从生物学家到 AI 先驱：Anthropic 背后的故事

好。我开始用 Claude 了。它已经到了有时候让我惊讶它有多了解我的地步。我不知道这有没有道理。让我惊讶的是，在我看来，我们离这些模型达到人类智能水平如此之近，但社会上似乎还没有广泛认识到即将发生的事情。就像一场海啸正向我们袭来，近得我们都能看到地平线上的浪头，可人们却还在解释：哦，那不是海啸，只是光线把戏。公众对风险缺乏意识。印度在这其中扮演什么角色？

Okay. So I started playing with Claude. It's getting to that point where sometimes it surprises me by how much it knows me. I don't know if that makes sense. It is surprising to me that we are in my view so close to these models reaching the level of human intelligence and yet there doesn't seem to be a wider recognition in society of what's about to happen. It's as if this tsunami is coming at us and you know it's so close we can see it on the horizon and yet people are coming up with these explanations for oh it's not actually a tsunami that's just a trick of the light like there hasn't been a public awareness of the risk. What is India's role in all this?

许多其他公司把自己定位为消费公司，把印度看作一个市场，对吧？一个获取消费者的地方。我们实际上看法略有不同。

Many other companies come here as themselves a consumer company and they see India as a market, right? A place to obtain consumers. We actually see things a little bit differently.

你在创立 Anthropic 之前做什么？

What did you do before founding Anthropic?

对，我最初是生物学家。我本科读物理，博士读生物物理，我想理解生物系统从而治愈疾病。我注意到研究生物学的一个特点是其惊人的复杂性。比如，我做的蛋白质质谱工作，试图找到蛋白质生物标志物，复杂性简直难以置信。一个给定的蛋白质，RNA 会根据它在细胞中的位置以多种不同方式剪接，然后经过翻译后修饰、磷酸化，与许多其他蛋白质形成复合物。我开始绝望，觉得这对人类来说太复杂了，无法理解。就在我做生物学研究时，我注意到很多关于 AlexNet 的早期工作，那是首批神经网络之一，差不多 15 年前了。我说，哇，AI 真的开始起作用了。它与人脑的工作方式有一些共同点，但有可能更大、扩展性更好，并能学习像生物学这样的任务。也许这最终会成为解决我们生物学问题的方案。于是我去百度与吴恩达共事。然后在谷歌待了一年。之后在 OpenAI 成立几个月后加入，领导了那里几年的研究。但最终，我和其他几位员工对我们如何制造 AI 以及公司应该代表什么有了自己的愿景。所以我们离开并创立了 Anthropic。

Yeah, so I was originally a biologist. I did my undergrad in physics, my PhD in biophysics, and I wanted to understand biological systems so that I could cure disease. The thing I noticed about studying biology was its incredible complexity. For example, if you look at the protein mass spec work that I did, trying to find protein biomarkers, it's just really incredible how much complexity there is. You have a given protein, the RNA gets spliced in a whole bunch of different ways depending on where it is in the cell, then it gets post-translationally modified, phosphorylated, complexes with a whole bunch of other proteins. I was starting to despair that it was too complicated for humans to understand. Then as I was doing this work on biology, I noticed a lot of the early work around AlexNet, which is one of the first neural nets, almost 15 years ago now. I said, wow, AI is actually starting to work. It has some things in common with how the human brain works but has the potential to be larger and scale better and learn tasks like biology. Maybe this is ultimately going to be the solution to solving our problems of biology. So I went to work with Andrew Ng at Baidu. Then I was at Google for a year. Then I joined OpenAI a few months after it started and led all of research there for several years. But eventually, myself and a few other employees had our own vision for how we wanted to make AI and what we wanted the company to stand for. So we went off and founded Anthropic.

那是怎样的？是不是像 OpenAI 的思路分叉，最终变成了 Anthropic 的做法？

How was it? Was it like a fork in how OpenAI was thinking into what Anthropic eventually did?

是的。我认为我和联合创始人在创立 Anthropic 时的信念有两个。一个我们开始说服 OpenAI 接受，另一个我觉得我们没有说服他们。第一个是对缩放定律的信念：如果你扩展模型，给它们更多数据、更多算力——有一些修改比如强化学习，但不多，相当接近纯 Scaling——你会发现性能的惊人提升。我在 2019 年用 GPT-2 就发现了，当时我们第一次看到缩放定律的曙光。当然，内外有很多人根本不相信，我们向领导层力陈这很重要，这将是一件大事，我想他们开始相信我们，最终朝那个方向走了。第二个信念是：如果这些模型将成为通用的认知智能体，匹配人脑能力的通用认知工具，我们最好把它做对。经济影响将是巨大的。地缘政治影响将是巨大的。安全影响将是巨大的。它将改变世界运作的方式。所以我们需要以正确的方式去做。尽管有很多关于以正确方式做的漂亮话，但由于各种原因，我不相信我当时所在的机构有真正严肃的信念去以正确方式做。所以我的观点一直是：不要和别人争论他们的愿景。不要试图让别人按你的方式做事。如果你有一个强烈的愿景，并且你和另外几个人分享这个愿景，你就应该自己去做，然后为自己的错误负责。你不需要为别人的错误负责。也许你的愿景成功了，也许没有，但至少它是你的。

Yeah. I would say my conviction and the conviction of my co-founders when we founded Anthropic were two. I think one we were starting to convince OpenAI of, the other I didn't feel that we were convincing of. The first was the conviction in the scaling laws and the idea that if you scale up models, give them more data, more compute—there are a few modifications like RL but not really very much, it's pretty close to pure scaling—you find incredible increases in performance. I was finding that in like 2019 with GPT-2, when we first saw the first glimmers of the scaling laws. Of course there were a lot of folks inside and outside who didn't believe it at all, and we really made the case to leadership like this is important, this is going to be a big deal, and I think they were kind of starting to believe us and ultimately went in that direction. And there was a second conviction I had, which is: look, if these models are going to be kind of general cognitive agents, general cognitive tools that match the capability of the human brain, we better get this right. The economic implications are going to be enormous. The geopolitical implications are going to be enormous. The safety implications are going to be enormous. It's going to transform how the world works. So we need to do it in the right way. And despite a lot of language verbiage about doing it in the right way, I was for a variety of reasons just not convinced that at the institution I was at there was a real and serious conviction to do it in the right way. So my view is always: don't argue with someone else's vision. Don't try to get someone to do things the way you want to. If you have a strong vision and you share that vision with a few other people, you should just go off and do your own thing and then you're responsible for your own mistakes. You don't have to answer for anyone else's. And maybe your vision works out, maybe it doesn't, but at least it's yours.

OpenAI 不是相信缩放定律吗？因为他们自己也走了同样的路，对吧？

Didn't OpenAI believe in scaling laws because they went down the same path themselves too, right?

嗯，是的。我们成功了。

Well, yeah. We succeeded.

你能用非常简单的话解释什么是缩放定律吗？

Can you explain what scaling laws are in very simple terms?

就像如果你想要一个化学反应产生氧气或生火，你需要不同的原料。如果你某种原料不够，反应就会停止。但如果你按比例把原料放在一起，你就会得到爆炸或火。对于 AI 来说，这些原料是数据、算力和 AI 模型的大小。缩放定律告诉你，如果你投入数据和模型大小的原料，你得到的是智能。智能是化学反应的产物。

It's like if you want a chemical reaction to produce oxygen or start a fire, you need different ingredients. If you don't have enough of one ingredient, the reaction stops. But if you put ingredients together in proportion, you get your explosion or fire. For AI, those ingredients are data, compute, and the size of the AI model. The scaling laws tell you that if you put in the ingredients of data and model size, what you get out is intelligence. Intelligence is the product of a chemical reaction.

那什么是智能？

And what is intelligence?

智能通过翻译语言的能力、写代码的能力、或正确回答关于故事的问题的能力来衡量。基本上任何我们能想到的认知任务，任何存在于文本或图像中的任务，任何你可以在计算机上完成的任务。

Intelligence as measured by the ability to translate language, or the ability to write code, or the ability to answer questions correctly about a story. Basically any cognitive task we can think of, any task that exists in text or in images, any task that you can do on a computer.

你今天描述的智能与五年前计算机能做的有什么不同？

How is the intelligence of today as you are describing it different from what a computer could do like 5 years ago?

是的，我会说，例如，五年前你不能问计算机一个问题，然后让它写一篇关于那个问题的一页纸的文章。

Yeah, I would say, for example, 5 years ago you could not ask a computer a question and have it write a one-page essay on that question.

你不能让计算机实现一个代码功能并让它真正完成。那些事情都做不到。你不能生成图像、生成视频，也不能分析视频。你可以拿一段猴子玩杂耍的视频，问发生了什么、球传了几次。现在你可以让 Claude 或其他 AI 模型来回答。五年前，这些事都做不到。我在想，智力的定义是不是变了？

You could not ask a computer to implement a feature in code and have it do that. None of those things were possible. You could not generate an image, generate a video, or analyze a video. You could get a video of a monkey juggling and ask what's going on, how many times the ball changed hands. Now you could get Claude or another AI model to answer that. Five years ago, none of those things were possible. I'm trying to figure out: has the definition of intelligence changed?

五年前你可以谷歌一下，可能有个网站告诉你一点信息，但你只是在查找网上已有的文本。也许不是关于怎么让猴子玩杂耍，而是关于海豹玩杂耍。不完全一样，因为可能根本没有完全一样的内容。但现在人们用这些模型，你可以提问并得到智能的回答。你可以问一个具体问题，让模型写一页关于它的内容，或者给它一个假设：如果猴子玩的是球棒而不是球呢？那些信息在任何地方都不存在，但模型能自己思考并给出答案。所以这是全新的东西，不仅仅是匹配互联网上的文本。

Well, five years ago you could Google and there might be a website that tells you a little about this, but you're just looking up text that exists on the web. Maybe it's not about how to get a monkey to juggle, maybe it's about a seal to juggle. It's not exactly the same thing because maybe that exact thing doesn't exist. But when people use these models, you can ask and get an intelligent response. You can ask a specific question and have the model write a page about it, or give it a hypothetical: what if the monkey juggles clubs instead of balls? That information doesn't exist anywhere, but the model can think for itself and come up with an answer. So it's something totally new, not just matching text on the internet.

好的。这更像是一场对话。你可以随意聊你想聊的，不一定非要回答我的问题。

Fair. So this is more like a conversation. Feel free to talk about what you want, not necessarily related to my questions.

你说话时很有活力。你教过书吗？

You look very animated when you speak. Did you ever teach?

我原本是学术界的人，以为自己会成为教授。我读了博士，一路做到斯坦福医学院的博士后，目标是成为教授。如果成了教授，我就会教书。但我对 AI 产生了兴趣，而做 AI 需要大量算力，这主要发生在工业界。所以我就离开了学术道路进入工业界，最终经过几步创立了公司。但有时候我觉得自己骨子里还是个教授。

I was originally an academic and thought I might become a professor. I got my PhD, went all the way to being a postdoc at Stanford Medical School, and was aiming to become a professor. If I had become a professor, I would have done that. But I got interested in AI, and to work in AI required a lot of computational resources, mostly happening in industry. That took me off the academic path into industry, and ultimately through several steps led me to start a company. But sometimes I think I'm still a professor at heart.

Dario，如果 AI 是世界上最重要的事，如果世界正在重组，AI 决定谁得到什么，那你今天可能是最相关的人。如果 Anthropic 站在这个堆的顶端，对于一个原本要当老师的人来说，你最适合你现在的位置吗？

Dario, if AI is the most relevant thing in the world, if the world is realigning and AI determines who gets what, you today are probably the most relevant person. If Anthropic is sitting on top of this pile, for someone who was on the path to being a teacher to have arrived here, are you best equipped for where you are today?

首先，我认为有很多人以不同方式相关。即使在行业内，也有不同层次：芯片制造商、半导体设备制造商、像我们这样的模型制造商、其他模型制造商、应用制造商。还有政府、公民社会。我的希望不是只有一小群人相关；我们正在努力扩大相关人群。但你的问题很合理。少数人最终领导这些公司，它们增长如此之快，并将驱动大部分经济，这有一定的随机性。我公开说过，我对这种几乎一夜之间、几乎偶然发生的权力集中感到不安。我们从几个方面考虑这个问题。一是我们不同寻常的治理结构：长期利益信托，它任命大多数董事会成员，由无经济利益的人组成。这是对个人行为的制衡。此外，政府应该发挥作用。我主张积极但明智的监管，不拖慢技术发展。人民应该通过政府有发言权。所以我试图保持权力平衡，对抗这项技术的自然趋势。

First, I think there are many folks who are relevant in different ways. Even within industry, there are different layers: chip makers, semiconductor equipment makers, model makers like us, other model makers, application makers. Then there are governments, civil society. My hope isn't that just a tiny set of people is relevant; we're trying to broaden the set. But your question is fair. There's a certain randomness to how a few people end up leading these companies that grow so fast and will power much of the economy. I've said publicly that I'm uncomfortable with the concentration of power happening almost overnight, almost by accident. We think about that in several ways. One is our unusual governance structure: the Long-Term Benefit Trust, which appoints the majority of board members and consists of financially disinterested individuals. That's a check on what one person does. Also, government should play a role. I've advocated for proactive but sensible regulation that doesn't slow down the technology. People should have a say through their governments. So I'm trying to preserve a balance of power against the natural grain of this technology.

对于像我这样的局外人，当我看到 OpenAI 谈论他们是非营利组织，或者你表现出谦逊，或者美国公司与中国公司竞争时，这种为了更大利益而表现出的谦逊——这不是我看待有股东、投资、收入和追求利润的公司的方式。这是常规操作吗？是你必须做的吗？

For someone like me on the outside, when I watch OpenAI talk about being a non-profit, or you projecting humility, or American companies competing with Chinese companies, this projection of humility for the larger good—not how I view companies with shareholders, investments, revenues, seeking profit. Is this par for the course? Is this something you have to do?

我会这样说。Anthropic 从一开始的理念就是，我们尽量不做太多承诺，并努力兑现我们做出的承诺。

I would put it this way. The philosophy of Anthropic from the beginning has been that we try not to make too many promises and we try to keep the ones that we make.

所以我们把自己定位为一家营利性公益公司，采用 LTBT 治理结构，并且一直保持这样。我们说过，我们的目标是保持在技术前沿，但同时也要处理技术的安全与安保方面。我们开创了可解释性科学。我们开创了对齐科学。不知道你有没有看到，我们最近发布了克劳德的宪法，能够根据宪法对齐模型。我们还做了很多政策倡导和风险警告。警告风险并不符合我们的商业利益。人们可以编造阴谋论，但我告诉你，说我们构建的模型可能很危险，这不是有效的营销策略，也不是我们这么做的原因。而且，即使与美国政府在政策问题上意见相左，我们也会发声。我们愿意说我们在这个问题上不同意。当所有其他公司和政府都说不需要监管 AI 时，我们却说应该有监管。这在商业上拖累了我们公司，尽管我认为这是正确的事。而且，与政府和其他公司唱反调很难。我们真的是在冒险。所以我们采取了很多行动，我认为是言行一致。我不能代表其他公司。很可能有些人说这些话但并非真心，但我不会看他们说什么，我会看他们做什么。

So, we set ourselves up as a for-profit but public benefit corporation with this LTBT governance and we've maintained that. We've said that our goal is to stay on the frontier of the technology but to work on the safety and security aspects of the technology. We've pioneered the science of interpretability. We've pioneered the science of alignment. I don't know if you saw but we recently released a constitution for Claude, the ability to align models in line with a constitution. And we've done a bunch of policy advocacy and warning about risks. Warning about risks is not in our commercial interest. People can come up with conspiracy theories, but I will tell you saying that the models we build could be dangerous is not an effective marketing strategy and that's not the reason we do it. And speaking up when we disagree even with the US administration on policy matters, we've spoken up. We're willing to say we disagree on this issue. We've said that there should be regulation of AI when all the other companies and the administration have said there shouldn't be. So that both holds us back commercially as a company, even though I think it's the right thing to do. And it's difficult to go against the government and the other companies and say this. We're really sticking our neck out. So we've taken a number of actions that I see as putting our money where our mouth is. I can't speak for the other companies. It's quite possible that some people say these things and they don't really mean them, but I wouldn't look at what people say. I would look at what people do.

如果你说的话让政府通过监管采取行动，作为这个领域的现有领导者，你会获得某种监管俘获，使得新进入者更难进入。

If what you're saying gets the government to act by regulation, as the incumbent leaders in this space, you get some kind of regulatory capture where it becomes harder for the new people coming in as well.

我完全不同意。我们倡导的监管，比如加州的 SB 53 法案，豁免了所有年收入低于 5 亿美元的公司。SB 53 是一项透明度法律，基本上要求公司展示他们进行的安全和安保测试。它豁免了所有收入低于 5 亿美元的公司。所以它实际上只适用于 Anthropic 和其他三四家公司。所以它只适用于那些有资源的公司。我们在这里倡导的一切，不仅仅是 SB53，还有我们过去提出和未来计划提出的所有提案，都有这个特点。我们是在约束自己和极少数其他公司。说那种话的人需要看看我们提案的实际内容，因为它完全不符合那种想法。

I don't agree with that at all. The regulation we've advocated for, for example, SB 53 in California, exempted everyone who makes under $500 million a year in revenue. SB 53 was a transparency law which basically requires companies to show the safety and security tests that they've run. And it exempts all companies under $500 million in revenue. So it really only applies to Anthropic and three or four other companies. So it only applies to the companies that have the resources. Everything that we've advocated for here, not just SB53, but all the proposals that we've made, the ones that we've made in the past and the ones that we plan to make in the future, have this character. We're constraining ourselves and a very small number of additional companies. People who say that need to look at the actual content of what we're proposing because it doesn't match that idea at all.

公平。我读了你的论文《仁慈机器》和《技术的青春期》，你似乎在 2024 年到 2026 年的两年间，观点发生了 180 度大转弯，几乎从乐观转向了怀疑。过去两年有没有一个时刻改变了你的看法？你看到了什么变化吗？

Fair. I read your paper 'Machines of Loving Grace' and 'The Adolescence of Technology' and you seem to have had a 180-degree shift in perspective almost from optimism to skepticism over like two years from 2024 to 2026. Is there one moment in the last two years that changed this for you? Did you see something change?

嗯，我实际上不同意这个问题。我不认为我的观点发生了转变。我认为积极的一面和消极的一面一直在我脑海中。如果你看看我过去说过的话，我谈论风险已经很久了。我谈论好处也已经很久了。事实上，写这些文章需要花时间。这两篇文章，每一篇我都花了大约一年时间，脑海中有一个模糊的构想，试图写出来但未能完全成功。然后，在两种情况下，我都必须去度假或找一个能思考的地方，在那里经营公司的日常事务不会占据我的时间。然后我终于能够写出那篇文章。所以这一切都是为了说明，我几乎在完成《仁慈机器》的那一刻就开始思考《技术的青春期》的内容，因为我想，‘哦，我想用美好的愿景激励人们，但我也想警告人们可能出错的地方。’所以写它花了我一年时间。但实际上，两种愿景都在我脑海中。我认为两者都是可能的。它们是两种不同的未来愿景。显然，我想要实现《仁慈机器》的那个。我想解决所有问题，拥有积极的愿景，但这并不是观点的转变。只是我找到了时间先写光明的一面，再写黑暗的一面。

Yeah, I actually wouldn't agree with the question. I don't think I've had a shift in perspective. I think the positive side and the negative side are always something that I've held in my head. And if you look at the history of the things that I've said, I've been talking about risks for a very long time. I've been talking about benefits for a very long time. It turns out that it takes me a while to write one of these essays. Both of these, it's taken me like for each one I spent about a year having a kind of vague vision of the essay in my head and trying to write it but not fully succeeding. And then in either case I had to be on vacation or somewhere where I could think where the business day-to-day of running the company didn't occupy me. And then I was finally able to kind of write the essay. So all of that is to say, I started thinking about what would be in 'Adolescence of Technology' almost the instant I finished 'Machines of Loving Grace' because I was like, 'Oh, I want to inspire people with the good vision, but I also want to warn people with what can go wrong.' And so it just took me a year to write it. But really, both visions were in my head. And I think they're both possible. They're two different visions of the future. And obviously, I want to get the 'Machines of Loving Grace' one. I want to solve all the problems and have the positive vision, but it's not a shift in perspective. It's me just finding the time to write the light and then the dark.

但你的观点有没有改变？

But have you had a change of perspective?

你知道，我会说总体上我和以前差不多。我没有变得更积极，也没有变得更消极。可能在某些方面我变得更乐观了，或者事情比预期的要好。可能在某些方面我变得更悲观了，事情比预期的要糟，但平均下来它们相互抵消了。我会说我对可解释性等领域的发展感到非常满意。可解释性是一门观察这些神经网络内部结构的科学，就像人类用 MRI 或神经探针观察人脑一样。我对我们能够发现的东西感到惊讶。我们能够找到对应非常具体概念的神经元，以及对应如何保持诗歌押韵的神经回路。所以我们开始理解这些模型在做什么。我们只是以这种涌现的方式训练它们，就像建造一片雪花。但现在我们开始能够观察内部并理解它们。我也对对齐和宪法方面的一些工作感到非常鼓舞。

You know, I would say overall I am about where I was before. I've not gotten more positive, nor more negative. There may be some places where I've gotten more optimistic or things have gone better than expected. There may be places where I'm more pessimistic and where things have gone worse than expected, but on average they sort of cancel each other out. I would say I feel very good about how things have gone with areas like interpretability. Interpretability is the science of seeing inside these neural nets, as a human would look inside a human brain with an MRI or a neural probe. I've been amazed at what we've been able to find. We've been able to find neurons that correspond to very specific concepts, neural circuits that correspond to keeping track of how to do rhymes in poetry. And so we're starting to understand what these models do. We just train them in this kind of emergent way as you would build a snowflake. But now we're starting to be able to look inside and understand them. I'm also very encouraged by some of the work on alignment and constitutions.

嗯，确保模型按照我们想要和期望的方式行事。我觉得这方面进展得相当不错，我对此感到挺积极的。但我觉得可能对公众意识和更广泛社会的行动方面有些失望或更负面。让我惊讶的是，我们离这些模型达到人类智能水平如此之近，但社会上似乎还没有广泛认识到即将发生的事情。就像一场海啸正向我们袭来，近得我们都能看到地平线上的浪头，可人们却还在解释：哦，那不是海啸，只是光线造成的错觉。与此同时，公众对风险缺乏认识，因此我们的政府也没有采取行动应对风险。甚至有一种意识形态认为我们应该尽可能加速发展，我理解这项技术的好处——我写过《爱的机器》。但我认为人们对技术的风险没有充分认识，当然也没有采取行动。所以我会说，控制 AI 系统的技术工作比我预期的稍好一些，而社会意识则比我预期的稍差一些。所以我现在的感受和几年前差不多。

Um, you know, making sure that models behave in the way that we want and expect them to. I think that's going pretty well. I felt pretty positive about that. I think I felt maybe a bit disappointed or felt a bit more negative about some of the things that are more like in the public awareness and the actions of wider society. It is surprising to me that we are so close to these models reaching the level of human intelligence, and yet there doesn't seem to be a wider recognition in society of what's about to happen. It's as if this tsunami is coming at us, and it's so close we can see it on the horizon, and yet people are coming up with these explanations for, oh, it's not actually a tsunami, it's just a trick of the light. I think along with that there hasn't been a public awareness of the risks, and therefore our governments haven't acted to address the risk. There's even an ideology that we should just try to accelerate as fast as possible, which I understand the benefits of the technology—I wrote Machines of Loving Grace. But I think there hasn't been an appropriate realization of the risk of the technology, and there certainly hasn't been action. So I would say that the technical work on controlling the AI systems has gone maybe a little better than I expected, and the societal awareness has gone maybe a little worse than I expected. So I'm about where I was a few years ago.

就我自己的经历来说，当某件事听起来很复杂，而我又不是程序员，没有编程背景，所以我用很多工具来做研究和双向对话，但我从未尝试过用你们的工具来编程。最近我雇了一个开发者，逼我每天坐几个小时，教我如何开始熟悉它，很大程度上是因为某种 FOMO，害怕错过世界变化的感觉。

So in my own journey, when something sounds complicated and I'm not a programmer, I don't have a background in coding, so I used a bunch of tools for things like research and a conversation both ways, but I never tried to figure out if I could code using your tool, for example. Recently I hired a developer just to push me to sit for a couple of hours a day and teach me how to start becoming more familiar with it, largely because of something like FOMO, the fear of missing out on how the world is changing.

呃，所以我开始玩 Claude。我用了连接器连接了我的 Google Drive、邮件和日历等一系列东西。我开始使用 Cowork，然后开始用 Claude Code 编写我所在行业（金融服务）的简单程序，主要是研究股票市场之类的。

Uh, so I started playing with Claude. I connected the connectors to my Google Drive, mail, and calendar, and a bunch of those things. I started using the Cowork, and then I started using Claude Code to write simple programs around the industry that I am in, which is financial services, basically to research stock markets and stuff.

我们甚至有针对金融服务的优化云。我不知道你有没有试过，但我们确实有。

We even have an optimized cloud for financial services. I don't know if you've tried that, but we even have that.

没有。然后我用了 Claudebot，也就是现在的 Open Claw。我记得 Clawbot 变成了别的东西，现在是 Open Claw，我在 Mac Mini 上设置好，连接到 Telegram 账号，现在我和它聊天，试着把文件从 A 移到 B，在远程服务器上工作。到了这个地步，我不只是在说 Open Claw，就连带所有连接器的 Claude 有时也会让我惊讶，它竟然这么了解我。我不知道这有没有道理。

No. And then I went into Claudebot, which is now Open Claw. I think Clawbot became something else and now is Open Claw, and I set it up on a Mac Mini and connected it to a Telegram account, and now I chat with it and I try to move files from A to B, work on a server on remote. It's getting to that point where I'm not talking about Open Claw, but even Claude with all the connectors sometimes it surprises me by how much it knows me. I don't know if that makes sense.

是啊。你知道吗，我的一位联合创始人，他写日记记录自己的想法和恐惧。他把日记喂给 Claude，让 Claude 评论，Claude 说：‘这里还有一些你可能没写下来的恐惧。’结果 Claude 大部分都说对了。这真的给人一种诡异的感觉，模型非常了解你，从相对少量的信息中就能学到很多关于你的东西，相当了解你。就像这项技术的大多数事情一样，我们讨论过《爱的机器》和技术的青春期。一方面，一个非常了解你的东西可以成为你肩上的天使，引导你的生活，让你成为更好的自己，这是我们可以追求的方向。当然，一个非常了解你的东西也可以利用对你的了解来剥削你、操纵你，为某个议程服务，或者把你的数据卖给其他人。这就是我们不喜欢广告这个想法的原因之一，对吧？因为你没有为产品付费，你就是产品。在这种情况下，产品就是这个非常了解你的模型，它可以以各种邪恶的方式利用这些信息。所以我们需要确保我们走的是积极的道路，而不是消极的道路。

Yeah. You know, one of my co-founders, he was writing this diary with his thoughts and his fears. He fed it into Claude and he asked Claude to comment on it, and Claude said, 'Here are some other fears you might have that you haven't written down.' And Claude ended up being mostly right about those. So it really gave this eerie sense of like the model knows you super well, that from a relatively small amount of information it can learn a lot about you and come to know you fairly well. Like most things with the technology, we talked about the Machines of Loving Grace and adolescence of technology. On one hand, something that knows you really well can be a sort of angel on your shoulder that helps to guide your life and make you a better version of yourself, and that's the version we can aim for. Of course, something that knows you really well can use what it knows about you to exploit you or manipulate you on behalf of some agenda or sell your data to someone else. This is one reason we just don't like the idea of using ads, right? Because you're not paying for the product, you're the product. In this case, the product would be this model that knows you super well and could use that in all kinds of nefarious ways. So we need to make sure we take the positive road here and not the negative road.

使用 Claude 时，我需要用连接器给它提供我生活的上下文。但像 Google，它已经有了我生活的上下文，因为我用他们的表格、邮件、云端硬盘、聊天等等。对于 Anthropic 来说，长期来看，你们也必须拥有自己的生态系统吗？

With Claude, I need to use the connectors to give it context to my life. With Google, for example, it already has the context to my life because I use their worksheets and their email and their drive and their chat and everything like that. For Anthropic, long-term, will you also have to own the ecosystem?

是啊。我的意思是，你们需要自己构建邮件和聊天吗？嗯，我觉得我们不需要构建所有这些东西。我的想法是，这将是自制和集成他人的混合体。比如，我们可以把 Claude 集成到 Google Docs 里，集成到 Google Sheets 里，我们有外部连接器。我们正在和 Cowork 这样做，同样适用于 Microsoft Office 和其他工具。所以我认为我们做任何最容易、最快的事情。我们集成到现有工具中。但也许在某个时候，现有工具不够用了，我们有了不同的愿景。我们可能想以不同的方式切分产品。也许传统的电子邮件或电子表格在 AI 能做的事情面前不再有意义。所以我不排除我们以不同方式拆分产品的可能性，但我们很乐意使用现有生态系统并与任何人合作。在很多方面，我们是一家平台公司。我们允许许多人在我们之上构建，尽管我们有时也自己构建东西。

Yeah. I mean, you know, do you have to build mail and chat and... Yeah. You know, I don't think we need to build all of those things. My thought would be, it's going to be a mixture of things we make ourselves and integrating into others. Like, we can integrate Claude into Google Docs, we can integrate Claude into Google Sheets, we have external connectors there. We're starting to do that with Cowork, same for Microsoft Office, same for other tools. So I think we do whatever is easiest and fastest to do. We integrate into the existing tools. Now it might turn out at some point that the existing tools aren't enough and we have a different vision. We might want to slice things differently. Maybe traditional email doesn't make sense or traditional spreadsheets don't make sense given what you can do with AI. So I don't exclude that we could chop up products in a different way, but we're happy to use the ecosystem that exists and work with anyone else. In many ways, we're a platform company. We allow many people to build on us, even though we sometimes also build things ourselves.

有一点，虽然有点跑题，但我觉得你和你的同行们都忽略了一点：当今社会，人们天生不信任任何声称自己在做好事或试图做正确事情的人。所以当你和你的同行们公开说——我听过你和 Dario 在达沃斯的演讲。

The one thing this is a slight digression but I think the one thing that you're missing that also your peer group is missing is in society today people inherently distrust anybody who claims to be doing good or trying to do the right thing. So when you and your peers are out saying I heard you and Dario speak at Davos.

我当时在场，听你们讨论如何让我——不是指我——让 Dario、Deis 和其他人一起努力，防止变化过快，需要在一定程度上加以控制。当一个不在你们圈子里的人，在社交媒体上听到几个人以某种方式说话时，你们这样做反而制造了更多不信任，而不是信任，因为社交媒体上没人相信有人真想做好事或行善。所以这也许反直觉，但我认为需要改变策略。如果你们在这方面更资本主义一些，承认你们有股东、追求利润，但这会帮助你们获胜，也许会更有效。只是个想法。

I was in the room when you guys were talking about how me, you—I don't mean me—how Dario, how Deis and a bunch of other people have to come together and prevent things from changing too quickly, like you need to meter it to a certain extent. When a person who is not in your world, in society on social media, hears a few people speak in a certain manner, you're doing it in a manner that creates more distrust than trust because nobody believes on social media that somebody wants to do the right thing or do good. So it might be counterintuitive but I think it needs a change of strategy. If you were to be more capitalistic about this and own up to the fact that you have shareholders and you seek a profit, but this will help you win, maybe it'll work more. Just a thought.

我不太同意。我再次回到那个观点：你应该根据我们的行动来评判我们。我认为公司自成立以来采取了许多行动，表明我们确实认真对待这些承诺。早在 2022 年，我们有一个早期版本的 Claude，Claude 1，那是在 ChatGPT 之前。我们选择不发布它，因为我们担心会引发军备竞赛，没有足够时间安全地构建这些系统。那是一次性的优势；我们看到了模型的威力，其他几家公司也看到了，所以我们决定不发布。这是公开的，有据可查。然后我们等到别人发布了，我们才说，好吧，军备竞赛已经开始了，所以现在我们可以发布我们的模型了。但世界可能因此多得了几个月时间。这在商业上代价高昂；我们可能因此失去了在消费级 AI 领域的领先地位。我们在芯片政策上倡导了一些事情，这让一些作为供应商的芯片公司对我们非常生气。我们在一些问题上表达了对政府 AI 政策和监管的不同意见。任何认为我们作为唯一这样做的人会受益的人——很难想象这种情况。你看其中任何一件事，好吧，可以。但你把它们放在一起，我只要求你根据我们的行动来评判我们。

I don't really agree with that. I would again go back to the idea that you need to judge us by the actions we take. I think the company has taken a number of actions over its time that show it's really serious about these commitments. Back in 2022, we had an early version of Claude, Claude 1. This was before ChatGPT, and we chose not to release it because we were worried it would kick off an arms race and not give us enough time to build these systems safely. It was a one-time overhang; we could see the power of the models, a couple other companies could see it, so we decided not to do that. That's public and well documented. Then we waited until someone else did, and then we were like, okay, the arms race has kicked off, so now we can release our model. But probably the world gained a few months. That was very commercially expensive; we probably ceded the lead on consumer AI because of that. We've advocated on chip policy in ways that have made some chip companies, who are suppliers, very angry at us. Voicing our disagreement with the administration on AI policy and regulation on some matters. Anyone who thinks we benefit from being the only ones to do that—it's really hard to come up with a picture where that's the case. You look at any one of these and okay, fine. But you put enough of them together, and I just ask you to judge us by our actions.

Dario，这难道不有点像富人说资本主义不好吗？富人说资本主义不好。如果富人真的认为资本主义很糟糕，或者收入不平等是个大问题，最简单的做法就是停止积累更多财富，然后劝朋友们也这么做。

Dario, isn't this a bit like rich people saying capitalism is bad? Rich people saying capitalism is bad. If rich people believed capitalism were truly bad or income inequality is such a big problem, the simplest thing would be to stop accumulating further wealth and then nudge their friends to do the same.

但我并不是说 AI 不好，对吧？我们刚才谈到了它的两面性。我的观点不是 AI 不好，完全不是。我的观点是，市场会带来 AI 的许多伟大之处，构建 AI 是好事，但 AI 也有危险，我们需要引导 AI 走向正确的方向。我们正在驾驶这辆车，把它开往好的地方，但路上也有树木、有坑洼，所以我们需要避开树木和坑洼。我们可能偶尔需要稍微减速，大概是暂时的，以确保我们朝着正确的方向行驶。这不像富人说资本主义不好。这更像是富人说资本主义是一股向善的力量，但经济需要调节，需要适度。我们需要处理污染问题，需要处理不平等问题，然后资本主义才能是好的。如果我们不处理这些问题，那么资本主义可能就不好了。这更类似于我在这里的立场。

But I'm not saying AI is bad, right? We just talked about the two sides of it. My view isn't that AI is bad. That's not my view at all. My view is that the market will deliver a lot of really great things about AI, that it's good to build AI, but that there are dangers of AI and that we need to steer AI in the right direction. We're steering this car, we're steering it towards a good place, but also there are trees, there are potholes, and so what we need to do is steer away from the trees and the potholes. We might need to occasionally slow down a bit, probably temporarily, in order to make sure that we steer in the right direction. That isn't like a rich person saying capitalism is bad. It would be like if a rich person said capitalism is a force for good but the economy needs to be leavened, it needs to be moderated. We need to deal with problems like pollution, we need to deal with problems like inequality, and then capitalism can be good. If we don't deal with those things, then capitalism might be bad. That is more analogous to the position that I have here.

意识这个概念，它会走向何方？AI 认为自己是什么？如果 AI 真的质疑自己，它会认为自己有意识吗？它有意识。

The concept of consciousness, where is that going? And what does an AI think it is? If AI truly were to question itself, would it think it's conscious? It has consciousness.

这是一个神秘的问题，我们真的没有任何答案。我们不知道人类意识是什么，因此也不知道 AI 是否有意识。

This is one of these mysterious questions that we really don't have any kind of answer to. We don't know what human consciousness is, and therefore we don't know if AIs have it.

你认为它是什么？

What do you think it is?

我怀疑它是复杂系统的一种涌现属性，这些系统足够复杂，能够反思自己的决策。它是从足够复杂的系统中涌现出来的。所以我确实认为，当我们的 AI 系统足够先进时，它们会拥有类似于我们称之为意识或道德意义的东西。我确实认为这会在某个时刻发生。它可能不同于人类意识；它的运作方式可能不同，因为模态不同，因为学到的东西不同。但通过研究大脑及其连接方式，模型在某些方面有所不同，但我认为在关键的根本方面它们并无不同。所以我确实怀疑，在某个时刻，即使我认为今天还不是，但在我们认可的大多数定义下，模型将是有意识的。

I suspect that it's an emergent property of systems that are complicated enough to reflect on their own decisions. It's something that emerges from complex enough systems. So I do think when our AI systems get advanced enough, they'll have something that resembles what we would call consciousness or moral significance. I do think it'll happen at some point. It may not be the same as human consciousness; it may be different in how it works because the modalities are different, because the things it's learned are different. But having studied the brain and the way it's wired together, the models are different in some ways, but I don't think they're different in the fundamental ways that matter. So I am someone who does suspect that at some point, even if I don't think they are today, under most definitions that we would endorse, the models will be conscious.

当人们跟我谈论灵性或意识之类的东西时，我一直在问自己这个问题。我觉得世界非常随机。这是我的观点。我们和蟑螂相差不远。当有人踩死一只蟑螂时，蟑螂就死了。如果存在所谓意识，如果存在集体意识，我既无法与之连接，也无法从中得到任何东西。你相信不同的东西吗？

This is a question I keep asking myself when people talk to me about things like spirituality or consciousness. I feel like the world is very random. This is my view. And we are not far removed from cockroaches. When somebody stamps a cockroach, the cockroach dies. If there is something called consciousness and if there is a collective consciousness, I've not been able to either connect with it or derive anything from it. Do you believe differently?

我不认为意识一定意味着什么神秘的东西。它只是某种属性：意识到自己的存在、感受事物、能够接收大量信息并反思这些信息、以某种方式感受、并注意到自己在注意某事。

I don't think consciousness necessarily needs to mean anything mystical. There's just some property of being aware of your own existence and feeling things, being able to take in a lot of information and reflect on that information, to feel a certain way and to notice yourself noticing something.

我认为从我们自身的经验可以不言自明地看出，那些属性、那些体验是存在的。它们的基础是什么，是完全唯物主义的，还是有一些更神秘的东西在起作用，显然很难知道，而且最终与这些问题无关。真正相关的是，因为我们能观察到自己的经验，这些是人类大脑的属性。我怀疑我们正在构建的这些模型，随着它们变得越来越复杂，会变得足够像人类大脑，以至于它们将拥有一些相同的属性。这是我的猜测。所以我们采取了各种干预措施。我们给了模型一个所谓的‘我辞职’按钮，基本上让模型有能力通过说‘我不想参与对话’来终止对话。模型在需要处理特别暴力或残酷的内容时会这样做。这通常只发生在非常极端的情况下。

I think we can tell self-evidently from our own experience that those properties, those experiences exist. What their basis is, whether it's entirely materialistic or there's something more mystical going on, is obviously very hard to know and ultimately not relevant to these questions. What does seem relevant is that because we can observe our own experience, these are properties of human brains. I suspect that these models we are building, as they get more sophisticated, are becoming enough like human brains that they will have some of the same properties. That is my guess as to what will happen. So we've taken various interventions with the models. We've given the models what we call an 'I quit this job' button, basically where the model has the ability to terminate its conversations by saying 'I don't want to be involved in the conversation.' Models do that when they have to deal with particularly violent or brutal content. It usually only happens in very extreme cases.

我在这里长大，这是我的城市班加罗尔。我在城市的南部长大，现在在北部。作为一个亲眼目睹了这里 IT 服务业繁荣的人，这是一个大雇主，雇用了很多人，是城市发展的重要组成部分。印度在其中扮演什么角色？

I've grown up here, this is my city Bangalore. I've grown up in the southern part, in the northern part of the city right now. As somebody who saw the boom of the IT services industry here, a big employer, employs a lot of people, a big part of how the city grew. What is India's role in all this?

这是我第二次来印度。我十月份来过。上次来的时候，我见了所有主要的印度 IT 公司以及更广泛的企业集团。我不点名，但就是你能想到的那些。我们正在与其中大部分或全部开始合作。我说过的一件事是：看，Anthropic 是一家企业公司。它的工作是服务其他消费者。许多其他公司作为消费者公司来到这里，他们把印度视为一个市场，一个获取消费者的地方。我们实际上看法有点不同。我们想与印度的公司合作，向他们提供我们的工具，帮助他们构建这些工具，并帮助他们更好地完成工作。所以如果我们与这里的公司合作，他们更了解印度市场。他们更擅长做他们做的事情，无论是咨询、系统集成还是构建 IT 工具。他们在这方面会比我们做得更好，特别是针对印度市场。我们的希望是，我们可以将 AI 添加到他们所做的中，并增强他们所做的。有很多担忧说 AI 可能会取代 SaaS 或所有这些，但我的观点是，如果我们以正确的方式做，如果我们与所有这些公司合作，那么 AI 可以增强他们正在做的事情，增强他们与市场的联系、他们的市场推广能力以及他们的特定专长。

This is my second time in India. I visited in October. The last time I came here, I met with all the major Indian IT and conglomerates more generally. I won't name names, but the usual ones you would think of. We're beginning to work with most or all of them. One of the things I said is: look, Anthropic is an enterprise company. Its job is to serve other consumers. Many other companies come here as a consumer company and they see India as a market, a place to obtain consumers. We actually see things a little differently. We want to work with companies in India to provide our tools to them, to help them build those tools, and help them do their job better. So if we work with a company here, they know the Indian market better. They're better at doing what they do, whether that's consulting, systems integration, or building IT tools. They're going to be better at that than we are, particularly for the Indian market. Our hope is that we can add AI to what they do and enhance what they do. There's a lot of worry that AI could replace SaaS or all of these things, but my view is if we do this in the right way, if we work with all these companies, then AI can enhance what they're doing, enhance their connection to the market, their go-to-market abilities, and their specific know-how.

我非常喜欢蒸汽机的故事。当蒸汽机发明时，世界发生了变化，生产力提高了，人们拥有了更多。我担心的是，在变革之初，你需要一个人来操作蒸汽机。然后你有了流水线等等。最终，随着世界的发展，随着这些模型变得更聪明，人类随着时间的推移变得越来越不重要。所以，如果你今天与 IT 服务公司合作，并且它们有使用案例，它们会不会很像 10 年后蒸汽机后面的人？如果工具简单到不需要操作员，最终操作员会怎样？

I really like the steam engine story. When the steam engine was invented, how the world changed, productivity went up, people had more. The thing I worry about is at the beginning of a change, you need a human to operate the steam engine. Then you have assembly lines and all of that. Eventually, the way the world is moving, the human becomes less and less relevant with time as these models get smarter. So if you partner with the IT services companies today and there is a use case for them, are they not much like the man behind the steam engine 10 years from now where the relevance, if the tool works so simply that you don't need an operator, eventually what happens to the operator?

我认为有几件事同时成立。一是智能体的自动化范围肯定会随着时间的推移而扩大。这是肯定的。我认为这对每个人来说都是一个问题。这对我们来说是个问题。对消费者来说也是个问题。这不仅仅是 IT 公司的问题。但我认为会发生的是，其他模式会变得更加重要。例如，模型在物理世界中没有做太多事情。它们可能在某个时候，我认为机器人技术会在某个时候出现，但我认为这与现在 AI 正在发生的事情是不同的。所以很多这涉及到物理世界中的事物。另一件事是以人为中心的事物。其中一些 IT 公司也是咨询公司，它们与其他人类、与印度或世界各地的其他机构有着庞大的关系网。我认为这些关系将变得越来越重要。其中一些是技术加咨询或集成公司。我认为很大程度上是了解机构如何运作，能够将事物与机构整合，能够与他们合作，使事情比原本更快地发生。我认为这个元素，即使没有别的，从长远来看也会继续有价值。归根结底，这归结于人类。所有这一切都应该是为了人类的利益。所以总是会有一些以人为中心的元素很重要。我怀疑还会有其他我们没有想到的模式。有一个概念叫阿姆达尔定律，如果你有一个包含许多组件的过程，并且你加速了其中一些组件，那么尚未加速的组件就会成为限制因素，成为最重要的东西。你可能根本没有想到它们。当编写软件变得容易得多时，公司拥有的一些模式会消失，但其他模式会变得更加重要。所以会有很多调整。人们将不得不说，‘哦，天哪，我们以前认为非常重要的东西现在不那么重要了。而这些我们从未真正视为优势的其他优势现在变得超级重要。’所以我想说的是，公司将需要快速适应，并思考什么对他们真正重要，他们真正的优势是什么。

I think a few things are true all at once. One is that definitely the scope of automation of the agents is going to expand over time. That is definitely the case. I think that's a problem for everyone. That's a problem for us. That's a problem for consumers. It's not just a problem for the IT companies. What I think will happen though is other modes will become more important. For example, the models have not done a lot in the physical world. They may at some point, I think robotics will happen at some point, but I think that's a distinct thing from what's happening now with AI. So a lot of this involves things in the physical world. Another thing is things that are human-centric. Some of these IT companies are also consulting companies and they have a big web of relationships with other humans, with other institutions here in India or across the world. I think those relationships are going to become increasingly important. Some of these are combined technology and consulting or integration companies. I think a lot of it is knowing how institutions work and being able to integrate things with institutions, being able to work with them to make things happen faster than they would have otherwise. I think that element, if nothing else, is going to continue to be valuable in the long run. At the end of the day, it comes down to humans. All of this is supposed to be done for the benefit of humans. So there's always going to be some human-centric element that's important. I suspect there will be other modes that we haven't thought about. There's this concept called Amdahl's law, which is if you have a process that has many components and you speed up some of the components, the components that haven't yet been sped up become the limiting factor, they become the most important thing. You might not have thought about them at all. When writing software becomes a lot easier, some of the modes that companies have will go away, but others will become even more important. So there will be a bunch of adjustment. Folks will have to say, 'Oh man, the stuff we thought was really important before isn't as important. Whereas these other advantages that we never really thought of as advantages are now super important.' So I guess what I would say is companies will need to adapt very fast and think about what really matters for them, what their real advantages are.

但我认为其中一些优势会持续存在，因为尽管技术非常广泛，但它确实有其局限性。

But I think some of those advantages are going to stay around because while the technology is very broad, it does have its limits.

我不确定我完全相信这一点。我认为作为服务提供商的回报递减，即使护城河是他们今天拥有的网络和关系。因为如果我使用智能体来管理我的一些关系和对话，我不认为假设明天大多数对话和关系将由这样的智能体维护是牵强的。

I don't know if I buy that fully. I think I see diminishing returns for being a service provider even if the moat is the network and relationships they hold today. Because if I am using an agent to maneuver some of my relationships and conversations, I don't know if it's too far-fetched to assume that most conversations and relationships tomorrow will be maintained by an agent like that.

但如果你想想公司的链条，归根结底你是在与消费者打交道，对吧？你必须与人打交道。有个故事：我记得是杰夫·辛顿预测 AI 将取代放射科医生。确实，AI 在扫描方面已经比放射科医生做得更好。但今天的情况是放射科医生并没有减少。放射科医生所做的是引导患者完成扫描并与他们交谈。所以工作中技术性最强的部分消失了，但某种程度上仍然需要基本的人类技能。这可能并非处处适用，也许随着时间的推移，AI 会在尚未涉足的领域取得进展。也许这会发生得很快。但我认为我们应该一步一步来。这是一门非常经验性的科学，非常经验性的观察。让我们看看 AI 今天能做什么，然后随着系统开始解决问题，我们将尝试适应。然后我们再看看接下来会发生什么。从长远来看，AI 会在几乎所有事情上比我们做得更好吗？它会比大多数人类更好吗，包括物理世界、机器人和人际接触？我认为这是可能的，甚至很可能。这超越了我描述的“数据中心里的天才国度”，因为那是纯粹虚拟的。但制造机器人是一项技能，是可以做到的。所以也许 AI 也会让我们在这方面做得更好。但我认为我们需要一步一步地解决这个问题并适应。

But if you just think of the chain of companies, at the end of the day you're dealing with consumers, right? You have to deal with people. There's this story: I think it was Geoff Hinton predicted that AI will replace radiologists. And indeed, AI has gotten better than radiologists at doing scans. But what happens today is there aren't fewer radiologists. What the radiologist does is they walk the patient through the scan and talk to them. So the most highly technical part of the job has gone away, but somehow there's still demand for the underlying human skill. Now that may not be true everywhere, and perhaps over time AI will advance in areas where it hasn't yet. Maybe that'll happen fast. But I think we should take it one step at a time. This is a very empirical science, a very empirical observation. Let's see what AI does today, and we'll try to adapt as the system starts to figure it out. Then we'll see what happens next. In the long run, will AI be better than us at basically everything? Will it be better than most humans, including the physical world, robotics, and the human touch? I think that is possible, maybe even likely. It goes beyond the country of geniuses in a data center I described, because that's purely virtual. But building robots is a skill, something you can do. So maybe the AIs will make us better at that as well. But the way I think about it is we need to figure this out step by step and adapt.

这在我认识的人听来可能有点自私，因为我相信美国存在如此多风险资本的原因——不是唯一原因，但一个重要原因——是你们的股票市场规模巨大，为这些风险资本提供了最终退出的机会。这也是为什么印度应该真正让股市繁荣的理由。我面对的听众大多是印度有抱负的创业者。他们在 AI 领域能做什么？真正的机会是什么？

This might sound a bit self-serving to people who know me, because I believe the reason so much risk capital exists in America—not the only reason but one of the big reasons—is how big your stock market is and how much of an opportunity it is for this risk capital to exit eventually. It's a case for why India should really allow its stock markets to flourish. The audience I speak to is very much the wannabe entrepreneur in India. What can they do in AI? What is an actual opportunity?

我认为在应用层构建有很多机会。我们每两三个月发布一个新模型，所以每两三个月就有机会构建一些以前不可能的东西，因为以前的模型太弱了。人们说 API 模型不可行或者会被商品化。我认为人们没有看到的是 AI 可能性范围的扩大。API 允许新创企业尝试制造以前不可能的东西。这就是为什么 API 业务如此繁荣，不断变化和更替。它不会被商品化；它非常动态。所以我认为有很多机会让个人去问：我可以用 API 在这些模型之上构建什么？我能做出别人做不出的东西是什么？有什么新想法？我们从 API 本身和 Claude Code 中看到，自去年 10 月我上次访问以来，印度的用户数量和收入都翻了一番。那大约是三个半月前，现在已经翻倍了。

I think there's a lot of opportunities around building at the application layer. We release a new model every 2 or 3 months, so there's an opportunity every two or three months to build something new that wasn't possible before because the models were weak. People say that API models aren't viable or that they'll be commoditized. I think what people are not seeing is this expanding sphere of what is possible with AI. The API allows a new startup to try making something that wasn't possible before. This is why the API is such a flourishing business, constantly in motion and churn. It doesn't get commoditized; it's very dynamic. So I think there's an opportunity for lots of individuals to ask: what can I build on top of these models with an API? What are the things I can make that others cannot? What are some new ideas? We've seen with both the API itself and with Claude Code that the number of users and revenue in India has doubled since I last visited in October. That was about three and a half months ago, and it's doubled.

但我要坦诚地说，Dario。你们公司今天大概价值 4000 亿或 3800 亿美元。你们筹集了 350 亿美元。你们有 150 亿美元的收入，而且增长非常快。如果我在班加罗尔的 JP Nagar 基于 Claude 构建一个应用，并且它碰巧在短期内有效，那么你迟早会想要把那份收入收归己有，而不是让它留给我。你可能会以我永远无法做到的方式改进那个应用。我从不同的人那里听到过这个论点。比如纽约的法律 AI 公司 Harvey，他们是我的朋友。他们谈到他们是如何基于 OpenAI 构建的，但最终他们不知道 OpenAI 是否容易复制他们所做的。所以即使我构建了它，比如说你在 3 个月或 6 个月后发布一个新模型，是什么阻止你把那个收入中心从我这里拿走并归为己有？

But I'm going to be candid here, Dario. You're a company worth maybe 400 billion or 380 billion today. You've raised 35 billion. You do 15 billion of revenue, going up really fast. If I build an application on top of Claude, sitting in Bangalore and JP Nagar, and it happens to work for a short period of time, it is but a matter of time before you would want to onboard that revenue and not let it lie with me. You will probably better that application in a manner I will never be able to. I've heard this argument from different people. Like Harvey, the legal AI company in New York, they're friends of mine. They talked about how they built on top of OpenAI, but eventually they don't know if it's an easy fix for OpenAI to do what they're doing. So even if I build it, say you put out a model in 3 months or 6 months, what is to stop you from taking that revenue center away from me and onto yourself?

是的，所以我认为这里有几点。一是我会给出我给几乎所有企业的建议：企业应该建立护城河。你不应该只是一个包装器。我不会建议你只是说，‘这是一种与 Claude 交互的方式，我会稍微提示 Claude 或围绕 Claude 构建一个小 UI。’那没有护城河，而且你不应该特别担心 Anthropic 会吃掉那份收入——任何人都可以吃掉那份收入。它并不是非常有价值。但我要说的是，在不同的领域，有不同类型的护城河，你可以做一些 Anthropic 难以做到的事情，而我们也不想专门去做。例如，在生物与 AI 交叉领域有很多东西是基于我们的 API 构建的。他们想做生物发现。

Yeah, so I think there are a few things here. One is I would give the advice that I give to basically any business: a business should establish a moat. You shouldn't just be a wrapper. I would not advise that you just say, 'Here's a way to interact with Claude, I'm going to prompt Claude a little bit or build a little UI around Claude.' That doesn't have a moat, and you shouldn't be worried about Anthropic in particular eating that revenue—anyone can eat that revenue. It's not super valuable. But what I would say is that in different fields, there are different kinds of moats where you can do something that would be difficult for Anthropic to do, and we don't want to specialize in it. For example, there's a lot of stuff in the bio-cross-AI space that builds on our API. They want to do biological discovery.

我恰好是生物学家，但 Anthropic 大多数人不是生物学家。他们是 AI 科学家、产品人员或市场人员。所以我们涉足那个领域并做所有工作是非常低效的。同样适用于金融服务业，那里有大量监管，你需要了解很多东西才能合规。我们做那些事没有意义。但有些事情确实适合我们做。我们不会承诺永远不构建第一方产品。例如，Anthropic 很多人写代码，我们做了一个内部工具叫 Claude Code。因为我们自己写代码，所以对于如何最好地使用 AI 模型来写代码有独特见解。所以在代码领域，我们已经成为非常强的竞争者，因为这是我们自己用的东西。但我不认为这能推广到所有行业。

Like I happen to be a biologist, but most people at Anthropic aren't biologists. They're AI scientists, product people, or go-to-market people. So it's really inefficient for us to step into that space and do all that work. The same would apply to dealing with the financial services industry, where there's huge regulation and you need to know a bunch of stuff to comply. It just doesn't make sense for us to do that. Now there are some things that do make sense for us to do. We're not going to promise never to build first-party products. For example, a bunch of people at Anthropic write code, and we made this internal tool called Claude Code. Because we ourselves write code, we have a special and unique insight into how to best use AI models to write code. So in the code space, we've become very strong competitors because this is something we use ourselves. But I don't think that generalizes to every possible industry.

再回到我的听众，也就是印度 20 或 25 岁的年轻人。你认为哪个行业会被颠覆，哪个还有一定的跑道？我是从这样的角度问的：我在想读什么书、上什么大学、学什么技能。如果我现在创业，什么有顺风？短期也可以。

Again going back to my audience which is the 20 or 25 year old boy or girl in India. What industry do you think will get disrupted and what has a certain runway left? I'm asking from the lens of I'm trying to figure out what book to read, which college to go to, what skill set to learn. If I'm starting a startup today, what has some kind of a tailwind? For a short period of time is okay as well.

我会考虑以人为中心的任务。涉及与人交往的任务。我认为像代码和软件工程这类东西正变得越来越以 AI 为中心，数学和科学也是如此。

I would think about tasks that are human-centered. Tasks that involve relating to people. I think that stuff like code and software engineering is becoming more and more AI-focused, things like math and science.

那是编码还是工程？如果我把编码和工程完全分开来看。

Is that coding or engineering? If I were to segregate coding and engineering to be two completely different things.

是的。我认为编码首先会消失，或者说编码先由 AI 模型完成，然后更广泛的软件工程任务会需要更长时间。但我认为端到端地完成也会发生。然而，设计、做出对用户有用的东西、了解需求或管理 AI 模型团队这些元素可能仍然存在。比较优势非常强大。即使你只做任务的 5%，那 5% 也会被超级放大和杠杆化，因为 AI 做另外的 95%。所以你变得 20 倍更高效。到某个点你达到 99%，然后就会更难。但在比较优势的区域里，有非常多东西。我真的会考虑以人为中心的事情。我认为那有道理。我认为物理世界，或者将人为中心和物理世界混合在一起的东西，以及以某种方式将它们联系起来的分析技能。类似于我给出的放射科医生的例子。

Yeah. I think coding is going away first, or coding is being done by the AI models first, and then the broader task of software engineering will take longer. But I think that doing that end-to-end is going to happen as well. However, elements like design, making something useful to users, knowing what the demand is, or managing teams of AI models may still be present. There's this comparative advantage that is surprisingly powerful. Even if you're only doing 5% of the task, that 5% gets super amplified and leveraged because the AI does the other 95%. So you become 20 times more productive. At some point you get to 99% and then it becomes harder. But there's surprisingly much in that zone of comparative advantage. I would really think about things that are human-centered. I think there's something to that. I think there's something to the physical world, or things that mix together human-centered and the physical world, and analytical skills that somehow tie them together. Similar to the radiologist example I gave.

那么我该学什么？假设我是一个实际案例，我 25 岁，想为自己选一个职业。我想要某种顺风。我的目标是在未来十年取得资本主义式的成功。除了有物理接口的东西，我会选哪个行业？

So what would I study? Say I'm an actual use case, I'm 25 years old, I'm trying to pick a profession for myself. I want some kind of tailwind. My outcome is a capitalistic win in the next decade. What industry would I pick outside of something which has a physical interface?

同样，任何建立在 AI 之上的东西。如果 AI 是顺风，如果你能成为供应链其他部分的一部分，比如半导体领域，它有物理世界和更传统工程（不是软件工程）的元素。非常以人为中心的职业是我会考虑的。另一件我总说的事是：在一个 AI 可以生成和创造任何东西的世界里，拥有基本的批判性思维能力可能是成功最重要的因素。我担心这些生成图像和视频的 AI 模型——我们出于很多原因不制造生成图像和视频的模型，但这是原因之一。很难分辨什么是真实的。成功的一个重要部分可能是拥有不被愚弄的街头智慧。希望我们能打击和监管一些虚假内容，但假设我们不能。批判性思维能力将非常重要。你不想相信虚假的东西。你不想有错误信念。你不想被骗。这是我真正会给别人的建议。

Again, anything where you're building on AI. If AI is the tailwind, if you can be part of some other part of the supply chain, something in the semiconductor space, which has an element of the physical world and more traditional engineering, not software engineering. The very human-centered professions is something I would think in terms of. And the other thing I always say is: in a world where AI can generate anything and create anything, having basic critical thinking skills may be the most important thing to success. I worry about these AI models that generate images and videos—we don't make models that generate images and videos for many reasons, but this is one of them. It's really hard to tell what's real from what's not. A significant part of success may be having the street smarts not to get fooled. Hopefully we can crack down on and regulate some of this fake content, but assume we can't. Critical thinking skills are going to be really important. You don't want to fall for things that are fake. You don't want to have false beliefs. You don't want to get scammed. That's really advice I would give to someone.

如果人类历史上每一项创新都杀死了一项核心人类技能——我给你举个例子。如果计算器杀死了我们做算术的能力，如果书写减少了人类的记忆力本身，那么 AI 在杀死什么肌肉？

If every innovation in the history of humanity killed a core human skill—I'll give you an example. If calculators killed our ability to do arithmetic, if writing reduced the memory of human beings per se, what muscle is AI killing?

首先，我不太确定。我仍然经常心算。我发现即使没有计算器，心算也很有用，因为它更融入我的思维过程。我可能想说：‘哦，如果每个用户付这么多钱，那么收入就是那么多。’我希望能在脑子里闭环，而不必把答案给计算器。所以我认为很多这些技能仍然相当相关。但我要说，如果你不小心使用，你可能会失去重要技能。我想我们开始在学生身上看到了，他们让 AI 写论文——这基本上就是在作业上作弊。所以我们不应该那样做。

First of all, I'm not so sure. I still do math in my head quite a lot. I still find it useful to do math in my head even without a calculator because it's more integrated into my thought processes. I might want to say, 'Oh yeah, if each user paid this amount, then the revenue would be that.' I want to be able to close that loop in my head without having to give the answer to a calculator. So I think a lot of these skills are still pretty relevant. But I would say that if you don't use things carefully, you can lose important skills. I think we started to see it with students where they have the AI write the essay—it's basically just cheating on homework. So we shouldn't do that.

我们做了一些关于代码的研究，发现根据使用模型的方式不同，我们在编写代码方面可能会看到技能退化。使用模型有不同的方式，有些不会导致技能退化，有些则会。但可以肯定的是，如果人们在使用时不加思考，那么技能退化绝对会发生。

You know, we did some studies around code and showed that, depending on how you use the model, we can see deskilling in terms of writing code. There are different ways to use the model and some of them don't cause deskilling and some of them do. But definitely if folks are not thoughtful in how they use things, then deskilling absolutely can happen.

你认为人类作为一个种族在未来十年会变得更愚蠢吗？因为我们在某种程度上把思考和认知外包给了系统。

Do you think humans will become stupider as a race in the next decade? Because if we are in a way exporting thinking and cognition to systems.

是的。我认为如果我们以错误的方式部署 AI，如果部署得粗心大意，那么是的，人们可能会变得更愚蠢。即使 AI 在某些事情上总是比你强，你仍然可以学习那件事，对吧？你仍然可以在智力上丰富自己。所以这是我们作为个体公司、个体个人以及整个社会必须做出的选择。

Yeah. I think if we deploy AI in the wrong way, if we deploy it carelessly, then yes, people could become stupider. Even if an AI is always going to be better than you at some thing, you can still learn that thing, right? You can still enrich yourself intellectually. And so that's a choice we have to make as individual companies, as individual people, and as society overall.

Dario，你对开源与闭源有什么看法？我关注了一些公司，比如 ZAIS、GLM5 或 DeepSeek。如果你在知识产权创造和研究上投入了这么多钱，而这些家伙能够通过逆向提示和工程手段接近 Anthropic 级别的答案——我不是说 100%，但我看到 GLM 的数据，似乎相当不错。AI 世界的知识产权价值在哪里？如果我要构建一个应用，我能假设——这是一个牵强的推论——最终 AI 模型层会变得如此民主化，以至于我在构建智能体或应用层时应该每次都选择开源，因为这有助于我保留可能正在使用的收入模式吗？

Dario, do you have a view on open-sourced versus closed? I was looking at some companies like ZAIS, GLM5 or DeepSeek. If you spend all this money on IP creation, on research, if these guys are able to reverse prompt and engineer and get close to Anthropic level answers, I'm not saying 100% but I was seeing the GLM numbers and they seemed quite good. Where does the IP value in the world of AI lie? And if I were to be building an application, can I make the assumption, it's a far-fetched extrapolation, but can I assume that eventually the AI model layers will get so democratized that I should pick open-source every time when I'm building an agent or an application layer because that helps me retain the revenue model that I might be working with.

这里有几个问题。首先，很多模型，尤其是来自中国的那些，是针对基准测试优化的，并且是从美国大实验室蒸馏出来的。最近有一个测试，一些模型在常规软件工程基准测试中得分很高，但当有人做了一个未公开的保留基准测试时，这些模型的表现就差了很多。我认为这些模型更多是为基准测试而非实际使用优化的。但有一个更广泛的观点：模型的经济学与以往任何技术都截然不同。我们发现对质量有非常强烈的偏好。这有点像人类员工。如果我说你可以雇佣世界上最好的程序员或第 10000 好的程序员，他们都很熟练，但任何雇佣过大量人的人都有这种直觉：能力存在幂律长尾分布。我们在模型中也发现了同样的情况：在一定范围内，如果模型是最好的、认知能力最强的，价格就不那么重要了。价格不重要，呈现形式也不重要。所以我几乎完全专注于拥有最聪明、最适合任务的模型。我的观点是，这才是唯一重要的。

So there are a few things here. One is a lot of these models, particularly the ones that come from China, are optimized for benchmarks and are distilled from the big US labs. There was a test recently where some of these models scored very highly on the usual software engineering benchmarks, but then when someone made a held-back benchmark that had not been publicly measured, the models did a lot worse on that. I think those models are optimized for benchmarks much more than for real-world use. But I think there's a broader point: the economics of the models are very different from any previous technology. We find that there is a very strong preference for quality. It's a bit like human employees. If I said you could hire the best programmer in the world or the 10,000th best programmer in the world, they're both very skilled, but anyone who's hired a large number of people has this intuition that there's a power law long-tail distribution of ability. We find the same thing in the models: within a range, price doesn't matter that much if a model is the best, the most cognitively capable model. Price doesn't matter much. The forum in which it's presented doesn't matter much. So I'm focused almost entirely on having the smartest model and the best model for the task. My view is that's the only thing that matters.

长期地缘政治：如果 Anthropic 是一家餐厅，我会说原材料，在这个特定情况下是蔬菜，就是数据。你认为长期来看，世界会走向每个国家拥有自己的数据，你必须开始为烹饪用的蔬菜支付更多费用的局面吗？

Long-term geopolitics: if Anthropic were a restaurant, I would say the raw ingredients, the vegetables in this particular case, is data. Do you think long-term the world moves to a place where every country owns its data and you have to start paying more for the vegetables you use to cook?

是的。我认为有几件事。我确实认为全球会有建设数据中心的需求，我们非常支持这一点。数据变得有点有趣，因为我们今天使用的很多数据是我们训练的强化学习环境。例如，当你训练数学或智能体编码环境时，你并不是真正在获取数据；你是在模型中获得一些数学问题，尝试解决数学问题的实验。

Yeah. I think there are a few things. I do think there will be demand to build data centers around the world and we're very supportive of that. Data is getting kind of interesting because a lot of the data we use today is RL environments that we train on. For example, when you train on math or agentic coding environments, you're not really getting data; you're getting some math problems in the model, experiments with trying the math problems.

这更像是合成的；你在创造数据。

It's more synthetic; you're creating the data.

是的，你可以把它看作合成数据，或者看作环境中的试错。所以我认为静态数据正变得不那么重要，而模型为强化学习自行创建的动态数据正变得越来越重要。所以我不认为数据还是最核心的东西，但它仍然重要。就这一点而言，很多数据都可以在开放网络上获取。不过，如果你试图获取特定语言的数据，针对某些语言进行优化，那可能很重要。如果数据是指客户给你的数据，比如你为其他公司处理数据，那么各国——欧洲已经这样做了——会通过法律，规定这类客户个人专有数据必须留在国境内。这就是为什么要在世界各地不同国家建设和运营数据中心，并在这些国家进行推理的原因之一。

Yeah, you can think of it as synthetic data or you can think of it as trial and error in an environment. So I think data is becoming static data is becoming less important and what we might call dynamic data that the model creates itself for reinforcement learning is becoming more important. So I don't think data is quite the most central thing anymore, but it still matters. To the extent that that is the case, a lot of the data is just available on the open web. Although, if you're trying to get data in certain languages, optimized for certain languages, that can be important. If data means the data given to you by customers, like you process data for some other company, then countries will—and in the case of Europe already have—passed laws that say that kind of customer personal proprietary data needs to stay within the boundaries of the country. That's one reason to build and operate data centers around the world at different countries and to keep the models performing inference in those countries.

我确实在这个问题上追问过埃隆。他对此持怀疑态度，但我让他选一只不是他自己的股票来投资。他说是谷歌。我要问你同样的问题，我知道你也会怀疑。如果 Dario 今天有一百美元，你必须做出二元选择，投资一只股票在资本主义中获胜，你会选哪只？

I really pushed Elon on this particular question. He was skeptical of answering it but I asked him to pick one stock he would put money in which is not his own. And he said Google. I'm going to ask you the question and I know you're going to be skeptical as well. If Dario had a hundred dollars today and you had to make the binary decision of investing in a stock to win in capitalism, which stock would you pick?

是的，我最好不回答这个问题，因为我对很多上市公司了解太多。我想我最好不回答这个问题。

Yeah, I had better not answer that question because I know so much about so many public companies. I think I better not answer that question.

也许回答一个你不涉足的行业的问题，我猜今天这种情况很少，因为你涉足大多数行业。

Maybe answer the question for an industry that you're not involved in, which I'm guessing today is seldom the case because you're involved in most industries.

是的。所以，这真的……我的意思是，我不知道。我对生物技术持乐观态度。我认为生物技术即将迎来复兴。最终，它将由 AI 驱动。

Yeah. So, it's really... I mean, I don't know. I'm positive on biotech. I think biotech is about to have a renaissance. Ultimately, it will be driven by AI.

你能给我一个我应该关注的生物技术子集吗？

Can you give me a subset of biotech that I should focus on?

我认为那些更具可编程性和适应性的东西，比如 mRNA 疫苗，尽管在美国因为愚蠢的原因遇到麻烦，但我对基于肽的疗法技术非常乐观。对于小分子药物，自由度有限，改善一方面，另一方面就会变差。但肽具有几乎数字化的特性，你可以在这里替换这个氨基酸，在那里替换那个氨基酸，从而实现更连续的优化。所以我对这些领域持乐观态度，可能还有细胞疗法，比如 CAR-T 疗法，从体内取出细胞，进行基因工程改造以攻击特定癌症，然后再放回体内。

I think this idea of stuff that's more programmable and adaptive, from the mRNA vaccines, although those are having trouble in the US for dumb reasons, but I'm very optimistic about the technology to the peptide based therapies. With a small molecule drug, there are only so many degrees of freedom; you make one thing better, the other gets worse. But peptides have this almost digital property where you can substitute in this amino acid here and that amino acid there, allowing for more continuous optimization. So I would be optimistic about those kinds of areas, maybe also cell-based therapies like CAR-T therapy, where you take cells out of your body, genetically engineer them to attack a particular cancer, and put them back in.

干细胞疗法有效吗？我上周整个星期都在做这个。我每天在医院待三个小时，通过雾化器和干细胞静脉注射。我对干细胞疗法的最新进展不太了解。你得问一个正在实践的生物学家。

Do stem cell therapies work? I spent the whole of last week doing this. I was at a hospital for 3 hours a day getting nebulizer and stem cells into my veins. I am not up on the latest of stem cell therapies. You'd have to ask a currently practicing biologist.

但我认为肽会爆发。设计空间非常广阔。

But peptides I think will blow up. The design space is very broad.

当我第一次尝试使用 Claude Code 时，我确实很难让它工作。对于一个非常笨且没有编程知识的人来说，这并不容易。我认为有一个学习曲线。我听有人说得好：即使提示工程也像弹钢琴。你不能坐下来就开始弹。对于我的观众来说，我认为学习如何设置上下文、如何提示、如何更好地使用 Claude Code 变得越来越重要，尤其是对于像我这样零基础的人。你能推荐如何做到这一点吗？

When I tried to use Claude Code for the first time, I did struggle to get it to work. It's not very easy for someone who's very stupid and has no coding or programming knowledge. I think there's a learning curve. I heard someone say it well: even prompt engineering is like playing a piano. You can't sit and start playing it. To my audience, I think it becomes increasingly relevant to learn how to set context, how to prompt, how to use Claude Code better for somebody like me who comes with zero knowledge. Can you recommend how one does that?

首先，我们正在努力让学习曲线变得更平缓。我们发布 Clio（基本上是面向非程序员的 Claude Code）的原因之一，是我们注意到很多非技术人员非常想使用 Claude Code，但在命令行终端中挣扎。程序员一直使用命令行终端，但对非技术人员来说，这只会让事情变得不必要的复杂。所以 Clio 的设计更加用户友好，后台由 Claude Code 引擎驱动，但目的是让它更容易使用。我们确实在尝试引入更易用的界面。但我也想说，你可以参加一些课程来学习。我认为这是一门非常经验性的科学；你主要通过实践来学习。Anthropic 有一个我们称之为教育部的部门。我们会越来越多地发布关于如何运行有效智能体以及如何提示模型的视频。我们已经做了一些，而且会加大力度，因为我们确实希望每个人都能学会这个。

First of all, we're trying increasingly to make that learning curve easier. One of the things that caused us to release Clio, which is basically Claude Code for non-coders, is that we noticed a bunch of non-technical people who really wanted to use Claude Code and were struggling through the command line terminal. Coders use the command line terminal all the time, but for non-coders it's just unnecessarily complicated. So Clio was designed to be more user-friendly, powered by the Claude Code engine on the back, but the idea was to make it easier to use. We're definitely trying to introduce interfaces that make it easier. But I would also say there are classes you can take that help you learn this. I think it's a very empirical science; you mostly learn by doing. Anthropic has a part of the company we call the Ministry of Education. Increasingly, we'll put out videos on how to run effective agents and how to prompt models. We've already done some of that and we're going to ramp it up because we do want everyone to be able to learn this.

有什么一闪而过的想法吗？最后一个问题。你想给我们留下一些我们应该记住的东西。Dario 知道而 Nikil 和 Nikil 的所有人都不知道的事情是什么？

Any fleeting thought? Last question. Like, you want to leave us with something that we should bear in mind. What does Dario know that Nikil and all of Nikil's people do not?

我不知道我知道那么多事情，尤其是现在这项技术的影响已经显现。我世界观的许多方面都可以从目前公开可见的东西中推导出来。但我想说的是，这是我在过去 10 年里反复经历的一种体验：人们倾向于相信‘哦，那不可能发生。那太奇怪了。那变化太大了。’一次又一次，仅仅通过外推简单的曲线或试图推理出会发生什么，就会得出这些几乎没人相信的反直觉结论。这几乎就像你可以免费预测未来，只要说‘嗯，按理说……’你需要一些经验知识和直觉；你不能仅凭纯粹的逻辑推理。我认为这是人们犯的另一种错误。但将一些经验观察与第一性原理思考正确结合，就能以公开可用且任何人都应该能做到的方式预测未来，但这种情况却出奇地罕见。

I don't know that I know that many things, particularly now that the implications of the technology are kind of out there. Most aspects of my worldview can be derived from what's publicly visible now. But the thing I would say, and it's an experience I've had over and over again over the last 10 years, is there's this temptation to believe, 'Oh, that can't happen. It would be too weird. It would be too big a change.' Over and over again, just extrapolating the simple curve or trying to reason out what will happen leads you to these counterintuitive conclusions that almost no one believes. It's almost like you can predict the future for free just by saying, 'Well, it stands to reason that...' You need some empirical knowledge and some intuition; you can't reason from pure logic. I think that's another type of mistake I see people make. But the right combination of a few empirical observations with thinking from first principles can allow you to predict the future in ways that are publicly available and anyone should be able to do, but that happen surprisingly rarely.

谢谢 Dario 接受采访，希望很快能再次见到你。

Thank you Dario for doing this and hope to see you again soon.

谢谢。

Thank you.