Modelos Fundacionais São uma Commodity | Benedict Evans na a16z

Mobile didn't need to wait for the internet. 移动领域不需要等待互联网。 The internet didn't need [music] to wait for PCs and PCs didn't need to wait for consumer electronics and semiconductors and so on. 互联网不需要等待PC，PC不需要等待消费电子和半导体，以此类推。 So, you've always got this accelerating adoption. 所以，技术采用始终在加速。 Benedict Evans is a tech analyst known for his presentation AI eats the world. Benedict Evans是一位科技分析师，以其演讲《AI吃掉世界》而闻名。 He sees AI differently than the world, spotting patterns others miss, and dives into how people really use AI. 他对AI的看法与众不同，善于发现别人错过的规律，深入探讨人们实际使用AI的方式。 They built this amazing piece of incredibly sophisticated very expensive global infrastructure [music] with enormous growth in use all the time and it changed all of our lives and we all pay for it and they didn't make any money from it because all the value moved up stack. 他们建造了这套令人叹为观止的、极其复杂、耗资巨大的全球基础设施，使用量不断攀升，改变了我们所有人的生活，我们都在为此付费，可他们自己却没赚到钱，因为所有价值都向上层迁移了。 The place that's got product market fit right now is coding and swap's gone from whatever it was 9 billion run rate at the end of last year to $47 billion run rate now. 眼下真正实现产品市场契合的领域是编程，Anthropic的年化营收已从去年年底的约90亿美元飙升到如今的470亿美元。 That's all software isn't it? 那全都是软件，不是吗？ So what happens when someone else in some other field gets something working? 所以，当其他领域的人也把某些东西跑通之后，会发生什么？ One of the characteristics of tech is that the moment that you understand something and you know what's going to happen is the moment you should move on to something else. 科技行业有个特点：当你真正理解了某件事、知道接下来会发生什么的那一刻，就是你应该转向下一个问题的时刻。 Yo, Google said that the risk of underinvesting is riskier than overinvesting. 谷歌说过，低估投资的风险比过度投资的风险更大。 Investors are kind of looking at all these companies and saying 投资者正在审视所有这些公司，然后说…… Benedict welcome back to the Asz podcast. Benedict，欢迎回到a16z播客。 Thank you. 谢谢。 Last time you were here we were discussing the first iteration of your presentation AI the world. 上次你来，我们讨论的是你那篇演讲《AI吃掉世界》的第一个版本。 uh you know you since wrote it almost a you know year and a half ago 嗯，你写那篇差不多已经是一年半前了 at at this point we're get 到现在这个阶段我们 you always begin your presentation with your 你每次演讲都以你的 you know what are the big questions but I'm curious this time first before getting to the what are the questions going forward 你知道，你会先列出那些大问题，不过这次我好奇，在我们聊接下来有哪些问题之前， I want you to reflect on what have we learned since you originally uh made the presentation what's played out um and let's reflect back on 我想让你先回顾一下：自从你最初做那场演讲以来，我们学到了什么，哪些判断已经验证，来，我们先回顾一下 what's changed in the last year so I think we have much more of a sense of diverging product strategy 过去一年发生了哪些变化，我觉得我们更清楚地看到了产品策略的分化， we have much more a s of a sense of kind competitive tension that goes beyond just make a bigger model faster with more more compute. 我们也更清楚地感受到了竞争博弈，而不仅仅是做一个更大、更快、算力更多的模型。 Um we've had several iterations of open AI strategy in particular from sort of everything all at once yesterday to oops no maybe we should double down on coding. OpenAI的策略经历了好几次迭代，从某种昨天一口气全做，到糟糕，也许我们应该在编程上加大投入。 um clearly agentic coding started working and so all the focus in tech has kind of narrowed in massively onto that as something that has absolute product market fit in the sense that like the customers are pulling it out of your hands. 智能体化编程显然开始奏效了，于是整个科技圈的注意力大规模聚焦到这个方向，它是真正实现了产品市场契合的领域，用户简直是把产品从你手里夺走。 Um and um and of course that comes with the supply crunch around capacity and price imbalance imbalance to supply demand capacity capex pricing that we see at the moment. 当然，这也带来了产能短缺问题，围绕产能、定价、资本支出，供需之间存在明显失衡。 Um so that's kind of the big shift like we had a moment of like this is kind of sort of working and kind of exciting but we're not quite sure what we're going to do with it to like right it works for coding 所以这大概是最大的转变，我们从某种程度上似乎有点用、挺令人兴奋，但不太清楚能拿来做什么，转变为好，编程这块是真管用， um will it work for anything else like yes almost certainly but that's what's working right now and so that's become we've got this kind of much narrower focus um otherwise um you know the chart on numbers keep coming up the models keep getting bigger the capex keeps growing the usage keeps growing people using this more but most of the sort of fundamental questions you might have had two or few years ago didn't really have answers like we don't know if there'll be a winner in the models. 至于其他领域，几乎可以肯定也会有效果，但现在能跑通的就是编程，所以焦点大幅收窄。此外各项数字持续上升，模型越来越大、资本支出持续增加、使用量也在增长，但那些两三年前就摆在那里的根本性问题，基本上还没有答案：模型领域会不会出现赢家通吃？ We don't know if they can capture value up the stack. 我们不知道模型公司能否沿价值链向上捕获价值。 We don't know how much the models can do. 我们不知道模型究竟能做到什么程度。 Um we don't see a way that consumers will use this daily rather than weekly with the technology we have right now. 以现有技术来看，我们看不到消费者每天而非每周使用这个东西的路径。 So all of those questions are still open. 所以那些问题依然悬而未决。 Yeah. 是的。 And just on on the on the coding, how could could we have figured 对了，说回编程这块，我们能不能预见到， could we have foreseen that that would have been the the the use case that really would have taken off or what's sort of reflection on that? 或者说，我们原本能不能预见到，编程会是真正起飞的那个应用场景？对此有什么反思？ Well, um deterministically you could have said, well, look, who's messing about with this stuff? 嗯，从确定性的角度来看，你可以说，好，看看谁在摆弄这些东西？ Software developers. 软件开发者。 What are software developers going to try and make work software 软件开发者会努力让什么东西跑通？软件 [snorts] [笑声] Um, so you know at a very kind of simplistic naive level 嗯，所以从一个非常简单、朴素的层面来说， well yeah the stuff that should work is software develop first is software development just as like kind of I often compare this moment to like the internet in like 9798 but it's also like the PCs in the early 80s or the late '7s. 好，最先应该起效的是软件开发，就像我经常把这个时刻比作97、98年的互联网，也像八十年代初、七十年代末的PC时代。 It's incredibly exciting but it's not quite clear what it's for and it doesn't quite work yet and clearly the first thing that people did with PCs was make computers. 那是极其令人兴奋的时刻，但不太清楚究竟是用来做什么的，也还没有完全跑通，而显然，人们最先用PC做的事情是做电脑。 Um, and the first thing that people are doing with LLMs, in a sense, LLMs are computers, is to make more compute. 从某种意义上说，LLM就是电脑，而人们最先用LLM做的事情，是制造更多的计算能力。 Um, and so that's not terribly surprising. 嗯，所以这并不令人特别意外。 I think the shift is been that the beginning of this year clearly that agentic coding went from being kind of useful to really changing everything. 我认为转折点在于，今年年初，智能体化编程显然从有点用变成了真正颠覆一切。 And I don't sure you could have you clearly there were people who were going to say, well, this is going to be able to do absolutely anything. 我不确定，你显然会有人说，好，这东西终将无所不能。 And so they will say, well, yes, look, I told you. 所以他们会说，你看，我说得没错吧。 Um, but I don't think anyone kind of kind of kind of deterministly predicted exactly when that was going to happen and that it was going to be coding. 但我不认为有谁能以确定性的方式预测，具体会在什么时候发生，而且会是在编程领域 It would work first. 率先跑通。 And and what have we learned about sort of, uh, you know, say more about what this means for engineers, junior engineers, senior engineers, sort of the the jobs discussion, how teams are organized, uh, etc. 那么，在工程师，初级工程师、高级工程师的议题上，以及团队组织方式等方面，我们学到了什么？ What have we learned so far? 目前为止我们有哪些收获？ I don't think we've learned anything. 我认为我们什么都没学到。 I mean, you know, this this didn't this didn't this didn't work six months ago. 我的意思是，你知道，这六个月前还根本不起效。 Yeah. 是的。 and everyone is scrambling around trying to work out what it means and you know you can get very very into the noise and the detail and what did somebody say at a party yesterday. 所有人都在手忙脚乱地搞清楚这意味着什么，你可以陷入大量的噪音和细节，昨天派对上某人说了什么， So oh my god that's how it's all going to work. 天啊，原来一切都将这样运作。 Um you know it's going to take a couple of years for this all to settle down. 嗯，需要几年时间才能尘埃落定。 You know if nothing else because of the pricing you know you've got this enormous crunch between the demand and the supply and hence the pricing. 你知道，光是定价问题，就存在需求与供给、供给与定价之间的巨大张力。 Um so we don't know what you know what a team is going to look like. 嗯，所以我们不知道，一个团队最终会是什么样子。 I think people are asking new questions around, you know, the sort of the obvious one of, you know, do you hire junior people and if so, what are they doing? 我认为人们正在提出新的问题，比如那个显而易见的：你还招不招初级员工，如果招，他们在做什么？ And why were you hiring junior people in the past and were you actually hiring to do the thing that they did or were you hiring them to do something else? 过去你为什么要招初级员工？你是真的在招人做那件事，还是在招人做其他事情？ And so if you automate away a class of stuff that used to get done by people, then what will happen? 如果你把一类过去由人来做的事情自动化掉，接下来会发生什么？ And that's sort of becomes much more real now in software development because you actually are automating a bunch of stuff that used to be done by people. 这个问题在软件开发领域变得非常现实，因为你确实在自动化一堆原本由人来做的工作。 So those questions are kind of now rather than theoretical. 所以这些问题现在是当下时态，而不再是理论问题。 But I don't think anybody can possibly say they kind of know what the market structure is going to look like or what the career of a software engineer is going to be in three years time. 但我认为，没有人能够说他们已经知道市场格局会是什么样子，或者三年后软件工程师的职业轨迹会如何。 I think it would be you'd be insane to think that you could know that yet. 如果有人认为现在就能知道，我觉得那才是真的疯了。 Yeah. 是的。 the talk about uh 说说 Open AI uh talk about what's most uh surprised you or how have you kind of made sense of their sort of strategy development and and the questions that they have going forward? 聊聊OpenAI，你觉得最令你意外的是什么，或者你怎么看他们的策略演变以及他们面临的问题？ Well, you know, it's always been such a such a a tranquil drama-free environment. 嗯，你知道，他们一直是那么风平浪静、没有任何戏剧性的地方。 So, you know, it's [laughter] and you know, obviously they've had the the issue with with with Fiji Simo having to take a medical leave um which kind of shuffled things up a bit. 所以嗯，[笑声]，显然他们有Fiji Simo因健康原因不得不暂时离开的问题，这让局面稍微有些混乱。 Look, clearly the second half, last quarter of last year, this their question was, "Right, well, the models are the models, but what else? 好，很明显，去年下半年和最后一个季度，他们面临的问题是：好，模型就是模型，但除此之外呢？ And how do we get people to do other stuff with this? 我们怎么让人们用这个做其他事情？ So, we'll do ads, we'll do e-commerce, we'll do shopping carts, we'll do payments, we'll do a browser, we'll do a social video app, we'll, you know, everything. 那就做广告，做电商，做购物车，做支付，做浏览器，做社交视频应用，什么都做。 You know, ask Chat GPT GPT for 15 ideas for what we could do to build value on top of infrastructure and then we'll do all of them." 就是让ChatGPT列出15个在基础设施上构建价值的方向，然后全做。 It's almost literally what what it looked like. 这几乎就是当时的真实写照。 And then um um Anthropic with having less capital raised said no we're going to focus on coding and they got coding working. 然后Anthropic，融资相对少，说我们专注于编程，然后他们把编程做通了。 Um whether that was like a deliberate strategy or kind of they stumbled into it is you know for other people to say but like clearly that worked and then so opening I kind of swing around and like okay well clearly that's the thing. 这究竟是刻意为之的策略，还是某种误打误撞，这留给其他人去评说，但显然那套路子是有效的，然后OpenAI某种程度上转过身来，好，显然这才是重点。 Um but the question kind of still remains like the stuff that's working right now is software development and some things in some other fields and then there's a lot of people who are kind of excited about using this around the edges and using this for some things but it's very unclear um how it is that this is instantiated as product and taken to you know the other 90% of people. 但问题依然存在：现在能跑通的是软件开发，以及其他一些零散领域，然后有很多人对在边缘场景和某些事情上使用这个技术感到兴奋，但这些东西如何以产品形态落地、如何触达另外90%的人，还非常不明朗。 Um, you know, we still see in the data that sort of 10% of people are daily active users and 30 40% of people are weekly active users. 嗯，你知道，数据显示仍然大约10%的人是日活用户，30%到40%是周活用户。 And if you're only using this once a week, then you haven't like achieved nana yet. 如果你一周才用一次，那还没有实现纳纳目标。 And there's clearly this kind of very widespread between people in the valley who bought you know a cluster of Mac studios and are running Claude Code all day versus um you know those other 40% of people who say yeah it's kind of useful um I used it last week for something [laughter] and like how do you how do you bridge that and [snorts] I don't think that question you self software is a place where that's really really bridge jumped over that bridge and I don't think and then there's a lot of other places where people are kind of scratching their heads and using it up to a point. 而且显然，硅谷里那些买了一堆Mac Studio算力集群、整天跑Claude Code的人，和那40%说嗯确实有点用、上周用它做了个什么东西的人之间，存在巨大鸿沟，[笑声]，而且你怎么跨越这道鸿沟，[笑声]，我认为软件领域是真正跨越了这道鸿沟的地方，但其他很多地方，人们还在挠头，用到某个程度就停了。 And then there's a lot of places where corporations are using it to automate some like specific back office process where you're not asking the user to work out what they do with the new tool. 还有很多地方，企业在用它自动化某个具体的后台流程，但不是让用户自己摸索该怎么用新工具， Instead, you're saying, "Okay, here's a problem that we can solve." 而是说，好，这是一个我们能解决的问题。 And, you know, I go and talk to, you know, companies outside America and outside of tech and talk to consultants and um, you know, investors, they're looking at those one at a time point solutions. 你知道，我去跟美国以外的公司、科技圈以外的公司聊，跟顾问、跟投资人聊，他们关注的是那些一对一的点解决方案。 Um, so like I'm speak a couple of days ago to a commodities company and they want to use LLMs to get better predictions on their cash flow because they deal with all sorts of small producers and they don't necessarily know when their invoices are going to get paid and it's a very low low margin business. 比如，我几天前跟一家大宗商品公司谈，他们想用LLM优化现金流预测，因为他们和各种小型生产商打交道，不一定知道发票什么时候能收到，这是一个利润率极低的生意。 So that's a big deal and so they want to use LLMs to get better cash flow forecasting. 所以现金流预测对他们来说非常关键，他们想用LLM来做得更准。 That's very different thing from kind of going to chatbt or claude and saying hey you know give me a summary of my meetings this week. 这和去Claude或者ChatGPT说帮我总结一下本周的会议，是完全不同的两件事。 Yeah. 是的。 Can you share how uh how did this compare with mobile um or other sort of platform in terms of you 你能分享一下，这和移动互联网、或者其他平台转变相比，情况如何 use user early user adoption on on sort of the you know weekly or daily user? 在早期用户采用、每周或每日用户数量方面？ So I think this there there's there's there's a bunch of different ways to answer this. 我觉得有好几种方式可以回答这个问题。 One of them [snorts] is like we're always standing on the shoulders of giants and the growth is always compounding. 其中一个，[笑声]，是我们始终站在巨人的肩膀上，增长是复利式的。 So mobile didn't need to wait for um the internet or cellular networks like mobile data mobile internet didn't need to wait for it kind of needed to wait for cellular data but it didn't need to wait for like the internet to happen. 所以移动互联网不需要等待互联网的到来，蜂窝数据不需要等待有线互联网，手机互联网不需要等到互联网完全成熟。 and the internet didn't need to wait for PCs and PCs didn't need to wait for consumer electronics and semiconductors and so on. 互联网不需要等待PC，PC不需要等待消费电子和半导体，以此类推。 So you've always got this accelerating adoption and you know when when when your boss my old boss Mark Andre was working on Netscape there were like doubledigit millions of PCs on the entire planet. 所以技术采用始终在加速，你知道，当我的前老板Marc Andreessen在做Netscape的时候，全世界的PC数量只有几千万台。 So like no you couldn't have 900 million weekly active users because there weren't 900 million PCs. 所以当然不可能有9亿周活用户，因为压根没有9亿台PC。 So there's always that acceleration. 所以那种加速效应始终存在。 So that's one point. 这是第一点。 I think the second point is like at the early stage of any of these shifts, it's not really clear how it's going to work and nothing works. 我认为第二点是，在这类转变的早期阶段，一切都还不清晰，什么都跑不通。 So, you know, like I'm just about old enough to remember this. 你知道，我勉强算是老到还记得这些的人。