现在处于 AI 宇宙的中心。感觉如何？

You are at the center of the AI universe right now. What does that feel like?

有趣的是，我整个职业生涯，尤其是在 Anthropic 的这段时间，一直有一种平滑的指数增长体验：什么都没发生，什么都没发生，什么都没发生，然后一点点事情发生，接着突然爆发。这就是世界的体验，也是公司相对于其他公司和世界的规模体验。所以我看着这个图表一段时间，说：“哦，是的，我们大概会在某个时候成为收入最高、估值最高的 AI 公司。”确实发生了。所以从某种意义上说，我并不惊讶，因为这只是一条平滑的曲线；但当然，从另一种意义上说，当事情真正发生时，你会看到更多的细节和色彩。这确实令人惊讶，但我们只是牢记我们通常关注的所有问题：如何训练好模型？如何将它们放入好的产品？如何确保一切安全？如何帮助人们，同时管理技术带来的社会风险？所有问题都一样，只是现在被放在更大的显微镜下审视。

The interesting thing is that the experience I've had for my whole career and certainly the whole time at Anthropic is that there's this kind of smooth exponential and the experience of the smooth exponential is nothing's happening, nothing's happening, nothing's happening. A little things happen and then zoom it goes crazy. That's the experience of the world. That's the experience of the scale of the company compared to the other companies and compared to the world. So I was watching this graph for a while and I said, 'Oh, yeah, we'll probably become the AI company with the most revenue and the most valuation sometime around this time' and indeed it has happened. So in one sense, I'm not surprised because this is just a smooth line on the graph, but of course in another sense when things actually happen, you see so much more detail and color to it. And it definitely is surprising but we're just keeping in mind all the things we usually keep in mind which are just how do we train good models? How do we put them in good products? How do we make sure that everything's safe? How do we help people but also manage the societal risks around the technology? It's all the same questions just under a bigger microscope as it were.

你在旧金山长大时是什么样子的？我知道你父亲是皮匠，母亲在图书馆工作。这对你有什么影响？

What were you like as a kid growing up in San Francisco? I know your dad was a leather craftsman, your mom worked in libraries. How did that shape you?

第一次互联网革命就在我身边发生，但我对此毫无兴趣。我只对做数学和浏览东西感兴趣。我对理解宇宙感兴趣，对科幻感兴趣。那就是大环境。我想我只是对世界充满好奇。

The whole first internet revolution was happening around me and I had absolutely no interest in it. I was just interested in doing math and scrolling things. I was interested in understanding the universe. I was interested in science fiction. That was the general milieu. I think I just felt a lot of curiosity about the world.

你在这个科技中心、现在的 AI 中心长大。这个地方、这座城市有没有影响你的世界观？

You grew up in the town that is the center of technology and right now it's the center of AI. Is there anything about this place, this city here that informed your worldview?

是的，我认为那种不墨守成规、个人主义以及“疯狂也没关系”的精神。我觉得很大程度上影响了我。你听说过一些故事，在欧洲国家甚至美国其他地方，以不同的方式思考或有一些疯狂的想法会被劝阻或视为怪异。我对硅谷有很多批评，但有一点我认为很好：即使所有专家都反对你，也没关系。如果你有一个连贯的愿景和世界观，就应该去追求。也许根本行不通，但如果行得通，就有一种长尾效应——在某些地方，你可以探索某些矿脉，可能会发现巨大的金矿。我认为这种精神非常重要。

Yeah, I think the general spirit of nonconformism and individualism and it's okay to be crazy. I think a good deal of that probably rubbed off on me. You hear these stories about going to countries in Europe or even other parts of this country where it's just discouraged or considered weird to think about things in some different way, or have some set of crazy ideas. There's a lot of things I'm actually very critical about with Silicon Valley. But one thing that I think is good about it is this encouragement that it doesn't matter if all the experts are against you. It doesn't matter. If you have a coherent vision and a coherent worldview, you should go and pursue it. Maybe it just won't work at all. But if it does, there's this kind of long-tailedness to it where there are certain places you can search certain veins and you might find a huge gold mine there. I think that spirit is very important.

2016 年，你、丹妮拉、你姐姐和她的丈夫霍尔登·卡诺夫斯基一起住在一个合租屋里。你们当时在争论什么？

You, Daniela, your sister, and her husband Holden Karnofsky lived in a group house together back in 2016. What were you debating back then?

那时 Open Philanthropy Project 刚刚起步，霍尔登是负责人。我当时是一名生物科学家。所以我帮他们做一些关于发展中国家健康或生物研究的工作。我就哪些领域有前景、哪些不太有前景提供建议。

That was the time when Open Philanthropy Project was first being started, which Holden was the lead of. I was at that time a biological scientist. So I was helping them with some of the stuff they were doing around developing world health or biological research. I advised on what areas were promising and what areas were less promising.

你离开 OpenAI 的决定已成为硅谷的传奇。

Your decision to leave OpenAI has become Silicon Valley lore.

在叙事之外，到底发生了什么？问题是什么？你们在哪些方面有分歧？

What really happened like beyond the narrative? What were the issues? What did you disagree on?

听着，我会说得非常简单。在构建强大技术时，你会面临许多困难问题，Anthropic 每天都要面对，我们不知道自己的决定是对是错。所以在安全问题上有很多合理的分歧。我们确实和他们有过一些分歧。但仅凭这点不足以离开。这里的人和我有过分歧，彼此之间也有分歧。但当你觉得无法信任某人，觉得他们的价值观并非如他们所说，觉得他们不诚实，觉得他们并非出于他们所说的原因，当你看到令人不安的行为模式、不诚实，那就很难继续与一家公司合作，继续信任这家公司。归根结底，当你和某人没有相同的愿景，也不信任他们时，为什么要和他们争论？解决之道就是你做你的事，他们做他们的事。我完全接受我们按自己的方式做事，他们按他们的方式做事。我们会在市场上见分晓，在舆论法庭上见分晓。我认为这些比任何关于谁离开的戏剧更有说服力。我们正在提供一个如何以我们认为负责任的方式部署这项技术的例子。如果他们不同意，他们应该提出论点。这就是我要说的全部。

Look, I'll say it very simply. There are many difficult issues you face when building powerful technology that Anthropic faces every day, where we don't know whether we're making the right decision or the wrong decision. So there are many valid disagreements to be had on safety. We certainly had some of those disagreements with them. But that alone is not sufficient to leave. People here have had disagreements with me. People here have disagreements with each other. But when you feel that you can't trust someone, when you feel that their values are not what they say they are, when you feel that they're not honest, when you feel that they're not in it for the reasons that they say, when you see disturbing patterns of behavior, dishonesty, that makes it very hard to continue to work with a company, to continue to trust the company. At the end of the day, why argue with someone when you don't have the same vision and you don't trust them? The way to resolve it is you go off and do your thing, they go off and do their thing. I am completely at peace with the idea that we're doing things our way and they're doing things their way. We'll see who wins in the market and who wins in the court of public opinion. I think those things speak louder than any drama about who left what. We're providing an example of how to deploy this technology in what we think is a responsible way. If they disagree, they should make that argument. That's really all there is to say about it.

在印度人工智能峰会上，你和 Sam Altman 似乎拒绝在舞台上牵手。发生了什么？

There was a moment at India's AI summit where you and Sam Altman appeared to refuse to hold hands on stage. What happened there?

事实是，峰会组织得非常混乱。我们都在最后一刻上台，然后他们改变了我们站立的顺序，接着给我们拍了照，然后命令我们所有人牵手。如果你参加过这类峰会——我不是特别针对印度——但所有这些有国家元首出席的国际峰会都超级混乱。

What happened is that the summit was extremely disorganized. We all came up at the last minute and then they changed the order in which we were standing and then they took a picture of us and then they ordered us all to hold hands. If you've ever been to one of these summits—I'm not saying anything bad about India in particular—but all of these international type summits with heads of state are super disorganized.

好吧。但其他人都牵手了。别这样。

Okay. But everyone else held hands. Come on.

我不知道该说什么。当时纳伦德拉·莫迪突然在那里让大家牵手。

I don't know what to tell you. There was Narendra Modi up there suddenly telling everyone to hold hands.

好吧。听着，Sam 和 Elon 在互相起诉。你不喜欢 Sam？如果构建世界上最重要技术的人都不能在舞台上牵手，我们怎么能相信你们会在存在风险上合作？

All right. Look, Sam and Elon are suing each other. You don't like Sam? It seems if the people building the most important technology in the world can't hold hands on stage, how can we trust you'll cooperate on existential risk?

所以我要告诉你的是。构建这项技术的人，其质量和可信度存在很大差异。我认为那种没人互相信任的说法是不对的。我认识 Demis Hassabis 15 年了，他构建的 Gemini 模型是 Claude 模型的竞争对手。我们在许多问题上合作过。我们从 Google 购买算力。我们一直在交流安全想法。所以我的看法是，首先，有些参与者比其他人更值得信任。我认为 Anthropic 之外也有我信任、认为值得信任的参与者。我认为需要发生的是，值得信任的参与者需要团结起来，让不值得信任的参与者不得不采用同样的标准。凭借大量经验，我了解到有些人不会自觉做正确的事。但如果行业中的大多数都在做正确的事，那么剩下的行业参与者就没什么选择，只能跟上来。有积极的一面，你可以激励他人。就像 Demis 和我互相激励。他做了 AlphaFold。我们也在生物领域尝试做一些事情。我们做可解释性研究，他们也开始了可解释性研究。这甚至不是竞争。只是一家公司做了很酷的事，另一家公司觉得那很酷，也想做，并看看能否在其中找到新东西。这就是奔向顶峰的胡萝卜一面。还有大棒一面，你会想，好吧，这些人在做正确的事。那些人如果不做正确的事就会显得不好。我们经常看到他们勉强做正确的事，同时试图假装他们在做不同的事，而我们有什么不好的或邪恶的。这是意料之中的。但我认为这就是我们让整个行业团结起来、让行业合作的方式。

So here is what I will tell you. There is a wide variance in the quality and the trustworthiness of the people building this technology. I think this meme that no one trusts each other is not right. I've known Demis Hassabis, who builds the Gemini models that are a competitor to Claude models, for 15 years. We've worked together on a number of issues. We buy compute from Google. We swap safety ideas all the time. So my view is that one, there are some players who are more trustworthy than others. I think there are players outside Anthropic who I trust, who I see as trustworthy. What I think needs to happen is that the trustworthy actors need to get together and put the untrustworthy actors in a position where they have to adopt the same standards. With a lot of experience, I've learned that there are some folks who don't do the right thing on their own. But if there's a majority of the industry doing the right thing, then the rest of the industry is left in a position where there's not much they can do but come along. There's the positive version where you inspire other people. That's like Demis and me inspiring each other. He does AlphaFold. We're trying to do something in bio as well. We do interpretability research, they start interpretability research. It's not even competition. It's just each company does something cool and the other company thinks that's cool and wants to do it too and see if there's something new within that. So that's the carrot side of the race to the top. Then there's the stick side where you're like, okay, these guys are doing the right thing. Those guys will look bad if they don't do the right thing. Often we see behaviors where they grudgingly do the right thing while trying to pretend they're doing something different and there's something bad or sinister about us. That is to be expected. But I think that's the way we get the industry together and that's the way we get the industry to cooperate.

早期其他人专注于有趣、炫目的消费类应用。你押注于编程、企业和云代码，结果大获成功。Claude 协作工具也大获成功。你为什么做出这个赌注？是价值观决定还是商业决定？

Now early on others focused on fun splashy consumer apps. You made a bet on coding and enterprise and Claude Code is a hit. Claude co-work is a hit. Why did you make that bet? Was it a values decision or a business decision?

当我们创办 Anthropic 时，最根本的事情，始终重要的事情，是我们想把这件事做好。但你必须问自己，为了资助这些非常昂贵的模型创建，它需要是一家有商业模式的公司。商业模式会妨碍价值观吗？这个问题一直存在。但我从在其他公司工作以及观察其他公司中学到的一件事是，如果你选择一种与你的价值观根本冲突的商业模式，你会遇到困难。要么你背叛自己的价值观，要么你变得无关紧要。你会陷入两难境地。有出路，但情况很艰难。选择一种与你的价值观兼容的商业模式要好得多。所以当我们思考时，我们说，你看，我们已经看到了社交媒体的世界，消费世界，它似乎确实鼓励参与，甚至成瘾。

When we started Anthropic, the base thing that mattered, the thing that always matters, is we want to do this right. But then you have to ask yourself, okay, in order to fund the very expensive creation of these models, it needs to be a company that has a business model. Does the business model get in the way of the values? There's always this question. But I think one of the things I learned from being at other companies and watching other companies is that if you pick a business model that fundamentally conflicts with your values, you're going to have a hard time. Either you betray your own values or you become irrelevant. You end up in a catch-22 situation. There are ways out, but it's a hard situation. It's far better to pick a business model that is compatible with your values. So when we thought about it, we said, look, we've seen the world of social media, the consumer world, it really seems to encourage engagement, even addiction.

你知道我们在 AI 视频模型上看到的那些垃圾内容，到底是怎么回事？它想最大化你关注的分钟数，因为那是广告收入驱动的。而如果我们看企业领域，我们想让这些模型对人们有用。我警告过很多负面的事情，但最终我们认为正面的事情会超过负面的事情。其中很多都属于企业的范畴。我们想用 AI 来治愈以前无法治愈的疾病。这需要与生物技术、制药和学术研究团体合作。这些都是企业。我们想用 AI 让能源更便宜、更高效。那都是企业。我们想用 AI 来帮助教育。大部分也是企业。我们想用 AI 来解决发展中国家的健康问题。那些是非营利组织，但基本上也是企业。我们想促进经济增长。那基本上也是企业。然后我认为还有另一个因素：企业非常关心信任和长期关系。消费者这边可能有点花哨。对企业来说，重要的是建立一种关系，你和一家公司合作多年，你兑现你的承诺，他们兑现他们的承诺，他们基本上信任你。所以这与我们以积极和安全的方式部署这些模型的目标非常协同。我认为这种商业模式在很大程度上符合我们的价值观，这对我们很有好处。不是说没有冲突，不是说没有艰难的选择，但我认为这样的选择数量比否则要少得多。

You know the slop we've seen with AI video models, it's like what's going on? It's wanting to maximize the number of minutes that you're paying attention to because that's the advertising revenue driven incentive. Whereas if we look at enterprise, we want to make these models useful to people. I warn a lot about the negative things, but ultimately we think the positive things will outweigh the negative things. Many of those fall under the banner of enterprise. We want to use AI to cure diseases that we couldn't cure before. That's working with biotech, pharma, and academic research groups. All of those are enterprises. We want to use AI to make energy cheaper and more efficient. That's all enterprise. We want to use AI to help with education. Most of that is enterprise. We want to use AI to address health in the developing world. Those are nonprofits, but they're basically enterprises. We want to increase economic growth. That is basically enterprise as well. And then I think there's another factor: enterprises care a lot about trust and long-term relationship. Consumer can have this almost gimmicky aspect to it. With enterprise, what matters is you build a relationship where you work with a company for many years, you deliver on what you say, they deliver on what they say, and they basically trust you. So it's very synergistic with our goal of deploying these models in a positive and safe way. I think it serves us well to have this business model that largely aligns with our values. Not that there aren't conflicts sometimes, not that there aren't hard choices we have to make, but I think the number of such choices is much lower than it would be otherwise.

一个开发者可以在一个下午从 Claude 切换到 GPT 或 Gemini。在这个行业真的有可能保持长期领先吗？一个认真的竞争对手要复制你建立的东西需要多长时间？

A developer can switch from Claude to GPT or Gemini in an afternoon. Is it really possible to have a long-term lead in this industry? And how long would it take a serious competitor to replicate what you've built?

模型质量是最重要的。我们现在在模型质量上遥遥领先。有一定的惯性，但我从未依赖过这一点。Anthropic 从未依赖过'这东西有粘性，人们不会换'。我认为你想要一个更好的模型。你想要一个更好的产品。我们看到增长率根本没有下降。如果有的话，它们还上升了，至少在这次采访录制时是这样。所以我倾向于认为那是最重要的。

Model quality is the most important thing. We're very far ahead right now on model quality. There is some amount of inertia, but I've never relied on that. Anthropic has never relied on 'this is sticky and people won't switch.' I think you want to have a better model. You want to have a better product. We see the growth rates haven't inflected at all. If anything, they've gone up, at least at the time of taping this interview. So I tend to think that is the most important thing.

Claude Co-work 发布后不久，2850 亿美元市值一夜蒸发。交易员称其为 SAS 末日。如果 AI 继续以这种速度改进，传统软件有多少会被取代，速度有多快？

Soon after Claude Co-work was released, $285 billion in market value vanished overnight. Traders called it the SAS apocalypse. If AI continues improving at this pace, how much of traditional software gets replaced and how fast?

这是一个很难提前预测的问题。如果你能完美预测，那么人们就会去做，并在市场上赚大钱，而且总是对的。所以没人确切知道会发生什么。但我要指出几点。所有这些传统软件公司都有一些护城河。我认为会发生的是，其中一些护城河会消失，但其他的会保留下来。快速编写软件的能力，我绝对认为那会消失。如果你的护城河是'我们编写了别人写不出来的复杂软件'，那祝你好运。你无法守住它。但我认为人们有客户关系。人们有对行业运作方式的了解。人们有独特的领域知识。所以我给所有这些人的建议是，显然不要自满。不要忽视它。列出你所有的护城河，并非常清楚其中一些会消失，而其他的会变得相对更重要，因为它们是限制因素。还可能出现新的护城河。我认为那些灵活应对、利用仍然存在的护城河以及新护城河的人会做得很好。我认为那些自满、自欺欺人地认为过去有效的方法将来还会继续有效的人，他们的日子不会好过。这就是我要给出的建议。归根结底，我猜这取决于你称什么为 SAS，不称什么为 SAS，但我猜测软件行业会变大而不是变小，尽管会有一些大的输家。

This is one of these questions that is very hard to predict in advance. If you could predict it perfectly, then people would and they'd make a huge amount of money on the market and they'd always be right. So no one knows exactly what's going to happen. But I would note a few things. All of these traditional software companies have a number of moats. I think what's going to happen is some of these moats are going to go away, but others are going to stay around. The ability to quickly write software, I definitely think that's going away. If your moat is 'we wrote this complex software that no one else can write,' good luck. You're not going to be able to defend that. But I think folks have customer relationships. Folks have knowhow of how the field works. Folks have unique domain knowledge. So my advice to all of these folks is obviously, don't be complacent. Don't ignore it. Make a list of all your moats and be very aware that some of them are going to go away, while others are going to become relatively more important because they're the limiting factors. And there may also be new moats. I think those that deftly respond, that lean into the list of moats that are still present as well as the new ones, will do well. I think those that are complacent, that delude themselves that what worked in the past will continue to work, they're not going to have a good time. So that is the advice I would give. At the end of the day, I would guess it depends what you call SAS and what you don't call SAS, but I would guess that the software industry gets larger, not smaller, although there will be some big losers.

解释一下。

Explain that.

我只是觉得蛋糕在变大。有了 AI，蛋糕在变大。现有的老牌企业可能在相对规模上变小。其中一些可能会贬值。有些甚至可能破产，如果他们不正确地适应。但你在增长非常快的时候经常看到这种情况。如果 AI 能做的事情增长了 10 倍，现有的老牌行业很容易增长 1.5 倍，只是没有整个大蛋糕增长得那么多。所以我认为这可能会发生。这并不是说我们不会有大的输家。我认为那些不适应、把头埋在沙子里、看不到即将发生的事情、不识别自己护城河的人，他们会非常艰难。

I just think the pie is getting bigger. With AI, the pie is getting bigger. The existing incumbents may be smaller in relative terms. Some of them may go down in value. Some of them may even go out of business if they don't adapt in the right way. But you see this often when growth is really fast. If what's possible with AI grows by 10x, it's very easy for an existing incumbent industry to go up by 1.5x, just not as much as the whole big pie is growing. So I think that may happen. That's not to say we won't have some big losers. I think those who don't adapt, who put their heads in the sand, who don't see what's coming, who don't identify the moats they have, they're going to have a really hard time.

你最大的支持者是像亚马逊、谷歌、微软和英伟达这样的公司。这些公司都有自己的议程。他们是合作伙伴也是竞争对手。你有巨大的商业里程碑与资金挂钩。谁真正说了算？

Your biggest backers are companies like Amazon, Google, Microsoft, and Nvidia. These are companies that all have their own agendas. They are partners and rivals. You have huge commercial milestones tied to funding. Who's really calling the shots?

有很多次我们真的说出了我们的想法。我一直非常直言不讳地主张对中国的芯片实施出口管制。我这么说是因为我认为中国在 AI 能力上领先对美国、对世界的民主状况非常不利。一些芯片制造商显然不同意这个观点，但这并没有阻止我说出来。我现在又说了，即使在我们签署了更多合作伙伴关系之后。他们知道的是我们一直与他们合作。我们一直是好伙伴。我们可以一起工作。我确定他们希望我们不说这些，但这些是我相信的。

There have been a number of cases where we've really spoken our minds about what we think. I've been very outspoken about the need for export controls on chips to China. I say this because I think it would be really bad for America, for the state of democracy in the world, for China to be ahead in AI capabilities. Some of the chip makers obviously don't agree with that view, but it hasn't stopped me from saying it. I'm saying it again now, even after we've signed more partnerships. What they know is that we always work with them. We've been good partners. We can work together. I'm sure they wish we didn't say these things, but these things are what I believe.

你打算怎么办？说到底，他们和我们一样从这些交易中获益。我们都是成年人了，可以在某件事上合作，同时在另一件事上持不同意见。彭博社报道说，你们的估值比 OpenAI 还高。一家成立五年的初创公司估值近万亿美元。你怎么理解这个数字？为什么需要那么多钱？如果你在算力上更自律，盈利路径会更快。

What are you going to do? They're at the end of the day, they want the, they benefit from these deals as much as we do. Look, we're all adults here. We can work together on one thing while disagreeing about another. Bloomberg's reported that you're at valuations that are higher than OpenAI. We're talking nearly a trillion dollars for a 5-year-old startup. How do you make sense of that number? And why do you need that much money? If you're more disciplined on compute, you have a faster path to profit.

算力增长非常快，对吧？所以可能出现的情况是，业务基本面看起来不错，但一年后你的算力会是现在的三倍或四倍——我不说具体数字——这些算力增长非常迅速，我们完全预期收入增长会赶上并超过它们。但融资是对这种不确定性锥的一种缓冲。所以这是完全理性的做法。对业务的稀释非常小。而且逻辑上这与基本面有问题完全不是一回事。事实上，它恰恰与基本面有问题相反。

The compute is ramping up very quickly, right? So it can both be the case that the fundamentals of the business look good but in a year you'll have three times as much compute, or three times or four times, I'm not going to give exact numbers, but these compute ramps are very fast and we have every expectation that the revenue ramp will meet and exceed those. But raising money is kind of the buffer against this cone of uncertainty. So it's a totally rational thing to do. It's a very small dilution to the business. And it logically is not at all the same thing. In fact, it's compatible with the opposite that there's anything wrong with the fundamentals of the business.

有报道称服务器压力大、可靠性问题、用户抱怨 token 用尽。你说其他公司在基础设施上盲目冒险。你们真的拥有所需的东西，还是在追赶？

There have been reports of server strain, reliability issues, people complaining about running out of tokens. You've said other companies are yoloing on infrastructure. Do you actually have what you need or are you playing catch-up?

关于算力的一点是，算力存在市场，对吧？所以我的观点是，在超过几个月的时间里，我们可以获得大量算力。这里值得说的是，我认为按任何合理标准，我们都没有买太少算力。我们原本计划算力每年增长 10 倍。每年 10 倍是我们的预期。但这并不是我们在 2026 年第一季度看到的情况。我们看到了季度收入增长超过 3 倍——只是季度内非年化的 3 倍，当然 3 的 4 次方是一年 80 倍。我们没有计划年化 80 倍的增长。计划年化 80 倍增长是不理性的，因为如果你只得到 10 倍，你就少了 8 倍。所以我们正处于一个局部极端的算力爆发中。这不会持续。如果持续下去，到年底你的收入将达到地球上任何公司都没有的水平。我不认为这会发生。这不可能。但你可以有这些短暂的时期，比如‘天哪，这增长比我们预想的快得多’。但你知道，你看到了与谷歌的算力交易，与亚马逊的算力交易。我们还能做更多。市场是流动的。如果你能很好地使用算力，并且有需求，你就会得到算力。可能只需要一两个月。

One of these things about compute is there's a market in compute, right? So my view is that over a period of time even longer than a couple months, we can get large amounts of compute. One thing worth saying here is I don't think we bought too little compute by any reasonable standard. We were planning for a 10x a year growth in compute. 10x a year is what we expect. That isn't what we've seen over the first quarter of 2026. We saw a greater than 3x growth in revenue quarterly, just in a not annualized 3x in the quarter, which of course three to the fourth power is 80x over the course of the year. We didn't plan for 80x annualized growth. It would not have been rational to plan for 80x annualized growth because that means if you only get 10x, you have eight times less. So we're in a locally extreme explosion of compute. That's not going to continue. If that continued, you get to revenue by the end of the year that no company on Earth has. I don't think that's going to happen. It just can't. But you can have these short periods where it's like, oh my god, this is faster growth than we ever possibly anticipated. But I don't know, you saw the compute deals with Google, you saw the compute deals with Amazon. There are more that we can and will do. The market's liquid. If you're able to use compute really well and there's the demand, you'll get your compute. It might just take a month or two.

超越你的主要对手感觉好吗？

Does it feel good to surpass your arch rival?

我们面前有很多艰巨的挑战。有一种‘竞相向上’的理念，我们试图带动其他公司一起前进。我认为我们已经看到我们带动了它们。有时它们不承认自己在这么做。有时它们在攻击我们的同时抄袭我们。但这种带动非常有价值。所以我认为，在商业和模型方面成为卓越公司的价值，不在于为了击败对手而击败对手。而在于有能力带动整个生态系统。我们希望未来能做得更多。

Look, we have a lot of difficult challenges in front of us. There's this race to the top idea that we're trying to pull other companies along with us. And I think we've seen that we have pulled them along with us. Sometimes they don't admit that that's what they're doing. Sometimes they copy us while they're attacking us. But this pull is very valuable. And so I think the value of being the preeminent company both commercially and in terms of models, it's not about beating rivals for the sake of beating rivals. It's about having the ability to pull the ecosystem along with us. And we hope that we can do more of that in the future.

但赢的感觉总得有点好吧。我是说，我们总是努力成功，对吧？我们不是来失败的。我不是那种认为应该关闭这项技术、不应该构建它的人。我们存在于自由企业体系中。这没什么错。我们只需要减轻模型的风险，对吧？所以一直是两者之间的平衡。在 Anthropic 的大部分历史中，你们是弱势方。我想，当你没什么可失去时，占据道德高地更容易。到了这个规模，坚持你们的价值观有多难？

But winning has to feel just a little bit good. I mean, look, we're always trying to succeed, right? We're always trying to, we're not trying to fail here, right? I'm not someone who believes we should shut this technology down. We shouldn't build it. We exist within a free enterprise system. And there's nothing wrong with this. We just have to mitigate the risks of the models, right? And so it's always been the balance between the two. Now, for most of Anthropic's history, you were the underdog. I imagine it's easier to take the moral high ground when you have nothing to lose. At this scale, how hard is it to stay true to your values?

我想说的是，我花了很多时间思考这个问题。随着公司规模扩大，我在每个规模都保持警惕。在公司的每个规模，都有新的挑战，新的方式让公司失去商业上的求胜意志或价值观的核心。我两者都担心，因为我认为它们是协同的。我实际上认为，我们能够制造出如此好的模型这一事实，使我们能够在公司成长、变大的过程中以有效的方式坚持我们的价值观。这里有很多陷阱。有很多出错的方式。不是因为我和联合创始人或公司领导层的价值观变了，而是因为公司的组成变化非常快。所以我大概花了一半的时间与公司谈论 Anthropic 的文化以及文化如何运作。当你增长这么快时，你从大型科技公司招聘很多人。如果你不告诉他们 Anthropic 如何运作，他们只会重复他们唯一知道的东西，即他们来自的公司如何运作。所以这是一场持续的斗争和持续的挑战。我和 Daniela 的第一要务可能就是弄清楚如何保持这一点，因为我们认识到从长远来看，这是我们的核心。

What I would say is that I've put a lot of time into thinking about how that's the case. As companies scale, I've been paranoid at every scale. At every scale of the company, there's some new challenge, some new way the company can lose either its will to win just commercially or the core of its values. I'm worried about both because I see them as synergistic. I actually see the fact that we've been able to make such good models as the thing that allows us to assert our values in a way that works as the company grows, as it gets bigger. There are lots of pitfalls here. There are lots of ways to go wrong. Not because me or the co-founders or the company's leaders' values change, but because the composition of the company changes very fast. So I spend probably half of my time just talking to the company about the culture of Anthropic and how the culture works. When you're growing this fast, you're hiring a bunch of people from big tech companies. If you don't tell them how Anthropic operates, they'll simply recapitulate the only thing they know, which is how to operate at the companies that they came from. And so this is a constant struggle and a constant challenge. And it's like me and Daniela's maybe number one top priority is figuring out how to preserve this because we recognize that this is the core of who we are in the long run.

你们的产品速度太疯狂了。你们出货这么快。你们是怎么做到的？

Your product velocity is insane. You're shipping so much so fast. How are you doing it?

我想说两点。第一，我们有一个统一的公司。我们有统一的文化。我认为我们在规模扩大的同时仍然非常高效，每个人步调一致，就是文化和组织的统一。我认为这是最大的因素。第二，是 Claude 本身，我们现在用 Claude 来帮助开发我们的模型，让它们更高效，并快速开发产品。你需要开发各种新的实践。

I would say two things. The first is, we have a unified company. We have a unified culture. I think we've grown larger while still being incredibly efficient, everyone still being on the same page, just the cultural and organizational unity. I would say that's the biggest factor. And I would say the second biggest factor is Claude itself that we're now using Claude to help develop our models and make them more efficient and quickly develop products. There's all kinds of new practices you have to develop.

你能告诉我你见过 AI 做过的最疯狂的事吗？

Will you tell me the most wild thing you've seen AI do?

我认为我见过最疯狂的一些事是在生物学和医学领域。我见过不少案例，包括丹妮拉本人，克劳德诊断出了一些名医漏诊的医疗问题。在生物学方面，模型开始在药物设计或计算化学等任务上变得出奇地好。作为一个曾经的生物学家，我看着这些，心想，哇，这很难——你需要大量训练才能做到——而克劳德正在变得擅长。这是我认为我们将获得巨大收益的一个领域。这是 AI 的积极面：我们将获得这些巨大的好处。生活会变得更好；人类体验的质量会提升。

I think some of the wildest stuff I've seen is around biology and medicine. I've seen a number of cases, including Daniela actually, where Claude diagnosed a medical problem that a bunch of fancy doctors had missed. And on the biology side, the models are starting to get surprisingly good at tasks like drug design or computational chemistry. As someone who used to be a biologist, I look at it and think, wow, that's hard—you need a lot of training to do that—and Claude is getting good at it. That's one area where I think we're going to get a hell of a lot of benefit. That's the positive for AI: we're going to get these huge enormous benefits. Life is going to get better; the quality of human experience is going to get better.

一个世纪的科学进步。

A century of scientific progress.

一个世纪的科学进步，一个世纪的人类体验进步。回到 1900 年。想想我们在 20 世纪面临的所有问题，人们过早死亡的所有原因，他们不得不忍受的所有痛苦，我们今天不必面对的所有物质匮乏。然后再想想另一个一百年。我坚信，这个世纪的科学和医学进步，如果我们能挺过去——我认为我们会，我越来越乐观——我们将拥有一个更美好得多的世界。

A century of scientific progress and a century of progress in what it's like to be human. Go back to 1900. Think of all the problems we had in the 1900s, all the reasons people died prematurely, all the problems they had to suffer, all the material deprivation that we don't have to deal with today. Then think of another hundred years of that. I really believe this century of scientific and medical progress, if we can get through this—and I think we will, I'm increasingly optimistic—we're going to have a much, much better world.

我知道你有多热爱写作。你以你的文章闻名。你会用克劳德来帮助写作吗？

I know how much you love writing. You're known for your essays. Do you use Claude to help write?

是的。我还没有到允许克劳德直接写文字的程度，因为我的风格非常独特，我对它有点挑剔。但我基本上用克劳德来帮我头脑风暴，帮我梳理主题，帮我找可以用的参考资料。所以它扮演的是辅助角色。我不知道我们离克劳德能写得比我好还有多远。我们还没到那一步，但我认为这一天会来的。

I do. I have not gotten to the point where I actually allow text directly written by Claude in, because I have such a specific style that I'm a little picky about it. But I basically use Claude to help me brainstorm, to help me think through the themes, to help me with references I could use. So it plays a supportive role. I don't know how far we are from Claude being able to write better than me. We're not quite there yet, but I think it's coming.

我也热爱写作，我觉得写作能帮你挣扎着理清思路。其中涉及很多批判性思考。如果我们让克劳德替我们做，我们会失去这些吗？

I love writing, too, and I feel like writing helps you struggle through ideas. There is a lot of critical thinking involved in that. Do we lose that if we let Claude do it for us?

我有点担心这个。事实上，这本身是我自己写作的一半原因。当然，这是为了外部读者——很多人读我写的东西——但同样也是为了理清我自己的思路，这样我就知道下一步该做什么，并在我和他人之间建立一个共同的参考点。我认为我们仍在努力解决一个问题：如何以保留这些好处的方式使用 AI。我认为我现在所做的就是这样，我用克劳德做研究，帮助我组织自己的想法。如果我们只是端到端地使用它——比如写一篇关于 AI 风险的文章——首先，它不会写出我所想的东西，而且我也会恰恰失去那个好处。随着模型变得更好，我认为也许我们可以更直接地将它们用于写作，同时仍然保留这些好处，但这会很微妙。不会是一刀切。我们需要随着时间的推移来摸索。

I'm a little worried about that. In fact, that's half the reason I write myself. It certainly is for external audiences—many people read what I write—but it is just as much to clarify my own thinking so that I know what to do next and to create a common reference point across me and others. I think we're still grappling with the question of how exactly do we use AI in a way that preserves those benefits. I think the thing I'm doing now does that, where I use Claude for research and to help organize my own thoughts. If we just used it end to end—like write an essay about the risks of AI—first of all, it wouldn't write the things that I think, but also I would exactly lose that benefit. As the models get better, I think probably we can use them much more directly in the writing and yet still preserve those benefits, but it's going to be subtle. It won't be all one thing. We'll have to figure it out over time.

我认为我们可能会出现一种非常不寻常的组合：非常快的 GDP 增长和高失业率，或至少是就业不足，或者大量低薪工作、高不平等。

I think we could have this very unusual combination of very fast GDP growth and high unemployment or at least underemployment, or a lot of low-wage jobs, high inequality.

他对失业问题一直非常直言不讳。AI 可能在未来 1 到 5 年内消除一半的初级白领工作。那是一年前说的。AI 发展得难以置信地快。现在还是 50%吗，还是更高了？

He's been really direct about job loss. AI could eliminate half of all entry-level white collar jobs in the next 1 to 5 years. That was a year ago. AI has moved incredibly fast. Is it still 50% or is it higher?

我一直说——如果你回去看那些原始片段，它们总是被截断成 3 秒的断章取义——但真正的陈述一直是：我不知道会发生什么，但这是事情可能变得多疯狂的一个数量级。而且，我一直在谈论我们可以对此做些什么。我谈过代币税和与企业合作调整人员。我对再培训项目有点怀疑，但我们应该把它们纳入考虑。宏观经济政策。即使从一开始，我一直在谈论解决方案，但不知何故，人类心理中有一种倾向，只截取那 3 秒的‘末日来临’。所以我的信息绝对不是‘末日来临’。我的信息是：这是我们应该预见到的，我们担心的，并且我们需要积极应对的事情。我不确切知道，但我仍然相当担忧。我的担忧程度和以前一样。我们现在看到 AI 让人们更有效率。但这是通常的瓶颈。如果你回顾工业革命——我在《技术的青春期》中写过——你自动化了 90%的工作。太好了。人们在另外 10%的工作中效率提高了 10 倍，因为他们被杠杆化了 10 倍。但最终它会接近 100%。那么接下来的问题是：你必须为他们找到别的事情做。我不确定长期会怎样。我对此真的不确定。但我确实认为有各种适应方式。我要谈的一件事是 Anthropic 内部的软件工程师。我们现在正在经历这个转变，AI 让软件工程师更有效率，尽管 AI 写了所有或几乎所有的代码，但它仍然让人们更有效率。但我们已经开始看到一些苗头：有些人并没有变得更有效率——让 AI 直接做事情更好。所以这是一方面。另一方面是：我们需要更多的需求在哪里？有一种我们称之为前向部署工程师或应用 AI 解决方案架构师的职位，他们的工作是技术工作和与客户沟通的结合。对此需求很大，因为客户很多，我们增长非常快。

I've always said—and if you go back to those original clips, they always get cut out of context in like 3 seconds—but the real statement was always: I don't know what's going to happen, but this is an order of magnitude for how crazy things could be. Also, I always talk about all the things we can do in response to this. I've talked about token tax and working with enterprises to adjust people. I'm a little skeptical of retraining programs, but we should throw them in the mix. Macroeconomic policy. Even from the beginning, I always talked about solutions, but somehow there's this tendency in human psychology to clip the 3 seconds of 'doom is coming.' So my message is definitely not 'doom is coming.' My message is: this is something we should see coming, that we're worried about, and that we need to actually respond to positively. I don't know exactly, but I'm still pretty concerned. I'm still the same order of concern. We are seeing right now that AI is making people more productive. But that's the usual hump. If you go back to the Industrial Revolution—I wrote about this in 'The Adolescence of Technology'—you automate 90% of the job. Great. People are 10 times more productive in the other 10% because they're 10 times more leveraged. But eventually it gets close to 100%. Now the sequel to that is: then you have to find something else for them to do. I don't know about the long run. I'm truly uncertain about that. But I do think there are types of adaptation. One thing I'll talk about is software engineers within Anthropic. We're going through this transition right now where AI makes the software engineers more productive even though AI writes all the code or almost all the code, but still it makes people more productive. But we're already starting to see the beginning of some people that it's not making more productive—that it's better for the AI to just do the thing. So that's one side of it. The other side is: what do we need more demand for? There's something we call a forward deployed engineer or an applied AI solutions architect, where their job is a mix of technical work and talking to customers. There's a lot of demand for that because there's a lot of customers and we're growing very quickly.

现在，每个从事纯软件工程的人都会为此工作吗？不，这不完美，不是一一对应的，但它让你感受到将会出现巨大的颠覆。事情也会调整。哪种会胜出？我不知道。但警告这一点很重要，因为这样我们才能应对，才能制定政策，对吧？无论是在 Anthropic 内部还是在全球宏观经济层面。我们想要发布经过深思熟虑的想法。我们不想说人们不相信我们会实际去做的事情。我们不想说半生不熟的话。我们想仔细思考应该对这些实际问题采取什么措施。

Now, does every person who is in pure software engineering work for that? No, it's not perfect, it's not one-to-one, but it gives you a flavor that there's going to be a hell of a lot of disruption. Things will also adjust. Which wins out? I don't know. But the reason it's important to warn about it is that that's how we can respond, that's how we can make policy, right? Both within Anthropic and macroeconomically for the whole world. We want to put out carefully considered thoughts. We don't want to say things that people don't believe we'll actually do. We don't want to say things that are half-baked. We want to think carefully about what should actually be done about these problems.

你发布了一张图表，显示了潜在的就业颠覆，比如销售、金融，你知道的，哪些工作会消失，谁会被取代，以及会创造哪些新工作。

You put out this chart showing potential job disruption like sales, finance, you know, which jobs go away, who gets replaced, and what new jobs are created.

所以，没有人确切知道。因为经济是不可预测的，对吧？就像股市一样。这些去中心化的过程，你无法提前知道人们还能做哪些工作。但我要广泛地说，在任何有这类初级白领工作的地方，无论是银行、金融，AI 都有很大潜力首先让人们更高效，但随后就会出现 AI 可以完成工作的全面替代。然后我们就得思考人们能做什么。我认为我们需要提前规划。我们在与企业客户交谈时已经在做这件事了。我们看到他们面临的选择：是节省成本，这通常意味着减少招聘，用更少的资源做同样的事情；还是用同样的资源做更多的事情？我们总是尽可能推动他们用同样的资源做更多的事情，因为这意味着雇佣相同数量甚至更多的人，但只是做新的事情，推动他们走向正和。我们这里拥有的有利条件是蛋糕会大大扩大。所以因为蛋糕会大大扩大，很可能会有地方让人们去。只是要足够快地找到它们。这是颠覆的规模。它会很大，这就是我在警告人们的事情。但我们得解决那个匹配问题。

So, no one knows for sure. Because the economy is unpredictable, right? It's the same as the stock market. These decentralized processes, you don't really know ahead of time what are the pieces of the job that people are still going to be able to do. But what I would say broadly is that anywhere you have these kind of entry-level white collar jobs, whether it's banking, finance, there's going to be a lot of potential for AI to first make people more productive, but then there's going to be a wholesale sale where AI can do the job. Then we're going to have to think about what is it that people can do. I think we need to plan about that ahead of time. We're already doing it when we talk to enterprise customers. We see choices that they face: should I save cost, which often means hiring less people, basically do the same thing with less resources; or should we do more things with the same amount of resources? We always, when we can, try to push them to doing more with the same amount of resources, because that means hire the same number of people or maybe even more people, but just do new things, pushing them towards the positive sum. The thing we have going for us here is the pie is going to expand a lot. So because the pie is going to expand a lot, there are probably going to be places where people can go. It's just a matter of finding them fast enough. It's the size of the disruption. It's going to be big, and that's what I'm warning people about. But we kind of have to solve that matching problem.

那么，给我稍微推演一下。你知道，五年后你醒来。这个国家会是什么样子？那些人在做什么？因为如果有那么高的失业率，那不就是革命开始的方式吗？

So play this out for me a little bit. You know, you wake up in 5 years. What does this country look like? What are those people doing? Because if there's that much unemployment, is that not how revolutions start?

是的。不，这是我们要防止的结果。这绝对是我们想要防止的结果。我认为有几个领域。没有一个是有保证的。我们不确定。但有物理世界，对吧？物理世界中的事物。是的，还有机器人革命，但它比 AI 领域发生的事情慢得多。人们总是谈论建设数据中心，但当处理任何类型的信息变得容易得多时，限制可能就在物理世界中的事物。所以我们需要更多人来制造、建造、生产物理世界中的东西。任何以人为中心的事情，我认为那都会很重要。我听到所有这些故事，说 AI 发现了我医生找不到的东西，但人们真的很想和其他人类交谈，尤其是在重要的事情上。也许 AI 能提供更好的客户服务，但尽管如此，人们，或者至少有些人，想和人类交谈。所以这些以人际关系为驱动的工作，我认为它们会很重要。我认为人类会努力去指导 AI。在某种程度上，它必须符合某人的价值观和意图。所以我认为那里会有一些角色，尽管我不知道这个角色会多薄或多厚。很难说。

Yeah. No, this is the outcome we want to prevent. This is absolutely the outcome we want to prevent. I think there are a few places. None of them are guaranteed. We're not sure. But there's the physical world, right? Things that are in the physical world. Yes, there's a robotics revolution as well, but it's a lot slower than what's happening in AI. People always talk about building data centers, but when processing information of any type becomes a lot easier, maybe the restriction is going to be things in the physical world. So we need a lot more people to make, build, manufacture things in the physical world. Anything that's human-centered, I think that's going to be a big deal. I hear all these stories about AI found something that my doctor couldn't find, but people really want to talk to other humans, particularly over kind of important things. Maybe AI can do better customer service, but nevertheless people, or at least some people, want to talk to humans. So these kind of human relationship driven jobs, I think those are going to be important. I think there will be some effort by the humans to kind of direct the AIs. At some level, it has to be in line with someone's values and someone's intentions. So I think there's going to be some role there, although I don't know how thin versus how thick it will be. It's very hard to say.

有很多反对声音，我知道你说过你在试图警告人们，但是，黄仁勋说你混淆了任务和工作。其他人说这是一种有利于 Anthropic 的末日营销。

There has been a lot of push back and I know you've said you're trying to warn people, but you know, Jensen Huang said you're conflating tasks with jobs. Other folks have said this is sort of doom marketing that benefits Anthropic.

所以我想非常明确地并强烈反驳这一点。整个图景是存在失业风险，这里有一些想法。我的意思是，我们还没有完全充实这些想法，因为我想把它们做对。但 Anthropic 提出了很多想法。我们有经济资助。我们有经济指数。我谈到了应对这些风险的可行方法，从税收和宏观经济政策到新工作是什么。在《技术的青春期》中，我用了大约五页篇幅阐述了任务和工作之间的区别，为什么这次与以往不同，列出了六件我们可以做的事情，从私人慈善到政府行动。我谈到了问题。我谈到了解决方案。但社交媒体，我厌恶它，我厌恶作为一个类别的它，人们有一年前的三秒剪辑。他们实际上不读文章，或者他们利用社交媒体……我关于这些风险写得更仔细。认为这是廉价营销的想法本身就是廉价营销。这是懒惰。这是未能参与严肃的智力工作。我认为这是问题的一部分。再次，我认为这是硅谷疾病的一部分。它陷入了这个三秒的社交媒体世界，所以人们只回应它，或者他们认为他们只需要回应它。再次，我认为这非常危险，我们未能进行成熟的对话。相反，人们只是懒惰地看到这个三秒剪辑，然后说，‘哦，这就是 Dario 说的。’这太愚蠢了。太不严肃了。每当有人这么说，我就对他们不那么认真了。

So I want to be really clear and push back hard against this. The whole picture of there are risks to job loss and here are some ideas. I mean, we haven't fully fleshed out the ideas because I want to get them right. But Anthropic has come up with lots of ideas. We've had economic grants. We have the economic index. I talk about the possible ways to address these risks from tax and macroeconomic policy to what the new jobs are. In 'The Adolescence of Technology', I lay out, I have like five pages where I lay out the difference between tasks and jobs, why this time is different than other times, a list of six different things we can do from private philanthropy to government action. I talk about the problems. I talk about the solutions. But social media, which I detest, which I detest as a category, people have these 3-second clips from a year ago. They don't actually read the essays or they prey on the idea that social media... I've written much more carefully about these things where I talk about the risks. The idea that this is cheap marketing is itself cheap marketing. This is laziness. This is failure to engage with serious intellectual work. And I think that is part of the problem. Again, I think it's part of the disease of Silicon Valley. It's been caught up in this social media world of 3 seconds, and so people only respond to it or they think they only have to respond to it. Again, I think it's very dangerous and we've failed to have a mature conversation. Instead, people just lazily see this 3-second clip and they're like, 'Oh, this is what Dario was saying.' It's so stupid. It's so unserious. And whenever someone says something like that, I take them less seriously.

然而，你们却是首批与美国国防部签约、在美军用于作战的机密网络上运行的人工智能公司之一。请解释一下。

And yet, you were one of the first AI companies to sign a contract with the Department of Defense to operate on classified networks that the US uses to fight wars. Explain that.

是的。我想说的是，世界在变化。我对这项技术的看法——当我看到俄罗斯入侵乌克兰，看到中国入侵台湾的风险时，我担心我们面临一个复兴的威权集团，他们非常咄咄逼人，我们需要自卫。这是我长期以来一直相信并继续相信的事情。这就是为什么，在两届政府期间——我可能不同意每一届政府的每一项政策——但这就是为什么我们总体上支持。我们不希望出现这样一个世界：中国和俄罗斯可以用人工智能分析所有情报，用人工智能攻击台湾和乌克兰，而我们却无法防御。所以我们与他们合作。我们当然不是为了钱。这非常麻烦——即使撇开法律战不谈，为了不多的钱接入政府网络也是一件非常麻烦的事。我们这样做是因为我们在乎。但同样，因为我们在乎，所以技术的使用必须有限制。我在技术青春期时提出的表述是：我们应该以各种方式使用这项技术，除了那些破坏我们自身价值观的方式——我们的红线是大规模监视和完全自主武器。这些是破坏我们价值观的东西。如果民主国家做了那些事，那么民主胜利就不值得。这就是我看到的平衡，也是我们采取的立场。这既解释了为什么我们是第一个与国防部合作的，也解释了为什么有些事别人愿意做我们却不做。我认为你需要选择一个立场并坚持到底。这种公司摇摆不定，从‘我们不会与政府做任何事’到‘我们与政府做所有事’——我不理解。你应该选择你的原则并坚持下去。

Yeah. So, what I would say is, look, the world changes. My view of this technology—when I see Russia invading Ukraine, when I see the risk of China invading Taiwan, it worries me that we have a resurgent authoritarian bloc that is very aggressive and that we need to defend ourselves. That is something I have believed for a while now and continue to believe. That's why, across both administrations—I may not agree with every policy of either administration—but that's why we've generally been supportive. We don't want a world where China and Russia can analyze all intelligence with AI, use AI for attacking Taiwan and Ukraine, and we can't defend them. So that's why we worked with them. We certainly don't do it for the money. It's a huge pain—even putting aside the lawfare, it's just a huge pain to get on government networks for not that much money. So we did it because we cared about it. But similarly, because we care about it, there need to be limitations on the use of the technology. The formulation I used in adolescence of technology: we should use this technology in every way except the ways that undermine our own values—our red lines of mass surveillance and fully autonomous weapons. Those are things that undermine our values. It's not worth democracies winning if democracies do those things. So that's the balance I see, and that's the stand we took. It explains both why we were the first to work with the Department of Defense and why there were some things we wouldn't do when others were willing. I think you need to pick a stand and stand your ground. This idea of companies that seesaw from 'we won't do anything with the government' to 'we're doing absolutely everything with the government'—I don't get it. You should pick your principles and stick with them.

你们从 2024 年开始与 Palantir 合作。没错。他们的技术被 ICE、警察部门以及加沙使用。Claude 是否被用于其他方式的监控？

You've been working with Palantir since 2024. That's right. Their technology is used by ICE, police departments, in Gaza. Is Claude being used for surveillance in other ways?

我们不通过 Palantir 或任何其他人与 ICE 合作。我们不与 CBP 合作。我不认为我们在加沙有业务。我们非常谨慎地将合作范围限定在我们相信的事情上。

We don't work with ICE either through Palantir or anyone else. We don't work with CBP. I don't believe we work in Gaza. We are very careful about scoping our engagements to things we believe in.

那么，你划定了红线。总统禁止你进入联邦政府。五角大楼将你列为供应链风险。OpenAI 介入并签下了你不愿签的合同。赢得这场斗争实际上是什么样的？

So, you drew your red lines. The president banned you from the federal government. The Pentagon labeled you a supply chain risk. OpenAI jumped in and signed the contract that you wouldn't. What does winning this fight actually look like?

我认为对于一家私营公司来说，这场斗争没有赢家。这不是 Anthropic 试图赢或考虑输赢的斗争。这更像是一场关于政府如何恰当使用人工智能的辩论。人工智能是一项新兴技术。我们不了解它在哪些方面可靠或不可靠。我们不了解它在哪些方面促进或破坏我们的价值观。所以我认为重要的一件事是，在我们认为好的用例（坦率地说，大多数都是好的）和我们担心的用例上建立先例。正如我所说，我们已经看到，合同能做的只有这么多。别人可以签一份不尊重你同样红线的合同。但这样做提高了对这个问题的认识。然后，国会中有严肃的两党努力，试图禁止一些我们担心的事情，并试图设置护栏。再次强调，我不想把这说成是一场斗争，但这在某种程度上是胜利——让我们的国家更仔细地思考什么是对这项技术的恰当使用。

I don't think there's any winning this fight for a private company. This isn't a fight Anthropic is trying to win or thinks about winning or losing. This is more a debate about what the proper use of AI by the government is. AI is an emerging new technology. We don't understand the ways in which it's reliable or unreliable. We don't understand the ways in which it promotes our values or undermines our values. So one of the things I thought was important was to establish a precedent on some use cases we think are good—which frankly is most of them—and some use cases we're concerned about. As I've said, we've already seen that you can only do so much with a contract. Someone else can sign a contract that doesn't respect your same red lines. But what it has done is raised awareness for the issue. And then we have serious bipartisan efforts in Congress attempting to ban some of the things we're concerned about and attempting to set guardrails. Again, I don't want to talk about this as a fight, but that's kind of winning—the effort to get our country to think more carefully about what is appropriate use of this technology.

你介意被称为意识形态疯子或一群左翼疯子吗？

Do you mind being called an ideological lunatic or a bunch of leftwing nut jobs?

我一直被骂得更难听。人们可以随便叫我或 Anthropic 什么。重要的两件事是：我们作为一家公司取得成功，并且我们坚持我们的价值观。从某些方面来说，我的生活真的很简单，因为当这两件事是你努力的目标时，一切都很简单——你总是知道自己的立场。

I've been called worse things than that all the time. People can call me or Anthropic whatever they want. The two things that matter are that we're successful as a company and we stand up for our values. In some ways my life is really easy because when those are the two things you're trying to do, it's really simple—you always know where you stand.

一位美国官员表示，在大语言模型的帮助下，美军从每天能打击一千个目标增加到五千个目标。这意味着 Claude 可以帮助更快地杀死更多人。你对此感到安心吗？

A US official has said with the help of LLMs, the US military has gone from being able to hit a thousand targets a day to 5,000 targets a day. That means Claude can help kill more people more quickly. Are you comfortable with that?

我认为这里有两件事。一是美国在军事上更有效的能力。我支持这种能力。我认为拥有更强的能力不会引发战争，而是威慑战争。你基本上是在问，你相信这个国家吗？你希望这个国家在世界舞台上成为一个更强大的角色，而不是一个更弱小的角色吗？是的，我希望。我是一个爱国者。另一个问题是：美国政府正在推行的某些政策，我是否支持？显然，我支持一些，不支持另一些。如果我们提供一项技术，这不由我决定。国防部提出了这一点，我们实际上同意他们的看法。如果我们提供一项技术，我们不能说你可以进行这项军事行动，而不能进行那项军事行动。我私下可能认为这项军事行动有意义，那项是坏主意，但我们不会拒绝提供技术。你必须把政策决定留给军事决策者。你能做的是设定一些高层边界，对我们来说，这些边界防止那些似乎与我们的价值观和国家价值观不一致的用例，并促进我们认为鼓励我们价值观的用例。这就是我们的想法。

I think there are two things here. There is the ability of the United States to be more effective militarily. I am supportive of that ability. I think having that ability be stronger doesn't cause wars; it deters wars. You're basically asking, do you believe in this country? Do you want this country to be a more powerful actor rather than a less powerful actor on the world stage? I do. I'm a patriot. There's a separate question: are there particular policies that the US government is engaged in that I might support or not support? Obviously I support some and not others. It's not up to me if we provide a technology. The DoD made this point and we actually agree with them. If we provide a technology, it's not up to us to say you can do this military operation and you can't do that military operation. Now I might privately believe that this military operation makes sense and that one is a bad idea, but we're not going to deny the technology. You have to leave policy in the hands of the military decision makers. What you can do is assert some high-level boundaries that, for us, prevent the use cases that seem inconsistent with our values and our country's values, and promote the use cases that we think encourage our values. So that's how we think about it.

彭博社报道称，Claude 正被美军用于伊朗战争，通过 Palantir 制造的 Maven 智能系统进行 AI 辅助瞄准。今年二月，一枚美国导弹据报击中伊朗一所女子学校，造成超过 150 人死亡，其中大部分是儿童。Claude 在那次袭击中扮演了角色吗？

Bloomberg has reported that Claude is being used by the US military in the war in Iran to do AI-assisted targeting via a platform made by Palantir, the Maven Smart System. In February, a US missile reportedly hit a girls' school in Iran, killing more than 150 people, most of them children. Did Claude play a role in that strike?

我们无法确切知道这些模型是如何使用的。显然，战争中的错误是可怕的。这是一件可怕的事情。如果这还不能说明为什么我们必须坚持反对我们不支持的使用场景——我们愿意冒着公司未来的风险来限制这些模型的使用方式。你提到的这个使用场景甚至没有违反我们的红线。我们担心的是，会有成百上千倍违反我们红线的使用场景。总的来说，我认为这些模型的使用是合适的，总体上是有益的。但军事决策者即使在最好的时候也会犯可怕的错误。我们可以讨论制定红线，防止那些更可能导致问题的使用方式。如果我们向完全自主武器让步——现在几乎所有其他公司都这样做了——而这是一个案例，Claude 提供辅助，但人类做出最终决定。是人类做出了那个最终决定，而不是 Claude。想象一下，如果别人家的 AI 模型直接做出决定，人类根本看不到。那才是我们反对的。另外，我们需要确保军事决策者不犯这些错误，他们能可靠地运作。政府大量使用微软 Excel；如果我说你不能用 Excel 来做这个军事行动，那是不现实的。希望这能让你了解我们的想法。

We don't have access to exactly how these models were used. Obviously, mistakes in warfare are terrible. This is a terrible thing. If that doesn't make clear why we have to stand up for use cases we don't support—we were willing to risk the future of our company to limit how these models are used. What you're talking about is a use case that doesn't even violate our red lines. We're worried there will be a hundred times as many use cases that do violate our red lines. Overall, I think the use of these models is appropriate; it's good on net. But military decision-makers make terrible mistakes even at the best of times. We can talk about making red lines that prevent uses more likely to lead to those problems. If we had given in to fully autonomous weapons—which almost every other company now has—this is a case where Claude assists, but a human makes the final call. A human made that final call, not Claude. Imagine if someone else's AI model just makes the decision and the human never sees it. That's what we were standing up against. Also, we need to ensure that military decision-makers don't make these mistakes, that they operate reliably. The government uses Microsoft Excel a lot; if I said you can't use Excel for this military operation, you can't realistically do that. Hopefully that gives you a sense of how we think about it.

这所学校有网站，你在谷歌搜索就能找到。Claude 难道不应该发现这一点吗？他们使用的 AI 或其他技术难道不应该发现吗？这是否说明了一个更可怕的问题，即在战争中把技术当作捷径？

This school had a website; you could have found it in a Google search. Shouldn't Claude have spotted that? Should an AI or whatever technology they used have spotted that? And does it speak to a scarier issue about using technology as a shortcut in war?

我不知道；这依赖于我可能没有的机密知识。但我们确立的原则，我认为这里遵守的原则是，人类做出最终决定。我不知道 Claude 或其他 AI 扮演了什么角色，但如果这还不能说明为什么这个原则如此重要，那我就不知道什么能说明了。

I don't know; this relies on classified knowledge I may not have. But the principle we have established, and I think the principle obeyed here, is that a human makes the final decision. I don't know what role Claude or any other AI had, but if this isn't an illustration of why that principle is so important, I don't know what is.

AI 战争更有可能阻止第三次世界大战（美中之间的战争），还是更有可能引发它？

Is AI warfare more likely to stop World War III, a war between the US and China, or is it more likely to make it happen?

总的来说，它更有可能阻止战争。但如果我们对其使用不加限制，那么它更有可能引发战争。你看过《奇爱博士》吧？前提是一个末日装置，当它认为核武器正在射向它时，会自动发射核武器。这能不出问题吗？我指的是这种致命的完全自主武器。我认为冲突的发生是因为双方互相攻击、互相误解。如果没有对这项技术的适当监督，这类事故更有可能发生。如果 AI 被恰当地使用——甚至不是在战争中，而是在情报收集方面——如果我们能预测对台湾的入侵或乌克兰的新动向，我们的对手在知道我们了解他们的一切时，就会三思而后行。卓越的情报可以威慑冲突。卓越的响应能力可以威慑冲突。我仍然相信这些。

On balance, it is more likely to stop it. But if we have no limits on how it's used, then it could be more likely to cause it. You've seen Dr. Strangelove, right? The premise is a doomsday device that automatically fires nuclear weapons when it thinks nuclear weapons are being fired at it. What could go wrong? I get to this lethal fully autonomous weapons thing. I think conflicts happen when two sides jump at each other and misunderstand each other. Without proper oversight of this technology, those kinds of accidents are more likely. If AI is used appropriately—not even in warfare, but in intelligence collection—if we can predict an invasion of Taiwan or a new movement in Ukraine, our adversaries will think twice about conducting an invasion if we know everything they're doing. Superior intelligence can deter conflict. Superior ability to respond can deter conflict. I continue to be a believer in these things.

Anthropic 几乎每周都上头条，最引人注目的是关于 Mythos。这是 Anthropic 最新、最强大的模型，能够自主遍历网络杀伤链的所有环节。你说过 Mythos 太强大了，不能向公众发布。它最让你惊讶的是什么？

Anthropic is making headlines almost weekly, most notably around Mythos. This is the latest and greatest Anthropic model, capable of going through all the links of the cyber kill chain and doing so autonomously. You said Mythos was too powerful to release to the public. What surprised you most about it?

最让我惊讶的是，模型在发现漏洞以及将漏洞转化为利用代码的能力上一直在提升——人们只谈论漏洞，很少谈论将其转化为利用代码，而它在这方面非常擅长。我们看到了一个巨大的飞跃，一个特别大的飞跃。在我们几乎没有提示的情况下，一些早期获得该模型的公司说：‘这是一种超级武器。你应该需要持枪执照才能使用它。请不要发布它。’要求不发布的声音来自我们提供模型的公司，他们发现了大量关键漏洞和可利用性，以至于他们基本上要求我们不要发布。需要明确的是，目标不是永远锁住它。我们正在逐步向更广泛的人群开放，最终我们认为应该向普通受众发布 Mythos，但要配备强大的网络安全防护措施。今天的网络安全防护措施——我们在 Opus 4.7 上发布的，那是一个不错的网络安全模型，但弱得多——是一个令人担忧的问题。

The thing that surprised me most was the models had been climbing in their ability to find vulnerabilities and, importantly, turn those vulnerabilities into exploits—people only talk about vulnerabilities, not often about turning them into exploits, which it was quite good at. We saw a huge jump, a particularly large jump. Without us really prompting them, some of the early companies we gave this to said things like, 'This is a super weapon. You should have to own a gun license to use it. Please don't release this.' The demand to not release was coming from the companies we gave it to, who were finding so many critical vulnerabilities and exploitability that they were basically asking us not to release it. To be clear, the goal isn't to keep it locked up forever. We're gradually trying to open it up to a wider set of people, and eventually we believe we should release Mythos to a general audience but with strong cyber safeguards. Today's cyber safeguards, which we did release on Opus 4.7—a good cyber model but substantially weaker—are a concern.

这些可以被越狱，我们有点担心其他一些公司认为这是一种足够的防御，因为是的，它有时有效，但你知道我们都知道这些分类器可以被越狱或绕过，我们自己的测试以及坦率地说，我们对其他公司已部署的防御措施的评估表明，这些防御还不够强大，这就是我们在等待的——让防御达到我们真正有信心的程度。

These can be jailbroken and we're a little concerned about some of the other companies who think this is a sufficient defense because yeah it works sometimes but you know we all know that these classifiers can be jailbroken or gone around and our own testing as well as frankly our assessment of the models that the other defenses that other companies have put in place suggest that these defenses are not strong enough yet and that's what we're waiting for getting the defenses is to the point where we really have confidence in them.

对此有很多反对意见。你知道，有研究人员说他们能够使用更便宜的开源模型复制它。有些人说 OpenAI 已经拥有这些能力。你对那些说这是大型公关的人怎么说？

There was a lot of push back on it. You know, you have researchers saying they were able to replicate it using cheaper open source models. Some folks say OpenAI has these capabilities already. What do you say to folks who say this is a grand PR?

声称可以用开源模型复制，这完全是错误的。所以，想法是 Mythos 扫描整个代码库并找到某些东西。有个人在推特上说，‘如果你把开源模型指向 Mythos 找到的那一行代码，它也会发现同样的问题。’那不是提示。那不是问题，对吧？那不是一回事。最终的测试是，我们去公司，去开源仓库。我们在 Firefox 中发现了 271 个新漏洞。我们在私人公司中发现了数千个，他们尚未修复或无法披露。之前的模型没有发现那 271 个漏洞。所以实际有效的工作流程是，而不是‘好吧，我找到了 Mythos 发现的那一行，我在干草堆里找到了针。现在别的东西可以捡起那根针。’

The claim that it could be replicated with open-source models, that's just incredibly false. So the idea is Mythos looks across the whole codebase and finds something. Some guy went on Twitter and said, 'Well, if you point an open source model at exactly the line of code that Mythos finds, then it finds the same issue.' That isn't the prompt. That isn't the question, right? That is not the same thing. The ultimate test of this is like we go to companies, we go to open source repos. We found 271 new vulnerabilities in Firefox. We've found many thousands within the private companies who haven't fixed them yet or can't disclose them yet. No one found those 271 vulnerabilities with the previous models. So the actual workflow of what actually works in practice as opposed to, okay, I find the exact line that Mythos found, I found the needle in the haystack. Something else can now pick up the needle.

但那些说这只是好营销的人呢？

But what about the folks who say this was just good marketing?

我们没有发布这个模型，在商业上遭受了巨大损失。这个模型极大地加速了 Anthropic 内部的研究、生产和下一代模型。如果我们发布它，它会在外部世界做同样的事情。这在商业上对我们造成了巨大伤害。

We have suffered enormously commercially from not releasing this model. This model has incredibly accelerated research within Anthropic and production and next models. It would do the same in the outside world if we were to release it. This has hurt us enormously commercially.

如果这有助于防御者，它也有助于攻击者。我们还能防御什么吗？

If this helps defenders, it also helps attackers. Can we defend anything anymore?

我要说的是，我们之所以在给攻击者之前先给防御者 Mythos，是为了修补所有漏洞。我不知道随着模型变得更好，可能会发现越来越多的漏洞，但漏洞是有限的，对吧？就像你有一个表面，上面只有有限数量的洞。你修补所有洞，然后表面变得很难攻击，而且代码本身是用强大的模型编写的，所以很难找到缺陷或入侵。所以我认为，在另一边，希望 6 个月或一年后，我们会有一个比过去更安全的互联网生态系统。我们正努力达到那个世界，并尽最大努力向新的网络防御者开放 Mythos。我们一直在与政府对话。我们非常尊重他们的建议。他们放慢了我们开放的速度，因为他们担心反情报风险。我认为这是明智的。我认为所有认真的人都明白这里有真正的权衡。我们看到很多来自推特和其他 AI 公司的冷嘲热讽。你看看他们说的话和他们做的事之间的不一致。他们不是认真的人。他们没有认真对待我们这里面临的严肃权衡。看，我每天都有客户打电话说，‘我想访问 Mythos。’有国家打电话说，‘我想访问 Mythos。’而美国政府和安全团队说，‘不，等等。这里有风险。’好吧，我不是说哪一方是对的。我认为介于两者之间。双方都有合理的观点。但这里有一个真正的挑战，我们需要作为一个社会共同面对。不要指责这是廉价营销，不要用廉价营销来试图对立，就像其他一些公司正在做的那样。这只会显示出令人难以置信的缺乏庄重和成熟。我们需要共同面对这一刻。

What I would say is that the reason that we're giving Mythos to defenders before we give it to attackers is to patch all the bugs. I don't know as the models get better there may be more and more bugs to be found but there's only so many they're finite right? It's like you have this surface and there's only so many holes in it. You patch all the holes and then the surface becomes very hard to attack as well as the code itself is written with the powerful models so it becomes very hard to find flaws in or break into. So I think on the other side of this hopefully 6 months or a year from now we have a much more secure internet ecosystem than we had in the past. We're trying to get to that world and we're doing the best we can to open up Mythos to new cyber defenders. We've been talking to the government. We're very respectful of their recommendations. They're slowing the pace at which we open it up because they're worried about counter intelligence risk. I think that's sensible. I think all serious people here understand that there's real trade-offs here. We see a lot of sniping from people on Twitter and from other AI companies. You look at what they're saying and the inconsistency with what they're doing. They're not serious people. They're not seriously engaging with the serious trade-offs that we have here. Look, I have customers calling me up every day saying, 'I want access to Mythos.' I have countries calling me up saying, 'I want access to Mythos.' And I have the US government and my security team saying, 'No, wait a minute. There's risk to it.' Well, I'm not saying one side or the other is right. I think it's somewhere in between. Both sides have valid points. But there's a real challenge here and we need to face it together as a society. Not accuse things of being cheap marketing, not use cheap marketing to try and counterposition, which some of the other companies are doing. It just shows an incredible lack of gravitas and maturity. We need to all face this moment together.

你是否已经做出了你并不完全满意的权衡？

Have you had to make trade-offs already that you're not entirely comfortable with?

Anthropic 的整个历史都是权衡，对吧？整个 Anthropic 的历史，对吧？在某个理想世界里，你希望在发布第一个聊天机器人之前，可以花几年时间研究它可能出错的每一件事。现在，我们确实推迟了 Claude 的初始发布，但我们只推迟了几个月。所以我要说的是，一切都是权衡。极端情况是完全疯狂的，对吧？所以一切都是权衡。我要说的是，既然我们现在处于我所说的商业领先地位，我和 Daniela 实际上正在尽一切努力将天平进一步向谨慎方向倾斜。这就是 Mythos 发布的意义，对吧？如果你不是领先者，很难做这样的事情。所以我认为你会看到更多类似的事情。

Throughout the entire history of Anthropic has been trade-offs, right? The entire history of Anthropic, right? Where, in some ideal world, you would prefer to before you release the first chatbot, you could spend years studying every possible thing that could go wrong with it. Now, we did delay the initial release of Claude, but we did it for a few months. So what I'm saying is everything is a trade-off. The extreme ends of the spectrum are completely insane, right? And so everything is a trade-off. What I would say is that now that we're in what I would describe as a commercially leading position, I and Daniela are actually doing all we can to move the dial even further towards being careful. That's what the Mythos release was about, right? It's very hard to do something like that if you're not the leading player. And so I think you're going to see more things like that.

有一种论点，为什么政府不接管你？为什么他们会让一家私营公司控制如此强大的技术？

There's this argument, why wouldn't the government take you over? Why would they let a private company control technology that's so powerful?

所以，我实际上认为这是一个非常严肃的问题，我也有同样的担忧。我不认为政府应该直接接管我们。但我会这样说。我想退一步描述一下情况，历史上我们见过的每一项强大技术要么是由政府建造的，要么起源于政府。所以核武器显然，最初由政府建造，之后也基本上由政府建造。但即使是互联网、GPS、手机，所有的研发都是在实验室、联邦实验室和大学里完成的。人工智能是第一个在私营部门建造的技术，政府没有真正扮演重要角色，而且进入游戏较晚。我认为这实际上是一个危险且不稳定的局面。这不是我会选择的情况。没有真正的替代方案，你知道这项技术是可以建造的。我们的对手正在建造它，它具有经济价值，它会被建造出来。问题在于政府没有做，而不是私营部门在做。我认为我们需要考虑权力的制衡。

So, I actually think that's a very serious question and I share those concerns. I don't think the government should outright take us over. But I would put it this way. I would say just to back up and describe the situation, every previous powerful technology we've seen in history was either built by the government or originated with the government. So nuclear weapons obviously, initially built by the government and pretty much built by the government after that. But even like the internet, GPS, cell phones, all the R&D was done in the labs and the federal labs in the universities. AI is the first technology that's been built in the private sector and where government has not really had a serious role and is coming in late to the game. I think that's actually a dangerous and unstable situation. It is not the situation I would have chosen. There's not really an alternative like you know this technology is possible to build. Our adversaries are building it has economic value like it's going to get built. The issue is the government not doing it not the private sector doing it. I think we need to think about checks and balances on power.

所以，我认为需要对 AI 公司的权力进行制衡。对吧？我们有一个叫长期利益信托的机制。它基本上是一个可以任命多数董事会成员并罢免多数董事会成员的机构。所以，如果追溯下去，它实际上有权解雇我。我们正在引入一些元素，虽然不是全部，但我们在引入一些公共治理的元素，对吧？也就是说，你要对不仅仅持有公司股票的人负责。这非常重要，而且这个结构无论公司发生什么都会继续存在。这是 AI 方面的，我们鼓励其他公司也有类似的结构。嗯，在政府方面，我认为我们需要制衡。你知道，国会已经宣布了制定那些红线的努力。所以，我真的认为立法和司法部门需要发挥作用，因为这项技术，我害怕公司拥有它，但也害怕政府拥有它。然后公司需要制衡政府，政府也需要制衡公司。我们需要对这项技术进行基本监管。我认为我们需要开始进行发布前测试，强制性的发布前测试，对模型进行测试和审计。我觉得很有趣的是，硅谷科技界有一群人，他们一开始的立场是，即使是对这项技术的透明度、出口管制，都会彻底摧毁我们创造技术的潜力，会扼杀创新。然后一旦他们看到第一个真正的危险——我一直预料到的——就出现了国有化的讨论，政府应该直接没收。拜托，各位。你们从最极端的反监管，比如看我们一眼就是在摧毁行业，一下子跳到完全共产主义，政府应该全部拿走。我们需要一个更明智、更温和的方法。这是我们一直推崇的，因为我们理解这项技术的力量。我们没有恐慌，也没有否认。我们看到了平滑的指数增长，并且正在做出适当的回应。

So, I think there need to be checks and balances on the power of the AI companies. Right? We have this thing the long-term benefit trust. What that is is it's a set of basically it's a body that can appoint the majority of the board members and remove the majority of the board members. So, it essentially if you thread it through has the power to fire me. And what we're looking at is we're introducing some elements, you know, nowhere near all the elements, but we're introducing a little bit of the elements of like public governance, right? Where it's like, you know, you're accountable to someone who doesn't just have stock in the company. So that's very important and that structure is going to continue no matter what happens to the company. That's on the AI and we encourage other companies to have similar structures. Um, on the government side, I think we need checks and balances. You know, there are efforts in Congress that have been announced to enact those red lines, right? So, I really think the legislative branch and the judicial branch need to exert themselves because this technology, I'm scared of companies having it, but I'm also scared of government having it. And then the companies need to provide checks on government and the government needs to provide checks on companies. You know, we need basic regulation of the technology. You know, I think we need to start doing pre-release testing, required pre-release testing, testing and auditing of the models. You know, it's very funny to me how there's a particular group of people in the tech world in Silicon Valley who started with a position of like even having transparency around this technology, even export control, you know, this is all just totally it'll apocalyptically destroy our potential to create the technology. It'll kill innovation. And then as soon as they see the first real danger, which I've been expecting all along, there's all this talk of like nationalization and the government should just seize it. Come on folks here. You're yo-yoing from the most extreme anti-regulatory, you know, if you look at us the wrong way, you're destroying the industry to this completely communist, the government should grab it all. We need a more sensible moderate approach. That's the one we've been favoring all along because we've understood the power of this technology. We're not panicking. We're not denying it. We see the smooth exponential and we're responding to it appropriately.

那么，你这次回白宫访问感觉如何？

So, how was your visit back to the White House?

你知道，我们总是尽量与政府中任何可以合作的人合作。我说过，我们有一个简单的方法，就是有一套原则。我们遵循这些原则，并希望对方是理性的。说实话，政府非常重视 Mythos。我们与贝森特部长、幕僚长苏西·威尔斯进行了很好的对话。我认为他们真正理解了这里的风险本质。Mythos，我认为，帮助他们更具体地感受到了这些风险所在。所以，和任何一届政府一样，有些部门我们相处得很好，他们也理解；而有些部门则比较难相处。我认为这很正常。任何一届政府都会这样，我们只能尽力应对。

You know, we always try to work together with whoever we can in government. You know, I said we have this simple approach like we have a set of principles. We like follow those principles and we hope that folks on the other side are reasonable. And you know, honestly, the government has taken Mythos very seriously. Like we've had good conversations with Secretary Bessant, with Chief of Staff Susie Wilds. I think they really understand the nature of the risks here. Mythos has, I think, helped them to feel much more concretely where these risks are. So, you know, again, as with any administration, there are parts who we get along with very well and who understand it. And you know there are other parts that are harder to get along with. I think that's normal. That would be the case in any administration and we just try to navigate it as best we can.

你职业生涯早期在百度工作过，那是一家中国大型科技公司。你在它的硅谷分部工作过，而且你对中国的看法一直很明确。中国正在推出强大的开源模型，美国公司免费基于它们进行开发。这是威胁吗？

You worked at BYU earlier in your career, big Chinese tech company. You worked at the Silicon Valley outpost of it and you've been clear on your views on China. Strong open source models are coming out of China and you have US companies building on them for free. Is that a threat?

所以，你知道，我们在这项技术中看到的一点是，模型的智能程度确实有溢价。我们很少看到人们更愿意使用智能较低的模型。需要说明的是，有一个繁荣的生态系统。有很多挑战和问题比我们需要前沿模型来解决的要容易得多。但同样，这是指数级的，对吧？这些远离前沿的模型可能具有与 2023 年和 2024 年相当的经济价值。但同样，我们每年有 10 倍的增长。所以我们发现，前沿的东西总是比非前沿的东西大得多。我认为这是习惯于在之前时代构建产品的人不太理解的。对吧？作为一个之前没有经营过公司的人，从未考虑过之前的产品时代，尤其是社交媒体时代，我觉得自己是那个世界的外来者，并且我认为人们的直觉是错误的。他们有各种产品启发式方法，我认为每年 10 倍的模型指数增长打破了这一点。智能是一个如此巨大的因素，它压倒了一切。所以我们一再看到，价值在于前沿。现在，我担心的一些落后模型的风险是，它们可能具备 Mythos 级别的网络能力。12 个月后，我们将拥有更好的网络能力，但 Mythos 级别的网络能力可能任何人都可以下载。希望到那时我们已经修补好了一切。我认为我们无法阻止它，但我认为这是一个严重的问题。

So, you know, one of the things we've seen with this technology is that there's really a premium to how intelligent the models are. Um, we very very rarely see that people would prefer to use models with lower intelligence. Now, to be clear, there's a thriving ecosystem. There are lots of challenges and problems that are much easier than the ones we need frontier models for. But again, it's an exponential, right? Like it's possible that these far from frontier models have economic value comparable to what we saw in 2023 and 2024. But again, we have this 10x a year growth. And so what we find is that what's on the frontier is always much much larger than what is away from the frontier. I think this is something that people who are used to building products in the previous era don't quite understand. Right? As someone who came in who hadn't run a company before, who has never thought about the previous product era, particularly to test the social media era. Um, I feel like an outsider to that world and I feel that people's instincts are wrong. Um, they have all these kind of product heuristics and I think the 10x per year model exponential really breaks that. Like intelligence is just such a huge factor that it outweighs everything else. And so we're just seeing over and over again that the value is found on the frontier. Now what I do worry about with some of these laggard models is the risks of them where we have Mythos-class cyber capabilities. 12 months from now we'll have much better cyber capabilities, but the Mythos-class cyber capabilities may just be available for anyone to download. Now hopefully we'll have patched everything before then. I don't think there's anything we can do to stop it, but I think it's a serious concern.

你在百度的经历是否塑造了你对中国的看法？

Did what you saw at BYU shape your views on China?

并不完全是。不。我在那里工作了一年。我想我可能学到了更多关于语音识别之类的东西。唯一让我担心的是，我们获取所有语音识别数据的一部分方式是，他们不祥地说，在中国我们不在乎隐私，所以我们有所有这些语音识别数据。但除此之外，我的担忧是地缘政治方面的。我认为最让我担心的是我们在中国看到的事情，比如对维吾尔人的所作所为，对批评的压制，甚至在美国，香港发生的事情，对吧？中国共产党能够渗透到美国的商业网络并压制批评。那是一个威权国家，一个高科技威权国家。当我看到这与 AI 结合时，你真的会得到一个像《1984》或更糟的反乌托邦。我的重点是试图阻止这种情况，我认为我们有机会阻止它。

Not really. No. Um I worked there for a year. You know, I think I probably learned more about speech recognition and all of that. Maybe the only thing that concerned me was, you know, part of how we got all the speech recognition data was, they said ominously, we don't care about privacy in China, so we have all this speech recognition data. Um, but I think aside from that, my worries here are geopolitical. You know, I think the things that most worried me about what happened in China are, what we saw happen to the Uyghurs, what we saw with suppression of criticism even in the US with what happened with Hong Kong, right? The fact that the CCP could reach into the US business network and suppress criticism. That's an authoritarian state and a high-tech authoritarian state. And when I see how that combines with AI, you really get a dystopia here like 1984 or worse. And my focus is on trying to prevent that and I think we have an opportunity to prevent that.

我认为我们有机会让 AI 成为一项支持民主的技术，让人们更自由，实现人人平等的正义承诺，但也可能走向反面。走向何方取决于 AI 公司、政府和所有人的行动。所以我认为我们负有责任。

I think we have an opportunity for AI to be a pro-democracy technology, you know, that kind of makes people freer, that delivers on the promise of equal justice for all, or it could go the other way. And which way it goes depends on the actions of the AI companies, the government, and all of us. So I see us as having responsibility here.

你所在领域的人谈到一个时刻，AI 变得足够好可以自我改进，然后改进后的版本再自我改进，如此循环。你们的一些研究人员认为那个时刻很近了。它有多远？

There's a moment that people in your field talk about where AI gets good enough to improve itself and then the improved version improves itself and so on. Some of your researchers think that moment is close. How far away is it?

我不认为这是一个时间点。我认为这是一个持续的过程。我们已经看到 AI 能够为下一代 AI 提出架构。一年前，我们看到 AI 带来的全要素生产率提高了 10%到 15%，现在可能已经达到 20%到 30%，甚至可能翻倍。就像所有事物一样，我们处于指数曲线上。没有哪个时刻 AI 会自我改进、失控或变得不安全。我们面对的是一个加速的指数曲线。在指数曲线上的每一点，我们都必须评估是放慢速度还是对这项技术施加更多控制。我认为这越来越必要。这一切的罗塞塔石碑就是平滑的指数曲线。

I don't think it's a moment in time. I think it's a continuous process. We're already seeing it in some ways where the AI is able to suggest architectures for the next AI. I would say a year ago we were seeing 10 to 15% increase in total factor productivity due to AI. That's probably up to 20 or 30% now. Might be doubling. As with all things, we're on the exponential. There's no moment where AI improves itself or runs out of control or becomes unsafe. What we have is an accelerating exponential. At each point on the exponential, we have to assess whether to slow down or put more controls on this technology. I think more and more of that is going to be required. The Rosetta Stone to all of this is the smooth exponential.

那些反对所有 AI 监管的人，看到一件事后就想要国有化，这是一个教训。那些轻视 AI 力量的人，然后说‘天哪，它在自我改进，失控了，我们必须全部关闭’，这也是一个教训。在这些极端反应之间摇摆是极其无益的。正确的回应是：‘我们不会恐慌。我们的对策将随着技术的力量平稳升级。’如果你看到有人有这种疯狂的摇摆反应，那表明他们措手不及，并且不认真。

There's an object lesson in the people who were against all AI regulation and then saw one thing and wanted to nationalize. There's an object lesson in the people who dismissed the power of AI and then said, 'Oh my god, it's improving itself. It's running out of control. We have to shut it all down.' Yo-yoing between those extreme reactions is incredibly unhelpful. The right response is to say, 'We're not going to panic. Our counter measures will smoothly ratchet up with the power of the technology.' If you see someone having this kind of crazy yo-yo reaction, that's a sign that they were caught by surprise and that they're not serious.

我知道你最喜欢的书之一是《原子弹的诞生》。你觉得自己和奥本海默有相似之处吗？

I understand one of your favorite books is The Making of the Atomic Bomb. Do you see parallels between yourself and Oppenheimer?

我最认同的人物是利奥·西拉德，他是第一个提出链式反应想法的人。我的观点是，我们不会靠那些超凡脱俗的人物或试图成为一切中心的人物来度过难关。需要权力平衡。这里有很多有利益关系的强大参与者，唯一能让所有人都有好结局的办法是处处有制衡。在某种程度上，我实际上把奥本海默视为一个失败案例，是不应该发生的事情。

The figure I most identified with was Leo Szilard, who was the one who first had the idea of a chain reaction. My view is we're not going to get through this with larger-than-life personalities or figures who try to be at the center of everything. There needs to be a balance of power. There are a lot of powerful actors with interests here, and the only way it's going to end well for everyone is if there are checks and balances everywhere. In some ways, I actually see Oppenheimer as a failure case, as what should not happen.

你说过大约有 10%到 25%的几率出现文明崩溃。这不可忽视。有没有可能由 Anthropic 构建的东西导致这种情况？

You've said there's roughly a 10 to 25% chance of civilizational collapse. That is not insignificant. Is there a scenario where it's something that Anthropic built that caused that?

我当然希望不会。我的观点是，我们采取的行动降低了那个概率，而不是增加了它。那个概率来自技术非常直接的配方：世界上存在许多国家，经济体内存在许多公司，如果空白没有被填补，还会有新公司出现。这是我们面临的困境。我们正在努力降低那个概率。我认为我们降低的远多于增加的。但这项技术的固有属性是不可预测的。我们努力构建东西，并在发布前进行大量测试。今天发布的模型并不危险，至少在网络之外不危险。然后我们迭代和学习。有无数防御机制。我们在公司内部做的一半事情就是尽可能降低风险，但它永远不会是零。

I certainly hope not. My view is that the actions we have taken lower that probability rather than increasing it. That probability comes from the very straightforward recipe of the technology: the existence of many countries, many companies within an economy, and new ones created if the void isn't filled. That's a dilemma we're in. We are trying to act to lower that probability. I think we lower it a lot more than we raise it. But the inherent property of this technology is that it's unpredictable. We try to build something and test it a lot before release. The models released today are not dangerous, at least not outside of cyber. Then we iterate and learn. There are a zillion defense mechanisms. Half of what we do within the company is try to reduce the risk as much as we can, but it's never going to be zero.

假设有很多航空公司，你说‘我要开一家更安全的航空公司’。你的航空公司可能比其他所有公司安全 10 倍。但如果有人问你‘你能保证你的飞机永远不会坠毁吗？’你怎么可能保证？但如果飞机坠毁的概率是 25%，你就不会上那架飞机。

Suppose there are a bunch of airline companies and you say, 'I'm going to make an airline company that's safer.' It can both be the case that your airline company is 10 times safer than all the others. But if someone asks you, 'Can you guarantee that your airplane will never crash?' How could you possibly? But if there was a 25% chance of an airplane crashing, you wouldn't get on that plane.

没错。25%太高了。我们正努力让那个概率低得多。这就是目标。

That's right. 25% is too high. We're trying to make that probability much, much lower. That is the goal.

你在构建非常强大的东西，并且会从中获得巨大收益。我们为什么要相信你？

You are building something incredibly powerful and stand to gain enormously from it. Why should we trust you?

任何公司起步时，尤其是考虑到过去几年硅谷的行为，从怀疑的立场出发是合理的。如果你对我或 Anthropic 一无所知，那很合理。硅谷已经失去了世界很多信任，必须重新赢得。我们试图传达的信息是我们确实不同，这必须通过我们实际做的事情来赢得。你可以同意或不同意，但我们坚守了我们的价值观。关于 Mythos，不发布这个非常强大的模型在商业上确实阻碍了我们。在此之前还有一系列小事。我们在中国问题上言行一致。我们切断了模型的访问。我们本不必那样做。没人要求我们。那花费了我们数亿美元，当时那是我们收入的很大一部分。延迟 Claude 2——我们在这方面有很长的历史。我们不完美。我们会犯错。但我会请人们看看整体历史，然后说关于我们的哪个假设与那段历史最一致。人们必须自己决定。但我认为一致的假设是，我们真心努力做正确的事。我们不完美。组织总是功能失调。我们总是努力修复它们，让它们更好地运作。很多失误，很多事出错。但根本上，我们有一个诚实而认真的关于如何做正确事的图景，并且我们正在努力执行那个图景。

When any company starts out, and particularly with what we've seen from Silicon Valley's behavior over the last couple years, starting from a position of distrust is rational. If you don't know anything about me or Anthropic, that's pretty rational. Silicon Valley has lost a lot of the world's trust and has to re-earn it. The message we're trying to send is that we're actually different, and that has to be earned through things we actually do. You can agree or disagree, but we stood up for our values. With Mythos, it really hampered us commercially not to put this very powerful model out. There were a bunch of smaller things before it. We put our money where our mouth is on China. We cut off access to models. We didn't have to do that. No one told us to. That cost us several hundred million back when that was a significant fraction of our revenue. The delay of Claude 2—we have a long history of this. We aren't perfect. We make mistakes. But I would ask people to look at the overall history and say what hypothesis about us is most consistent with that history. People have to decide for themselves. But I think the hypothesis that is consistent is that we are genuinely trying to do the right thing. We're imperfect. Organizations are always dysfunctional. We're always trying to fix them and make them work better. Many footfalls, many things go wrong. But at basis, we have an honest and earnest picture of how to do the right thing, and we're trying to execute on that picture.

我们将在指数曲线的另一端见。搞定。

We will see you on the other side of the exponential. Done.

呃，希望如此。

Uh, hopefully.

你一直想成为好莱坞明星。

You always wanted to be a Hollywood star.

有一件关于 CEO 工作我没想到的事，就是你得经常化妆。呃，这不在我的预料之中。

That's one surprising thing that I didn't understand about the CEO job is how often you have to wear makeup. Uh, that was not on my bingo card.

就一点粉。

Just a little powder.