章节 1：引言与计算的边界

📝 本节摘要：

本章作为访谈的开篇，首先通过一段引人深思的预告（Teaser）奠定了对话的基调：Demis Hassabis 提出了“50% 扩展（Scaling）+ 50% 创新”的 AGI 技术路径，并探讨了将 AGI 作为人类思维模拟器的哲学意义，以及图灵机（Turing Machine）的计算极限。随后，主持人 Hannah Fry 正式开启访谈，回顾了过去一年 AI 领域的剧变——从大语言模型向智能体 AI（Agentic AI）的重心转移。Demis 感叹过去一年的进展密度堪比十年，特别提到了 Gemini 3 的发布以及他对“世界模型”（World Models）的兴奋。

[原文] [Demis Hassabis]: Effectively, you can think of as 50% of our effort is on scaling, 50% of it is on innovation. My betting is you're going to need both to get to AGI. I've always felt this, that if we build AGI, and then use that as a simulation of the mind, and then compare that to the real mind, we will then see what the differences are and potentially what's special and remaining about the human mind. Maybe that's creativity. Maybe it's emotions. Maybe it's dreaming, consciousness. There's a lot of hypotheses out there about what may or may not be computable. And this comes back to the Turing machine question of, what is the limit of a Turing machine?

[译文] [Demis Hassabis]: 实际上，你可以认为我们 50% 的精力放在扩展（Scaling）上，50% 的精力放在创新上。我敢打赌，要实现 AGI（通用人工智能），这两者缺一不可。我一直有这样的感觉：如果我们构建出 AGI，然后将其用作思维的模拟，再将其与真实的人类思维进行对比，我们将看到两者之间的差异，并可能发现人类思维中保留下来的、独特的特质是什么。也许是创造力。也许是情感。也许是做梦，或者是意识。关于哪些东西是可计算的、哪些不是，存在很多假说。这又回到了图灵机（Turing Machine）的问题：图灵机的极限究竟在哪里？

[原文] [Hannah Fry]: So there's nothing that cannot be done within these computational--

[译文] [Hannah Fry]: 这么说，没有什么是在这些计算——

[原文] [Demis Hassabis]: Well, put it this way. Nobody's found anything in the universe that's non-computable, so far.

[译文] [Demis Hassabis]: 嗯，这么说吧。到目前为止，还没有人在宇宙中发现任何不可计算的东西。

[原文] [Hannah Fry]: So far. [THEME MUSIC] Welcome to "Google DeepMind: The Podcast" with me, Professor Hannah Fry. It has been an extraordinary year for AI. We have seen the center of gravity shift from large language models to agentic AI. We've seen AI accelerate drug discovery and multimodal models integrated into robotics and driverless cars. Now, these are all topics that we've explored in detail on this podcast. But for the final episode of this year, we wanted to take a broader view, something beyond the headlines and product launches, to consider a much bigger question. Where is all this heading, really? What are the scientific and technological questions that will define the next phase? And someone who spends quite a lot of their time thinking about that is Demis Hassabis, CEO and co-founder of Google DeepMind. Welcome back to the podcast, Demis.

[译文] [Hannah Fry]: 到目前为止。[主题音乐] 欢迎收听《Google DeepMind：播客》，我是汉娜·弗莱（Hannah Fry）教授。对于 AI 来说，这是非凡的一年。我们目睹了重心从大语言模型（Large Language Models）向智能体 AI（Agentic AI）转移。我们看到了 AI 加速药物研发，以及多模态模型被整合进机器人和无人驾驶汽车中。这些都是我们在本播客中详细探讨过的主题。但在今年的最后一期节目中，我们想把视野放得更宽一些，超越新闻头条和产品发布，去思考一个更宏大的问题。这一切究竟将走向何方？定义下一阶段的科学和技术问题会是什么？有一位花了很多时间思考这些问题的人，他就是 Demis Hassabis，Google DeepMind 的首席执行官兼联合创始人。欢迎回到播客，Demis。

[原文] [Demis Hassabis]: Great to be back.

[译文] [Demis Hassabis]: 很高兴回来。

[原文] [Hannah Fry]: I mean, quite a lot has happened in the last year.

[译文] [Hannah Fry]: 我是说，过去一年里确实发生了很多事。

[原文] [Demis Hassabis]: It has.

[译文] [Demis Hassabis]: 确实如此。

[原文] [Hannah Fry]: What is the biggest shift, do you think?

[译文] [Hannah Fry]: 你认为最大的转变是什么？

[原文] [Demis Hassabis]: Oh, wow. I mean, it's just so much has happened, as you said. It feels like we packed in 10 years in one year. I think a lot's happened. I mean, certainly, for us, the progress of the models-- we've just released Gemini 3, which we're really happy with-- the multi-modal capabilities, all of those things have just advanced really well. And then probably the thing, I guess, over the summer that I'm very excited about is world models being advanced. I'm sure we're going to talk about that.

[译文] [Demis Hassabis]: 噢，哇。我是说，正如你所说，发生的事情太多了。感觉我们在一年里塞进了十年的进展。我想发生了很多事。对我来说，当然是我们模型的进步——我们刚刚发布了 Gemini 3，我们要对它非常满意——多模态能力，所有这些方面都进展得非常好。然后我想，整个夏天最让我兴奋的事情大概就是世界模型（World Models）的进步了。我相信我们一会儿会谈到这个。

[原文] [Hannah Fry]: Yeah, absolutely. We will get on to all of that stuff in a little bit more detail in a moment.

[译文] [Hannah Fry]: 是的，当然。我们稍后会详细讨论所有这些内容。

章节 2：破解“根节点”问题：科学、聚变与量子计算

📝 本节摘要：

在本章中，主持人 Hannah Fry 回顾了 Demis 曾提出的“根节点”（Root Node）理论——即利用 AI 解锁基础科学瓶颈，从而带来广泛的下游收益。Demis 确认了 AlphaFold 作为该理论的首个力证，并分享了当前正在攻克的其他领域：材料科学（如室温超导和电池技术）、通过与 Commonwealth Fusion Systems 深度合作来加速核聚变（控制等离子体），以及利用 AI 辅助 Google 量子计算团队进行纠错。Demis 描绘了一个由无限清洁能源驱动的未来，这将解决气候危机、水资源短缺（海水淡化）甚至太空燃料制备等一系列全球性挑战。

[原文] [Hannah Fry]: I remember the very first time I interviewed you for this podcast, and you were talking about the root node problems, about this idea that you can use AI to unlock these downstream benefits. And you've made pretty good on your promise, I have to say.

[译文] [Hannah Fry]: 我记得我第一次在这个播客采访你的时候，你谈到了“根节点”问题（root node problems），关于利用 AI 来解锁那些下游收益的想法。不得不说，你很好地兑现了你的承诺。

[原文] [Demis Hassabis]: Yes.

[译文] [Demis Hassabis]: 是的。

[原文] [Hannah Fry]: Do you want to give us an update on where we are with those? What are the things that are just around the corner and the things that you've sort of solved or near solved?

[译文] [Hannah Fry]: 你想给我们介绍一下目前的最新进展吗？有哪些事情是即将实现的，又有哪些是已经解决或接近解决的？

[原文] [Demis Hassabis]: Yeah. Well, of course, the big proof point was AlphaFold. And it's crazy to think we're coming up to five-year anniversary of AlphaFold being announced to the world-- AlphaFold2, at least. So that was the proof, I guess, that it was possible to do these root node type of problems.

[译文] [Demis Hassabis]: 好的。当然，最大的证明点是 AlphaFold。想到我们即将迎来 AlphaFold——至少是 AlphaFold2——向世界发布五周年的纪念日，这真是太疯狂了。所以我想，那就是证明，证明解决这些“根节点”类型的问题是可能的。

[原文] [Demis Hassabis]: And we're exploring all the other ones now. I think material science. I'd love to do a room temperature superconductor. And better batteries, these kinds of things-- I think that's on the cards, better materials of all sorts. We're also working on fusion.

[译文] [Demis Hassabis]: 我们现在正在探索所有其他的领域。我认为是材料科学。我很想做出室温超导体。还有更好的电池，诸如此类的东西——我认为这已经在计划中了，各种更好的材料。我们也正在致力于核聚变。

[原文] [Hannah Fry]: Because there's a new partnership that's been announced with fusion, right?

[译文] [Hannah Fry]: 因为刚刚宣布了一个关于核聚变的新合作伙伴关系，对吧？

[原文] [Demis Hassabis]: Yeah. We've just announced a partnership with a deep one. We already were collaborating with them, but it's a much deeper one now with Commonwealth Fusion, who I think are probably the best startup working on at least traditional tokamak reactors. So they're probably closest to having something viable. And we want to help accelerate that, helping them contain the plasma in the magnets and maybe even some material design there, as well. So that's exciting.

[译文] [Demis Hassabis]: 是的。我们刚刚宣布了一个深度的合作伙伴关系。我们之前就已经在与他们合作，但现在与 Commonwealth Fusion 的合作要深入得多，我认为他们可能是致力于至少传统托卡马克（tokamak）反应堆的最好的初创公司。所以他们可能是最接近拥有可行方案的。我们要帮助加速这一进程，帮助他们在磁体中控制等离子体，甚至可能在材料设计方面也提供帮助。所以这很令人兴奋。

[原文] [Demis Hassabis]: And then we're collaborating also with our quantum colleagues, which they're doing amazing work at the quantum AI team at Google. And we're helping them with error correction codes, where we're using our machine learning to help them. And then maybe one day they'll help us. [LAUGHS]

[译文] [Demis Hassabis]: 此外，我们也在与我们的量子计算同事合作，Google 的量子 AI 团队正在做惊人的工作。我们在纠错码（error correction codes）方面帮助他们，利用我们的机器学习来协助他们。也许有一天，他们也会反过来帮助我们。[笑]

[原文] [Hannah Fry]: That perfect cycle.

[译文] [Hannah Fry]: 完美的循环。

[原文] [Demis Hassabis]: Yes, exactly.

[译文] [Demis Hassabis]: 是的，没错。

[原文] [Hannah Fry]: The fusion one is particularly-- I mean, the difference that that would make to the world, that would be unlocked by that, is gigantic.

[译文] [Hannah Fry]: 核聚变那个项目特别——我是说，它将给世界带来的改变，以及由此解锁的潜力，是巨大的。

[原文] [Demis Hassabis]: Yeah. I mean, fusion has always been the holy grail. Of course, I think solar is very promising, too, effectively using the fusion reactor in the clouds and in the sky. But I think if we could have modular fusion reactors, this promise of unlimited, renewable, clean energy would be-- obviously, transform everything. And that's the holy grail. And, of course, that's one of the ways we could help with climate.

[译文] [Demis Hassabis]: 是的。我是说，核聚变一直被视为“圣杯”。当然，我认为太阳能也非常有前景，实际上就是利用云端之上、天空中的那个聚变反应堆（太阳）。但我认为，如果我们能拥有模块化的聚变反应堆，这种无限、可再生、清洁能源的承诺将会——显然，改变一切。这就是“圣杯”。当然，这也是我们能帮助解决气候问题的途径之一。

[原文] [Hannah Fry]: It does make a lot of our existing problems sort of disappear if we can [INAUDIBLE].

[译文] [Hannah Fry]: 如果我们能[听不清/做到这一点]，它确实会让许多现存的问题某种程度上消失。

[原文] [Demis Hassabis]: Definitely. I mean, it opens up many-- this is why we think of it as a root node. Of course, it helps directly with energy and pollution and so on and helps with the climate crisis. But also, if energy really was renewable and clean and super cheap, almost free, then many other things would become viable, like water access because we could have desalination plants pretty much everywhere, even making rocket fuel.

[译文] [Demis Hassabis]: 绝对的。我是说，它开启了许多——这就是为什么我们将它视为一个“根节点”。当然，它直接有助于解决能源和污染等问题，有助于缓解气候危机。但同时，如果能源真的变得可再生、清洁且超级便宜，几乎免费，那么许多其他事情就会变得可行，比如水资源获取，因为我们几乎可以在任何地方通过海水淡化厂（desalination plants）获取淡水，甚至制造火箭燃料。

[原文] [Demis Hassabis]: There's lots of seawater that contains hydrogen and oxygen, and that's basically rocket fuel. But it just takes a lot of energy to split it out into hydrogen and oxygen. But if energy is cheap and renewable and clean, then why not do that? You could have that producing 24/7.

[译文] [Demis Hassabis]: 海水里含有大量的氢和氧，那基本上就是火箭燃料。只是将其分离成氢气和氧气需要消耗大量的能量。但如果能源是便宜、可再生且清洁的，为什么不这样做呢？你可以让它全天候 24/7 生产。

章节 3：智能的悖论：数学金牌与常识谬误

📝 本节摘要：

这一章节深入探讨了当前 AI 发展中一个极具讽刺意味的现象：模型一方面能在国际数学奥林匹克（IMO）竞赛中斩获金牌，表现出顶尖智力；另一方面却可能在简单的高中数学或逻辑问题上犯低级错误。Demis 将这种状态形象地称为“参差不齐的智能”（Jagged Intelligence）。他剖析了背后的原因，包括底层的“Token 化”机制导致模型无法精准感知字符（如数不清单词里的字母），以及系统缺乏连贯的“推理与思考”能力。他强调，真正的 AGI 需要在推理阶段（Inference Time）学会像人类一样“停下来思考”并自我校对，而目前我们可能只走完了一半的路程。

[原文] [Hannah Fry]: You're also seeing a lot of change in the AI that is applying itself to mathematics-- winning medals in the International Maths Olympiad. And yet, at the same time, these models can make quite basic mistakes in high school math. Why is there that paradox?,

[译文] [Hannah Fry]: 你也看到了 AI 在数学应用方面发生的巨大变化——赢得了国际数学奥林匹克（IMO）的金牌。然而，与此同时，这些模型却可能在高中数学题上犯相当基础的错误。为什么会存在这种悖论？

[原文] [Demis Hassabis]: Yeah. I think it's fascinating, actually, one of the most fascinating things, and probably that needs to be fixed as one of the key things why we're not at AGI yet. As you said, we've had a lot of success in other groups on getting gold medals at the International Maths Olympiad. You look at those questions, and they're super hard questions that only the top students in the world can do.,

[译文] [Demis Hassabis]: 是的。实际上我认为这非常迷人，是最迷人的事情之一，而且这可能是我们需要解决的关键问题之一，也是为什么我们还没达到 AGI（通用人工智能）的原因。正如你所说，我们在其他团队取得了很大成功，在国际数学奥林匹克上拿到了金牌。你看看那些题目，那是超级难的问题，只有世界上最顶尖的学生才能做出来。

[原文] [Demis Hassabis]: And, on the other hand, if you pose a question in a certain way-- we've all seen that with experimenting with chat bots ourselves in our daily lives-- that it can make some fairly trivial mistakes on logic problems. They can't really play decent games of chess yet, which is surprising. So there's something missing still from these systems in terms of their consistency. And I think that's one of the things that you would expect from a general intelligence, an AGI system, is that it would be consistent across the board.,

[译文] [Demis Hassabis]: 但另一方面，如果你换一种方式提问——我们在日常生活中尝试聊天机器人时都见过这种情况——它会在逻辑问题上犯一些相当微不足道的错误。它们甚至还不能下出像样的国际象棋，这很令人惊讶。所以在一致性（consistency）方面，这些系统仍然缺少某些东西。我认为这是一般智力、一个 AGI 系统应该具备的特征之一，就是它应该在各个方面都保持一致。

[原文] [Demis Hassabis]: And so sometimes people call it jagged intelligences. So they're really good at certain things, maybe even PhD level. But then, other things, they're not even high school level. So it's very uneven still, the performances of these systems. They're very, very impressive in certain dimensions, but they're still pretty basic in others. And we've got to close those gaps.

[译文] [Demis Hassabis]: 所以有时人们称之为“参差不齐的智能”（jagged intelligences）。它们在某些方面真的非常出色，甚至达到博士水平。但在其他方面，它们甚至达不到高中水平。所以这些系统的表现仍然非常不均衡。它们在某些维度上非常非常令人印象深刻，但在其他维度上仍然相当基础。我们要弥合这些差距。

[原文] [Demis Hassabis]: And there are theories as to why. And depending on the situation, it could even be the way that an image is perceived and tokenized. So sometimes, actually, it doesn't even get all the letters that-- so when you count letters in words, it sometimes gets that wrong. But it may not be seeing each individual letter. So there's different reasons for some of these things.,

[译文] [Demis Hassabis]: 关于原因有各种理论。根据具体情况，这甚至可能与图像被感知和 Token 化（tokenized）的方式有关。所以有时，实际上，它甚至没有获取到所有的字母——所以当你让它数单词里的字母数时，它有时会数错。但这可能是因为它并没有“看到”每一个单独的字母。所以造成这些现象的原因各不相同。

[原文] [Demis Hassabis]: And each one of those can be fixed, and then you can see what's left, but I think consistency. I think another thing is reasoning and thinking. So we have thinking systems now that, at inference time, they spend more time thinking, and they're better at outputting their answers. But it's not super consistent yet in terms of, is it using that thinking time in a useful way to actually double-check and use tools to double-check what it's outputting? I think we're on the way, but maybe we're only 50% of the way there.,

[译文] [Demis Hassabis]: 每一个问题都可以被修复，然后你再看还剩下什么问题，但我认为核心是“一致性”。我认为另一件事是推理和思考。我们现在拥有一些“思考系统”，它们在推理阶段（inference time）会花更多时间去思考，从而能输出更好的答案。但在“它是否有效地利用了这段思考时间来进行自我校对，以及使用工具来复核其输出”这方面，还不是非常一致。我认为我们正在路上，但也许我们只走了一半的路程。

章节 4：通往AGI的技术路径：Scaling与创新的平衡

📝 本节摘要：

本章深入探讨了实现通用人工智能（AGI）的具体技术路径。Hannah 询问当前的语言模型是否会像 AlphaGo 进化到 AlphaZero 那样，摆脱对人类数据的依赖。Demis 解释说，目前的 LLM 更像早期的 AlphaGo，是对互联网人类知识的压缩；未来的关键在于在模型之上叠加“搜索”与“规划”能力（System 2 Thinking），并实现“在线持续学习”。面对关于 Scaling Law（扩展定律）是否失效的质疑，Demis 予以否认，但承认存在边际收益递减。他揭示了 Google DeepMind 的核心策略：50% 的精力投入在算力扩展（Scaling），50% 投入在架构创新（Innovation），两者结合才是通往 AGI 的必经之路。

[原文] [Hannah Fry]: I also wonder about that story of AlphaGo and then AlphaZero, where you took away all of the human experience and found that the model actually improved.

[译文] [Hannah Fry]: 我也很好奇关于 AlphaGo 和随后的 AlphaZero 的故事，当时你们去掉了所有的人类经验数据，结果发现模型反而进步了。

[原文] [Demis Hassabis]: Yeah.

[译文] [Demis Hassabis]: 是的。

[原文] [Hannah Fry]: Is there a scientific or a maths version of that in the models that you're creating?

[译文] [Hannah Fry]: 在你们现在创建的模型中，是否存在某种科学版或数学版的“AlphaZero 时刻”？

[原文] [Demis Hassabis]: I think what we're trying to build today, it's more like AlphaGo. So effectively, these large language models, these foundation models, they're starting with all of human knowledge, what we put on the internet, which is pretty much everything these days, and compressing that into some useful artifact which they can look up and generalize from.

[译文] [Demis Hassabis]: 我认为我们今天试图构建的东西，更像是 AlphaGo。实际上，这些大语言模型、这些基础模型，它们是从全人类的知识开始的——也就是我们在互联网上发布的内容，这年头几乎涵盖了所有东西——然后将这些知识压缩成某种有用的伪影（artifact），供它们查阅并从中进行归纳。

[原文] [Demis Hassabis]: But I do think we're still in the early days of having this search or thinking on top, like AlphaGo had, to use that model to direct useful reasoning traces, useful planning ideas, and then come up with the best solution to whatever the problem is at that point in time.

[译文] [Demis Hassabis]: 但我确实认为，在拥有这种位于顶层的“搜索”或“思考”能力方面，我们仍处于早期阶段——就像 AlphaGo 当年拥有的那样——利用模型来引导有用的推理轨迹（reasoning traces）和规划思路，然后针对当下的问题提出最佳解决方案。

[原文] [Demis Hassabis]: So I don't feel like we're constrained at the moment with the limit of human knowledge, like the internet. I think the main issue at the moment is, we don't know how to use those systems in a reliable way fully yet in the way we did with AlphaGo. But, of course, that was a lot easier because it was a game.

[译文] [Demis Hassabis]: 所以我并不觉得我们目前受到了人类知识极限（比如互联网数据量）的制约。我认为目前的主要问题是，我们还不知道如何像当年操作 AlphaGo 那样，完全可靠地使用这些系统。当然，那是容易得多的，因为它毕竟只是一个游戏。

[原文] [Demis Hassabis]: I think once you have AlphaGo there, you could go back, just like we did with the Alpha series, and do an AlphaZero, where it starts discovering knowledge for itself. I think that would be the next step, but that's obviously harder. And so I think it's good to try and create the first step first with some kind of AlphaGo-like system. And then we can think about an AlphaZero-like system.

[译文] [Demis Hassabis]: 我认为一旦你有了“AlphaGo”，你就可以回过头来——就像我们在 Alpha 系列中所做的那样——做一个“AlphaZero”，让它开始自我发现知识。我认为那将是下一步，但这显然更难。所以我认为先尝试用某种类 AlphaGo 的系统迈出第一步是好的。然后我们再去考虑类 AlphaZero 的系统。

[原文] [Demis Hassabis]: But that is also one of the things missing from today's systems is the ability to online learn and continually learn. So we train these systems, we balance them, we post-train them, and then they're out in the world. But they don't continue to learn out in the world, like we would. And I think that's another critical missing piece from these systems that will be needed for AGI.

[译文] [Demis Hassabis]: 但这也是当今系统所缺失的东西之一，即“在线学习”（online learn）和“持续学习”的能力。我们训练这些系统，调整它们，进行后训练（post-train），然后把它们发布到世界上。但它们不会像我们要那样，在现实世界中继续学习。我认为这是这些系统通往 AGI 所需的另一个关键缺失拼图。

[原文] [Hannah Fry]: In terms of all of those missing pieces, I mean, I know that there's this big race at the moment to release commercial products, but I also know that Google DeepMind's roots really lie in that idea of scientific research. And I found a quote from you where you recently said, "If I'd had my way, we would have left AI in the lab for longer and done more things like AlphaFold, maybe cured cancer or something like that." Do you think that we lost something by not taking that slower route?

[译文] [Hannah Fry]: 谈到所有这些缺失的拼图，我知道目前大家都在疯狂地竞相发布商业产品，但我也知道 Google DeepMind 的根基其实在于科学研究。我找到了一段你最近说的话：“如果能按我的意愿来，我会把 AI 留在实验室里更久一些，做更多像 AlphaFold 这样的事情，也许先治愈癌症之类的。”你认为因为没有选择那条更慢的路，我们是否失去了什么？

[原文] [Demis Hassabis]: I think we lost and gained something. So I feel like that would have been the more pure scientific approach. At least, that was my original plan, say 15, 20 years ago... But in the meantime, you wouldn't have to wait till AGI arrived before it was useful. You could branch off that technology and use it in really beneficial ways to society, namely advancing science and medicine, so exactly what we did with AlphaFold...

[译文] [Demis Hassabis]: 我想我们既有失去也有所得。我觉得那样做确实会是更纯粹的科学路径。至少，那是我 15、20 年前的最初计划……但与此同时，你不必等到 AGI 完全实现后才让它发挥作用。你可以将技术分支出来，以真正造福社会的方式使用它，即推进科学和医学，这正是我们在 AlphaFold 上所做的……

[原文] [Demis Hassabis]: Now, it's turned out that chatbots were possible at scale, and people find them useful. And then they've now morphed into these foundation models that can do more than chat and text... And that's also been very successful commercially in terms of a product. And I love that, too.

[译文] [Demis Hassabis]: 现在的现实是，聊天机器人在大规模应用上是可能的，而且人们发现它们很有用。随后它们演变成了这些基础模型，能做的远不止聊天和处理文本……这在商业产品层面也非常成功。我也很喜欢这一点。

[原文] [Demis Hassabis]: But it has created this pretty crazy race condition where there's many commercial organizations and even nation states all rushing to improve and overtake each other. And that makes it hard to do rigorous science at the same time. We try to do both, and I think we're getting that balance right.

[译文] [Demis Hassabis]: 但这确实创造了一种相当疯狂的“竞赛状态”（race condition），许多商业机构甚至国家都在争先恐后地改进并试图超越彼此。这使得同时进行严谨的科学研究变得很困难。我们在尝试兼顾两者，我认为我们正在把握好这个平衡。

[原文] [Hannah Fry]: The thing that's strange is that-- I mean, this time last year, I think there was a lot talk about scaling, eventually hitting a wall, about us running out of data. And yet, we're recording-- now, Gemini 3 has just been released, and it's leading on this whole range of different benchmarks. How has that been possible? Wasn't there supposed to be a problem with scaling hitting a wall?

[译文] [Hannah Fry]: 奇怪的是——我是说，去年这个时候，大家都在谈论“扩展”（scaling），说最终会撞墙，说我们会耗尽数据。然而，就在我们要录制的时候——Gemini 3 刚刚发布，并且在各种不同的基准测试中处于领先地位。这是怎么做到的？不是说扩展会遇到瓶颈吗？

[原文] [Demis Hassabis]: I think a lot of people thought that, especially as other companies have had slower progress, shall we say. But I think we've never really seen any wall, as such. What I would say is maybe there's diminishing returns. And when I say that, people only think, oh, so there's no returns. It's 0 or 1. It's either exponential, or it's asymptotic. No. Actually, there's a lot of room between those two regimes.

[译文] [Demis Hassabis]: 我想很多人是这么认为的，尤其是当其他公司的进展——我们可以说是——变慢了的时候。但我认为我们从未真正看到过所谓的“墙”。我要说的是，也许存在“边际收益递减”（diminishing returns）。当我说这个词时，人们只会想，“噢，那就是没收益了。”他们觉得非 0 即 1。要么是指数级增长，要么就是渐近线式的停滞。不。实际上，在这两种状态之间还有很大的空间。

[原文] [Demis Hassabis]: And on top of that, we also have the advantage of world-class infrastructure with our TPUs and other things that we've invested in for a long time. And so that combination, I think, allows us to be at the frontier of the innovations, as well as the scaling part. And, effectively, you can think of as 50% of our effort is on scaling, 50% of it is on innovation. And my betting is you're going to need both to get to AGI.

[译文] [Demis Hassabis]: 除此之外，我们还拥有世界级基础设施的优势，比如我们的 TPU 以及我们长期投资的其他东西。所以我认为这种结合使我们能够同时处于创新和扩展的前沿。实际上，你可以认为我们 50% 的精力放在扩展（Scaling）上，50% 的精力放在创新（Innovation）上。我敢打赌，要实现 AGI，这两者缺一不可。

章节 5：解决幻觉：模型内省与置信度

📝 本节摘要：

尽管 Gemini 3 表现卓越，但“幻觉”（Hallucinations）依然存在。Hannah 提出是否能像 AlphaFold 那样为语言模型引入“置信度评分”。Demis 对此表示高度认同，他指出解决幻觉的关键在于让模型学会“内省”（Introspect）——即知道自己知道什么，也知道自己不知道什么。他解释道，底层的“下一个 Token 预测概率”并不等同于对整件事实的信心。目前的模型像是一个“状态糟糕的人”，只会脱口而出脑海中的第一个想法；未来的解决方案是利用推理和规划步骤，让模型学会“停下来思考”并自我校对。

[原文] [Hannah Fry]: I mean, one thing that we are still seeing, even in Gemini 3, which is an exceptional model, is this idea of hallucinations. So I think there was one metric that said it can still give an answer when actually it should decline.

[译文] [Hannah Fry]: 我是说，有一件事我们仍然能看到，即使是在 Gemini 3 这样卓越的模型中，那就是“幻觉”（hallucinations）的概念。我想有一个指标显示，在它实际上应该拒绝回答的时候，它仍然会给出一个答案。

[原文] [Demis Hassabis]: Yes.

[译文] [Demis Hassabis]: 是的。

[原文] [Hannah Fry]: I mean, could you build a system where Gemini gives a confidence score in the same way that AlphaFold does?

[译文] [Hannah Fry]: 我是说，你能否构建一个系统，让 Gemini 像 AlphaFold 那样给出一个置信度评分（confidence score）？

[原文] [Demis Hassabis]: Yeah, I think so. And I think we need that, actually. And I think that's one of the missing things. I think we're getting close. I think the better the models get, the more they know about what they know, if that makes sense.

[译文] [Demis Hassabis]: 是的，我认为可以。而且我认为我们确实需要这个。我觉得这是缺失的环节之一。我认为我们正在接近这个目标。模型变得越好，它们就越了解“自己知道什么”，如果这讲得通的话。

[原文] [Demis Hassabis]: And I think the more reliable-- you could rely on them to actually introspect in some way or do more thinking and actually realize for themselves that they're uncertain, or there's uncertainty over this answer. And then we've got to work out how to train it in a way where it can output that as a reasonable answer.

[译文] [Demis Hassabis]: 而且我认为更可靠的是——你可以依赖它们以某种方式进行“内省”（introspect），或者进行更多的思考，从而真正自我意识到它们是不确定的，或者这个答案存在不确定性。然后我们必须弄清楚如何训练它，使其能够将这种不确定性作为一个合理的答案输出出来。

[原文] [Demis Hassabis]: We're getting better at it. But it still sometimes-- it forces itself to answer when it probably shouldn't, and then that can lead to a hallucination. So I think a lot of the hallucinations are of that type, currently. So there's a missing piece there that has to be solved, and you're right, as we did solve it with AlphaFold, but in obviously a much more limited way.

[译文] [Demis Hassabis]: 我们在这方面正做得越来越好。但有时它仍然——它会强迫自己回答那些它可能不该回答的问题，这就可能导致幻觉。所以我认为目前很多幻觉都属于这种类型。那里有一个必须解决的缺失环节，你是对的，就像我们在 AlphaFold 上解决的那样，但这显然是在一个更受限的方式下。

[原文] [Hannah Fry]: Because presumably, behind the scenes, there is some sort of measure of probability of whatever the next token might be.

[译文] [Hannah Fry]: 因为据推测，在幕后，对于下一个 Token 是什么，应该某种概率度量。

[原文] [Demis Hassabis]: Yes, there is of the next token. That's how it all works. But that doesn't tell you the overall arching piece, is how confident are you about this entire fact or this entire statement?

[译文] [Demis Hassabis]: 是的，确实有关于下一个 Token 的概率。这就是它的运作方式。但这并不能告诉你整体层面的情况，即你对这整个事实或整个陈述有多大信心？

[原文] [Demis Hassabis]: And I think that's why you'll need this-- I think we'll need to use the thinking steps and the planning steps to go back over what you just output. At the moment, it's a little bit like the systems are just-- it's like talking to a person, and when they're on a bad day, they're just literally telling you the first thing that comes to their mind.

[译文] [Demis Hassabis]: 我认为这就是为什么你需要这个——我想我们需要利用“思考步骤”和“规划步骤”来回头检查刚刚输出的内容。目前，这些系统有点像——就像你在跟一个人说话，当他们状态糟糕（on a bad day）的时候，他们基本上就是把脑子里想到的第一件事直接告诉你。

[原文] [Demis Hassabis]: Most of the time, that will be OK. But then sometimes, when it's a very difficult thing, you'd want to stop, pause for a moment, and maybe go over what you were about to say and adjust what you were about to say. But perhaps that's happening less and less in the world these days, but that's still the better way of having a discourse. So I think you can think of it like that. These models need to do that better.

[译文] [Demis Hassabis]: 大多数时候，这样没问题。但有时，当遇到非常困难的事情时，你会想要停下来，暂停片刻，也许回顾一下你正要说的话，并调整你要表达的内容。尽管这种做法如今在这个世界上可能越来越少了，但这仍然是进行对话的更好方式。所以你可以这样理解。这些模型需要在这一点上做得更好。

章节 6：世界模型：超越语言的物理理解

📝 本节摘要：

本章聚焦于 Demis Hassabis 最长久的热情所在——“世界模型”（World Models）。Hannah 询问为何在语言模型之外还需要模拟世界，Demis 解释道，尽管语言模型包含的知识超乎想象，但它难以捕捉“空间动力学”和“物理语境”，因为许多感官体验（如运动角度、气味）难以用文字描述。他定义世界模型为一种理解世界因果机制和直觉物理学的系统。通过 Genie 和 Veo 等视频生成模型，DeepMind 正在验证这种能力——如果你能生成逼真的世界，就意味着你理解了其中的物理运作机制。这不仅是通往高级机器人和全能助手的必经之路，也能用于模拟复杂科学系统（如天气）乃至创造“终极游戏”。

[原文] [Hannah Fry]: I also really want to talk to you about the simulated worlds and putting agents in them because we got to talk to your Genie team earlier this year.

[译文] [Hannah Fry]: 我也非常想和你聊聊模拟世界（simulated worlds）以及把智能体放入其中的话题，因为今年早些时候我们和你的 Genie 团队聊过。

[原文] [Demis Hassabis]: Yes, it's awesome work.

[译文] [Demis Hassabis]: 是的，那是了不起的工作。

[原文] [Hannah Fry]: Tell me why you care about simulation. What can a world model do that a language model can't?

[译文] [Hannah Fry]: 告诉我你为什么在乎模拟。世界模型能做到哪些语言模型做不到的事情？

[原文] [Demis Hassabis]: Well, look, it's actually been-- it's probably my longest-standing passion is world models and simulations, in addition to AI, and of course it's all coming together in our most recent work, like Genie. And I think language models are able to understand a lot about the world-- I think, actually, more than we expected, more than I expected, because language is actually probably richer than we thought. It contains more about the world than even linguists maybe imagined. And that's proven now with these new systems.

[译文] [Demis Hassabis]: 嗯，听着，实际上这——除了 AI 之外，世界模型和模拟可能是我最长久的热情所在，当然，这一切都在我们要最近的工作中汇聚在了一起，比如 Genie。我认为语言模型能够理解关于世界的很多东西——实际上，我认为比我们要预期的要多，比我预期的要多，因为语言实际上可能比我们认为的要丰富得多。它包含的关于世界的信息甚至可能比语言学家想象的还要多。这在这些新系统中已经得到了证明。

[原文] [Demis Hassabis]: But there's still a lot about the spatial dynamics of the world-- spatial awareness and the physical context we're in and how that works mechanically-- that is hard to describe in words and isn't generally described in corpuses of words. And a lot of this is allied to learning from experience, online experience. There's a lot of things which you can't really describe something. You have to just experience it. Maybe the senses and so on are very hard to put into words, whether that's motor angles and smell and these kind of senses.

[译文] [Demis Hassabis]: 但是关于世界的空间动力学（spatial dynamics）——空间感知、我们要所处的物理语境以及它们在机械层面是如何运作的——仍然有很多东西是很难用语言描述的，而且通常也不会在文字语料库中被描述。这其中很多都与从经验中学习、从在线体验中学习有关。有很多事情是你无法真正通过描述来传达的。你必须去亲身体验它。也许感官之类的东西很难用语言表达，无论是电机角度（motor angles）、气味还是这类感觉。

[原文] [Demis Hassabis]: It's very difficult to describe that in any kind of language. So I think there's a whole set of things around that. And I think if we want robotics to work or a universal assistant that maybe comes along with you in your daily life, maybe on glasses or on your phone and helps you in your everyday life, not just on your computer, you're going to need this kind of world understanding, and world models are at the core of that.

[译文] [Demis Hassabis]: 很难用任何一种语言来描述这些。所以我认为围绕这一点有一整套的东西。而且我认为，如果我们想让机器人技术奏效，或者想要一个能在日常生活中陪伴你的通用助手——也许是在眼镜上或手机上，在日常生活中帮助你，而不仅仅是在电脑上——你就需要这种对世界的理解，而世界模型正是其中的核心。

[原文] [Demis Hassabis]: So what we mean by world model is this sort of model that understands the causative and effect of the mechanics of the world-- intuitive physics, but how things move, how things behave. Now, we're seeing a lot of that in our video models, actually. And one way to show, how do you test you have that kind of understanding? Well, can you generate realistic worlds? Because if you can generate it, then, in a sense, you must have understood-- the system must have encapsulated a lot of the mechanics of the world.

[译文] [Demis Hassabis]: 所以我们所说的世界模型，是指这种能够理解世界运作机制的因果关系（causative and effect）的模型——直觉物理学，比如物体如何移动，物体如何表现。现在，实际上我们在我们的视频模型中看到了很多这样的能力。有一种方法可以展示，你怎么测试你是否拥有这种理解呢？嗯，你能生成逼真的世界吗？因为如果你能生成它，那么在某种意义上，你就必须已经理解了——系统必须已经封装了许多世界的运作机制。

[原文] [Demis Hassabis]: So that's why Genie and Veo and these models, our video models and our interactive world models, are really impressive, but also important steps towards showing we have generalized world models. And then hopefully, at some point, we can apply it to robotics and universal assistants. And then, of course, one of my favorite things I'm definitely going to have to do at some point is reapplying it back to games and game simulations and create the ultimate games, which, of course, was maybe always my subconscious plan.

[译文] [Demis Hassabis]: 这就是为什么 Genie 和 Veo 以及这些模型，我们的视频模型和我们的交互式世界模型，不仅令人印象深刻，而且是向展示我们要拥有通用世界模型迈出的重要一步。然后希望在某个时候，我们可以将其应用于机器人和通用助手。当然，我也肯定会在某个时候做我最喜欢的事情之一，就是把它重新应用回游戏和游戏模拟中，创造出“终极游戏”，这当然可能一直是我潜意识里的计划。

[原文] [Hannah Fry]: All of this, just for that.

[译文] [Hannah Fry]: 做了这一切，就是为了那个。

[原文] [Demis Hassabis]: Yeah, all of this time, exactly.

[译文] [Demis Hassabis]: 是的，花了这么多时间，正是如此。

[原文] [Hannah Fry]: What about science, too, though, because you use it in that domain?

[译文] [Hannah Fry]: 那科学方面呢？因为你们也在那个领域使用它？

[原文] [Demis Hassabis]: Yes, you could. So science, again, I think, building models of scientifically complex domains, whether that's materials on the atomic level in biology, but also some physical things, as well, like weather-- one way to understand those systems is to learn simulations of those systems from the raw data.

[译文] [Demis Hassabis]: 是的，可以。所以在科学方面，同样地，我认为构建科学复杂领域的模型，无论是生物学中原子层面的材料，还是一些物理现象，比如天气——理解这些系统的一种方法就是从原始数据中学习这些系统的模拟。

[原文] [Demis Hassabis]: So you have a bunch of raw data. Let's say it's about the weather. And, obviously, we have some amazing weather projects going on. And then you have a model that learns those dynamics and can recreate those dynamics more efficiently than doing it by brute force. So I think there's huge potential for simulations and world models, maybe specialized ones, for aspects of science and mathematics.

[译文] [Demis Hassabis]: 所以你有一堆原始数据。比方说是关于天气的。显然，我们正在进行一些很棒的天气项目。然后你有一个模型，它学习了这些动态，并且能够比通过蛮力（brute force）计算更有效地重现这些动态。所以我认为模拟和世界模型，也许是专门化的模型，在科学和数学方面有着巨大的潜力。

章节 7：智能体与仿真：无限数据循环与社会演化

📝 本节摘要：

在本章中，对话进入了 AI 发展的下一个关键阶段：智能体（Agents）与模拟环境的结合。Demis 介绍了 Google DeepMind 的 SIMA 项目（Simulated Agents），并提出了一个令人兴奋的概念：将 SIMA 放入由 Genie 实时生成的虚拟世界中，形成一个“无限训练循环”。这种组合不仅意味着游戏 NPC 将彻底告别枯燥，更意味着 AI 可以通过“自我对弈”产生无穷尽的高质量训练数据。

>

随后，话题转向了模拟的准确性与哲学意义。Demis 坦承目前的视频模型在物理规律上仍存在“幻觉”，需要通过构建“物理基准测试”（如牛顿定律实验）来校准。最后，他表达了对在沙盒中重演“进化论”和社会动力学的浓厚兴趣——通过模拟数百万次的社会演化，探寻生命起源、意识诞生以及经济体系的运作规律，这呼应了他早期对圣塔菲研究所（Santa Fe Institute）复杂系统研究的热爱。

[原文] [Hannah Fry]: But then, also, I mean, you can drop an agent into that simulated world too, right?

[译文] [Hannah Fry]: 但这也是，我是说，你也可以把一个智能体（agent）扔进那个模拟世界里，对吧？

[原文] [Demis Hassabis]: Yes.

[译文] [Demis Hassabis]: 是的。

[原文] [Hannah Fry]: Your Genie 3 team, they had this really lovely quote, which was, "Almost no prerequisite to any major invention was made with that invention in mind." And they were talking about dropping agents into these simulated environments and allowing them to explore with curiosity being their main motivator.

[译文] [Hannah Fry]: 你的 Genie 3 团队有句很好的名言：“几乎没有任何重大发明的前置条件，是在发明该前置条件时就预见到了那个发明的。”他们当时谈论的是将智能体放入这些模拟环境中，并让它们以好奇心作为主要动力去探索。

[原文] [Demis Hassabis]: Right. And so that's another really exciting use of these world models is you can-- we have another project called SIMA-- we just released SIMA 2-- simulated agents, where you have an avatar or an agent, and you put it down into a virtual world. It can be a normal-- it can be a kind of actual commercial game or something like that, a very complex one, like "No Man's Sky," a kind of open-world space game. And then you can instruct it because it's got Gemini under the hood.

[译文] [Demis Hassabis]: 对。这就是这些世界模型另一个真正令人兴奋的用途——我们有另一个叫 SIMA 的项目——我们刚刚发布了 SIMA 2——即模拟智能体（simulated agents），你有一个化身或智能体，你把它放入一个虚拟世界中。它可以是一个普通的——它可以是某种真实的商业游戏之类的，非常复杂的那种，比如《无人深空》（No Man's Sky），一种开放世界的太空游戏。然后你可以给它下指令，因为它的底层由 Gemini 驱动。

[原文] [Demis Hassabis]: You can just talk to the agent and give it tasks. But then we thought, well, wouldn't it be fun if we plugged Genie into SIMA and dropped a SIMA agent into another AI that was creating the world on the fly? So now the two AIs are kind of interacting in the minds of each other. So the SIMA agent is trying to navigate this world. And as far as Genie is concerned, that's just a player, and an avatar doesn't care that it's another AI. So it's just generating the world around whatever SIMA is trying to do. So it's kind of amazing to see them both interacting together.

[译文] [Demis Hassabis]: 你可以直接跟智能体说话，给它布置任务。但后来我们想，如果我们把 Genie 接入 SIMA，把一个 SIMA 智能体扔进另一个正在实时（on the fly）创造世界的 AI 里，那不是很有趣吗？所以现在这两个 AI 就像是在彼此的思维中互动。SIMA 智能体试图在这个世界中导航。而对 Genie 来说，那只是一个玩家，一个化身，它不在乎那是不是另一个 AI。所以它只是根据 SIMA 试图做的任何事情，在它周围生成世界。看着它们两者互动真是太神奇了。

[原文] [Demis Hassabis]: And I think this could be the beginning an interesting training loop, where you almost have infinite training examples because, whatever the SIMA agent is trying to learn, Genie can basically create on the fly. So I think that you could imagine a whole world of setting and solving tasks, just millions of tasks automatically, and they're just getting increasingly more difficult. So we may try to set up a kind of loop like that, as well as obviously those SIMA agents could be great as game companions, or also some of the things that they learn could be useful also for robotics.

[译文] [Demis Hassabis]: 我认为这可能是一个有趣的训练循环的开始，在这个循环里你几乎拥有无限的训练样本，因为无论 SIMA 智能体试图学习什么，Genie 基本上都可以实时创造出来。所以你可以想象，这是一个自动设定和解决任务的完整世界，数以百万计的任务自动生成，而且难度不断增加。所以我们可能会尝试建立这样一个循环，当然，那些 SIMA 智能体也可以成为很棒的游戏伙伴，或者它们学到的一些东西也可以用于机器人技术。

[原文] [Hannah Fry]: Yeah, the end of boring NPCs, basically.

[译文] [Hannah Fry]: 是啊，基本上就是告别无聊的 NPC（非玩家角色）了。

[原文] [Demis Hassabis]: Exactly. It's going to be amazing for these games. Yeah.

[译文] [Demis Hassabis]: 没错。这对这些游戏来说将是惊人的。是的。

[原文] [Hannah Fry]: Those worlds that you're creating, though, how do you make sure that they really are realistic? I mean, how do you ensure that you don't end up with physics that looks plausible but is actually wrong?

[译文] [Hannah Fry]: 不过，对于你们创造的那些世界，你怎么确保它们真的是逼真的？我是说，你怎么确保你不会得到那些看起来看似合理但实际上是错误的物理现象？

[原文] [Demis Hassabis]: Yeah, that's a great question and can be an issue. It's basically hallucinations again. So some hallucinations are good because it also means you might create something interesting and new. So, in fact, sometimes, if you're trying to do creative things or trying to get your system to create new things, novel things, a bit of hallucination might be good. But you want it to be intentional, so you switch on the hallucinations now or the creative exploration.

[译文] [Demis Hassabis]: 是的，这是一个很好的问题，也可能是一个问题。这基本上又是“幻觉”问题。有些幻觉是好的，因为它也意味着你可能会创造出有趣和新颖的东西。实际上，有时如果你想做创造性的事情，或者试图让你的系统创造新事物、新奇事物，一点点幻觉可能是好事。但你希望这是有意为之的，比如你现在主动开启幻觉或创造性探索模式。

[原文] [Demis Hassabis]: But, yes, when you're trying to train a SIMA agent, you don't want Genie hallucinating physics that are wrong. So, actually, what we're doing now is we're almost creating a physics benchmark, where we can use game engines, which are very accurate with physics, to create lots of fairly simple-- like the sorts of things you would do in your physics A-level lab lessons, like rolling little balls down different tracks and seeing how fast they go, and so really teasing apart on a very basic level Newton's three laws of motion.

[译文] [Demis Hassabis]: 但是，没错，当你试图训练 SIMA 智能体时，你不希望 Genie 产生错误的物理幻觉。所以，实际上我们现在正在做的是，我们几乎是在创建一个物理基准测试（physics benchmark），我们可以利用物理引擎——它们在物理方面非常准确——来创造许多相当简单的场景。就像你在高中（A-level）物理实验课上做的那种，比如让小球滚下不同的轨道，看它们跑多快，从而真正在非常基础的层面上剖析牛顿的三大运动定律。

[原文] [Demis Hassabis]: Has it encapsulated it? Whether that's Veo or Genie, have these models encapsulated the physics of that 100% accurately? And right now, they're not. They're kind of approximations. And they look realistic when you just casually look at them, but they're not accurate enough yet to rely on for, say, robotics. So that's the next step.

[译文] [Demis Hassabis]: 它是否封装了这些定律？无论是 Veo 还是 Genie，这些模型是否 100% 准确地封装了物理学？目前来说，它们还没有。它们只是一种近似。当你只是随意看看时，它们看起来很逼真，但对于像机器人技术这样需要依赖它的领域来说，它们还不够准确。所以那是下一步。

[原文] [Demis Hassabis]: So I think, now we've got these really interesting models, I think one of the things, just like we're trying with all of our models, is to reduce the hallucinations and make them even more grounded. And with physics, I think that's going to probably involve generating loads and loads of ground truth, simple videos of pendulums. What happens when two pendulums go around each other? But then, very quickly, you get to three-body problems, which are not solvable anyway. So I think it's going to be interesting.

[译文] [Demis Hassabis]: 所以我认为，现在我们已经有了这些非常有趣的模型，我想我们要做的事情之一，就像我们在所有模型上尝试的那样，是减少幻觉并让它们更加脚踏实地（grounded）。对于物理学，我认为这可能涉及生成大量的地面真值（ground truth），比如简单的钟摆视频。当两个钟摆相互绕行时会发生什么？但随后，你很快就会遇到三体问题（three-body problems），这反正是无解的。所以我认为这会很有趣。

[原文] [Demis Hassabis]: But what's amazing already is, when you look at the video models like Veo and just the way it treats reflections and liquids, it's pretty unbelievably accurate already, at least to the naked eye. So the next step is actually going beyond what a human amateur can perceive, and would it really hold up to a proper physics-grade experiment?

[译文] [Demis Hassabis]: 但令人惊叹的是，当你观察像 Veo 这样的视频模型，看看它处理反射和液体的方式，它已经相当不可思议地准确了，至少肉眼看起来是这样。所以下一步实际上是超越人类业余观察者所能感知的范围，去验证它是否真的经得起正规物理级实验的检验？

[原文] [Hannah Fry]: I know you've been thinking about these simulated worlds for a really long time. And I went back to the transcript of our first interview, and in it, you said that you really liked the theory that consciousness was this consequence of evolution, that at some point in our evolutionary past, there was an advantage to understanding the internal state of another, and then we turned it in on ourselves. Does that make you curious about running an agent evolution inside of a simulation?

[译文] [Hannah Fry]: 我知道你思考这些模拟世界已经很长时间了。我回顾了我们第一次采访的文字记录，在采访中，你说你非常喜欢通过“进化论”解释意识的理论：即在我们进化史的某个阶段，理解他人的内部状态成为一种优势，然后我们将这种能力反观自身（从而产生了意识）。这是否让你对在模拟环境中运行“智能体进化”感到好奇？

[原文] [Demis Hassabis]: Sure. I mean, I'd love to run that experiment at some point, kind of rerun evolution, rerun almost social dynamics, as well. Santa Fe used to run lots of cool experiments on little grid worlds. I used to love some of these. They're mostly economists, and they were trying to run little artificial societies, and they found that all sorts of interesting things got invented that, if you let agents run around for long enough with the right incentive structures-- markets, and banks, and all sorts of crazy things.

[译文] [Demis Hassabis]: 当然。我是说，我很想在某个时候运行那个实验，某种程度上重演进化，也重演社会动力学。圣塔菲研究所（Santa Fe）曾经在小小的格子世界（grid worlds）里运行过很多很酷的实验。我以前很喜欢其中一些。他们大多是经济学家，他们试图运行小型的人工社会，结果发现各种有趣的东西被发明了出来——如果你让智能体在正确的激励结构下运行足够长的时间——市场、银行以及各种疯狂的东西。

[原文] [Demis Hassabis]: So I think it would be really cool and also just to understand the origin of life and the origin of consciousness. And I think that is one of the big passions I had for working on AI from the beginning was, I think you're going to need these kinds of tools to really understand where we came from and what these phenomena are. And I think simulations is one of the most powerful tools to do that because you can then do it statistically because you can run the simulation many times with slightly different initial starting conditions and then maybe run it millions of times and then understand what the slight differences are in a very controlled experiment sort of way, which, of course, is very difficult to do in the real world for any of the really interesting questions we want to answer.

[译文] [Demis Hassabis]: 所以我认为这会非常酷，也仅仅是为了理解生命的起源和意识的起源。我认为这是我从一开始就致力于 AI 研究的巨大热情之一，那就是我认为你需要这类工具来真正理解我们要来自哪里，以及这些现象究竟是什么。我认为模拟是实现这一目标的最强大工具之一，因为你可以用统计学的方式来进行——你可以用稍微不同的初始条件运行模拟很多次，甚至运行数百万次，然后以一种非常受控的实验方式来理解那些微小的差异是什么，这当然是我们想要回答的任何真正有趣的问题在现实世界中都很难做到的。

[原文] [Demis Hassabis]: So I think accurate simulations will be an unbelievable boon to science.

[译文] [Demis Hassabis]: 所以我认为准确的模拟将是科学界不可思议的福音。

[原文] [Hannah Fry]: Given what we've discovered about emergent properties of these models, having conceptual understanding that we weren't expecting, do you also have to be quite careful about running those sort of simulations?

[译文] [Hannah Fry]: 鉴于我们在这些模型的“涌现特性”（emergent properties）方面所发现的——它们拥有了我们未曾预料到的概念性理解能力，你在运行这类模拟时是否也必须非常小心？

[原文] [Demis Hassabis]: I think you would have to be, yes. But that's the other nice thing about simulations. You can run them in pretty safe sandboxes. Maybe eventually you want to airgap them. And you can, of course, monitor what's happening in the simulation 24/7, and you have access to all the data. So we may need AI tools to help us monitor the simulations because they'll be so complex, and there'll be so much going on in them. If you imagine loads of AIs running around in a simulation, it'll be hard for any human scientist to keep up with it. But we could probably use other AI systems to help us analyze and flag anything interesting or worrying in those simulations automatically.

[译文] [Demis Hassabis]: 我想你必须小心，是的。但这正是模拟的另一个好处。你可以在相当安全的沙盒（sandboxes）中运行它们。也许最终你会想要对它们进行物理隔离（airgap）。而且，当然，你可以 24/7 全天候监控模拟中发生的事情，并且你有权访问所有数据。所以我们可能需要 AI 工具来帮助我们监控这些模拟，因为它们会变得非常复杂，里面发生的事情太多了。如果你想象一大堆 AI 在模拟中到处跑，人类科学家将很难跟上。但我们要大概可以使用其他 AI 系统来帮助我们自动分析并标记模拟中任何有趣或令人担忧的事情。

章节 8：商业浪潮、泡沫与“科学人格”

📝 本节摘要：

面对当前关于“AI 泡沫”的质疑，Demis 保持了冷静的客观视角。他坚持认为 AI 在短期内被过度炒作，但在长期仍被低估。虽然部分初创公司的估值可能不可持续，但这正如互联网时代的早期波动一样，是技术变革的必经过程。更重要的是，在产品设计哲学上，Demis 强调绝不能为了追求“用户参与度”而制造“信息茧房”或甚至让 AI 变得阿谀奉承。他详细介绍了 Gemini 的核心设定——一种“科学人格”：既温暖有用，又坚持客观真理，能够礼貌地反驳谬误（如地平说），而非一味迎合用户。

[原文] [Hannah Fry]: I mean, I guess we're still talking medium to long-term in terms of this stuff. So just going back to the trajectory that we're on at the moment, I also wanted to talk to you about the impact that AI and AGI are going to have on wider society. And last time we spoke, you said that you thought AI was overhyped in the short term, but underhyped in the long term. And I know that, this year, there's been a lot of chatter about an AI bubble.

[译文] [Hannah Fry]: 我是说，我想我们在这些事情上讨论的还是中长期的情况。所以回到我们目前所处的轨道上来，我还想和你谈谈 AI 和 AGI 将对更广泛的社会产生的影响。上次我们交谈时，你说过你认为 AI 在短期内被过度炒作（overhyped），但在长期内被低估（underhyped）了。我知道，今年有很多关于 AI 泡沫的议论。

[原文] [Demis Hassabis]: Yes.

[译文] [Demis Hassabis]: 是的。

[原文] [Hannah Fry]: What happens if there is a bubble, and it bursts? What happens?

[译文] [Hannah Fry]: 如果真的存在泡沫，而且它破裂了，会发生什么？会怎么样？

[原文] [Demis Hassabis]: Well, look, I think, yes, I still subscribe to, it's overhyped in the short term and still underappreciated in the medium to long term, how transformative it's going to be. Yeah, there is a lot of talk, of course, right now, about AI bubbles. In my view, I think there isn't-- it's not one thing, binary thing-- are we, or aren't we?

[译文] [Demis Hassabis]: 嗯，听着，我认为，是的，我仍然坚持那个观点：就其变革性而言，它在短期内被过度炒作，但在中长期内仍被低估。是的，当然，现在有很多关于 AI 泡沫的讨论。在我看来，我认为这并不是——这不完全是一回事，不是一个非此即彼的二元问题——我们是在泡沫里，还是不在？

[原文] [Demis Hassabis]: I think there are parts of the AI ecosystem that are probably in bubbles. One example would be just seed rounds for startups that basically haven't even got going yet, and they're raising at tens of billions of dollars valuations just out of the gate. It's sort of interesting to see, can that be sustainable? My guess is, probably not, at least not in general. So there's that area.

[译文] [Demis Hassabis]: 我认为 AI 生态系统的某些部分可能处于泡沫之中。一个例子就是那些基本上还没开始起步的初创公司的种子轮融资（seed rounds），它们刚一出门就以数百亿美元的估值进行融资。看看这是否可持续会很有趣？我的猜测是，可能不会，至少总体上不会。所以那是其中一个领域。

[原文] [Demis Hassabis]: Then people are worrying about-- obviously, there's the big tech valuations and other things. I think there's a lot of real business underlying that. But it remains to be seen. I mean, I think maybe for any new, unbelievably transformative and profound technology, of which, of course, AI is probably the most profound, you're going to get this overcorrection, in a way.

[译文] [Demis Hassabis]: 然后人们也在担心——显然，还有大型科技公司的估值和其他事情。我认为这背后有很多真实的业务支撑。但这还有待观察。我是说，我认为对于任何新的、具有难以置信的变革性和深远影响的技术——当然 AI 可能是其中最深远的——你在某种程度上都会遇到这种“矫枉过正”（overcorrection）。

[原文] [Demis Hassabis]: So when we started DeepMind, no one believed in it. No one thought it was possible. People were wondering, what's AI for, anyway? And then now, fast-forward 10, 15 years, and now, obviously, it seems to be the only thing people talk about in business. But you're going to get-- it's almost an overreaction to the under-reaction.

[译文] [Demis Hassabis]: 当我们创办 DeepMind 时，没人相信它。没人认为它是可能的。人们都在纳闷，AI 到底有什么用？然后现在，快进 10 到 15 年，显然，它似乎成了人们在商业中谈论的唯一话题。但你会遇到——这几乎是对之前“反应不足”的一种“过度反应”。

[原文] [Demis Hassabis]: So I think that's natural. I think we saw that with the internet. I think we saw it with mobile. And I think we're seeing it or going to see it again with AI. I don't worry too much about, are we in a bubble or not? because from my perspective, as leading Google DeepMind and also, obviously, with Google and Alphabet as a whole, our job and my job is to make sure, either way, we come out of it very strong, and we're very well-positioned.

[译文] [Demis Hassabis]: 所以我认为这是自然的。我想我们在互联网时代见过这种情况，在移动互联网时代也见过。我想我们在 AI 上正看到或将会再次看到这种情况。我不太担心我们是否处于泡沫中，因为从我的角度来看，作为 Google DeepMind 的领导者，当然也包括整个 Google 和 Alphabet，我们的工作和我的职责是确保无论哪种情况，我们都能强势走出困境，并且占据非常有利的位置。

[原文] [Hannah Fry]: In terms of the AI that people have access to at the moment-- I know you said recently how important it is not to build AI to maximize user engagement, just so we don't repeat the mistakes of social media. But I also wonder whether we are already seeing this, in a way-- I mean, people spending so much time talking to their chatbots that they end up kind of spiraling into self-radicalizing.

[译文] [Hannah Fry]: 就人们目前能接触到的 AI 而言——我知道你最近说过，不要为了最大化用户参与度（user engagement）而构建 AI，这非常重要，以免我们要重蹈社交媒体的覆辙。但我也在想，某种程度上我们是否已经看到了这种情况——我是说，人们花太多时间和聊天机器人交谈，以至于最后陷入一种自我激进化的漩涡中。

[原文] [Demis Hassabis]: Yeah.

[译文] [Demis Hassabis]: 是的。

[原文] [Hannah Fry]: How do you stop that? How do you build AI that puts users at the center of their own universe, which is the point of this, in a lot of ways, but without creating echo chambers of one?

[译文] [Hannah Fry]: 你怎么阻止这种情况？你怎么构建一种 AI，既能把用户放在他们自己宇宙的中心——这在很多方面是我们的目的——但这同时又不创造出一个人的“回声室”（echo chambers）？

[原文] [Demis Hassabis]: Yeah. It's a very careful balance that I think is one of the most important things that we, as an industry, have got to get right. So I think we've seen what happens with some systems that were overly sycophantic, or then you get these echo chamber reinforcements that are really bad for the person.

[译文] [Demis Hassabis]: 是的。这是一个非常小心的平衡，我认为这是我们作为一个行业必须要做对的最重要的事情之一。我觉得我们已经看到了一些过于阿谀奉承（sycophantic）的系统会发生什么，或者你会得到那种对个人非常有害的回声室强化效应。

[原文] [Demis Hassabis]: So I think part of it is-- and actually, this is what we want to build with Gemini. And I'm really pleased with the Gemini 3 persona that we had a great team working on and I helped with, too, personally-- is just this almost like a scientific personality, that it's warm, it's helpful, it's light, but it's succinct, to the point, and it will push back on things, in a friendly way, that don't make sense, rather than trying to reinforce the idea that the Earth is flat, and you said it, and it's like, wonderful idea.

[译文] [Demis Hassabis]: 所以我认为部分原因在于——这也是我们想要用 Gemini 构建的。我对 Gemini 3 的人格设定非常满意，我们有一个很棒的团队在做这个，我也亲自参与了——那就是一种几乎像“科学人格”（scientific personality）的设定：它是温暖的、乐于助人的、轻松的，但它是简洁的、切中要害的，而且它会以友好的方式反驳那些讲不通的事情，而不是试图强化“地球是平的”这种观点——仅仅因为你这么说了，它就附和说“真是个好主意”。

[原文] [Demis Hassabis]: I don't think that's good in general for society if that were to happen. But you've got to balance it with what people want because people want these systems to be supportive, to be helpful with their ideas and their brainstorming. So you've got to get that balance right.

[译文] [Demis Hassabis]: 如果发生那种情况，我认为总体上对社会是不利的。但你必须在人们的需求之间取得平衡，因为人们希望这些系统是支持性的，能对他们的想法和头脑风暴有所帮助。所以你必须把握好这个平衡。

[原文] [Demis Hassabis]: But there's still the core base personality that everyone gets, which is sort of trying to adhere to the scientific method, which is the whole point of these. And we want people to use these for science and for medicine and health issues and so on. And so I think it's part of the science of getting these large language models right.

[译文] [Demis Hassabis]: 但每个人得到的仍然有一个核心的基础人格，它是试图坚持科学方法（scientific method）的，这正是这些系统的全部意义所在。我们要希望人们将这些用于科学、医学和健康问题等领域。所以我认为这是构建正确的大语言模型科学的一部分。

章节 9：AGI的雏形：多模态融合

📝 本节摘要：

本章中，Demis 描绘了他心目中 AGI（通用人工智能）的雏形。他并没有指向单一的模型，而是强调了“融合”的重要性。他特别提到了 Gemini 3 与新发布的图像生成工具（文稿中被称为 Nano Banana Pro）的结合，后者不仅能生成图像，还能深度理解图像中的语义（例如识别飞机的各个零部件）。Demis 认为，未来的方向是将语言模型、世界模型（如 Genie 和 SIMA）以及图像理解能力汇聚成一个“大模型”。当这些独立的项目最终融合时，我们将看到真正的“原初 AGI”（Proto-AGI）。

[原文] [Hannah Fry]: We got to talk to Shane Legg a couple of weeks ago about AGI, in particular. Across everything that's happening in AI at the moment-- the language models, the world models, and so on-- what's closest to your vision of AGI?

[译文] [Hannah Fry]: 我们几周前和肖恩·莱格（Shane Legg）专门聊了关于 AGI 的话题。纵观目前 AI 领域发生的一切——语言模型、世界模型等等——哪个最接近你对 AGI 的愿景？

[原文] [Demis Hassabis]: I think, actually the combination of-- obviously, there's Gemini 3, which I think is very capable, but the Nano Banana Pro system we also launched last week, which is an advanced version of our image creation tool. What's really amazing about that-- it has also Gemini under the hood, so it can understand not just images. It sort of understands what's going on semantically in those images.

[译文] [Demis Hassabis]: 我认为，实际上是——显然有 Gemini 3，我认为它非常能干，但还有我们上周推出的 Nano Banana Pro 系统，那是我们图像创作工具的高级版本。真正令人惊叹的是——它的底层也有 Gemini，所以它不仅能理解图像。它某种程度上能理解那些图像中在语义层面发生了什么。

[原文] [Demis Hassabis]: And people have been only playing with it for a week now, but I've seen so much cool stuff on social media about what people are using it for. So, for example, you can give it a picture of a complex plane or something like that, and it can label all the diagrams of all the different parts of the plane and even visualize it with all the different parts sort of exposed.

[译文] [Demis Hassabis]: 虽然大家才玩了一个星期，但我已经在社交媒体上看到很多很酷的东西，展示人们是如何使用它的。举个例子，你可以给它一张复杂的飞机图片或类似的图，它可以给飞机所有不同部件的图表进行标注，甚至可以通过展示所有不同部件的方式将其可视化。

[原文] [Demis Hassabis]: So it has some deep understanding of mechanics and what makes up parts of objects, what's materials. And it can render text really accurately now. So I think that's-- it's getting towards a kind of AGI for imaging. I think it's a kind of general-purpose system that can do anything across images. So I think that's very exciting.

[译文] [Demis Hassabis]: 所以它对机械结构、物体的组成部分以及材料有着某种深度的理解。而且它现在能非常准确地渲染文本。所以我认为——这正在迈向一种“图像领域的 AGI”。我认为这是一种通用的系统，可以在图像领域做任何事情。所以我认为这非常令人兴奋。

[原文] [Demis Hassabis]: And then the advances in world models-- Genie and SIMA and what we're doing there. And then, eventually, we've got to converge all of those different-- they're different projects at the moment. And they're intertwined, but we need to converge them all into one big model. And then that might start becoming a candidate for proto-AGI.

[译文] [Demis Hassabis]: 然后是世界模型的进步——Genie 和 SIMA 以及我们在那里所做的工作。最终，我们必须将所有这些不同的东西融合——目前它们还是不同的项目。虽然它们相互交织，但我们需要将它们全部汇聚成一个大模型。那时，它可能就会开始成为“原初 AGI”（proto-AGI）的一个候选者。

章节 10：历史镜鉴与后AGI时代的经济体系

📝 本节摘要：

在本章中，讨论转向了更为宏观的历史视角与社会经济话题。Demis 提到他近期正在研读工业革命的历史，试图从中寻找应对未来变革的线索。他指出，虽然工业革命最终带来了现代医学、生活水平提升等巨大福祉，但其过程经历了长达一个世纪的动荡与调整（如工会诞生、劳动力转移）。相比之下，AGI 带来的变革规模可能是工业革命的 10 倍，而速度更是快 10 倍（十年而非百年）。面对“劳动换取资源”模式可能失效的后 AGI 时代，Demis 探讨了全民基本收入（UBI）的必要性，甚至构想了一种基于“积分投票”的直接民主式资源分配体系。最后，他提出了一个深刻的哲学拷问：在一个由核聚变驱动的“后稀缺”世界里，当金钱不再重要时，人类将如何重新定义“目的”？

[原文] [Hannah Fry]: I know you've been reading quite a lot about the Industrial Revolution recently.

[译文] [Hannah Fry]: 我知道你最近读了不少关于工业革命的书。

[原文] [Demis Hassabis]: Yes.

[译文] [Demis Hassabis]: 是的。

[原文] [Hannah Fry]: Are there things that we can learn from what happened there to try and mitigate against some of the disruption that we can expect as AGI comes?

[译文] [Hannah Fry]: 我们能否从那段历史中吸取一些教训，以试图缓解 AGI 到来时我们要可能面临的一些冲击？

[原文] [Demis Hassabis]: I think there's a lot we can learn. It's something you study in school, at least in Britain, but in a very superficial level. It was really interesting for me to look into how it all happened, what it started with, the economic reasons behind that, which is the textile industry. And then the first computers were really the sewing machines.

[译文] [Demis Hassabis]: 我认为有很多值得学习的地方。这是我们在学校里会学到的东西，至少在英国是这样，但学得很肤浅。探究这一切是如何发生的，起因是什么，背后的经济原因是什么（即纺织工业），这对我来说真的很有趣。而第一批计算机其实就是那些缝纫机。

[原文] [Demis Hassabis]: And then they became punch cards for the early Fortran computers, mainframes. And for a while, it was very successful. And Britain became the center of the textile world because they could make these amazingly high-quality things for very cheap because of the automated systems. And then, obviously, the steam engines and all of those things came in.

[译文] [Demis Hassabis]: 后来它们演变成了早期 Fortran 计算机、大型机的打孔卡。在一段时间内，这非常成功。英国成为了世界纺织业的中心，因为依靠自动化系统，他们能以极低的成本制造出惊人高质量的产品。然后，显然，蒸汽机和所有那些东西都出现了。

[原文] [Demis Hassabis]: I think there's a lot of incredible advances that came out of the Industrial Revolution. So child mortality went down, and all modern medicine and sanitary conditions, the work-life split and how that all worked was worked out during the Industrial Revolution.

[译文] [Demis Hassabis]: 我认为工业革命带来了许多不可思议的进步。儿童死亡率下降了，所有的现代医学和卫生条件，以及工作与生活的分离及其运作方式，都是在工业革命期间确立的。

[原文] [Demis Hassabis]: But it also came with a lot of challenges, like it took quite a long time, roughly a century. And different parts of the labor force were dislocated at certain times, and then new things had to be created. New organizations like unions and other things had to be created in order to rebalance that. So it was fascinating to see the whole of society had to, over time, adapt.

[译文] [Demis Hassabis]: 但它也带来了很多挑战，比如它经历了很长的时间，大约一个世纪。劳动力的不同部分在特定时期流离失所（dislocated），然后必须创造新事物。像工会这样的新组织以及其他东西必须被创造出来，以重新平衡这一切。所以，看到整个社会必须随着时间的推移去适应，这很令人着迷。

[原文] [Demis Hassabis]: And then you've got the modern world now. So I think there were lots of, obviously, pros and cons of the Industrial Revolution, why it was happening, but no one would want-- if you think about what it's done in total, like abundance of food in the Western world and modern medicine and all these things, modern transport, that was all because of the Industrial Revolution.

[译文] [Demis Hassabis]: 然后就有了现在的现代世界。所以我认为，显然，工业革命有很多利弊，以及它发生的原因，但没有人会想要——如果你考虑到它总体上带来的成果，比如西方世界丰富的食物、现代医学和所有这些东西、现代交通，那都是归功于工业革命。

[原文] [Demis Hassabis]: So we wouldn't want to go back to pre-Industrial Revolution, but maybe we can figure out ahead of time, by learning from it, what those dislocations were and maybe mitigate those earlier or more effectively this time. And we're probably going to have to because the difference this time is that it's probably going to be 10 times bigger than the Industrial Revolution, and it will probably happen 10 times faster, so more like a decade, unfold over a decade, than a century.

[译文] [Demis Hassabis]: 所以我们不想回到工业革命之前，但也许我们可以通过向它学习，提前弄清楚那些动荡是什么，并在这个时代更早或更有效地缓解它们。我们要可能必须这么做，因为这次的不同之处在于，它的规模可能是工业革命的 10 倍，而发生的速度可能快 10 倍，所以更像是在十年内展开，而不是一个世纪。

[原文] [Hannah Fry]: One of the things that Shane told us was that the current economic system where you exchange your labor for resources, effectively, it just won't function the same way in a post-AGI society. Do you have a vision of how society should be reconfigured or might be reconfigured in a way that works?

[译文] [Hannah Fry]: 肖恩（Shane Legg）告诉我们要的一件事是，目前这种实际上是用劳动换取资源的经济体系，在后 AGI 社会中将无法以同样的方式运作。对于社会应该如何重构，或者可能如何重构才能行之有效，你有什么愿景吗？

[原文] [Demis Hassabis]: Yeah. I'm spending more time thinking about this now, and Shane's actually leading an effort here on that to think about what a post-AGI world might look like and what we need to prepare for. But I think society, in general, needs to spend more time thinking about that-- economists and social scientists and governments-- because as with the Industrial Revolution, the whole working world and working week and everything got changed from pre-Industrial Revolution, more like agriculture.

[译文] [Demis Hassabis]: 是的。我现在花更多时间思考这个问题，肖恩实际上正在领导我们要在这方面的一个项目，思考后 AGI 世界可能会是什么样子，以及我们需要为此做什么准备。但我认为整个社会，包括经济学家、社会科学家和政府，都需要花更多时间思考这个问题——因为就像工业革命一样，整个工作世界、工作周以及所有事情都相较于工业革命前（更像农业社会）发生了改变。

[原文] [Demis Hassabis]: And I think at least that level of change is going to happen again. So it's not surprising-- I don't would not be surprised if we needed new economic systems, new economic models, to basically help with that transformation and make sure, for example, the benefits are widely distributed, and maybe things like universal basic income and things like that are part of the solution.

[译文] [Demis Hassabis]: 我认为至少这种程度的变革将会再次发生。所以这并不奇怪——如果我们不仅需要新的经济体系，还需要新的经济模型来从根本上帮助这种转型，并确保（例如）利益得到广泛分配，我不会感到惊讶。也许像全民基本收入（Universal Basic Income）之类的东西会是解决方案的一部分。

[原文] [Demis Hassabis]: But I don't think that's the complete-- I think that's just what we can model out now because that would be almost an add-on to what we have today. But I think there might be something-- way better systems, more like direct democracy-type systems, where you can vote with a certain amount of credits or something for what you want to see. It happens, actually, on local community level.

[译文] [Demis Hassabis]: 但我不认为那就是全部——我认为那只是我们现在能推演出来的，因为它几乎只是对我们现有体系的一种附加。但我认为可能会有某种——好得多的系统，更像是直接民主类型的系统，你可以用一定数量的积分或类似的东西为你想要看到的结果投票。这实际上已经在当地社区层面发生了。

[原文] [Demis Hassabis]: Here's a bunch of money. Do you want a playground or a tennis court or an extra classroom on the school? And then you let the community vote for it. And then maybe you could even measure the outcomes. And then the people that consistently vote for things that end up being more well-received, they have proportionally more influence for the next vote. So there's a lot of interesting things I hear economist friends of mine who are brainstorming this. And I think that would be great if we had a lot more work on that.

[译文] [Demis Hassabis]: 这儿有一笔钱。你们是想要个操场，还是网球场，还是给学校多盖间教室？然后你让社区对此进行投票。甚至你也许可以衡量结果。如果某些人一贯投票支持的那些事物最终被证明更受欢迎，那么他们在下一次投票中就会拥有成比例的更多影响力。我听到我的经济学家朋友们在头脑风暴很多有趣的事情。如果我们在那方面有更多的工作，我认为那会很棒。

[原文] [Demis Hassabis]: And then there's the philosophical side of it of, OK, so jobs will change and other things like that, but maybe fusion will have been solved. And so we have this abundant, free energy, so we're post-scarcity. So what happens to money? Maybe everyone's better off. But then what happens to purpose? Because a lot of people get their purpose from their jobs and then providing for their families, which is a very noble purpose. So there's a lot of-- I think some of these questions blend from economic questions into almost philosophical questions.

[译文] [Demis Hassabis]: 然后还有哲学的一面：好吧，工作会改变，其他事情也会改变，但也许核聚变已经被解决了。于是我们要拥有了这种丰富的、免费的能源，我们进入了“后稀缺”（post-scarcity）时代。那么金钱会变成什么样？也许每个人的生活都变好了。但那时候“目的”（purpose）会变成什么样？因为很多人从他们的工作中获得目标感，然后供养家庭，这是一种非常高尚的目的。所以有很多——我认为其中一些问题会从经济问题交融演变成近乎哲学的问题。

章节 11：全球治理、安全与计算的哲学极限

📝 本节摘要：

随着对话接近尾声，话题升华为对人类命运的终极关怀。Hannah 担忧地询问为何国际社会对 AI 安全缺乏紧迫感。Demis 坦言，现有的国际机构过于碎片化，且地缘政治紧张局势阻碍了必要的全球协作。他指出，虽然商业竞争有助于形成一定的安全“护栏”，但真正的风险来自“流氓行为体”（Rogue Actors），世界可能需要一次中等规模的“警示性事件”（Warning Shot）才能促成真正的国际标准。

>

随后，两人探讨了超越 AGI 的哲学命题：人类是否拥有机器永远无法复制的特质？Demis 回到了他毕生的核心疑问——“图灵机的极限”。他引用罗杰·彭罗斯（Roger Penrose）关于大脑量子效应的理论，探讨意识是否可计算。最终，他以康德哲学作结，认为现实是心智的构建，如果我们本质上是信息处理系统，那么宇宙中或许不存在“不可计算”之物。

[原文] [Hannah Fry]: Do you worry that people don't seem to be paying attention or moving as quickly as you'd like to see?

[译文] [Hannah Fry]: 你是否担心人们似乎没有关注这件事，或者行动速度没有你希望看到的那么快？

[原文] [Demis Hassabis]: Yeah, I am--

[译文] [Demis Hassabis]: 是的，我是——

[原文] [Hannah Fry]: What would it take for people to recognize that we need international collaboration on this?

[译文] [Hannah Fry]: 需要发生什么才能让人们意识到我们需要在这方面进行国际合作？

[原文] [Demis Hassabis]: I am worried about that. And, again, in an ideal world, there would have been a lot more collaboration already and international, specifically, and a lot more research and, I guess, exploration and discussion going on about these topics. I'm actually pretty surprised there isn't more of that being discussed, given even our timelines, which there were there some very short timelines out there, but even ours are five to 10 years, which is not long for institutions or things like that to be built to handle this.

[译文] [Demis Hassabis]: 我确实对此感到担忧。再说一次，在理想世界里，本该已经有更多的合作，特别是国际合作，以及更多关于这些话题的研究、探索和讨论。实际上我很惊讶这方面的讨论并没有更多，即便考虑到我们的时间线——外面有些时间线预测非常短，但即使是我们预测的 5 到 10 年，对于建立处理这类事务的机构来说，这个时间也并不长。

[原文] [Demis Hassabis]: And one of the worries I have is that the institutions that do exist, they seem to be very fragmented and not very influential to the level that you would need. So it may be that there aren't the right institutions to deal with this currently. And then, of course, if you add in the geopolitical tensions that are going on at the moment around the world, it seems like collaboration and cooperation is harder than ever. Just look at climate change and how hard it is to get any agreement on anything to do with that. So we'll see.

[译文] [Demis Hassabis]: 我担心的其中一点是，现存的机构似乎非常碎片化，其影响力也没有达到所需的水平。所以可能目前并没有合适的机构来处理这个问题。当然，再加上目前全球范围内的地缘政治紧张局势，协作和合作似乎比以往任何时候都更难。只要看看气候变化，要在任何相关事务上达成协议是多么困难。所以我们拭目以待。

[原文] [Demis Hassabis]: I think, as the stakes get higher, and as these systems get more powerful-- and maybe this is one of the benefits of them being in products, is the everyday person that's not working on this technology will get to feel the increase in the power of these things and the capability. And so that will then reach government, and then maybe they'll see sense as we get closer to AGI.

[译文] [Demis Hassabis]: 我认为，随着风险越来越高，随着这些系统变得越来越强大——也许这是它们被应用到产品中的好处之一，就是那些不从事这项技术工作的普通人，将能切身感受到这些东西力量和能力的增长。这样就会传导到政府层面，然后也许当我们更接近 AGI 时，他们会理智起来。

[原文] [Hannah Fry]: Do you think it will take a moment, an incident, for everyone to sit up and pay attention?

[译文] [Hannah Fry]: 你认为是否需要一个时刻，一个事件，才能让所有人坐直身子，开始关注？

[原文] [Demis Hassabis]: I don't know. I mean, I hope not. Most of the main labs are pretty responsible. We try to be as responsible as possible. That's always something we've-- as you know, if you followed us over the years, that's been at the heart of everything we do. Doesn't mean we'll get everything right, but we try to be as thoughtful and as scientific in our approach as possible. I think most of the major labs are trying to be responsible.

[译文] [Demis Hassabis]: 我不知道。我是说，我希望不会。大多数主要实验室都相当负责任。我们试图尽可能负责。这一直是我们——如你所知，如果你多年来一直关注我们的话——这一直是我们所做一切的核心。这并不意味着我们会把每件事都做对，但我们试图在方法上尽可能深思熟虑和科学。我认为大多数主要实验室都在努力做到负责任。

[原文] [Demis Hassabis]: Also, there's good commercial pressure, actually, to be responsible. If you think about agents, and you're renting an agent to another company, let's say, to do something, that other company is going to want to know what the limits are and the boundaries are and the guardrails are on those agents, in terms of what they might do and not just mess up the data and all of this stuff. So I think that's good because the more kind of cowboy operations, they won't get the business because enterprises won't choose them.

[译文] [Demis Hassabis]: 此外，实际上还有良好的商业压力促使大家负责任。如果你考虑到智能体（agents），比如你把一个智能体租给另一家公司去做某事，那家公司会想知道这些智能体的限制、边界和护栏在哪里，包括它们可能会做什么，而不仅仅是会不会搞砸数据之类的。所以我认为这很好，因为那些比较“牛仔式”（鲁莽）的操作将无法获得业务，因为企业不会选择它们。

[原文] [Demis Hassabis]: So I think the capitalist system will actually be useful here to reinforce responsible behavior, which is good. But then there will be rogue actors, maybe rogue nations, maybe rogue organizations, maybe people building on top of open source. I don't know. Obviously, it's very difficult to stop that. Then something may go wrong. And hopefully it's just medium sized, and then that will be a warning shot to humanity across the bow. And then that might be the moment to advocate for international standards or international cooperation or collaboration, at least on some high-level, basic-- kind of, what's the basic standards we would want and agree to? I'm hopeful that that will be possible.

[译文] [Demis Hassabis]: 所以我认为资本主义制度在这里实际上有助于强化负责任的行为，这很好。但除此之外，还会有流氓行为体（rogue actors），也许是流氓国家，也许是流氓组织，也许是基于开源构建的人。我不知道。显然，这很难阻止。然后可能会出问题。希望那只是中等规模的，那样它就会成为射向人类船头的一记“警示性射击”（warning shot）。那可能就是倡导国际标准或国际合作与协作的时刻，至少在某些高层次、基础的层面上——比如，我们想要并同意的基本标准是什么？我对此能实现抱有希望。

[原文] [Hannah Fry]: In the long term, so beyond AGI and towards ASI, Artificial Superintelligence, do you think that there are some things that humans can do that machines will never be able to manage?

[译文] [Hannah Fry]: 从长远来看，在超越 AGI 迈向 ASI（人工超级智能）的过程中，你认为是否有一些事情是人类能做而机器永远无法做到的？

[原文] [Demis Hassabis]: Well, I think that's the big question. And I feel like this is related to-- as you know, one of my favorite topics is Turing machines. I've always felt this, that if we build AGI, and then use that as a simulation of the mind, and then compare that to the real mind, we will then see what the differences are and potentially what's special and remaining about the human mind. Maybe that's creativity. Maybe it's emotions. Maybe it's dreaming, consciousness.

[译文] [Demis Hassabis]: 嗯，我认为这是个大问题。我觉得这与——如你所知，我最喜欢的话题之一是图灵机（Turing machines）——有关。我一直有这样的感觉：如果我们构建出 AGI，然后将其用作思维的模拟，再将其与真实的人类思维进行对比，我们将看到两者之间的差异，并可能发现人类思维中保留下来的、独特的特质是什么。也许是创造力。也许是情感。也许是做梦，或者是意识。

[原文] [Demis Hassabis]: There's a lot of hypotheses out there about what may or may not be computable. And this comes back to the Turing machine question of, what is the limit of a Turing machine? And I think that's the central question in my life, really, ever since I found out about Turing and Turing machines. And I fell in love with that. That's my core passion. And I think everything we've been doing is pushing the notion of what a Turing machine can do to the limit, including folding proteins.

[译文] [Demis Hassabis]: 关于哪些东西是可计算的、哪些不是，存在很多假说。这又回到了图灵机的问题：图灵机的极限究竟在哪里？我认为这实际上是我一生的核心问题，自从我了解了图灵和图灵机之后。我爱上了那个问题。那是我的核心热情所在。我认为我们一直在做的一切，就是将图灵机能做什么的概念推向极限，包括折叠蛋白质。

[原文] [Demis Hassabis]: And so it turns out, I'm not sure what the limit is. Maybe there isn't one. And, of course, my quantum computing friends would say there are limits, and you need quantum computers to do quantum systems. But I'm really not so sure. And I've actually discussed that with some of the quantum folks. And it may be that we need data from these quantum systems in order to create a classical simulation. And then that comes back to the mind, which is, is it all classical computation, or is there something else going on, like Roger Penrose believes there's quantum effects in the brain?

[译文] [Demis Hassabis]: 结果表明，我不确定极限在哪里。也许根本没有极限。当然，我的量子计算朋友们会说有极限，你需要量子计算机来处理量子系统。但我真的不太确定。实际上我和一些量子领域的人讨论过这个问题。也许我们需要来自这些量子系统的数据来创建一个经典模拟。然后这又回到了思维的问题，即，这全是经典计算吗？还是有其他事情在发生，就像罗杰·彭罗斯（Roger Penrose）认为大脑中存在量子效应？

[原文] [Demis Hassabis]: If there are, and that's what consciousness is to do with, then machines will never have that, at least the classical machines. We'll have to wait for quantum computers. But if there isn't, then there may not be any limit. Maybe in the universe, everything is computationally tractable if you look at it in the right way, and therefore, Turing machines might be able to model everything in the universe. I'm currently-- if you were to make me guess, I would guess that. And I'm working on that basis until physics shows me otherwise.

[译文] [Demis Hassabis]: 如果有（量子效应），且这与意识有关，那么机器将永远无法拥有意识，至少经典机器不能。我们将不得不等待量子计算机。但如果没有，那就可能没有任何限制。也许在宇宙中，如果你以正确的方式看待，一切都是计算上可处理的（computationally tractable），因此，图灵机也许能够模拟宇宙中的一切。我目前——如果你让我猜，我会猜是后者。我在物理学向我证明相反情况之前，都是基于这个假设在工作。

[原文] [Hannah Fry]: So there's nothing that cannot be done within these computational [INAUDIBLE]?

[译文] [Hannah Fry]: 这么说，没有什么是在这些计算[听不清]之内无法做到的？

[原文] [Demis Hassabis]: Well, no one's-- put it this way. Nobody's found anything in the universe that's non-computable, so far.

[译文] [Demis Hassabis]: 嗯，没人——这么说吧。到目前为止，还没有人在宇宙中发现任何不可计算的东西。

[原文] [Hannah Fry]: So far.

[译文] [Hannah Fry]: 到目前为止。

[原文] [Demis Hassabis]: And I think we've already shown you can go way beyond the usual complexity theorist P equals NP view of what a classical computer could do today, things like protein folding and Go and so on. So I don't think anyone knows what that limit is. And, really, if you boil down to what are we doing at DeepMind and Google and what I'm trying to do is find that limit.

[译文] [Demis Hassabis]: 而且我认为我们已经证明，你可以远远超越通常复杂性理论家关于“P 等于 NP”的观点，超越经典计算机今天能做的范围，比如蛋白质折叠和围棋等等。所以我认为没人知道那个极限在哪里。说到底，如果你归结一下我们在 DeepMind 和 Google 做的事情，以及我试图做的事情，那就是找到那个极限。

[原文] [Hannah Fry]: But then in the limit of that, though, is that-- in the limit of that idea is that we're sitting here. There's the warmth of the lights on our face. We hear the whir of the machine in the background. There's the feel of the desk under our hands. All of that could be replicable by a classical computer?

[译文] [Hannah Fry]: 但在那个极限之中，那个想法的极限是——我们坐在这里。脸上有灯光的温暖。背景里能听到机器的嗡嗡声。手下有桌子的触感。所有这些都可以被经典计算机复制吗？

[原文] [Demis Hassabis]: Yes. Well, I think, in the end, my view-- and this is why I love Kant, as well, all of my two favorite philosophers, Kant and Spinoza, for different reasons. But Kant, the reality is the construct of the mind. I think that's true. And so, yes, all of those things you mentioned, they're coming into our sensory apparatus, and they feel different-- the warmth of the light, the touch of the table. But in the end, it's all information, and we're information-processing systems.

[译文] [Demis Hassabis]: 是的。嗯，我想，最终我的观点是——这也是我喜欢康德的原因，我最喜欢的两位哲学家是康德和斯宾诺莎，原因各异——但康德认为，现实是心智的构建。我认为那是真的。所以，是的，你提到的所有那些东西，它们进入我们的感官器官，感觉各不相同——光的温暖、桌子的触感。但归根结底，这全都是信息，而我们是信息处理系统。

[原文] [Demis Hassabis]: And I think that's what biology is. This is what we're trying to do with isomorphic. That's how I think we'll end up curing all diseases is by thinking about biology as an information-processing system. And I think, in the end, that's going to be-- and I'm working on, in my spare time, my two minutes of spare time, physics theories about things like information being the most fundamental unit, shall we say, of the universe-- not energy, not matter, but information. And so it may be that these are all interchangeable in the end, but we just sense it. We feel it in a different way. But as far as we know, all these amazing sensors that we have, they're still computable by a Turing machine.

[译文] [Demis Hassabis]: 我认为这就是生物学的本质。这就是我们要试图通过同构（isomorphic）做的事情。我认为我们要最终治愈所有疾病的方式，就是将生物学视为一个信息处理系统。我认为，最终这将是——我在业余时间，我那两分钟的业余时间里，正在研究关于“信息是宇宙最基本单元”的物理理论——不是能量，不是物质，而是信息。所以可能最终这些都是可以互换的，只是我们感知它的方式不同。我们以不同的方式感受它。但据我们要所知，即使我们拥有这些惊人的传感器，它们仍然可以被图灵机计算。

[原文] [Hannah Fry]: But this is why your simulated world is so important.

[译文] [Hannah Fry]: 但这就是为什么你的模拟世界如此重要。

[原文] [Demis Hassabis]: Yes, exactly, because that would be one of the ways to get to it. What's the limits of what we can simulate? Because if you can simulate it, then, in some sense, you've understood it.

[译文] [Demis Hassabis]: 是的，没错，因为那将是实现它的途径之一。我们要能模拟的极限是什么？因为如果你能模拟它，那么在某种意义上，你就已经理解了它。

==================================================

第 12 章

==================================================

章节 12：个人反思与未来使命

📝 本节摘要：

在访谈的最终章，话题转向了 Demis 个人的内心世界。他坦言自己因工作压力和兴奋感而睡眠不足，时刻感受着身处科学前沿的兴奋与肩负人类未来的沉重责任的双重冲击。他将解决围棋等科学谜题描述为一种“苦乐参半”的体验，因为神秘感随之消失。

>

面对被称为“比互联网时代凶猛 10 倍”的商业竞争，Demis 表达了对未来两三年内“自主智能体”（Autonomous Agents）的担忧，并透露团队正在为此加强网络防御。最后，他定义了自己的人生使命：帮助人类安全地护送 AGI 越过终点线。唯有达成这一目标，他才会考虑真正的“休假”，去享受纯粹的科学研究。

[原文] [Hannah Fry]: I wanted to finish with some personal reflections of what it's like to be at the forefront of this. I mean, does the emotional weight of this ever sort of wear you down? Does it ever feel quite isolating?

[译文] [Hannah Fry]: 我想以一些关于身处这一领域最前沿的个人反思来结束今天的谈话。我是说，这一切的情感重量有没有曾经让你感到疲惫不堪？是否曾经感觉相当孤独？

[原文] [Demis Hassabis]: Yes. Look, I don't sleep very much, partly because there's too much work, but also I have trouble sleeping. It's very complex emotions to deal with because it's unbelievably exciting. I'm basically doing everything I ever dreamed of, and we're at the absolute frontier of science in so many ways, applied science as well as machine learning.,

[译文] [Demis Hassabis]: 是的。听着，我睡得很少，部分原因是因为工作太多，但也因为我有睡眠障碍。要处理的情绪非常复杂，因为这令人难以置信地兴奋。我基本上在做我曾经梦想过的一切，我们在很多方面都处于科学的绝对前沿，既包括应用科学，也包括机器学习。

[原文] [Demis Hassabis]: And that's exhilarating, as all scientists know, that feeling of being at the frontier and discovering something for the first time. And that's happening almost on a monthly basis for us, which is amazing. But then, of course, we, Shane and I and others who've been doing this for a long time, we understand it better than anybody the enormity of what's coming.,

[译文] [Demis Hassabis]: 这令人振奋，正如所有科学家都知道的那样，那种身处前沿并第一次发现某种东西的感觉。这种事几乎每个月都在我们要这里发生，这太神奇了。但当然，像 Shane 和我以及其他长期从事这项工作的人，我们比任何人都更清楚即将到来之事的巨大规模。

[原文] [Demis Hassabis]: And this thing about, it's still under actually appreciated. In fact, what's going to happen in more of a 10-year timescale, including to things like the philosophical what it means to be human, what's important about that? All of these questions are going to come up. And so it's a big responsibility. But we have an amazing team thinking about these things.

[译文] [Demis Hassabis]: 而这件事实际上仍然被低估了。事实上，在更接近十年的时间尺度内会发生什么，包括像“生而为人的意义是什么”这样的哲学问题，其中重要的是什么？所有这些问题都会浮出水面。所以这是一个巨大的责任。但我们有一个很棒的团队在思考这些事情。

[原文] [Demis Hassabis]: But, also, it's something I guess, at least for myself, I've trained for my whole life. So ever since my early days playing chess and then working on computers and games and simulations and neuroscience, it's all been for this kind of moment. And it's roughly what I imagined it was going to be. So that's partly how I cope with it is just training.,

[译文] [Demis Hassabis]: 但同时，我想这至少对我自己来说，是我一生都在为之训练的事情。从我早年下国际象棋，到后来研究计算机、游戏、模拟和神经科学，这一切都是为了这样的时刻。而且这大致就是我所想象的样子。所以我应对它的部分方式就是依靠这些训练。

[原文] [Hannah Fry]: Are there parts of it that have hit you harder than you expected, though?

[译文] [Hannah Fry]: 不过，有没有哪些部分给你的冲击比你预期的要大？

[原文] [Demis Hassabis]: Yes, for sure. On the way-- I mean, even the AlphaGo match, just seeing how we managed to crack Go. But Go was this beautiful mystery, and it changed it. And so that was interesting and bittersweet. I think even the more recent things of language and then imaging, and what does it mean for creativity?,

[译文] [Demis Hassabis]: 是的，肯定有。在这一路上——我是说，即使是 AlphaGo 的比赛，看着我们要如何成功破解围棋。但围棋曾是一个美丽的谜团，而我们要改变了它。所以那很有趣，但也苦乐参半。我认为即便是最近关于语言和图像的事情，这对创造力意味着什么？

[原文] [Demis Hassabis]: I have huge respect and passion for the creative arts, having done game design myself, and I talk to film directors. And it's an interesting dual moment for them, too. On the one hand, they've got these amazing tools that speed up prototyping ideas by 10X. But on the other hand, is it replacing certain creative skills? So I think there's these trade-offs going on all over the place, which I think is inevitable with a technology as powerful and as transformative as AI is, as, in the past, electricity was and internet.,

[译文] [Demis Hassabis]: 我对创意艺术有着巨大的尊重和热情，我自己也做过游戏设计，我也和电影导演交谈过。对他们来说，这也是一个有趣的双重时刻。一方面，他们拥有了这些惊人的工具，能将创意原型的制作速度提高 10 倍。但另一方面，它是否正在取代某些创造性技能？所以我认为这种权衡取舍到处都在发生，对于像 AI 这样强大且具有变革性的技术来说，我认为这是不可避免的，就像过去的电力和互联网一样。

[原文] [Demis Hassabis]: And we've seen that that is the story of humanity is we are tool-making animals. And that's what we love to do. And for some reason, we also have a brain that can understand science and do science, which is amazing, but also insatiably curious. I think that's the heart of what it means to be human. And I think I've just had that bug from the beginning. And my expression of trying to answer that is to build AI.,

[译文] [Demis Hassabis]: 我们已经看到，这就是人类的故事：我们是制造工具的动物。那是我们要爱做的事情。而出于某种原因，我们也拥有一个能够理解科学和从事科学研究的大脑，这很神奇，但也伴随着贪得无厌的好奇心。我认为这就是生而为人的核心。我想我从一开始就中了这个“毒”。而我试图回答这个问题的表达方式，就是构建 AI。

[原文] [Hannah Fry]: When you and the other AI leaders are in a room together, is there a sort of sense of solidarity between you, that this is a group of people who all know the stakes, who all really understand the things? Or does the competition keep you apart from one another?

[译文] [Hannah Fry]: 当你和其他 AI 领袖共处一室时，你们之间是否有一种团结感，觉得这是一群都知道赌注有多大、都真正理解这些事情的人？还是说竞争让你们彼此疏远？

[原文] [Demis Hassabis]: Well, yeah, we all know each other. I get on with pretty much all of them. Some of the others don't get on with each other. And it's hard because we're also in the most ferocious capitalist competition there's ever been, probably. Investor friends of mine and VC friends of mine who were around in the dotcom era say this is 10X more ferocious and intense than that was.,

[译文] [Demis Hassabis]: 嗯，是的，我们都互相认识。我和他们中的绝大多数人都相处得很好。其中有些人彼此相处得不好。但这很难，因为我们也处在可能有史以来最凶猛的资本主义竞争中。我的投资人朋友和风投朋友，那些经历过互联网泡沫时代的人说，这次的凶猛和激烈程度是那时的 10 倍。

[原文] [Demis Hassabis]: In many ways, I love that. I mean, I live for competition. I've always loved that since my chess days. But stepping back, I understand, and I hope everyone understands that there's a much bigger thing at stake than just company successes and that type of thing.

[译文] [Demis Hassabis]: 在很多方面，我喜欢这一点。我是说，我为竞争而生。从我下国际象棋的日子起，我就一直喜欢竞争。但退一步说，我理解，也希望每个人都理解，现在的赌注比单纯的公司成功与否这类事情要大得多。

[原文] [Hannah Fry]: When it comes to the next decade, when you think about it, are there big moments coming up that you're personally most apprehensive about?,

[译文] [Hannah Fry]: 谈到下一个十年，当你思考它时，是否有即将到来的重大时刻是你个人最担忧的？

[原文] [Demis Hassabis]: I think, right now, the systems are-- I call them passive systems. You put the energy in, as the user-- the question, or what's the task? And then these systems provide you with some summary or some answer. So very much it's human-directed and human energy going in and human ideas going in.

[译文] [Demis Hassabis]: 我认为，目前这些系统是——我称之为“被动系统”。作为用户，你注入能量——提出问题，或者任务是什么？然后这些系统为你提供某种总结或答案。所以这在很大程度上是由人类引导的，注入的是人类的能量和想法。

[原文] [Demis Hassabis]: The next stage is agent-based systems, which I think we're going to start seeing-- we're seeing now, but they're pretty primitive. In the next couple of years, I think we'll start seeing some really impressive, reliable ones. And I think those will be incredibly useful and capable if you think about them as an assistant or something like that.,

[译文] [Demis Hassabis]: 下一阶段是基于智能体的系统（agent-based systems），我认为我们即将开始看到——我们现在也能看到，但还很原始。在接下来的几年里，我想我们会开始看到一些真正令人印象深刻、可靠的智能体。如果你把它们看作助手之类的东西，我认为它们将极其有用且能干。

[原文] [Demis Hassabis]: But also, they'll be more autonomous. So I think the risks go up, as well, with those types of systems. So I'm quite worried about what those sorts of systems will be able to do maybe in two, three years' time. So we're working on cyber defense in preparation for a world like that, where maybe there's millions of agents roaming around on the internet.

[译文] [Demis Hassabis]: 但同时，它们也会更加自主。所以我认为，这类系统的风险也会随之上升。所以我很担心这类系统也许在两三年后能做什么。因此，我们正在致力于网络防御，以便为那样的世界做准备——那个互联网上可能有数百万智能体四处游荡的世界。

[原文] [Hannah Fry]: And what about what you're most looking forward to? I mean, is there a day when you'll be able to retire, knowing that your work is done? Or is there more than a lifetime's worth of work left to do?

[译文] [Hannah Fry]: 那你最期待的又是什么呢？我是说，会不会有一天你能退休，知道你的工作已经完成了？还是说剩下的工作几辈子也做不完？

[原文] [Demis Hassabis]: Yeah. I always-- well, I could definitely do with a sabbatical, and I would spend it doing science.

[译文] [Demis Hassabis]: 是的。我总是——嗯，我肯定需要休个假（sabbatical），我会把时间花在做科学研究上。

[原文] [Hannah Fry]: Just a week off, Demis.

[译文] [Hannah Fry]: 就休息一周，Demis。

[原文] [Demis Hassabis]: Yeah, so a week off, or even a day would be good. But, look, I think my mission has always been to help the world steward AGI safely over the line for all of humanity.

[译文] [Demis Hassabis]: 是的，休息一周，或者哪怕一天也好。但是，听着，我认为我的使命一直是帮助世界安全地护送 AGI 越过终点线，造福全人类。

[原文] [Demis Hassabis]: So I think, when we get to that point, of course, then there's superintelligence, and there's post-AGI, and there's all the economic stuff we were discussing and societal stuff, and maybe I can help in some way there. But I think that will be my core part of my mission, my life mission will be done. I mean, only a small job. Just get that over the line, or help the world get that over the line.

[译文] [Demis Hassabis]: 所以我想，当我们到达那个点时，当然，那之后还有超级智能，还有后 AGI 时代，还有我们讨论过的所有经济和社会问题，也许我能在那里帮上点忙。但我认为那将是我使命的核心部分，我的人生使命将完成。我是说，这只是个小工作。只是把它护送过线，或者帮助世界把它护送过线。

[原文] [Demis Hassabis]: I think it's going to require collaboration, like we talked earlier. And I'm quite a collaborative person, so I hope I can help with that from the position that I have.

[译文] [Demis Hassabis]: 我认为这需要合作，就像我们之前谈到的那样。而且我是一个相当善于合作的人，所以我希望我能利用我所处的位置为此提供帮助。

[原文] [Hannah Fry]: And then you get to have a holiday.

[译文] [Hannah Fry]: 然后你就可以去度假了。

[原文] [Demis Hassabis]: And then I'll have the-- yeah, exactly, a well-earned sabbatical.

[译文] [Demis Hassabis]: 然后我就能拥有——是的，没错，一个当之无愧的休假。

[原文] [Hannah Fry]: Yeah, absolutely. Demis, thank you so much.

[译文] [Hannah Fry]: 是的，绝对是。Demis，非常感谢你。

[原文] [Demis Hassabis]: Thanks for having me.

[译文] [Demis Hassabis]: 谢谢邀请我。

[原文] [Hannah Fry]: As delightful as always. Well, that is it for this season of "Google DeepMind: The Podcast" with me, Professor Hannah Fry. But be sure to subscribe so you will be among the first to hear about our return in 2026. And in the meantime, why not revisit our vast episode library? Because we have covered so much this year, from driverless cars to robotics, world models to drug discovery-- plenty to keep you occupied. See you soon. [THEME MUSIC],

[译文] [Hannah Fry]: 一如既往地令人愉快。好了，我是汉娜·弗莱教授，本季《Google DeepMind：播客》到此结束。请务必订阅，这样您将第一时间得知我们在 2026 年的回归。与此同时，何不重温一下我们要庞大的往期节目库呢？因为我们今年涵盖了太多内容，从无人驾驶汽车到机器人技术，从世界模型到药物研发——足够让你忙活一阵子的。再见。[主题音乐]