This is AGI

"Thank you all for being here. We do this as a service to the community because we are living through important times."

"谢谢各位今天能来。我们办这个活动,是当作对整个社区的一项服务——因为我们正身处一段重要的时代。"

"Sonya, Konstantine and I are going to say a couple of words to start. I'll say a few words of overall calibration, then Sonya will say a bit about what we see today, and then Konstantine will say a bit about what we think might be coming tomorrow."

"开场,我、Sonya(黄之熹)和 Konstantine(布勒)会先说几句。我先做一个整体的校准(calibration),然后 Sonya 讲讲我们今天看到的,接着 Konstantine 讲讲我们认为明天可能会到来的东西。"

"So for calibration we're going to start by zooming out. Going back to the silicon-based transistors which gave this area its name."

"做校准,我们先把镜头拉远。一路回到硅基晶体管——这片地方(硅谷)的名字就是它给的。"

"They got built into systems connected by networks that went public in the form of the internet, supported applications like social media and the cloud, eventually showed up in our pockets in mobile devices that today are capable of doing something indistinguishable from magic, which is AI."

"晶体管被装进系统,系统又被网络连起来,以互联网的形态走向大众,撑起了社交媒体和云这样的应用,最后变成口袋里的移动设备;而今天,这些设备能做出一件几乎与魔法无异的事——那就是 AI。"

"The reason we like to show this slide is because it reminds us that all of these waves are additive."

"我们爱放这张图,是因为它提醒我们:这一波波浪潮是叠加(additive)的。"

"And we sort of needed all of these decades of evolution to have the compute, the bandwidth, the data, the talent to make the most of this moment."

"某种意义上,正是这几十年的演化,才攒下了算力、带宽、数据和人才,让我们能把眼下这一刻用到极致。"

"Now this AI wave is a little bit different in three ways. First, it's the biggest wave yet."

"不过这一波 AI 浪潮有三处不一样。第一,它是迄今最大的一波。"

"And that's generally true, but there is something more specifically true about this wave, which is it is the first one that is both software and services."

"这话泛泛地说成立,但这一波还有一层更具体的成立之处:它是第一波同时吃下软件和服务的浪潮。"

"The top row shows the first 15 years of the cloud transition where the TAM for software went from about 350 billion to 650 billion, and cloud grew to be about 400 billion of that."

"上面一行是云迁移的头 15 年:软件的 TAM(总可触达市场)从大约 3500 亿涨到 6500 亿美元,其中云占了约 4000 亿。"

"The bottom row is what is brand new. This is the services revenue that seems to also be available now."

"下面一行才是全新的东西:现在似乎也能拿到的服务收入。"

"$10 trillion is a conveniently round number. We don't know if it's 10 trillion or 5 trillion or 50 trillion."

"$10 万亿是个凑得很整的数。我们并不知道到底是 10 万亿、5 万亿还是 50 万亿。"

"We do know that legal services in the US alone is a $400 billion market. That is one vertical and one geo. And it's the same as all of software. So this opportunity is immense."

"但我们知道,光是美国的法律服务就是一个 4000 亿美元的市场。这只是一个垂直行业、一个地区,体量却和整个软件业相当。所以这个机会大得吓人。"

"Point number two: fastest wave yet. I think we can all feel this."

"第二点:迄今最快的一波。我想这一点大家都能切身感觉到。"

"What it means is that this white space — and I direct your attention to the AI side of this page — this white space is getting filled pretty fast."

"它的意思是,这块空白地带——请大家看这一页 AI 那一侧——正在被很快地填满。"

"These logos are the companies that got to a billion-plus of revenue as a result of the cloud, mobile, and now AI tectonic shifts. And at current course and speed, there are more coming soon."

"这些 logo,是借着云、移动、如今又是 AI 这几次地壳级变动,做到十亿美元以上收入的公司。按现在的方向和速度,很快还会冒出更多。"

"Point number three, which is probably the most interesting one — and I borrowed this from my partner Konstantine — is that there are two basic kinds of revolutions in technology."

"第三点,大概是最有意思的一点——这是我从合伙人 Konstantine 那儿借来的——技术革命基本上分两种。"

"There are revolutions of communication, which are about the way information is distributed. Most of the people in this room have only lived through revolutions in communication. The internet, the cloud, mobile — those are all about information distribution."

"一种是通信(communication)的革命,关乎信息怎么被分发。在座的多数人,经历过的只有通信革命。互联网、云、移动——它们讲的都是信息的分发。"

"AI is different. AI is a revolution in computation. It's about how information is processed."

"AI 不一样。AI 是计算(computation)的革命,关乎信息怎么被处理。"

"And that might sound like semantics, but these are fundamentally different shapes of waves."

"这听上去像是在抠字眼,但它们其实是形状根本不同的两种浪潮。"

"And maybe the most visceral way to feel this is to think about the fact that the floor keeps moving underfoot. The technology foundation on which everybody builds changes every day when new capabilities come out."

"也许最直观的体会方式是:脚下的地板一直在动。每当新能力放出来,所有人赖以搭建的技术地基,每天都在变。"

"And we've had three major inflection points over the last handful of years."

"过去这几年,我们经历了三个重大的拐点。"

"First one: the ChatGPT moment, November 2022. The world saw the power of pre-training."

"第一个:ChatGPT 时刻,2022 年 11 月。全世界看到了预训练(pre-training)的威力。"

"Second one, a couple years later: the o1 model — reasoning. All of a sudden a second scaling law emerges around inference-time compute."

"第二个,两年后:o1 模型——推理(reasoning)。一下子,围绕推理时算力(inference-time compute)冒出了第二条 scaling law。"

"Third one, just recently: Claude Code. The world saw the power of long-horizon agents."

"第三个,就在最近:Claude Code。全世界看到了长程 agent(long-horizon agents)的威力。"

"And while these look like three points on a continuum, it's kind of a heartbreak between two and three. It's a little bit of a discontinuous change."

"虽然这三个看上去像是一条连续曲线上的三个点,但二和三之间其实有一道裂口,是一段不太连续的突变。"

"And if we may be so bold, we would say that this is AGI."

"如果允许我们大胆一点,我们会说:这就是 AGI。"

"And look, I'm an econ major, we're venture capitalists, not about to propose a technical definition for AGI."

"听着,我是学经济出身的,我们是做风投的,可不打算给 AGI 提一个技术定义。"

"We study founders and markets and the collision thereof, which is businesses. But we do study businesses."

"我们研究的是创始人、市场,以及二者相撞的产物——也就是生意。但生意这件事,我们是真研究的。"

"And so from a commercial standpoint, from a practical standpoint, from a functional standpoint: if you can dispatch an agent to do a job and it can recover from failure and persist until that job is done — I don't know, that feels pretty much like AGI."

"所以,从商业的角度、实用的角度、功能的角度看:如果你能派出一个 agent 去干一件活,它能从失败中恢复、并坚持到把这件活干完——我说不好,那感觉就已经挺像 AGI 了。"

"Even if you don't think it's AGI, which is fine — Sonya will talk a lot more about this in her part — I think we can all see that the cars have arrived."

"就算你不认为这是 AGI,也没关系——Sonya 在她那部分会多讲很多——但我想,我们都能看到:汽车已经到站了。"

"Last few years we've had a lot of faster horses. Applications that made you 10 or 40% more productive, but didn't fundamentally change the way you work."

"过去几年,我们见到的多是更快的马。那些应用让你的效率提升 10% 或 40%,却没有从根本上改变你的工作方式。"

"Now we're starting to see cars. Applications that make you 10 or 40X more productive. And absolutely change the way that you work. Change the nature of your work. Change the nature of your organization. Cars have arrived."

"现在,我们开始看到汽车了。那些应用让你的效率变成 10 倍、40 倍,并且彻底改变你工作的方式、改变你工作的性质、改变你组织的形态。汽车已经到了。"

"This is the founder of Sequoia, Don Valentine. He was famous for asking one question: So what? Why does all this stuff matter?"

"这位是红杉的创始人 Don Valentine(唐·瓦伦丁)。他以爱问一个问题著称:那又怎样(So what)?这一切到底为什么重要?"

"Well, it matters because just in the last few months the race has begun. And it's a different kind of race than what we're used to."

"它重要,是因为就在过去这几个月,比赛已经鸣枪了。而且这是一种和我们习惯的不一样的比赛。"

"The way you drive a car is different than the way you ride a horse. The way you build a car is different than the way you take care of a horse. So it's a very different sort of race."

"开车的方式和骑马不一样,造车的方式和养马也不一样。所以这是一场截然不同的比赛。"

"And one of the reasons that we wanted to gather everybody here today is because nobody has all the answers. And the more time we can spend together, the more we can learn and hopefully figure out where all this stuff is headed."

"我们今天想把大家聚到一起,原因之一就是:没有人手握所有答案。我们待在一起的时间越多,就能学到越多,但愿能一起搞清楚这一切到底要往哪儿去。"

"And it's important that we do so as soon as possible because there's a lot at stake. Just from a commercial perspective there's $10 trillion up for grabs."

"而且越早这么做越重要,因为赌注很大。单从商业角度看,就有 $10 万亿等着被抢。"

"We've got labs coming at it from a tech-out approach. We've got startups building on top coming at it from more of a customer-back approach."

"一边是各家实验室(labs),从"技术往外推"(tech-out)的路子切入;另一边是在上面搭建的创业公司,更多走"从客户往回想"(customer-back)的路子。"

"We do have all of the labs represented in this room, but most of you are building on top."

"在座确实各家实验室都有人,但你们大多数是在模型之上做东西的人。"

"So our advice for those of you who are building on top of the labs is free advice, and so it's worth every penny you paid for it. Our advice would be to get mad."

"所以给你们这些在实验室之上搭建的人一点建议——免费的,所以一分钱一分货。我们的建议是:get mad(发飙)。"

"And we don't actually need you to be angry. You can be angry if you want. But this is just a convenient acronym for moats, affordance, and diffusion — three pillars of a strategy for building on top of the models."

"其实我们不是真要你生气。你想生气也行。但这只是个顺口的缩写,代表 moats(护城河)、affordance、diffusion(扩散)——在模型之上搭建时,一套策略的三根支柱。"

"As a reminder, this slide shows the merchandising cycle, which is the links in the value chain required to take something from an idea to a happy customer."

"先提个醒,这张图画的是"商品化循环"(merchandising cycle)——把一个点子变成一个满意客户,所需的价值链上一环又一环。"

"The point I want to make: if you approach things from a tech-out point of view, each link in the chain gets approached a little differently. If you approach from a customer-back point of view, each link you approach a little differently."

"我想说的是:你若从"技术往外推"的视角出发,链条上每一环的打法都会略有不同;你若从"从客户往回想"的视角出发,每一环的打法又会是另一种。"

"Now here's the part that's counterintuitive. In a revolution of computation, what you want to do is look down here because there's cool new stuff coming out all the time. What you should actually do for the sake of building moats is look up here, because your customers are not changing nearly as fast as the capabilities are changing."

"接下来这点很反直觉。在一场计算革命里,你的本能是往下看——因为底层天天都有酷炫的新东西冒出来。但为了真正筑起护城河,你该做的是往上看,因为你的客户变化的速度,远远赶不上能力变化的速度。"

"The things that you built might be irrelevant tomorrow. The degree to which you wrap yourself around your customers is going to be a bit more durable."

"你今天造的东西,明天可能就没用了。但你把自己紧紧缠绕在客户身上的那个程度,会更经得起时间。"

"That's not to say the product and technology is not important. It is insanely important. And generally speaking, best product wins."

"这不是说产品和技术不重要。它重要得不得了,而且一般来说,最好的产品会赢。"

"But in a world where product changes so fast because capabilities change so fast, in thinking about moats, we would encourage you to go as customer-back as possible and think about all the ways you can wrap yourself around those customers."

"但在一个因为能力变得太快、产品也跟着变得太快的世界里,谈到护城河,我们鼓励你尽可能往"从客户往回想"那一头走,去琢磨所有能把自己缠绕在客户身上的办法。"

"Okay, the A in MAD stands for affordance. This is a term that we borrow from the design world."

"好,MAD 里的 A 是 affordance(可供性)。这是我们从设计圈借来的词。"

"A hammer is an object that has affordance. I have a two-year-old son. If I give him a hammer, he would know what to do with it. He would grab it and start hitting stuff. That's why we don't give him hammers."

"锤子就是一个带 affordance 的物件。我有个两岁的儿子,把锤子递给他,他立刻就知道该拿它干嘛——抓起来,开始到处砸。所以我们不给他锤子。"

"An object with affordance is one that doesn't need to be explained. People just know what to do with it."

"一个带 affordance 的东西,是那种不需要解释的东西:人一看就知道该拿它怎么办。"

"Claude Code is insanely powerful. Go open up a terminal for the average Fortune 500 employee and see how far they get. While it is powerful, it does not offer that much affordance."

"Claude Code 强得离谱。但你给一个普通的财富 500 强员工打开一个终端,看看他能走多远。它很强,却没提供多少 affordance。"

"That's not a knock on Anthropic, but it is an opportunity for anybody who wants to build on top — to create paths of least resistance for your specific customers and their specific problems, so that it's just brain-dead simple for them to get to the outcome they need. That's the concept of affordance."

"这不是在贬 Anthropic,而是任何想在它之上搭建的人的机会——为你那群特定客户、他们那些特定问题,造出一条阻力最小的路,让他们闭着眼睛都能拿到自己生意要的结果。这就是 affordance 的概念。"

"And then finally, the D in MAD is diffusion. And the diffusion gap is the opportunity for companies building at the application layer."

"最后,MAD 里的 D 是 diffusion(扩散)。而这道"扩散鸿沟",正是应用层公司的机会所在。"

"The rate at which capabilities are diffusing out into the market is far shy of the rate at which those capabilities are being created. And every day that the foundation models move faster than your average Fortune 500 enterprise, that gap gets bigger and that opportunity gets bigger."

"能力扩散进市场的速度,远远落后于这些能力被创造出来的速度。基础模型每多跑赢一个普通的财富 500 强企业一天,这道鸿沟就更大一分,机会也就更大一分。"

"So for moats, try to think customer-back. For affordance, try to think about creating those paths of least resistance for your customers. And that diffusion gap, that represents your opportunity."

"所以,护城河,就尽量从客户往回想;affordance,就去为客户造那条阻力最小的路;而那道扩散鸿沟,就是你的机会。"

"Unless that slide from earlier with the white space starting to fill up was discouraging for anybody, may we remind you that no lead is safe."

"如果刚才那张"空白正在被填满"的图让谁有点泄气,请允许我们提醒一句:没有任何领先是安全的。"

"There's this expression in racing: You cannot pass 15 cars in the sun, but you can pass 15 cars in the rain."

"赛车圈有句话:大晴天里你超不过 15 辆车,但下大雨时你能一口气超 15 辆。"

"And right now there is a torrential downpour of new capabilities coming out of the foundation models. Which means that no lead is safe, but it also means that anybody can win. What a time to be alive. And with that, I'll hand it off to Sonya."

"而此刻,基础模型正下着一场新能力的倾盆暴雨。这意味着没有任何领先是安全的,但也意味着任何人都可能赢。能活在这个时代,真好。说到这儿,我把话筒交给 Sonya。"

"Thank you, Pat. And can I just say it's so nice to see so many friendly faces in the audience. There is an exceptional group of people here today, and I'm just really happy to be part of this ecosystem with all of you."

"谢谢 Pat。我得说,看到台下这么多熟悉友善的面孔,真好。今天在场的是一群非常出色的人,能和你们一起身处这个生态里,我真的很开心。"

"And so, the purpose of my section is to talk about what's happening in AI right now, which for 2026 is agents."

"我这一节的任务,是讲讲 AI 此刻正在发生什么——而 2026 年的关键词,就是 agent。"

"Okay, flashback to 2022. Show of hands, does anybody here remember AutoGPT or BabyAGI?"

"好,把时间倒回 2022 年。举个手,在座有人还记得 AutoGPT 或者 BabyAGI 吗?"

"So these projects were overnight hits on GitHub, and what they did was they took GPT-3, gave it some tools, wrapped it in a loop, and let it run towards a goal."

"这些项目当年在 GitHub 上一夜爆红。它们做的事是:拿来 GPT-3,给它配几样工具,套进一个循环里,然后放它朝一个目标跑。"

"And it was promising until you watched those agents just fail over and over and over again. Kind of cute, kind of endearing, but completely useless."

"一开始挺有希望,直到你眼看着那些 agent 一次又一次又一次地失败。挺可爱、挺招人疼,但完全没用。"

"And I put this slide here to remind us that we all knew agents were coming. We could have seen it years ago, but back in 2022, the models just weren't ready yet."

"我放这张图,是想提醒大家:我们其实都知道 agent 终会到来。几年前就该看出来了,只不过 2022 年那会儿,模型还没准备好。"

"Fast forward to today, something around the turn of the year really changed. Suddenly we have agents everywhere around us and they seem to actually be working."

"快进到今天,大概是年关前后,有什么东西真的变了。突然之间,我们身边到处都是 agent,而且它们看起来是真能干活了。"

"Two agents in particular have been home runs. Claude Code for the technical crowd, and open claw and all of its lobster brethren, which democratized agents to anybody with a phone."

"其中有两个 agent 堪称全垒打。一个是面向技术人群的 Claude Code;另一个是 open claw 以及它那一堆"龙虾兄弟",把 agent 普及给了任何一个有手机的人。"

"And so whether you are a hardcore engineer or a normie, the punchline is that anybody can create agents now."

"所以,不管你是硬核工程师还是普通人(normie),结论是:现在任何人都能造 agent 了。"

"And so what we're seeing is people are building agents for everything. There is silly stuff like an open claw agent that will literally snitch on your neighbors for tax fraud. Please don't do this. Or actually, maybe please do this."

"于是我们看到的是:人们在给一切事情造 agent。有些很无厘头,比如一个 open claw agent,真的会去举报你邻居偷税漏税。拜托别这么干。或者……其实,也许真该这么干。"

"There's entrepreneurial stuff. Agents running generative media campaigns to sell construction services."

"也有创业向的:用 agent 跑生成式的媒体投放,去卖建筑施工服务。"

"And then there's the professional layer. I can tell you there's a huge race internally at Sequoia for who can build the best agents to do our jobs better."

"再往上是专业级的。我可以告诉你,红杉内部正掀起一场大比拼:看谁能造出最好的 agent,把我们自己的活干得更漂亮。"

"So what does it mean to be an agent? Here is one possible definition. An agent is a system that perceives its environment, chooses actions, and progresses autonomously towards a goal."

"那么,所谓 agent 到底是什么?这里给一个可能的定义:agent 是一个能感知所处环境、选择行动、并朝着一个目标自主推进的系统。"

"By the way, guys, I made this in Sora by myself. I'm very proud of it. The video models have come a long way."

"顺便说一句,各位,这个(动画)是我自己用 Sora 做的,我可得意了。视频模型这一路真是进步神速。"

"And more specifically, I view agents as having three functional components. First is the ability to reason and plan. This is the baseline level of intuition and the ability to think on the fly."

"再具体一点,我把 agent 看作有三个功能组件。第一,推理和规划(reason and plan)的能力——这是底层的直觉,以及临场思考、随机应变的本事。"

"Second is the ability to take actions. This is tools, search, write, compile."

"第二,采取行动的能力——也就是工具:搜索、写入、编译。"

"And then finally, the ability to iterate towards a goal. This is the persistence that gives agents the ability to accomplish things over long time horizons."

"最后,朝目标反复迭代的能力——正是这份"持续性"(persistence),让 agent 能在很长的时间跨度上把事情办成。"

"And so agency combines these three things. It is simply the ability to get [stuff] done."

"所以,所谓"能动性"(agency),就是这三样东西的合体。说白了,就是把事情搞定的能力。"

"If we boil the agents down into their constituent components — the models, the tools, the harnesses — each component has progressed rapidly over the last year."

"如果把 agent 拆解成它的几个组成部件——模型(models)、工具(tools)、harness——过去这一年里,每个部件都飞速进步。"

"First, the models are the brain. This is the most important thing that's happened. The METR chart measures how long a model can sustain progress on a complex task without going off the rails. And we've gone from the order of tens of minutes a year ago to the order of hours today."

"首先,模型是大脑。这是最重要的一件事。METR 那张图衡量的是:一个模型在一项复杂任务上,能在不"翻车跑偏"的前提下持续推进多久。我们已经从一年前的"几十分钟"量级,走到了今天的"小时"量级。"

"And so this is the most important thing that's happened. The models are finally getting capable enough to sustain performance on long-horizon tasks."

"所以这真的是最重要的一件事:模型终于变得足够强,能在长程任务上维持住表现。"

"Second, the tools are the arms and the legs. These give models access to things that make us productive on the computer. The terminal for file systems and dev tools, iMessage, Slack, web search, computer use, you name it."

"其次,工具是手和脚。它们让模型能够触达那些让我们在电脑上高产的东西:管文件系统和开发工具的终端、iMessage、Slack、网页搜索、computer use,等等等等。"

"And the last two decades that we spent building tools for humans have ended up being able to port over to be incredibly useful for agents as well."

"过去二十年我们为人类打造的那些工具,结果发现可以原样迁移过来,对 agent 一样好用得不得了。"

"And there's a common refrain that SaaS is dead. I think to the contrary, the value of these tools is going to explode as the number of agents using them increases."

"现在流行一种说法:SaaS 已死。我反倒认为正相反——随着使用这些工具的 agent 越来越多,它们的价值会爆炸式增长。"

"Models and tools give agents capability. The harness is what gives them persistence — the ability to stay on task, adapt, and keep going. And that feedback loop is now really starting to crank."

"模型和工具,给了 agent 能力;而 harness,给的是持续性——盯住任务、随机调整、一直跑下去的能力。这个反馈循环,如今真的开始转起来了。"

"Especially now with reinforcement learning, we're taking them to driving school, training them in RL gyms, and we're pushing performance in different settings from mechanical engineering to design to finance."

"尤其是现在有了强化学习(reinforcement learning),我们等于是把它们送进驾校,在 RL gym(强化学习训练场)里调教,在从机械工程到设计再到金融的各种场景里把表现往上顶。"

"We're also seeing the early glimmers of self-improvement, or the machine building the machine. For example, Andrej's other research project improves research autonomously towards a GPT-2 level model in just 2 hours."

"我们也开始看到自我改进——也就是"机器造机器"——的早期苗头。比如 Andrej(Karpathy)的另一个研究项目,能自主地把研究往前推,仅用 2 小时就训练出一个 GPT-2 级别的模型。"

"So what does the world of agents everywhere look like? Agents exist on a sliding scale of agenticness."

"那么,一个"到处都是 agent"的世界长什么样?agent 存在于一条"agent 化程度"(agenticness)的滑动刻度上。"

"And so let's take coding as an example. In 2023, we had tab autocomplete, one AI assisting a human in line. This was incrementally useful, fundamentally not transformative."

"拿写代码举例。2023 年,我们有的是 tab 自动补全——一个 AI 在行内辅助一个人。这有点用,是增量式的,但本质上谈不上变革。"

"We now have agentic development, one human talking to an agent, instructing it what to do, maybe managing a team of agents."

"现在我们有了 agentic 开发:一个人对着一个 agent 说话、指挥它干什么,甚至管理一支 agent 小队。"

"But this paradigm is getting pushed further. We're now seeing background agents, async agents, agents spawning sub-agents."

"但这套范式正被往前推。我们已经看到后台 agent、异步(async)agent,以及 agent 自己派生出子 agent。"

"We think that async agents in this whole paradigm is likely to overtake the current paradigm in volume just because of the amount of leverage in the system."

"我们认为,光凭这套系统里的杠杆(leverage)之大,异步 agent 这一范式很可能在体量上盖过当下这套范式。"

"And then finally, pushing the bleeding edge of the frontier, what I call dark factories — taking human review out of the system completely."

"最后,顶在最前沿的刀尖上的,是我称之为"黑灯工厂"(dark factories)的东西——把人工审查彻底从系统里拿掉。"

"This sounds crazy, but I've seen it happen in production, including with cybersecurity companies. It is possible with good enough guardrails and good enough engineering."

"这听上去很疯,但我已经亲眼见它在生产环境里跑起来,包括在一些网络安全公司里。只要护栏(guardrails)足够好、工程足够扎实,它就是可行的。"

"So we're progressing up a scale of agenticness. Agents are going from little helpers that do a little amount by your side, to interns that need to be managed, to interns that manage themselves. And eventually to interns that can be trusted enough to push to prod without oversight."

"所以我们正沿着这条 agent 化刻度往上爬。agent 正从"在你身边帮点小忙的小助手",变成"需要被管的实习生",再变成"会自我管理的实习生",最终变成"靠谱到可以无人监督就直接推到生产环境(push to prod)的实习生"。"

"And so that's the evolution that's happening not just in coding, but across all of agents."

"而这场演化,不只发生在写代码这一件事上,而是横跨所有 agent 在发生。"

"The most important takeaway for the founders in this room is that services is the new software."

"对在座创始人来说,最重要的一条结论是:服务,就是新的软件(services is the new software)。"

"Pat's been saying this for as long as I've known him. And our partner Julian, who's in the audience today as well, published a great article on this. We've known this for a long time, but I think it's actually happening."

"从我认识 Pat 起,他就一直在说这句话。我们的合伙人 Julian——他今天也在台下——还为此写过一篇很棒的文章。这个判断我们早就有了,但我觉得它现在是真的在发生。"

"So in medicine, you're able to hire an agent that inspects your genome, gives you personalized recommendations, can prescribe you medication, recommend you clinical trials."

"比如医疗:你可以雇一个 agent,它会检视你的基因组、给你个性化建议、能给你开药、还能给你推荐临床试验。"

"And law, you'll be able to hire agents that can negotiate contracts on your behalf, even perform litigation and settle for you."

"再比如法律:你将能雇到替你谈合同的 agent,它甚至能替你打官司、替你和解。"

"In math and the sciences, we're seeing agents that can solve Erdős problems or discover new superconductors. Like how thrilling is that?"

"在数学和科学领域,我们看到 agent 能解 Erdős(埃尔德什)难题,或者发现新的超导体。这有多让人激动?"

"Or in the consumer world, personal agents that can manage your inbox for you, your calendar, your finances, file your taxes."

"又或者在消费端:个人 agent 能替你打理收件箱、日历、财务,还能替你报税。"

"And we expect there's going to be agents everywhere, and that's in part because hiring agents is so much easier than hiring employees."

"我们预计 agent 会无处不在,部分原因是:雇 agent 比雇员工容易太多了。"

"Humans are hard to scale. Agents are infinitely scalable with compute."

"人很难规模化。agent 只要有算力,就能无限扩展。"

"Humans are hard to keep happy — except for me, I'm always happy. Agents are low maintenance."

"人很难一直哄开心——除了我,我永远很开心。agent 则几乎不用伺候。"

"Humans are expensive. You pay them salaries. You pay agents tokens. Generally, it costs less to accomplish a task with tokens than the equivalent in salary."

"人很贵,你得给他们发工资;而你给 agent 付的是 token。一般来说,用 token 完成一件任务的成本,低于等量工资。"

"Today, humans are still generally smarter, but the bitter lesson presses on, and soon agents will be smarter at many things."

"今天,人整体上还是更聪明;但"惨痛的教训"(the bitter lesson)在持续应验,很快,agent 在许多事情上会比我们更聪明。"

"And so the point of this slide is not that we humans are out of a job. I think a uniquely human trait is adaptability."

"所以这张图想说的,不是我们人类要失业了。我认为人类有一项独有的特质,叫适应力(adaptability)。"

"But we do expect the deployment of agents across the application layer to be swift and at an unprecedented rate and scale, because the economics are so clear and because of the inherent scalability of bits."

"但我们确实预计,agent 在应用层的铺开会很快,而且其速率与规模前所未有——因为经济账太清楚了,也因为比特(bits)天生就能规模化。"

"So if you add all this up, the number of agents is ballooning on some sort of exponential, maybe super-exponential."

"所以把这些加在一起,agent 的数量正沿着某种指数曲线膨胀,也许是超指数。"

"And I think we're about to hit the point where things get genuinely strange. What happens when commerce happens between agents? Can they pay each other? What happens when agents can actually negotiate the terms of a transaction with each other?"

"我觉得我们就快撞上一个真正变得诡异的临界点。当商业行为发生在 agent 之间,会怎样?它们能彼此付款吗?当 agent 真的能互相谈一笔交易的条款,又会怎样?"

"Are we going to have swarms of agents policing us, preventing things like cybersecurity Armageddon?"

"我们会不会有成群结队的 agent 在监管我们、阻止类似"网络安全末日"(cybersecurity Armageddon)那样的事发生?"

"All we know is the world is getting weird extremely quickly."

"我们唯一确定的是:世界正以极快的速度变得怪异。"

"And so I'll close by channeling my inner Bene Gesserit. Long-horizon agents are here. The curve that they're on is very clear."

"那我就附身一下我内心的 Bene Gesserit(《沙丘》里的姐妹会)来收尾吧。长程 agent 已经来了,它们所在的那条曲线,非常清楚。"

"And for founders, I think everybody has examples of people that are accomplishing insanely hard timelines thanks to AI."

"对创始人来说,我想每个人手上都有这样的例子:有人靠着 AI,把难到离谱的时间线给做到了。"

"So Nathan from Zed accomplished a three-year moonshot project over the holidays by himself with Claude Code. Bret Taylor rebuilt Sierra over a weekend. The Notion team rewrote 8 million lines of code in just 6 weeks."

"比如 Zed 的 Nathan,假期里靠 Claude Code 一个人干完了一个原本三年的登月级项目;Bret Taylor 一个周末重建了 Sierra;Notion 团队仅用 6 周就重写了 800 万行代码。"

"And so everybody has these examples of compressed timelines."

"所以人人都有这类"时间线被压缩"的例子。"

"But I think very few people outside of the AGI labs have seen what happens when you take these compressed timelines and you stack them on top of each other. And that's what's possible now."

"但我认为,在 AGI 实验室之外,极少有人见过:当你把一条条被压缩的时间线,一层层叠起来,会发生什么。而这,正是现在能做到的事。"

"And so whatever you could imagine building over the next 100 years, we think is now possible in 100 days thanks to agents. I will pass it over to Konstantine."

"所以,任何你能想象在未来 100 年里建成的东西,我们认为,如今靠 agent,在 100 天里就能建成。我把话筒交给 Konstantine。"

"Thank you so much, Sonya, Pat, for the brilliant overview and analysis. In this section, we're going to talk a little bit about what's next."

"非常感谢 Sonya、Pat 这番精彩的总览和分析。在这一节,我们来聊聊接下来会怎样。"

"So the goal here is: we all know we're in an AI age. What's it going to look like? What's it going to feel like? How's it characterized?"

"我的目标是:我们都知道自己身处一个 AI 时代。那它会是什么样子?会是什么感觉?该怎么去刻画它?"

"Earlier in the presentation, Pat bifurcated technological revolutions between compute and communication. We're going to do another bifurcation here for types of work."

"刚才 Pat 把技术革命一分为二:计算与通信。这里我们再做一次"二分",针对工作的类型。"

"There is physical work. This is a package on the Pony Express. This is a satellite on a Falcon 9. Work equals force times distance — physical movement."

"一类是体力工作。这是 Pony Express(美国早期快马邮递)上的一个包裹,这是猎鹰 9 号(Falcon 9)上的一颗卫星。功等于力乘以距离——也就是物理上的移动。"

"And then there's cognitive work. This is Pythagoras coming up with his theorem. This is DeepMind solving the protein-folding problem. Conscious thinking."

"另一类是认知工作。这是 Pythagoras(毕达哥拉斯)想出他的定理,这是 DeepMind 攻克蛋白质折叠难题。是有意识的思考。"

"These are very different types of work. But we believe that they're going to follow a very similar pattern in revolution."

"这是两类很不一样的工作。但我们相信,它们的革命会遵循一个非常相似的模式。"

"So let's talk about physical work, because we've been through this revolution with the Industrial Revolution."

"那先说体力工作,因为这场革命我们已经经历过一次了——工业革命。"

"For the vast majority of human history, virtually all the work for serving humans was done by some sort of muscle. People or animals. People moving something, or an animal pulling the human along. This starts at 1700, but it goes back millennia."

"在人类历史的绝大部分时间里,几乎所有为人服务的工作,都靠某种肌肉完成——人或牲畜:人搬动东西,或牲畜拉着人走。这条线从 1700 年算起,但其实可以追溯到几千年前。"

"Then things started to change. Water and wind. Steam engines. And then things accelerated. Steam engines, combustion, electric motors."

"然后情况开始变了。水力和风力。蒸汽机。接着一切加速:蒸汽机、内燃机、电动机。"

"Today, 2026, you could estimate that 99-plus percent of all the physical work done on planet Earth for humans is done by machine."

"今天,2026 年,你可以估算:地球上为人完成的所有体力工作里,99% 以上是由机器干的。"

"The plane that brought you here, the manufacturing of all the goods in this room, all the transportation that sets up for the pinnacle of the human experience you're having right now."

"把你送来这儿的飞机、这屋里所有商品的制造、撑起你此刻这场巅峰人类体验的全部运输——都是。"

"Well, we think a similar pattern's going to happen in cognition. We're just a little earlier on."

"而我们认为,认知领域会上演一个相似的模式,只不过我们现在还处在更早一点的阶段。"

"So for most of human history, all the thinking on planet Earth for humans was done primarily by humans. Maybe a little bit for animals — the sheepdog chasing the sheep, right? And there was this sliver on top of mechanical work, the astrolabe or the clock."

"在人类历史的大部分时间里,地球上为人所做的全部思考,主要由人来完成。或许还有一丁点是为动物的——牧羊犬追羊,对吧?另外还有薄薄一层是机械式的运算,比如星盘(astrolabe)或时钟。"

"Now, over the past couple hundred years, there was not a lot of progress until electronic computation. And in the past 100 years, think about all the trillions of calculations that are happening at any given moment to serve you the human."

"在过去这两三百年里,直到出现电子计算之前,进展都不大。而在过去这 100 年里,想想此刻每一瞬间,为你这个人服务而正在进行的、数以万亿计的运算。"

"All of that work, all of that cognitive work that's happening to serve us at any given moment. Trillions of calculations."

"所有这些工作,所有这些此刻正为我们服务的认知工作。数以万亿计的运算。"

"We believe that the neural network is the next big wave. And that in the near future, 99.9% of cognition on planet Earth will be done by machines."

"我们相信,神经网络就是下一个大浪潮。而在不远的将来,地球上 99.9% 的认知,将由机器完成。"

"Well, the parallel is pretty stark. And the good news is we've been through a revolution like this. The cognitive revolution is going to be a lot like the Industrial Revolution. Just much, much bigger and much faster."

"这个类比相当鲜明。好消息是,这样的革命我们经历过。认知革命会很像工业革命,只是规模大得多、速度也快得多。"

"So what's it going to be like living in this future? I'd like to share some motivations for this future in the form of four short stories."

"那么,活在这样的未来里,会是什么样?我想用四个小故事,来讲讲这个未来值得期待的理由。"

"The first story. In the mid-1800s, America wanted to build a grand monument to our first president and our greatest war hero, George Washington."

"第一个故事。19 世纪中叶,美国想为我们的首任总统、最伟大的战争英雄 George Washington(华盛顿)建一座宏伟的纪念碑。"

"So we designed the tallest building in the world at the time, the Washington National Monument, and we wanted to cap it with the most precious metal in the world — 100 oz of the most precious metal in the world."

"于是我们设计了当时世界上最高的建筑——华盛顿纪念碑,并打算在顶端封上全世界最珍贵的金属:100 盎司全世界最珍贵的金属。"

"So precious, in fact, that we put it on display at Tiffany's in Manhattan. That metal was aluminum."

"珍贵到什么程度?我们把它放在曼哈顿的 Tiffany's(蒂芙尼)橱窗里展出。那种金属,是铝。"

"Within decades of the completion of the Washington National Monument, a young inventor came up with electrolysis, the process of separating aluminum from dirt. And within decades, aluminum was used to wrap our candies and our sandwiches, and then tossed into the trash."

"在华盛顿纪念碑落成后的几十年内,一位年轻的发明家想出了电解法(electrolysis)——把铝从泥土里分离出来的工艺。再过几十年,铝被用来包我们的糖果、三明治,然后随手扔进垃圾桶。"

"Aluminum is intelligence. Electrolysis is artificial intelligence."

"铝,就是智能。电解法,就是人工智能。"

"We're about to enter a world where some of the most precious skills that took decades to earn — PhD-level skills — are so instantly invoked that right after using them, you can crumple them up and throw them right in the trash."

"我们即将进入这样一个世界:一些最珍贵、要花几十年才能习得的技能——博士级别的技能——会变得唾手可即,以至于一用完,你就能把它揉成一团,直接扔进垃圾桶。"

"Story number two. We are entering a world of alien design. The world as we see it today is all about design for humans. It's been optimized in a way that makes sense to our brains because we are doing almost all the cognition in the world."

"第二个故事。我们正进入一个"异形设计"(alien design)的世界。我们今天所见的世界,全是为人而设计的。它被优化成我们大脑能理解的样子,因为世界上几乎所有的认知都是我们人在做。"

"Well, when machines do the cognition, it's going to be a little different."

"可一旦换成机器来做认知,事情就会有点不一样了。"

"In 2006, NASA was optimizing an antenna for a large satellite space mission. And traditionally, their antennas looked like this — a beautiful geometric, symmetrical pattern that optimized surface area for some power constraints."

"2006 年,NASA 为一项大型卫星太空任务优化一根天线。传统上,他们的天线长这样——一种漂亮的、几何对称的图案,在某些功率约束下优化表面积。"

"This time around, they said, 'We're going to hand it over to computer and we're going to have an evolutionary algorithm.' A lot like reinforcement learning. The result: this antenna right here. Dramatically more productive. Not intuitive to the human mind."

"这一次,他们说:"我们要把它交给计算机,用一套进化算法(evolutionary algorithm)。"——和强化学习很像。结果,就是眼前这根天线:效能高出一大截,却完全不符合人脑的直觉。"

"In this AI era, when we hand over cognition to machines, we're going to get results that are not intuitive to us. When AI's designing chips, cars, buildings, they might look dramatically different."

"在这个 AI 时代,当我们把认知交给机器,我们会得到一些不符合自己直觉的结果。当 AI 来设计芯片、汽车、建筑,它们的样子可能会大不相同。"

"The world that we enter into, we have to be open-minded, because the AI is not going to think like us. It's going to have alien design."

"对我们将要进入的这个世界,我们必须保持开放的心态,因为 AI 不会像我们一样思考。它有的是异形的设计。"

"The third motivation story is on emerging sciences. Not emerging science. We all know there's emerging science. I'm talking about emerging sciences."

"第三个故事,讲的是"新兴的那些科学"(emerging sciences,复数)。不是"新兴科学"(单数)——新兴科学谁都知道有。我说的是会有全新门类的科学诞生。"

"In the early Industrial Revolution, you had great engineers like Newcomen and Watt. And they perfected combustion engines. Basically, put a petrochemical into a piston, ignite it on fire, millions, billions of particles explode, move the piston, work."

"工业革命早期,有 Newcomen(纽科门)和 Watt(瓦特)这样伟大的工程师,他们把燃烧式发动机做到了极致。原理大致是:把石化燃料注入活塞,点火,数百万、数十亿颗粒子爆炸,推动活塞,做功。"

"For almost 100 years, all of that was tinkering. It was an engineer saying, 'Ah, that works a little bit better.' Maybe something you could see like a scaling law, but it was engineers playing with the product and seeing how they could improve it a little bit."

"差不多有 100 年,这一切都只是"瞎鼓捣"(tinkering):一个工程师说"啊,这样好像好一点点"。也许其中能看出某种类似 scaling law 的东西,但本质上,是工程师在摆弄产品,看怎么能改进一星半点。"

"Over 120 years later, Sadi Carnot came around and formalized this in a new science, thermodynamics. He said, 'Wait a second. There are millions or billions of particles. We can actually formalize what that all looks like.'"

"120 多年后,Sadi Carnot(萨迪·卡诺)出现了,把这一切形式化成一门新科学:热力学。他说:"等一下,这里有数百万、数十亿颗粒子,我们其实可以把这整个图景形式化地写出来。""

"In this case, there are billions of neurons, trillions of tokens. Right now, we're in the tinkering phase of AI. Even if we think it's an understood science, it's not."

"放到 AI 上,这里有数十亿个神经元、数万亿个 token。眼下,我们正处在 AI 的"瞎鼓捣"阶段。哪怕我们以为它已经是一门被理解的科学,其实并不是。"

"In the future, we will have a science as fundamental as thermodynamics introduced in the next couple decades. Someone in this room might come up with that science. And that science will be taught in high schools. It will be that fundamental. And it will help us master AI. It will even help us master consciousness."

"未来,在接下来的几十年里,会诞生一门像热力学一样根本的科学。也许就是在座的某个人想出来的。那门科学会被写进高中课本——它就有那么根本。它会帮我们驾驭 AI,甚至帮我们驾驭意识(consciousness)。"

"Fourth story. The art of unreason. So for the vast majority of human history, tens of thousands of years, art has been a progression towards realism."

"第四个故事。非理性之艺术(the art of unreason)。在人类历史的绝大部分时间——数万年里——艺术一直是一条朝着写实(realism)前进的路。"

"This is a cave painting from about 25,000 years ago. Egyptian hieroglyphs. Greek pottery. Renaissance paintings — a grand transformation toward realistic art. Just look at the difference over tens of thousands of years, the triumph of humanity."

"这是一幅大约 25000 年前的洞穴壁画。埃及象形文字。希腊陶器。文艺复兴绘画——一场朝着写实艺术的宏大演进。看看这数万年间的差别,这是人类的凯歌。"

"And then engineering came along. The daguerreotype, early photography. And all of a sudden, what was spent decades of life to perfect — the skill of getting every brushstroke perfect..."

"然后,工程登场了。银版照相法(daguerreotype),早期摄影。一夜之间,那项耗尽数十载人生才臻于完美的技艺——把每一笔笔触画得分毫不差的本事——(就被颠覆了)。"

"So how did the world react? They thought that painting was over. 'Oh, that's it. The machine can do it better than any human. Art is ended.'"

"那世界是怎么反应的?人们以为绘画完了:"哦,到头了。机器干得比任何人都好。艺术终结了。""

"Well, what happened? How did humans respond? Humans responded by saying, 'Was the purpose of this art to capture the moment in the way the eye sees it? Or was it to capture the moment in the way the heart and the soul sees it?'"

"结果呢?人类是怎么回应的?人类回应说:"这门艺术的目的,是按眼睛所见去捕捉那一刻吗?还是按心与灵魂所见去捕捉那一刻?""

"Impressionism, expressionism, cubism, neo-expressionism. All these new forms of art are how humanity responded to this dramatic change in science."

"印象派、表现主义、立体主义、新表现主义。所有这些新的艺术形式,都是人类对科学这场剧变的回应。"

"2,500 years ago, Greek philosopher Protagoras wrote, 'Man is the measure of all things.' What he meant is that nothing in a vacuum has value to humans. Not aluminum, not art, not intelligence."

"2500 年前,希腊哲学家 Protagoras(普罗泰戈拉)写下:"人是万物的尺度。"他的意思是:任何东西在真空里,对人都没有价值。铝不行,艺术不行,智能也不行。"

"It only has value because of the experience. AI can do the work. AI will do the work. But only the human connection can give you a reason to care."

"它之所以有价值,只因为那份体验。AI 能做这些活,AI 也将会做这些活。但唯有人与人的连接,才能给你一个"在乎"的理由。"

"That's why we're all in this room today. A decade from now, work is going to be dramatically different. Things are going to change so much, but the one thing that will be constant is the relationships that you form today with the person right next to you will endure."

"这正是我们今天齐聚一堂的原因。十年后,工作会大不相同,太多东西都会改变;但有一样东西恒定不变:你今天和身边这个人结下的关系,会长久留存。"

"That's what you're going to look back on. That's what's going to be valuable from today. So I encourage you to form those relationships with the people next to you. Enjoy your time together at this AI Ascent, and really lean into what makes us most human."

"那才是你日后会回望的东西,才是今天这一刻真正有价值的东西。所以我鼓励你,去和身边的人建立那些关系。好好享受在这场 AI Ascent 相聚的时光,并真正地、用力地拥抱那些让我们最像人的东西。"