OpenAI vs Google vs Meta: Business Model War
发布时间 2025-11-28 17:00:20 来源
摘要
Pre-Thanksgiving chatter from the Lessins' Surf Shack: Jess, Brit, Dave, and Sam pinball from holiday-card automation to ...
GPT-4正在为你翻译摘要中......
中英文字稿 
When Steve Jobs died, we like forgot how to market. And then there was this low. And then Elon came back with this incredible Jobsie and style marketing. And now everybody's adopted it. And marketing is cool again. It could be that the actual innovation of Silicon Valley is the greatest fucking marketing in the world. It's not that we're better at marketing. It's that we have the megaphone's to be good at marketing because marketing is only digital these days. Well, hello friends. Welcome to more or less. Pre-thanks giving edition. Hey, Jess. How's everyone doing? Good. I was working today. And then for the last 30 minutes, I was working on our holiday cards.
当史蒂夫·乔布斯去世时，我们似乎忘记了如何做市场营销。然后出现了一段低潮期。接着，埃隆以一种令人惊叹的“乔布斯风格”的营销方式回归。现在大家都采纳了这种方式，营销又变得很酷了。可能硅谷真正的创新是世界上最棒的营销。这并不是因为我们比别人更擅长营销，而是因为我们有平台可以展现好的营销，因为如今营销几乎全是数字化的。你好，朋友们。欢迎来到“或多或少”节目，感恩节特别版。嗨，杰西。大家过得怎么样？很好。我今天在工作，最后30分钟还在忙我们的节日贺卡。
And I didn't have you guys played with the Canva integration with GPT. It's actually so good. I made my holiday cards in minutes. I'm just psyched because I heard from a bunch of listeners today, which this never happens consistently. But I learned about a bunch of listeners this morning that I'm just giving a shout out to. We've got some great people in Brooklyn. I didn't know Ryan Hoover listens every week. We've got all kinds of good people listening. Hi, Ryan Hoover. Hey, Ryan. No pressure, folks, but are historically this episode, Time to Thin Exgiving Week, is one of our best. I got to say, I don't know if you remember the Altman years.
我不知道大家是否使用过Canva和GPT的整合功能。这个功能真的非常棒！我只用了几分钟就制作好了我的节日贺卡。今天听到很多听众的反馈让我感到很兴奋，这种情况不常发生。我今天早上了解到有很多听众，我要特别地致谢。布鲁克林有些很棒的人在收听，我还不知道Ryan Hoover每周都会收听。我们有很多优秀的听众。你好，Ryan Hoover。嘿，Ryan。大家不用有压力，不过历史上这期节目——感恩节周的时段，通常是我们最优秀的节目之一。我得说，不知道你是否还记得Altman时代。
This little Sam tell us. And Sam, hello from the Surfshack. I see you've traded in the pool house. I mean, I got to be the shack at all times. Got a lot of different types of houses. A house would be generous for this structure. We like structure. It's just a structure. We like structures. Well, they both have sonnism. So that connects them. I mean, come on. I've talked about my Surfshack before. I mean, it's a work of art. We don't have to go into it, but it is. No, I just, I remember when we was in two years ago, we were in bulliness doing our thing. Sorry, Wes Moran, we're supposed to call it anyway.
这小子山姆告诉我们。山姆，你好，我们在冲浪小屋这边。我看到你已经换了个地方，不再是在泳池小屋了。我必须一直待在这个小屋里。我们有很多不同种类的房子。把这个称作房子已经很宽宏了。我们喜欢结构，这只是一个结构而已。我们喜欢这样的结构。嗯，它们都有一点相似性，所以这让它们有了联系。拜托，我之前说过我的冲浪小屋。我是说，这可是件艺术品。我们不需要深入讨论，但确实是。我只是记得两年前我们在某个地方忙着自己事的时候，对不起，Wes Moran，我们应该把它叫...无论如何。
And this was like the Altman got kicked out of Open AI back in. That was a fun, that was fun. I remember that. That was really fun. That was it. And I think that just set off this pattern of amazing Thanksgiving week episodes, just whenever and dying to tune in. But yeah, so Brett, let's do our check-in. So you made a GPT, are you made your holiday card with Canva and GPT? Well, I always do a little twist on our Moran holiday card. And I'm not going to give away. Last year was Where's Waldo, right? I was searching for you guys.
这就像当年奥特曼被OpenAI踢出去一样，真有趣。我还记得那段经历，太有意思了。这就是开始了这个精彩的感恩节周期间的一系列事件，每一次都让人迫不及待地想收听。好的，Brett，让我们来更新一下近况。你用Canva和GPT做了一个节日贺卡吗？我总会在我们的Moran节日贺卡上搞点小创新，不会透露太多。去年的主题是《寻找沃尔多》，对吧？我一直在找你们。
Yeah. Right. But I used Fiverr last year to have this illustrator animate us hidden into this huge Christmas scene. And that was really fun. That is pretty symbolic. You replaced the freelance animator with the AI. No, no. I mean, I actually paid someone last year. They just were like in the Philippines. No, I know. But this year you didn't have to pay anyone. Yeah. And the point is that the only labor is cutting out as the Philippines labor. Like that's used as suffering. Oh, okay. This is going worse than I thought. Brains hasn't paid an American to do work for her in ages. I paid Sam Altman. Come on, guys.
是啊，对的。但是我去年用Fiverr请了一位插画师把我们设计成隐匿在一个巨大的圣诞场景中。这真的很有趣。这有点象征意义。您把自由职业的动画师替换成了AI。不，不是的。我去年确实付钱请了一个人，他们只是住在菲律宾。我明白，但今年你不需要支付给任何人了。是啊，重点是唯一减少的劳动力就是菲律宾的劳动力。这好像在利用别人的痛苦。哦，好吧，这比我想象的更糟糕。脑筋女士已经很久没有支付美国人做工作了。我把钱给了Sam Altman。拜托，大家。
But did you like, is you a free version? I mean, I paid 200 bucks a month for that shit. Maybe Claire, you actually cost Sam Altman money. He didn't pay Sam Altman. I did do a lot of brainstorming. He just underwrote part of your holiday cards, actually. It's really unclear what you can slice and dice. I find angels in a million ways to make a million arguments. So I'm sure there's some accounting calculation that shows he didn't, you didn't cost them money. But all I'm saying though is I feel a little bit bad for like minted and shutter fly in all these companies that have these amazing templates that people use for their holiday cards.
但你喜欢吗？你是一个免费版本吗？我的意思是，我每个月花了200美元买这个东西。也许克莱尔实际上给山姆·奥特曼带来了损失。山姆·奥特曼没有支付费用。我确实进行了很多头脑风暴。他实际上资助了你的一部分节日贺卡。很难清楚地划分细节。我能从无数种方式中找到理由支持我的观点。所以我确信有某种会计计算表明他并没有因此损失钱。但我想说的是，我对像Minted和Shutterfly这样的公司感到有点遗憾，因为它们有这些人们用来制作节日贺卡的精美模板。
And I use minted. But to change to Canva this year, Ala Chat GPT was like an anomaly for me. So I'm curious how much share those companies are going to lose for the holidays this year. Just saying. Well, the thing I was amazed by is I don't understand how holiday cards are made. And I asked Jess, you were like, who do you want to send the holiday cards to? And you sent me like a 600 line spreadsheet with everyone's addresses, which answered a question room which is like, where do people live?
我使用过Minted，但今年要换成Canva，像使用ChatGPT那样对我来说是一种不寻常的体验。所以我很好奇，这些公司今年在假期中会失去多少市场份额。只是随便说说罢了。令我惊讶的是，我不太明白节日卡片是怎么制作的。我问了Jess，你就像是在问：“你想把节日卡片寄给谁呢？” 然后你给我发了一张包含600个地址的电子表格，这也解答了我心中的一个疑问：人们住在哪里呢？
Like I had no idea you were keeping this for years. And I was actually quite impressed with it. This is what the women of the households do. I have to update this list every year and it's so annoying because I had to literally go through it and be like, oh, wait, I think those people might have moved. And then I have to figure out how to reach out to them. And I'm sure Jess does this too. I wait for the bounce back email. I wait for the bounce back card. And then I reach out to them. Why hasn't this part of the like why hasn't this been lost to the internet? It's been shot.
我真没想到你竟然保存了这么多年。这事让我印象深刻。这就是家庭主妇们要做的事情。我每年都必须更新这个名单，这真的很烦人，因为我得一个个查看，然后想，哦，等等，我觉得那些人可能搬走了。然后我还得想办法联系他们。我相信Jess也这样做。我等着电子邮件退回，等着贺卡退回，然后再去联系他们。为什么这部分东西没有被互联网取代呢？真是令人惊讶。
Everyone has tried this so many times. And it's one of those classic problems that really shouldn't be a problem and any product person looks at it like this is a ridiculous state of affairs. But no one cares enough to fix it is like the upshot and like consistent way, right? But at your right date, like from a first principles perspective, it's completely ridiculous that like addresses are not like you can't just like make an API call that turns an email into an e-tune to address and hit set. Well there ever be a future where we have like a authenticated social graph where my agent can talk to your agent. No, but this is never going to happen.
每个人都尝试过很多次。这个问题是非常经典的问题之一，真的不应该成为问题，任何产品经理都会觉得这是很荒谬的情况。然而，没有人关心到足以去解决它，这就是总结出来的一致情况，对吧？但是在你设置正确日期的时候，从最基本的角度来看，像地址问题完全是荒谬的：你不能简单地通过API调用把一个电子邮件转换成一个实际地址并设定它。我们是否会有一个未来，在那里我们拥有一个经过认证的社交图谱，使得我的代理可以与您的代理进行交流？不，这永远不会发生。
This is what Sam and I are talking about. I mean, the sheer number of identity projects between all of us, it's like ridiculous. It's just an impossible problem. My favorite is I've owned the domain openpreferences.com for like 20 years, right? Because you think about like it's like it's a classic thing. It's like always you are like, I wait like dietary, you're like, I don't want to answer any of these shit. It's all the same questions. One repository of all of my preferences on everything and never and just like shit. And the answer is no, you know you can't.
这是我和Sam正在讨论的内容。我的意思是，涉及我们所有人的身份项目数量实在太多了，简直荒谬。这真的是个棘手的问题。我特别喜欢的一件事是，我已经拥有openpreferences.com这个域名大约20年了。因为你会想到，这是一个经典的问题，就像是人们总是问你饮食习惯的事，而你根本不想回答那些啰嗦的问题。这些问题总是一样的。想象一下，有一个储存我所有偏好的统一平台，那该多好，但实际上根本不可能实现。
Why won't your agent have this for you, Sam? Like isn't this where the world's going? Because it just doesn't, it's just like no one gets, it doesn't care enough. It's like it's from a first principle perspective, absolutely, which is why we all own versions of this domain a thousand times over. And like it's obvious. And from a practical perspective, anyone who would care to work on this has higher leverage more interesting things to work on. And it's just not that important.
为什么你的代理人不会为你做到这一点，Sam？这不就是世界发展的方向吗？因为事实就是这样，似乎没有人真正重视它。理论上来说，这当然是正确的，所以我们都拥有这个领域的多个版本。这很明显。但是从实际的角度来看，任何愿意投入精力研究这个问题的人都有更高收益、更有趣的事情去做。这个问题并不那么重要。
It's like, yeah, you're just going to tell you again that like I'm whatever my size and weight is, deal with it. And even if you're a massively resourced company and you choose to work on it, you still don't solve it. That's what the really crazy thing is. But you do through like Gemini and GPT is what I'm saying. No. Why? This was like the entire identity focus of Facebook for like 20 years. I know, but that was so long ago where I didn't know era Dave. This is the future. There is no question that every few years you have to rebuild everything and realize the same things again about how the things play out.
这句话大致意思是：就像，我要再说一遍，我的体型和体重就是这样，接受它吧。即使是资源丰富的大公司，如果你决定去应对这个问题，你也解决不了。这才是最让人惊讶的。但是，你可以通过像Gemini和GPT这样的技术来解决。为什么不呢？这就像Facebook在过去20年的整个身份焦点。我知道，但那是很久以前的事了，那时我不了解大卫的时代。这才是未来。毫无疑问，每隔几年你都要重建一切，并再次意识到事情是如何发展的。
But yeah, whatever, maybe this time to finally build it again. Guys, I'm so bored. I just checking the circle port. So I just want everyone to know that's where I am in this conversation. But no, what are your surf preferences? And that's why it will never be built because it's too fucking boring, even though it's obvious. It just switched the Friday forecast on me, which I don't appreciate because I'm kind of making plans based on it. But anyway, OK, Brett, I wish you I know you're going to have an awesome holiday card.
好的，不管怎样，也许这次终于可以重新开始建造它。大家，我实在无聊。我只是检查了一下循环端口。所以我只是想让大家知道这是我现在谈话的背景。但不，你们喜欢哪种冲浪方式？那就是为什么它永远无法建成，因为实在太无聊了，尽管很明显。我不喜欢他们突然改变了周五的预报，因为我有点在根据它安排计划。不过总之，好吧，布雷特，我希望你节日贺卡会很棒。
I was actually arrived early, put very little thought into it. I took the picture in July. I was like, this works. And we're go. So. Well, I will say what you guys are going to be very excited about is my God. Do I have a holiday? Yes for you. Jessica, proofread it. No, I did not proofread this. Jessica read it. It was like, I read it. And I said, put that in Gemini. Is it AI's flop? Sam. It's so good. It's so good. It is, it is, it is our, we wrote a book, our guide to etiquette.
我其实早到了，没有怎么考虑，就在七月拍了这张照片。我当时觉得，这样就行。然后我们就开始了。我想说的是，你们会很兴奋，因为，天哪，我有一个假期。是的，给你，杰西卡，你校对了吗？没有，我没有校对这个。杰西卡读了一下，她说，我看过了。然后我说，放到Gemini上吧。这是AI的杰作吗？Sam，太好了，太好了。这是我们的，我们写了一本关于礼仪指南的书。
It's fabulous. It's awesome. You will get it at our holiday party. And it's going into wide release. Jessica, I made sure she read it before we sent it to Prince because even I caught one or two things in it that I'm like, if I'm going to cancel over that. And so I want to make sure that we weren't going to get hardcore canceled. No, well, I, here's the problem, Dave. I don't care what is on the internet because you can always recall it. There is a thing about getting canceled in Prince that you can't like take back, right?
这真是太棒了。这是个令人惊喜的事情。你会在我们的节日派对上得到它，而且它还会广泛发布。杰西卡，我确保她在我们送去印刷之前读过它，因为即便是我也在里面发现了一两处，如果因为那些而被取消就太糟糕了。所以我想确保我们不会因此被严重抵制。大卫，这里有个问题。我不在乎网上的内容，因为你总能撤回。但是，一旦在印刷品上被取消就没法挽回了，对吧？
Like it's a little harder to like, oh, like that is the wrong gender role for that character. Let's just flip it. You know, it's on the internet versus whatever. Oh, you wanted that level of reading. Did you not read it that way? Because like that's what I was worried about. I had to correct one or two of our comics, but I'm like, even I think that's not the correct tone to be taking. Well, good thing I like my friends canceled. Oh, nice. That's a Taylor Swift reference. Sam. Anyway, the book is amazing.
这段话的大意是这样：在网络平台上，有时候很难一下子意识到一个角色的性别设定是否不当，毕竟这是网上发表的内容。有些人在阅读时可能没有注意到这样的细节，而这是我所担心的。我曾经不得不纠正一两个漫画中的这些问题，即便我觉得这种处理方式的语气可能不太对。还好，我的朋友们取消了这个问题。哦，这里有个关于泰勒·斯威夫特的梗。另外，这本书真的很棒。
You're going to love it. It's going to be a wide release. More or less listeners, I will eventually post a link where you can get a copy either by financial eventors team member or by buying it using American dollars. But it's awesome. What? No, USD. I'm giving some crypto. There still isn't like a really wildly. There isn't like an easy way that I know of to like do that, which says something. It says something. It says something like crypto number go down right now. No, crypto number five.
你会喜欢它的。这次会是一个广泛的发布。不管听众多少，我最终会发布一个链接，你可以通过一些方式获得副本，比如通过金融活动团队成员或者用美元购买。不过，它真的很棒。什么？不，美元。我会用一些加密货币给出。现在还没有一个非常流行或者容易的方法去做到这一点，这说明了一些问题。这说明的问题就像现在加密货币的数值在下降。不是，加密货币是数字五。
Number five. Crypto number down. Crypto number up. Crypto number. It's just a number. What's it a number? Let's start. We've got. So we have a pretty despite the dawn of the holiday week. There's a lot of news. And I think I try and mix up the topics a little. We all try to. But I got to say I think today's looking like crypto, open AI and Google. That's all I got. These are what else is there? Wait, the Genesis act. We don't want to talk about that. Oh, yeah, Brad, you want to start by talking about the Genesis act. What about GROC? I thought you wanted to talk about GROC. Oh, right. And X. There's just so much news. I forgot. Okay. Great. Why don't you kick us off with a what is the Genesis act? Well, is the Genesis act new or old? It's not an act. It's called Genesis mission. Genesis mission. Well, the Genesis act was was created in July. And the Genesis mission was created a couple days ago. Waiter, these are related. Okay. I'm not following. You need to unpack this. I was wondering why it was called Genesis mission. It's the strangest name. I thought that Genesis story was written 4500 years ago. What are we talking about?
第五点。加密货币数字下降。加密货币数字上升。加密货币数字。这只是一个数字。这是什么数字？开始吧，我们来看看。尽管是假期周，但还是有很多新闻。我想尽量多谈几个话题，我们都在尝试这样做。但我得说，我觉得今天主要是围绕加密货币、OpenAI 和 Google 的内容。还有什么呢？等等，Genesis 法案，我们不想谈谈这个吗？哦，对了，布拉德，你想先谈谈 Genesis 法案。GROC 怎么样？我以为你想谈谈 GROC。哦，对，还有 X。新闻太多了，我都忘了。好的，你为什么不从什么是 Genesis 法案开始呢？嗯，Genesis 法案是新的吗，还是旧的？这不是一个法案，叫 Genesis 任务。Genesis 任务。嗯，Genesis 法案是在七月创立的，而 Genesis 任务是几天前创立的。服务员，这些有关联吗？我跟不上了。你需要详细解释一下。我在想为什么它叫 Genesis 任务，这是个奇怪的名字。我以为 Genesis 故事是4500年前写的。我们在谈论什么呢？
All right. Trump signed an executive order yesterday, which was. All the Genesis mission is a lot like the Manhattan project. He is calling a national effort to accelerate scientific discovering innovation using AI. So we are going to build a shared platform with scientific data leveraging all of our national laboratories and research institutions at universities for supercomputers and expertise and focus on national challenges, including American energy dominance, advancing scientific discovery and national security. Yeah. So here, this is all in the Department of Energy. So this is entirely focused on scientific discovery related to nuclear quantum, you know, that realm. The DOE is amazing. If you guys have read, there's a great book about the DOE and how crazy it is. Most people don't understand just how crazy the DOE is. And they don't, I don't even think they understand what it is. I mean, I don't even really understand what it is. And I even read the book on it. I mean, that's the whole point of the DOE is it's like this very shadowy, but very powerful organization, which by the way, we should go into conspiracy corner for some small part of this podcast because a new UFO documentary just released. Oh, my goodness. There's an amazing book called Area 51 and Uncensored History of Top Secret Military Base. Dave, who you read this book? It's fabulous.
好的。特朗普昨天签署了一项行政命令。这项命令类似于曼哈顿计划的全民任务，旨在通过人工智能加速科学发现和创新。他号召全国共同努力，建立一个综合科学数据共享平台，利用国家实验室、大学研究机构的超算和专长，解决包括美国能源主导地位、推进科学发现和国家安全在内的国家挑战。 是的，这一切都在能源部的范围内，完全关注于与核量子领域相关的科学发现。能源部非常了不起。大家如果读过关于能源部的书，就知道它的复杂程度。大多数人都不了解能源部究竟有多么惊人，我自己看过那本书后也不敢说完全理解。能源部就像一个既神秘又强大的组织。顺便说一下，我们在这个播客的一小部分时间里也许该聊聊阴谋论，因为最近有一个新的UFO纪录片刚发布。还有一本名为《51区：揭秘秘密军事基地的历史》的精彩书籍。 Dave，你读过这本书吗？真是杰作。
Yeah, I read it at your house. It's a great book. Anyway, there's a whole section about how powerful and unknown the DOE is. Oh, yeah. The DOE was so just to give everybody a, the DOE was created during the Manhattan Project or directly after. It was related to atomic energy and the entire management development, scientific effort around that and the ongoing matrix of activities. It is the Manhattan. I'm going to get this wrong. The whole thing is, the DOE is the Manhattan Project. It's just kept changing names. And like there's all these like special carve outs where they get to do things that are completely like inaccessible to the rest of the government. But it is like, and what happens is every time there's a scandal, they just rename it and it continues. That's basically the upshot, right? Exactly. And that one of the more interesting things is they have all these levels of top secret classification that exists outside of the realm of all of the rest of the government's classifications. And so it's frequently used to keep many things secret that otherwise would be accessible.
好的，我在你家读了那本书。这是一本很棒的书。书里有一个部分专门讨论了能源部（DOE）的强大和不为人知的地方。哦，对了，DOE是在曼哈顿计划期间或之后不久成立的，涉及到原子能及其相关的管理发展、科学研究以及持续的活动网络。可以说，DOE就是曼哈顿计划，只不过是换了名字。此外，他们还有一些特别的权限，可以做一些其他政府部门无法企及的事情。而且，每次有丑闻发生时，他们只需更换名字继续下去。基本上就是这样，对吧？没错。其中比较有趣的一点是，他们拥有一套比政府其他部门的分类更加严密的顶级机密系统，因此能将很多事情保密，否则这些信息会被外界获取。
Britt, do you want to, it seems like you have a reaction to this Genesis thing. What's coming up for you? Well, so last week Trump posted the Saudi crown prince, right? And then I guess he had committed a trillion dollars from the Arab nations. And now Trump signed an executive order. Guys, isn't it me or is it a trillion? Are we so like, inoculated to a trillion dollars now? I don't even know what trillions are anymore. Yeah, what comes after trillions? Anyway, I don't know. I just think that's relevant because we're making an even bigger effort to be the weeder in the world. I can't wait to see if someone announces some sort of funding for a quintillion dollars. Definitely. Saudi first, right? My bet is Sam Alman. My bet is Sam Alman. I'm going to need 15 quintillion dollars. I just back to the Genesis act like, or Genesis plan or my mission. I mean, innovation is great. Yeah, innovation. I think, and sure, there are national security reasons to stay ahead. That makes sense to me. But it's also completely true that the US and China are one is not going to best the other in this war.
布里特，你有什么想法吗？看起来你对这个Genesis计划有些反应。你有什么感受吗？ 好，上周特朗普接待了沙特王储，对吧？然后，我猜阿拉伯国家承诺了数万亿美元。现在特朗普签署了一项行政命令。大家，是我觉得奇怪，还是我们对“万亿”这个数字已经麻木了？我现在都不知道“万亿”是什么概念了。那么，“万亿”之后是什么呢？不管怎样，我觉得这很重要，因为我们正在更努力地成为世界的领导者。我很期待有没有人会宣布某种五千万亿美元的资金。肯定是沙特先来，对吧？我猜是萨姆·阿尔曼。我猜是萨姆·阿尔曼。我可能需要1500千万亿美元。我只是回到Genesis计划，或者我的使命上。我是说，创新是很好的，没错，创新。而且，出于国家安全的理由保持领先也是有道理的，这我理解。但美国和中国在这个较量中都无法彻底战胜对方，这也是完全真实的。
I just don't, there's just no doubt. I made this. Doesn't that depend on your ideology? Well, I mean, just in terms of, I was just referring to in terms of technical prowess, right? Like, I wasn't, that wasn't a political statement. That was the like, well, I know, but I'm saying ideologically in, in technical, land, there are a bunch of people who either believe that we're racing towards some sort of AI singularity where when you hit this point, you win and all everyone else loses, it's completely zero sum. Yeah, I don't believe that.
我只是觉得毫无疑问，这确实是我做的。这难道不取决于你的意识形态吗？我的意思是，我只是在谈技术能力方面，不是在发表政治观点。我知道，但我是说在技术领域，很多人相信我们正朝着某种AI奇点前进，在到达那个点时，有人赢了，其他人就输了，是完全零和的。但是我不相信那种观点。
Yeah, so that's what I'm saying is there's a whole ideology in the valley. I mean, everywhere around the valley right now that like this is the truth that there, it's 100% zero sum and whoever gets their first to super, super intelligence. Because everyone watched the most recent Tom Cruise Mission Impossible movie, right? Like it's like all the same thing, but it's just a reason to have a race, right?
是的，所以我想说的是，在山谷中，有一种普遍的意识形态。我是说，现在山谷周围的每个地方都认为这是绝对正确的，就是这个游戏是零和游戏，谁能率先达到超级智能就能赢。就像大家都看了最新的汤姆·克鲁斯的《碟中谍》电影一样，这感觉都是一样的，只不过这是一个竞争的理由，对吧？
I think part of the thing is like, people want reasons to have a zero sum race because they know what the fun part of that zero sum races are. You could do whatever you want, right? Like if all the stakes, if you basically have a narrative where there's some point or something you achieve that is the end of the race, then like nothing else matters and that's really fun, right? Because like you can just do anything whereas the reality is the more dirty real version, but you actually see out like Ukraine, Russia, like real conflict, friction, full world, is like, that's not how it works and it is kind of dirty and you don't get to just do whatever you want for the sake of a zero sum outcome.
我认为部分原因是，人们喜欢“零和竞赛”，因为他们知道这种竞赛的乐趣所在。你可以随心所欲地做任何事情，对吧？如果从一开始，大家就有一个明确的目标或终点，那么在达到目标前，其他一切都变得无关紧要了，这其实挺有意思的。相反，现实往往更加复杂和混乱。比如，在乌克兰和俄罗斯的真实冲突中，你会看到现实并不是这种简单的零和游戏。你不能只为了达到零和结果而为所欲为，因为世界充满了冲突和摩擦，事情远不如想象的那么简单。
Rich brings us back to the Department of Energy, which is we've been looking for a Manhattan project ever since the Manhattan project. And so here we are giving a Manhattan project back to the Manhattan project. But there are so many problems with that analogy. Like one, the technology is literally bunkered away and in the brains of like six people. Here the technology is literally open source. I mean, I don't know how many atomic. You're saying in the Manhattan project.
Rich让我们回到能源部，自从曼哈顿计划以来，我们就一直在寻找一个类似的项目。现在我们把一个“曼哈顿项目”献给了原来的曼哈顿计划。但这种类比存在许多问题。首先，当时的技术是被严格保密的，只掌握在大约六个人的大脑中。但现在的技术是开源的。我不清楚原来的曼哈顿计划中有多少与原子有关的内容。
Yeah. And now the technology, like, you know, these models are neck and neck and it's sort of the nature of the technology itself, right? I mean, I'm not saying that you can't have an advantage, but people were kind of neck and neck on nuclear technology. The Germans took a wrong turn and like I think that it was an open source. Oh, definitely not. But I'm saying there was like a highly competitive landscape to get there first. And the Germans like probably should have and they screwed up, right? And like, and then the Americans got there.
是的，现在这些技术（例如各种模型）几乎是不相上下的，这似乎就是技术本身的发展特性。我不是说没有可能取得优势，但就像当时在核技术领域，各国都在你追我赶。德国人当时走错了方向，尽管核技术并不是开放源代码的，但当时这的确是一个竞争非常激烈的领域，大家都在争取第一个实现突破。原本德国人可能应该领先，但他们搞砸了，结果美国人走在了前面。
And everyone will look at that. Like that was a zero sum game. It played out and it had massive world complications. Now what happened? What happened within a few years? What happened in a few years is that the Russians stole it. Everyone's the British, everyone stole it, right? From the Manhattan project and you end up with this kind of like, you know, multi-nation kind of relative balance of power again, right? So I don't know. It's like, I think the thing is like it's it's almost like too intellectual to think about which version of the world this is. I think people, the zero sum game of like race to AGI.
每个人都会看到这一点。就像这是一场零和游戏，它的结果对世界造成了巨大的影响。那么发生了什么呢？在几年之内发生了什么？几年后，苏联窃取了成果。不仅仅是他们，还有英国等其他国家都偷走了来自曼哈顿计划的技术，结果你又回到了一个多国之间相对力量平衡的局面。所以，我不清楚。我觉得这是一个让人费脑筋的问题——要去思考这是哪个版本的世界。我想人们对争夺通用人工智能的零和游戏就是这样的感觉。
The question is whether that's like the right analogy, but it's a good story either way, right? Well, I've noticed I heard Eric Schmidt say this privately and then I looked it up and he said it publicly because I and so I can repeat it. But he refers to this as the San Francisco consensus. And he and it and again, he's now said this multiple times. Global macrogeopolitical way to refer to things. I mean, it's such a Schmidt Schmidt, and Schmidt, and Schmidt, and yeah, that's like when people refer for the global south or these sort of, you know, Davosian.
问题在于这种比喻是否恰当，但无论如何，这都是一个好故事，对吧？我注意到埃里克·施密特私下里说过这句话，然后我查了一下，发现他公开场合也说过，所以我可以重复一下。他把这称为“旧金山共识”。他已经多次提到这一点，作为全球宏观地缘政治的一种说法。这种说法很有施密特个人的风格，就像人们提到“全球南方”或者类似“达沃斯”的概念一样。
I agree. But consensus is a great word for that. But to be clear, he is skeptical of it. He is skeptical of it. Yeah. Is it really this San Francisco consensus? Well, this is what he says. I don't know. I'm in a search chat. And he talks to more people about this than I do. So what's the search chat consensus? What's the West Marin consensus?
我同意。不过"共识"这个词用得很好。但是要说明的是，他对此持怀疑态度。他对此确实持怀疑态度。是的。这真是旧金山的共识吗？这是他说的。我不确定。我在搜索聊天中。他和比我更多的人谈过这个问题。那么，搜索聊天中的共识是什么？西马林的共识又是什么呢？
I do feel like we're talking about consensus and like Genesis missions. I do think it's time, guys. We need to like plan to flag and define something. Don't you think people are really good at branding? I actually, in some ways, in some ways, I feel like the real era here is like, I think the California and Silicon Valley has gotten unbelievably good at marketing. Right. And they're like, what's really changed in the last few years? It's just incredible marketing.
我确实觉得我们在讨论共识和类似于“创世纪”这样的任务。我认为是时候了，大家。我们需要计划、确立和定义一些东西。难道你们不觉得人们在品牌塑造方面真的很厉害吗？其实，在某种程度上，我觉得现在的时代就是这样的，我觉得加州和硅谷在营销方面变得非常出色。最近几年真正改变的，就是令人惊叹的营销能力。
Right. And like, there's a whole, I think there is a San Francisco consensus of marketing, right? That Elon, I think, is a major pioneer of this extremely effective. Yeah. Let's go think some problems for a free independent and accurate media. I think you can go back further, Sam. And you could say that when Steve Jobs died, we like forgot how to market. And then there was this low. And then Elon came back with this incredible Jobsie and style marketing. And now everybody's adopted it. And marketing is cool again. It could be that the actual innovation of Silicon Valley is the greatest fucking marketing in the world.
好的，我觉得有一个关于营销的“旧金山共识”。埃隆·马斯克就是这种极其有效的营销方式的主要倡导者之一。我们可以思考一下如何让媒体保持自由、独立和准确。其实你可以往前追溯，当史蒂夫·乔布斯去世时，我们似乎忘了怎么去做好营销，然后营销进入了一段低谷期。后来，埃隆以一种令人惊叹的乔布斯式的营销风格重新出现，现在大家都开始采用这种风格，营销又变得酷炫起来。也许硅谷真正的创新就是世界上最棒的营销。
But to be clear, what you're talking about in Silicon Valley, getting better at marketing is just saying what Silicon Valley hopes to be true or what's people to believe and saying it enough times. Guys, it's not that we're better at marketing is that we have the megaphone's to be good at marketing. Marketing is only digital these days. And like Elon owns one of the biggest channels for distribution and Zach owns one of the biggest channels for distribution. Sam, who's crushing marketing and tech land? Every CEO of every one of these companies who has the biggest megaphone.
需要明确的是，你所说的硅谷在营销方面变得更好，实际上只是反复强调硅谷希望成为现实或者希望人们相信的事情。朋友们，不是我们在营销上更有优势，而是我们拥有能做好营销的大喇叭。如今，营销几乎完全是数字化的。比如，埃隆拥有最大的分销渠道之一，而扎克也拥有最大的分销渠道之一。至于在科技领域谁在营销方面表现出色？答案是每一家拥有巨大喇叭的公司背后的CEO。
So just because you have a big megaphone doesn't mean you use it well. So who's who's winning? Well, I would argue like the Mag 7 is like crushing marketing right now because they have the biggest megaphone. The Mag 7 is crushing marketing. Okay. Yeah. And video is crushing marketing. Sam Altman's crushing marketing like the private, you know, Sequoia and as you can see, is crushing marketing like I would just say that like this is this maybe this is the real San Francisco consensus is like there's a set of people have like figured out marketing.
所以，仅仅因为你有一个大的扩音器，并不意味着你能很好地使用它。那么，谁在赢呢？我认为像“Mag 7”现在在营销方面表现非常出色，因为他们有最大的扩音器。“Mag 7”在营销上很成功。对，还有像NVIDIA这样的公司在营销上表现突出。Sam Altman在营销上也很出色，像私人投资公司Sequoia也是如此。因此，我想说，这可能就是所谓的旧金山共识：有一群人已经摸索出了成功的营销策略。
And you know what that means? You get all the resources and all the smart people come to the North Bay and all the money. And then you get all the resources. But there's are the people with all the resources. Tell me someone who doesn't have resources that's crushing marketing. No, but that's the thing you wouldn't you if you didn't have if you if in some ways, if you don't have the resources, you're bad at marketing by definition. I know. Here's what's happening. I just figured it out guys. Thank goodness.
你知道这意味着什么吗？所有的资源、聪明的人们和资金都会涌向北湾。你获得了所有的资源。不过，掌握资源的人就是那些拥有所有资源的人。告诉我，有谁在没有资源的情况下还能在市场营销上取得巨大成功？没有这样的例子。但如果没有资源，从某种意义上说，你注定在营销上表现不好。我明白了，现在我终于搞清楚这件事了，真是太感谢了。
No. I think what's happening is everyone's finding their tribes. So you're finding more messaging that you're like a men to, but it doesn't mean that the marketing is inherently good or bad. People are just going after their tribes. Tham believes all of his marketing is good. For my tribe, Jess, you're making my point, which is there is no San Francisco consensus. There is a San Francisco consensus about the like a I singularity, but it's not everyone's no, I agree.
不，我认为现在的状况是每个人都在寻找自己的圈子。所以你会发现有越来越多的与你的观念一致的信息出现，但这并不意味着这些营销本身是好还是坏。人们只是在追寻他们的圈子。Tham 认为他的所有营销都是好的。对于我的圈子来说，Jess，你是在证明我的观点，那就是并没有所谓的旧金山共识。关于人工智能奇点可能有一个旧金山共识，但那并不是所有人的看法。我同意这个说法。
I agree. I think that's true. But it's not you if you say a it doesn't pick out a headline. So Schmidt has say the first of all, calling something a concept, it sounds like it says incredible marketing, right? So you're like, I'm just asserting it's the consensus. It's great. You know, Eric Schmidt has always been good at a talking point. So he's a great, he's a great marketer. Is there anything that's a consensus in 2025? But that's why you just say the consensus to make it a consensus.
我同意。我认为这是真的。但如果你说它而不强调一个重点，那就不符合你的风格。所以施密特首先指出，称某个东西为概念听起来像是绝妙的市场营销，对吧？你就好像在说，这是共识，这很棒。你知道，埃里克·施密特一直擅长制造话题，他是个出色的营销者。在2025年有什么被公认为共识的吗？但这就是为什么你只要说这是共识，就能让它成为共识。
Well, which is also is why we think everyone's good at marketing because if people just say things that we like, we listen to them more and then we think they've nailed marketing. But Jess, you just made the key point, which is creator fund oriented. It's everything we're seeing, right? This is an era of cults and tribes, right? Where you want super fucking core committed group of fail-anks of people that are your true believer followers.
好啦，这也是为什么我们认为每个人都擅长市场营销。因为如果有人说我们喜欢听的话，我们就会更关注他们，然后觉得他们在市场营销方面很厉害。不过，Jess，你刚刚提到了一个关键点，那就是以创作者基金为导向。这正是我们所看到的一切，对吧？这是一个崇拜和部落的时代，你需要一个超级核心、坚定不移的追随者群体，真正的信徒们。
And what you, if you think about it, if last generation marketing was about mass, is that getting convincing everyone in America that like, you know, brown here was cool through like magazines and you had no targeting. The entire story of our era right now is screw that, right? It's go find your super deep tribe and there can be lots of them. And they fight with each other and they overlap. But being good at marketing means you have an army behind you, right?
如果你仔细想想，上一个时代的营销注重的是大众化，比如通过杂志让全美国的人都觉得棕色头发很酷，那时是没有精准定位的。而我们这个时代的整个故事是要抛开那些，去找到属于你的小众群体，他们可以有很多，而且会互相竞争、相互交错。但要做好营销就意味着你要有一个支持你的群体。
And you that means you have capital and customers and you pay them in stock go up and all the things which again, you gotta get Elon credit. He, I've always said this and I mean it to rocket totally and also not is like, he's not a great technologist. He is the greatest marketer of our generation, right? And that means he gets all the capital and the people and the things. He's also gotten rockets. You're sure the best marketer maybe of all time.
这句话的中文翻译可以是：
这意味着你拥有资本和客户，并通过股票上涨来回报他们，而这些都是你必须给予埃隆（Elon）认可的地方。我一直说，并且完全相信这一点，他可能不是最伟大的技术专家，但他是我们这一代最伟大的营销大师。这意味着他获得了所有的资本、人力和资源。他也确实取得了火箭的成功。他很可能是有史以来最出色的营销人员。
What you need for rockets is a ton of money and rope, right? And then people who like, we're like, I only have one option if I care about the stuff because like, by the way, last year we'll love rockets, which is like, I gotta go work for you on like, it's an incredible plan. You know, Blue Origin landed its rocket. It was, you know, on a barge. So I thought that was pretty, pretty interesting. Right. They're like the Russians. They like the ones who steal the nuclear product. And it also says negatively. It's like once it's proven, it's like the, it's the four minute mile. Once you prove you can do it, you're going to be able to shave some people out of the thing that did it and redo it. That's not that hard. It's hard. But like if you have $200 billion a play with you're going to get it done.
你需要大量资金和资源来制造火箭，对吧？然后那些对航天事业感兴趣的人可能会觉得，如果他们真的关心这方面的发展，就只有一个选择，那就是去为这方面的公司工作。去年，火箭爱好者们可能对Blue Origin成功让火箭降落在驳船上感到非常有趣。这有点像俄罗斯的做法，就像那些盗取核产品的人一样。这件事的消极面在于，一旦某事被证明可行，就像打破四分钟一英里的极限一样，一旦你证明了你可以做到，就会有能力把那些参与这个过程的人淘汰，然后重现这个成果。这虽然不简单，但也不是那么难。如果你有2000亿美元可以投入到这件事情上，一定能够成功。
But don't you think also, I, so I am still caught. It is so interesting to me how hard Google the valley, the markets, everyone has swung in the last 10 days. I'm so sad. I didn't have anybody on this. I know Sam. I'm sorry. I'm sorry. Again, every episode. Jess, I'm glad you brought this up because I had a chart ready for this. Thank God, Dave. Someone who keeps me honest. Which by the way, it has another great concept in it, which is the Google Complex versus the OpenAI Complex. Complex. Complex is not as cool as a consensus, but still pretty good. I agree. You kind of want both.
但你不觉得，我，所以我还是感到困惑。对于我来说，Google、硅谷、市场等在过去的10天里变化如此之大，真的很有意思。我很遗憾，我在这方面没有任何帮助。我知道Sam，我很抱歉。很抱歉。每次都是这样。Jess，我很高兴你提到这个，因为我正好为此准备了一张图表。感谢上帝，Dave，还有一个让我诚实的人。顺便说一下，它里面还有另一个很棒的概念，那就是“Google复合体”与“OpenAI复合体”。“复合体”虽然比不上“共识”那么酷，但仍然相当不错。我同意，可能两个都想要。
So like this is amazing. This is the, the difference in price performance between the Google Complex and the Open AI Complex. Co-2 put this out like two hours ago. Dave, explain what you're seeing for those that are only listening. There's just two lines. The, you know, the performance of the Google Complex, which is Google Broadcom, the momentum TTM technologies, and then the OpenAI Complex, which is Nvidia, Softbank, Oracle, AMD, Microsoft, CoreWeave, etc. So it's percent increase. It's one year price performance.
就像这样，真是太惊人了。这是Google集团和OpenAI集团之间价格表现的差异。Co-2大约在两小时前发布了这个信息。戴夫，给只是在听的人解释一下你所看到的。图中只有两条线，一条代表Google集团的表现，包括Google、Broadcom、Momentum、TTM Technologies，另一条代表OpenAI集团的表现，包括Nvidia、Softbank、Oracle、AMD、Microsoft、CoreWeave等。这是以百分比表示的增长幅度，即一年的价格表现。
Yeah. And it's tracking. It's sort of through the first half of the year through May are, are pretty in concert. Then they both start rising through August and then they diverge. They've urged rapidly within the last two weeks. And then the last two weeks. Yeah. Well, here. So this is relevant. You know, there's been a lot of speculation. We've talked about how Google's TPUs are going to challenge Nvidia and my colleagues at the information. We wrote a story very recently that had two new data points. One meta is considering and in very serious talks, there were billions of dollars, spending billions of dollars on TPUs in a future deal. And then there's also we had new details about pitches, Google's making to banks, about using them.
好的，它正在追踪。从今年上半年，到五月时，两者的趋势基本一致。然后它们都开始上升，到八月时它们的走向开始分化。在最近两周内，它们迅速地背道而驰。在这里，这一点是相关的。大家一直在推测关于谷歌的TPU（张量处理单元）将如何挑战英伟达。最近，我和信息公司的同事撰写了一篇故事，里面有两个新的重要信息点。首先，Meta正在考虑并认真洽谈一项涉及数十亿美元的未来交易，计划在TPU上投入大量资金。此外，我们还了解到谷歌正向银行推介使用它们的新细节。
And what's different about this is you wouldn't be accessing them through Google's infrastructure and Google Cloud. These would be models where you're selling or renting to other data centers more directly competitive. And you know, there's been a lot of chatter about this. But I think the story was the first to show two new data points really on how Google plans to get these to market in a different way. And I don't know if you guys saw this, but a day or a couple hours afterwards in video, tweeted out speaking of marketing a statement. And you're delighted by Google's success. They made great advances in AI and we continue to supply to Google.
这段话的大意是：这次的不同之处在于，您不会通过Google的基础设施和Google Cloud来访问这些模型。这些模型是以直接和其他数据中心竞争的方式出售或出租的。很多人对此议论纷纷，但我认为这个故事首次揭示了两个关于Google如何以不同方式将这些模型推向市场的新数据点。不知道你们看到没有，就在这之后几个小时，英伟达发布了一条和市场推广有关的视频，他们表示对Google的成功感到高兴。Google在人工智能领域取得了重大进展，而英伟达会继续向Google供应产品。
Nvidia is a generation ahead of the industry. It's the only platform that runs every model and does it everywhere computing is done. Nvidia offers greater performance, versatility and fungibility than ASICs, which are designed for specific AI frameworks or functions. You know that's not great marketing. I was going to say great or not great. I saw a tweet. Everyone's tweeting like, isn't video okay? Well, if that were Elon or Sam Altman, they would just be like, yeah, Google thinks it's winning, but we're going to be the real winners here. Like they would just say something ridiculous. They'd be like two shades. Stay tuned.
Nvidia在行业中领先一代。它是唯一一个可以运行每种模型并在所有计算环境中运作的平台。与为特定AI框架或功能设计的ASIC相比，Nvidia提供了更高的性能、更多的多功能性和可替代性。你知道这并不是很好的宣传。我本来想说好或不好。我看到一条推文。大家都在发推文，说视频不是挺好吗？如果换成是Elon或者Sam Altman，他们可能就会说，谷歌以为自己赢了，但我们才是真正的赢家。可能他们会说一些夸张的话，就像，“请继续关注。”
Yeah, exactly. But they're being classy with their PR. That's their first mistake or old school. I don't know. Is it classy or just old school? Look, you know what it sounds like? It sounds like there's two ways in versions of the world. The pre-Trumpian era, what you'd say is actually if you're in video, you kind of need one or two legit serious competitors. You're going to get antics trusted. Right? And so you're like, okay, like I actually want someone to fast follow me pretty closely so that I can point to them and be like, see, I'm not a monopoly.
是的，没错。不过他们在公关上的方式很优雅。这是他们的第一个错误，或者说是老派。我也不太确定，这到底是优雅还是只是老派？听起来就好像在这个世界上存在两种方式。在特朗普时代之前，如果你在视频行业，其实你需要一两个真正有实力的竞争者。这样一来你的行为就会被信任，对吧？所以你可能会想，我希望有人能紧跟我的步伐，这样我就可以指着他们说：“看，我不是垄断者。”
Right? And so like that would be like the old way of thinking. Right? I think the new way of thinking is they're like, I don't know. They're like, they need new marketing people because like, that's clearly not the story. Now look, the interesting thing is most, my understanding is like a lot of companies are already using TPUs as well through Google. Like a lot of the next, it's not like this isn't like a new, yeah. So like it's a change in business strategy or model from the say you can just like, but it's like, that's the thing is like everyone's like, Google is already very deeply in the game here and people are used to using TPUs, right?
对吧？可以说，这就是过去的思维方式。对吧？我觉得现在的新思维是，他们好像有点不确定。他们可能需要新的市场营销人员，因为显然这并不是现状。你看，有趣的是，据我了解，很多公司已经通过谷歌使用TPUs了。很多下一步发展不是全新的。所以这就像是业务策略或模式的变化，但问题在于大家都认为谷歌在这个领域已经身处其中，而且大家都习惯于使用TPUs，对吧？
Well, I also think that you got to caveat this a little bit in that they're familiar with some aspects of them. But not the old problem again of the software stack and CUDA and video. So and data centers have different designs. So this isn't like a, you flip a switch, but we reported that Google thinks it can get 10% of in video's business revenue. Right. Which means they actually think they can get 20% and they're going to undershoot it. It's what they actually think.
好的，我认为你需要稍微加一些说明，因为他们对这些有些方面比较熟悉。但并不是老问题，比如软件堆栈、CUDA和视频。因此，数据中心的设计也各不相同。所以这并不像按下开关那么简单。我们报道说，谷歌认为它可以获得NVIDIA业务收入的10%。实际上，这意味着他们觉得可以获得20%的份额，只是故意保守一点。这是他们真正的想法。
So it's a real, I just love, I mean, I love trying to analyze companies PR particularly because they deal with us one way and then you know, six hours later, they deal with the public another way. And part of me is like, okay, speak your mind of video, tell your story, put your statement out there. You should have a better story. It's like their meme game is weak. They didn't need to have meme game sound. Like they were winning the product race up until like two weeks ago.
所以，这确实是一个让我感兴趣的事情。我真的很喜欢分析公司们的公关手段，特别是因为他们对我们的处理方式是这样，而六个小时后，他们对公众的处理方式又是那样。我的一部分想法是，好吧，说出你的想法，拍个视频讲述你的故事，把你的声明公之于众。你应该有一个更好的故事，这就像他们在社交媒体上的传播能力不足一样。两周前，他们看起来还在产品竞争中处于领先地位。
This is what happens when you're too far ahead, you get lazy and you know, your weapons are not refined. I don't think they're lazy. Jensen is the most paranoid CEO of all of them, which is saying a lot. It's like, look, if you're if your marketing team has been like, I don't need to shift the last two years because like the product sells itself, then like, by like you get lazy, you just you don't have the war muscle.
这就是当你领先太多时可能发生的事情：你会变得懒散，而且你的武器（产品）也不够精炼。我并不认为他们懒惰。要知道，Jensen 是所有 CEO 中最有戒心的人，这已经说明了很多问题。就好比说，如果你的营销团队在过去两年里都觉得不用努力，因为产品自然而然就能卖出去，那么你可能就会变得懒散，不再有战斗的动力。
And then when something comes up, you're not like ready to go, right? And this is like not are ready to go tweet. Like that to me feels like what's going on here is that, you know, we've seen this narrative shift over the last two to three weeks where people are actually starting to ask questions about this entire supply chain, right?
然后，当出现问题时，你就没有准备好，对吧？这就像一个未经准备的推文。对我来说，这感觉就像最近两到三个星期里，我们观察到一个叙事的转变，人们开始对整个供应链提出质疑。
Like how much do these chips cost? Are they going to be able to have business models sitting on top of them? What's the depreciation, right? Like there's like a real sort of financial analysis starting to happen in the last. Really, it's like that chart. I want a t-shirt that says what's the depreciation to great t-shirt. I've told you guys, my Gavin Baker quote, we interviewed him in April and someone asked him the depreciation of a GPU and he said, what's the meaning of life?
这些芯片要多少钱？能在这些芯片基础上建立商业模式吗？折旧是多少，对吧？最近真的开始出现一波这样的财务分析。我真的觉得这就像那张图表。我想要一件印着“折旧是多少”的T恤，太棒了。我跟你们说过，我的那句来自Gavin Baker的名言。我们在四月份采访过他，有人问他关于GPU的折旧，他回答说：“生命的意义是什么？”
I thought that was a great quote. Oh, boy. Yeah, I mean, there's not a podcast that this hasn't been debated on in the last two weeks, right? I don't know. I think it's all vibes. Like I sort of ribbed the journal for like, you know, the day after Nvidia earnings, their lead story on the front page was Nvidia results put to rest bubble fears. And the next day it was exactly the opposite.
我觉得这句话很不错。天哪。是啊，我的意思是，在过去两周里，几乎每个播客都在讨论这个话题吧？我不知道。我觉得这完全是种感觉。就像我有点调侃《华尔街日报》，因为在英伟达财报发布后的第二天，他们头版的主导文章标题是“英伟达的业绩消除了泡沫担忧”，而第二天就完全写了相反的内容。
Like they had reversed every word in the headline. And that doesn't mean it just means you can't analyze this every 24 hours is what it means, right? But they're paying to analyze it. Which is how it's to sell out the newsstand. What's the paper boy? What's the paper boy going to sell? I know. I'm around together second print newspaper, everyone who's tracking the print newspaper.
就好像他们把标题里的每个词都倒过来了。这并不是说你不能每24小时分析一次，而是他们正在为分析这个付费。这样做的目的是为了让报纸在报摊上卖光。报童呢？报童能卖什么？我明白，我和大家一起关注第二版的报纸，所有关注印刷报纸的人。
I just think the numbers are getting extremely large and people are actually starting. Like, you know, I saw this letter from an LP letter from a fund that I'm an investor in that used the number of seconds to contextualize this. And I really liked it. One million seconds is roughly 12 days. One billion seconds is 32 years. 500 billion seconds is 15,000 years.
我只是觉得这些数字变得非常庞大，人们真的开始注意到这一点。比如，我看到了一封来自我投资的基金的有限合伙人信，其中使用秒数来帮助理解这些数字，这种方式我非常喜欢。 一百万秒大约是12天，一十亿秒是32年，而五千亿秒则是1.5万年。
How is this useful to you? It's just the scale of the dollars is insane. Like until until like if you look, if you think about the 2010s, right? Uber raised something on the order. It was like 40 billion or something like that. Which was insane. Which was insane at the time, right? And so the numbers that we're looking at are just really, really nuts.
这对你有什么用呢？这笔钱的规模实在是太疯狂了。如果你回想一下2010年代，比如说Uber，当时筹集了大约400亿美元左右。当时这已经是个非常夸张的数字。而我们现在看到的数字，真的令人难以置信。
And the revenue isn't there. And so I think people are actually like, that's what started to happen in the last two or three weeks. Because people are starting to be like, where are we going to get margin from? Is the inference going to cost come down? Is it not going to? Like is TPU better for that? Is GPU better for that? I think everyone's just looking over their shoulder and doing what they're doing.
收入并没有如预期增加。所以我认为实际上在过去的两三周里，人们开始意识到这个问题。大家都在考虑，我们要从哪里获取利润率？推理的成本会降低吗？还是不会？TPU在这方面更好还是GPU更好？我觉得现在每个人都在关注自己的状况，同时也在观察别人怎么做。
What the next guys do? I agree with everything you're saying about the uncertainty. But I think it's so uncertain that like you can't explain the market by like, and now people are looking at this and now people are looking at that. I think they're just reading the room. And when no one can read the room, they're panicking.
“下一步的人会怎么做？我同意你所说的不确定性。不过，我觉得市场的不确定性大到无法通过简单的分析来解释，比如说，现在人们关注这个，现在人们关注那个。我认为他们只是在‘察言观色’。但当没人能够读懂‘场合气氛’时，他们开始感到恐慌。”
I look at it. Can I give a different spin on it? Here's what I really, I actually think a lot of this gets back to politics. Right. Like from a below, here's what happened from my perspective. For the first, you know, whatever, like half of the year, we had a two-prob Trump in control, pretty doing just Trump land, right?
我来看一下这个问题。能不能从不同的角度解读？其实，我认为很多问题最终都跟政治有关。我的观点是，在今年上半年，特朗普还是牢牢掌握局势的，好像还挺成功地运营着他的“特朗普王国”。
Like no challenges to it. And everyone understands debt overhang must make GDP go up story. So I was like, look, this is all fucking crazy. But Trump will make sure we all, like, Trump will make sure we get paid, right? Like we're going to get paid somehow. Like I'm like, not stressed about it because like, eh, we'll just cut, it will cut interest rates. We'll just do what has to be done to make number go up. And so everyone's like, we don't, this analysis is all crazy, right? Like this whole, you know, whatever, we're going to have a Chinese hazard, whatever, but it kind of doesn't matter because if it works, it works and you get the gods scenario. If it doesn't work, we'll still get paid, right? Like I think we did.
就像没人对此提出质疑。而且大家都明白，债务负担最终会推动GDP增长。所以我当时就觉得，这一切太疯狂了。但是特朗普会确保我们都能拿到钱，对吧？无论如何，我们会得到报酬。我不太担心，因为我们会降低利率，做一切必要的事情让数字上涨。所以大家都觉得这个分析很疯狂，对吗？像是我们要迎来中国式风险之类的，但其实并不重要，因为如果成功了，那就是最好的结果。如果不成功，我们还是会拿到钱，对吧？我想我们做到了。
I think the narrative broke a little bit with the midterms where I was like, ooh, like Trump less in control, right? And like if Trump less in control, then all of a sudden the analysis of like how we have to evaluate this completely changes, right? Because always, and you're like, okay, this is a lot of money. And like there is downside risk here, as opposed to being like a free upside option, right? Is the way I would look at it. So I actually think this is a political economy story. It is playing out in tech more than anything else. And then you have like Google firing up and blah, blah, blah. So I just think you have to contextualize it in that context.
我觉得在中期选举后，这种叙述有点转变了，当时我心想，哦，特朗普的控制力似乎减少了，对吧？如果特朗普的控制力减弱了，那么我们必须重新评估整个情况的分析方式就会完全改变，对吗？以前大家可能觉得这是一种免费获得收益的机会，但现在就像是一笔有风险的投资。我认为这实际上是一个政治经济的故事，特别是在科技领域表现得尤为明显。与此同时，像谷歌这样的公司也在积极行动等等。我觉得我们需要在这样的背景下理解和分析这个问题。
Sam, I agree with you. And I think that was the crewgment take two or three days ago. And if you look at that chart, I should go find it. I mean, the fall in the crypto markets is directly related to the midterm elections. I mean, it happened almost in block step, which you could argue is the Trump trade unwinding. Quotum, quote unquote. Well, yeah, I just know it's, I don't even know if I'm winding. I just think it's like ultimately the irony of the crypto stuff is it's the ultimate risk on asset, right? And like, so it's going to be the first thing people pull from effectively.
萨姆，我同意你的看法。我认为那是两三天前的一个重要情况。如果你看一下那个图表，我应该去找找。我的意思是，加密货币市场的下跌直接与中期选举有关。可以说，这几乎是同步发生的，而这可能是“特朗普交易”的解散，所谓的“特朗普交易”。嗯，是的，我只是知道这情况，我甚至不知道是否是在解散。我只是觉得讽刺的是，最终加密货币是最高风险的资产，对吧？所以它就会是人们首先撤资的目标。
Yeah, totally. I think your case, I think your argument that you just made is, you know, that the Meg 7 follows crypto. And so I think if you overlay all of these charts, like you see the risk, well, it's just risk on this happening across the curve. Yeah. But, you know, it's just, it's not clear where it goes from here. I mean, I think I do just think the macro story is even more important than anything going on specifically in the tech world. And I think, you know, again, like if you, I've always said that if you want to invest in the AI narrative, the easiest bet in the world is meta because it's all upside, no downside. I stand by that.
是的，完全同意。我觉得你刚刚提到的观点，即Meg 7跟随加密货币的走向，非常有意思。所以如果你把这些图表叠加起来看，就会看到整个曲线上的风险变化。不确定的是，这一趋势会如何发展。在我看来，宏观经济比科技领域发生的具体事情更为重要。如果你想投资于人工智能的故事，我一直认为投资Meta是最简单明了的选择，因为它只有上涨的潜力，没有下跌的风险。我坚持这个观点。
You know, I think that's continued to continue the case in like that doesn't even matter if they spend tens of billions of dollars on the wrong stuff. If that happens, it doesn't affect the overall story. I think Google is the interesting one because it was, there was a serious downside case, right, which is the core, core search business. There was all these positives, but there was also this like big bogey to the negative. I think people got uncomfortable that the bogey to the negative probably isn't going to play out or at least not dramatically, right? That's the place where it's not going to be like a dramatic shift tomorrow. And they just have so many weapons.
你知道，我认为这种情况一直持续着，即便他们在错误的方向上花费数百亿美元，也无关紧要。如果发生这种事，也不会影响整体的发展。我觉得谷歌很有趣，因为它有一个严重的负面情况，那就是核心搜索业务。虽然有很多积极因素，但也有一个大的负面风险。我想人们不安的原因是，这个负面风险可能不会发生，或者至少不会戏剧性地发生，对吧？在这个领域，不会在明天突然有戏剧性的变化。而且他们有很多优势和资源。
I think they've gone more than comfortable, Sam. I think you look at that chart from co-tune and within the last two weeks, people's entire narrative on Google has shifted into the positive. Well, that was related to Gemini 3 though, right? Yeah. Yes. But narratives can ship back. I'm curious what Brit thinks because Brit, you have you found, what are you using, chat GPT4? Guys, I switched. In the last two weeks, I have become primarily a Gemini, not GPT. I know. Well, and again, I will say every episode that I called, oh, well, I'm using, here's the problem. Here's the problem of switching.
我觉得他们已经不仅仅是舒服，而是过于自满了，Sam。你看看来自Co-Tune的那张图表，在过去的两周里，人们对谷歌的整个叙述变得积极起来了。嗯，不过那是因为Gemini 3，对吧？是的。但是叙述可能会改变回去。我很好奇Brit的看法，因为Brit，你找到了什么，你用的是Chat GPT4吗？大家，我换了。在过去的两周里，我主要使用Gemini，而不是GPT。是的。而且，我会说每一集，我都叫，我在使用这个，这就是更换的问题所在。
GPT has the memory of all my preferences and all these things I've brainstormed about. And I've been brainstorming. Do you know that many preferences? I do. I have a lot. You should see the stuff that I talked to GPT about in Gemini. But like, so I feel like I'm in a transition period where I'm trying to get Gemini up to speed on who I am and how I like my stuff and all the things I'm thinking about and working on. You're going through an AI breakup is what you're saying? Well, I wish I would have like an intake form.
GPT记得我所有的偏好和我曾经讨论过的所有想法。而我一直在进行头脑风暴。你知道我有这么多偏好吗？我确实有很多。你应该看看我在Gemini中和GPT聊过的那些内容。不过，我觉得现在是一个过渡期，我正在努力让Gemini了解我是谁，我的喜好，以及我正在思考和从事的事情。你是不是在经历“和AI分手”的阶段？我希望能有一个信息录入表。
Or I guess I could export. Hello, it's your entire Google search history, all your Gmail, your calendar. Can I export my GPT into my Gemini? Okay. I might just. Well, no, you can't export it in, but you can get it out. But I'm sure they make the file really awful and not readable. Well, I want to do that. It really doesn't matter, right? It does, my. It does matter, dude. The format, not in the arrow of GPT. This is my thing. Twitter does. In the era of, there was this era where everyone was trying to implement GDPR, right? And like, download your data, which is ridiculous. And everyone was like, okay, cool. We'll give you the file format. We're going to make them so unusable and broken. We technically comply, but good luck doing anything with the data you've downloaded.
或许我可以导出这些数据。你好，这是你的整个谷歌搜索历史、所有的Gmail邮件和你的日历。我可以把我的GPT数据导出到我的Gemini吗？嗯，我可能可以这样做。不，其实不能直接导出，但你可以把它取出来。不过，我敢肯定他们会把文件弄得非常难以阅读。好吧，我还是想这样做。其实这很重要，对吧？的确很重要，兄弟。格式问题在GPT的时代并不显著。这就是我要说的。在那个大家都在努力实施GDPR的时代，每个人都能下载自己的数据，这实在是可笑。大家都说，好吧，没问题，我们会给你文件格式。但我们会让它们难以使用和破损。我们在技术上合规了，但下载的数据你很难有什么用。
That did work. But guess what? Because I've done this. In an era of chat, it's so easy. Because you're basically like, hey, Gemini, hey, whatever. Here's my entire Twitter download your data file, which is intentionally broken. Or my entire whatever file just fix it and use it. And it's great. I don't think that'll work, Sam. I also think the chat GPT memory is a liability. I understand how the concept of memory is important, but in practice, I'll ask it about some physical element. And it will be like, well, because you asked me about another physical element last week. I'm like, no, they're not related. Like, don't try and prove to me. You remembered last week. It's an entirely different issue.
这确实奏效了。但你猜怎么着？因为我已经这样做过了。在这个聊天时代，这太容易了。因为你基本上可以这样：嘿，Gemini，或者，不管是谁。这是我整个Twitter的数据下载文件，虽然它故意是有问题的，或者是我完整的某个文件，只要修复并使用它就好了。我觉得这不管用，Sam。我也认为聊天GPT的记忆功能是个隐患。我理解记忆这个概念很重要，但在实际操作中，我会问它一些关于某个物理元素的问题，然后它会说，因为上周你问过我另一个物理元素。我心想，它们根本没有关联。别试图证明你记得上周的事，那完全是不同的问题。
So I find it's not, I think you can move on. I do think this is the stuff, though. This is the battleground, because people are not going to, but the same is like, you don't switch between Apple photos and Google photos, because you've just got all of your photos over there, and it's a pain in the ass to move it. Like, it is actually a pain in the ass, Sam. Like, I get conceptually, I get conceptually where you're coming from. Like, you're right, technically, however, the devil is in the details in that, in order to export everything out of chat GPT, it's an enormous amount of text. It's like a huge file. Like, you can't just download that into Gemini. Like, it's not possible.
所以我觉得这并不是问题，我认为你可以继续前进。不过，我确实觉得这是一个重要的问题。这就像一个战场，因为人们往往不会更换服务。比如说，你不会轻易从 Apple 照片换到 Google 照片，因为你已经在一个地方存了所有照片，搬迁实在是麻烦。真的很麻烦，Sam。我概念上理解你的看法，你在技术上是对的，但是关键在于细节。如果要把所有的聊天记录从 Chat GPT 导出，那是一个巨大的文本文件。你无法只是简单地下载并导入到 Gemini，这是行不通的。
The text is not like, photos is interesting, because it's literally like trucking around the files. It's just like a lot of band. Like, that is an actually interesting lock-in. I just went and chat GPT, I said, I'm moving to Gemini. I can give me a full rundown of my preferences, you know, etc. So I can use Gemini as well as you. It's like, absolutely. Here's a clean, safe, high-single rundown of your preferences styles for any AI work with you. It contains, and- It's not like, such as my preferences, though. It's all the things I brainstormed. It's not, it's not the preferences. Yeah, it's all the data. It's so much shatter. It's not that I always did it. That's just small. Text is small. Just, you have to choose because this is the war, right?
这段话不太像文字，照片比较有趣，因为它就像是在处理文件。这就像是一堆资料。这是一个真正有趣的体验。我只是去了ChatGPT，对它说我要转到Gemini。我可以提供我所有的偏好设置等等。所以我可以像你一样用Gemini。它就说，当然可以。这里有一个清洁、安全、高级的偏好设置总结，适用于与任何AI合作。然而，这并不是我的偏好，更多是我头脑风暴出的东西，并不是我真正的偏好。是的，这是所有的数据。杂乱无章的信息太多了，这不代表我一直都是这样。这些只是小细节。你必须做出选择，因为这是一个竞争的战场，对吧？
Like, keeping people- No, that's the narrative of the war. I don't think so, Sam. Like, I'm the user of these things, and I do see that Gemini 3 is good, but chat GPT knows a lot more about how I think what I've been thinking about. The war will be won. What do you guys think? The war, I don't know if I believe this, but I say the war will be won with the easiest hardware access, which is where the world is going. I don't agree. With the war, distribution, undistribution. That's it. I just think that this is a real thing that being able to talk freely to these things with a good user interface that remembers who you are, understands as much about you as possible, and then gives you the best answer.
像，让人们——不，那是战争的叙述。我不这么认为，Sam。我是这些东西的用户，我确实觉得Gemini 3不错，但ChatGPT更了解我在想什么。战争会赢。你们怎么看？战争，我不知道我是否相信这一点，但我认为战争将通过最容易获取的硬件来赢得，而这正是世界的发展方向。我不同意。对战争来说，分发、非分发。这就是。我只是觉得，能够自由地与这些东西交流，并拥有一个良好的用户界面，它能够记住你是谁，尽可能多地了解你，然后给出最佳答案，这是真实存在的。
Like, Jess, the fact that it didn't understand that those two ailments were different is a problem, right? Like, but they're going to solve that problem, at least if you believe their narratives at all about their research stream. And so the better it knows you, the more it can answer questions for you. I listen, I literally built the original theory of when I went to Facebook to work on profiles to get over exactly this. So the only lock in is going to be memory when you get better at exposing memory because that was the whole thesis. I'll tell you, I don't think it's intellectual, it's correct. I think when you get down to the practicality of how people use these things and their actual preferences and whatever, it just doesn't play out that way.
比如，Jess，这个系统无法区分这两种疾病，这确实是个问题，对吧？不过，如果你相信他们关于研究进展的说法的话，他们肯定会解决这个问题。而且，系统越了解你，它就越能为你解答问题。我当初去Facebook工作，专注于个人资料，就是为了突破这个问题。实际上，唯一的壁垒就是记忆，当你更好地利用记忆功能时，这就会成为主导要素。这就是我的基本观点。但我要说，我不认为这仅仅是一个理论上的正确性问题。实际上，当你深入到人们如何使用这些东西、他们的实际偏好等层面时，发现情况并不是那样的。
It's not as much of a lock in as you want to be. I guess like, I'm just, I'm going off my own personal experience, which is that I don't want to switch because it knows more. Guys, it's 2025. It is so early. I really believe that. Like, and I do, to defend my hardware point, I just find increasingly the friction of like picking up my phone, clicking through to the app instead of just like talking to something. Whether it's my phone or it is, well, that's the, that's the medocase. That's your meta thesis is impeding my usage. And what I really want, you know, I was in downtown Burley game. Fun fact, there are two downtown Burley games. Who knew what? I learned this the other day. In old downtown Burley game. And I instantly just wanted to know like, what is the history of these two Burley games? Like, you know, as a mid-predence law resident, I need to know this information. I didn't want to know enough to like, click into my phone, decide if I was going to type, decide if I was going to use the automatic dictation. I just want to know.
这并不像你想的那么难以摆脱。我只是想分享一下我个人的体验，总感觉不想换，是因为它知道得更多。各位，现在才2025年，还很早，我真心相信这一点。为了证明我对硬件的看法，只是觉得拿起手机，点进应用，比起直接对着什么东西说话越来越繁琐。不管是对着手机说，还是，呃，这是个例子，这是你们的核心理念，阻碍了我的使用。我真正想要的，举个例子，我在伯灵格姆市中心，顺便说一句，那里有两个市中心，这我也是刚知道的。就在老市区，我突然想知道，这两个市中心的历史是什么。作为个中半岛的居民，我需要了解这些信息。但我并不想麻烦到去点开手机，考虑要打字还是用语音输入，我只想知道答案。
Like, I think those are going to be the use cases. That's a long meta case because basically that's why you have glasses, that's why you're trying to get in. And by the way, what you're saying is that information. It's a long new form factor. Yeah. The interesting thing about that is it's not, I don't know, man, I don't think it's long meta. Meta is like, keep going. I just say, like, look, to me, I'd say that information, which is super fucking commodity information. Like, no, no LLM is going to have a dramatic, correct. This does not require an LLM. Leg up on like wire their two Burley. This is easy. So if actually most of the story is effectively Alexa 2.0, where Alexa actually works, right, where it's like, these are not hard questions. Most of human questions are not hard questions, right? Everyone's willing to answer them. There's no technical mode from that perspective. It's literally just who's closest to the person's face.
我认为这些将是主要的应用场景。这是一种长期的元案例，因为基本上这就是你为什么需要佩戴眼镜的原因，也是你为什么想要进入这个领域的原因。顺便说一下，你所说的信息，就是这个新形态的意义所在。有趣的是，我不确定这是否属于长远的“元”概念。元的意思是不断前进。我只是觉得，这是一些普通的信息，就像非常普通的商品信息一样。不需要大型语言模型（LLM）来做出巨大改变。这不需要LLM。这很简单。如果大多数情况实际上类似于“Alexa 2.0”，那么Alexa真的能起作用，就像，这些问题并不难解决。大多数人的问题其实并不难，大家都愿意回答这些问题。从这个角度来看，没有技术壁垒，关键只是谁离用户更近。
Then like the correct strategy is to race to the face. I like that. Race to the face. Yeah. I just don't think this captures the nuance. Like, go ask all of these systems. Ask meta. What do you know about me? Question mark. Go ask all the four main systems or five, the same question. The context is wildly different. Meta knows literally nothing. Like they don't have any contactually relevant information about you. Dave, Sam is not arguing for the superiority of their AI. I think of Meta's AI. I think he's. No, I'm just saying if the race is just the race is just the interface, here's my thing though. And this is an interesting question. Is how nuanced really are humans, right? The easy answer is we're all snowflakes and we're all super special. And like therefore, Dave, by the way, Dave, you guys have specific tastes. I believe that you're in the 1% of special people, right?
翻译成中文：
那么正确的策略就是直奔主题。我喜欢这个，“直奔主题”。是的。但我只是觉得这没有抓住细微差别。比如，去问这些系统，问Meta：“你对我了解多少？” 去问所有的四大或五大系统相同的问题。上下文截然不同。Meta几乎对你一无所知。他们没有任何与你相关的契约性信息。戴夫，山姆并不是在为他们的AI优越性辩护。我觉得可能是Meta的AI，我觉得他是。不是，我只是在说，如果竞赛只不过是界面的话，这是我的看法。不过，有个有趣的问题是，人类到底有多微妙？简单的答案是，我们都是独特的，每个人都非常特别。顺便说一下，戴夫，你们有特定的口味。我相信你们是1%中特别的人，对吧？
But I actually think most people are like not that. Oh, thanks, Sam. I'll be a snowflake. I think you have a lot of preferences. Like the reality is it's like. Hey, K, we're aggro. No, it's like it's Coke or Pepsi. It's just not that complicated. And I think if you just like strip it back for the vast majority of cases of people, it's like all this personalization content doesn't matter. I think we're talking past each other. Yeah, you're talking about six different things. But can we talk about hardware for a sec because Emerson Collectives demo day, Lorraine Powell jobs is. They have a demo day? What? Domo day. I've been. How does everyone have a demo day now? Can we consolidate the demo days? I'm going to do a demo day. Oh fuck. I'm going to do a demo day after your demo day. There are no days left on the calendar between your etiquette classes, your demo days, your protests.
但我其实认为大多数人并不是那样的。哦，谢谢你，Sam。我会成为一个特立独行的人。我认为你有很多偏好。现实是这样的：嘿，K，我们很急躁。就像是可乐还是百事可乐，这并不复杂。而且我认为如果你在大多数情况下剥去表层，这些个性化内容根本不重要。我觉得我们有点鸡同鸭讲了。是的，你在谈论六件不同的事情。但是我们可以聊聊硬件吗，因为Emerson Collective的展示日，Lorraine Powell Jobs他们有一个展示日？什么？展示日。我在想，怎么现在大家都有展示日？能不能把这些展示日合并一下？我也要办一个展示日。哦，天哪，我会在你的展示日后再办一个展示日。在你的礼仪课、你的展示日、你的抗议活动之间，日历上已经没有空余的日子了。
What's your protest, by the way? Oh, the one for. Oh, I cannot believe I forgot. I forgot my shirt. I know, Sam. It's okay. We can do it next time. Okay. But guys, breaking news, and I'm going to give this one to break breaking news. Johnny and Sam have settled on a prototype, which they will tell us in two years, two years. No. Okay, guys, this is not what happened yesterday. I watched this. This is not what happened. That's not what happened? No. He said it would be ready in two years. No, they didn't. There was a bunch of Lorraine Powell jobs trying to get Johnny to pin down a timeline. He said, I can't talk about that. You can probably get more out of Sam than me. Lorraine said, then she said, well, is it five years? He's like, no, no, no, no, much earlier than that. She said two years, and he said probably earlier than that. And it was not a like, you're going to get this in two years kind of conversation.
你在抗议什么？哦，是那件事情。天哪，我居然忘了。我忘记带我的T恤了。我知道，Sam。没关系，我们下次再做。好吧，不过有一个重磅新闻，我要让Johnny和Sam来宣布他们已经敲定了一个原型，他们将在两年后告诉我们。两年？不，不是这样的。昨天的情况不是这样，我看了，这不是事实。没有人说两年后会准备好。实际上是Lorraine Powell Jobs在逼Johnny透露个时间表，他说：“我不能谈这个，你可能能从Sam那里得到更多信息。” 然后Lorraine问：“是五年吗？”他回答：“不，不，不，远早于那个时间。”她说：“两年？”Johnny说，可能会更早。并没有说两年后你们就能看到这样的对话。
Dave, I'm glad you went to the source. Thank you. So what do you think, Dave? Is this one year? When does this ship? No. You know, I thought that their conversation was quite cool, personally. And I thought that their conversation was very interesting. Like, I think that the way that Sam actually framed the problem was kind of cool. He was talking about what did he say? I didn't see this. He's a great marketer. He was just talking about that they went to Johnny and were like, hey, we don't really know what to do. We have this new type of computer that can reason and think. And it's clear that it should be better suited for different kinds of interfaces, but we have literally no idea what to make. And so could we work with you? And Johnny was like, give me $3 billion.
戴夫，我很高兴你去了解了来源。谢谢你。那么，你怎么认为，戴夫？这是一年吗？什么时候发布？不。我个人觉得他们的对话挺酷的，而且非常有趣。我觉得山姆提出问题的方式很不错。他说了什么？我没看到。他是个很棒的营销者。他提到他们去找了强尼，说我们真的不知道怎么做。我们有一种新的计算机，它能推理和思考，显然应该更适合不同类型的界面，但我们真的不知道该做什么。所以我们能和你合作吗？而强尼的回答是：给我30亿美元。
The other thing that I liked about what they talked about is that they talked about the state of the current internet and that the internet and modern technology has really led us into this kind of attention economy where, you know, advertising is the ruling model. Because of that, the feeling of using most of the modern internet is like Times Square. And they want to create something that's kind of the opposite that, you know, is more quiet, gives you more the feeling of like being in nature next to a lake or something like that. And I like that analogy that that is perhaps the opportunity of AI that we might be able to create a better computing feeling. I think we're just going to make way better ads. I thought it was interesting.
他们所谈到的另一点让我感兴趣的是，他们讨论了当前互联网的状况。他们认为互联网和现代科技让我们进入了一个注意力经济时代，在这个时代，广告是主要的模式。因此，使用现代互联网多数时候的感觉就像置身于时报广场。他们想创造一种完全相反的东西，一种更安静的环境，给人一种仿佛置身于湖边大自然的感觉。我喜欢这个比喻，因为它说不定是人工智能的新机会，我们或许能够创造出更好的计算体验。但我认为，最终我们可能只是会做出更好的广告。我觉得这个想法很有趣。
Well, I also think, and I don't want to belabor this point and obviously, but I am interested. Lauren is clearly in team open AI in this in this battle playing out. And that, you know, she has had long standing obviously relationships with all men and it's all relational. But I think it's interesting. And I do think that there's a time when it would have been awkward for her to, you know, be so openly endorsing what open AI is doing given it's aiming right for the heart of Apple. And I just think it's interesting that she's deciding it's not awkward for her to do that right now. And I have theories on why that I probably can't speculate about.
好的，我也这样认为，不想对此大做文章，但我确实感兴趣。在这场正在展开的竞争中，Lauren显然是支持OpenAI的。显然，她与所有人都有长期的关系，一切都是关系使然。但我觉得这很有趣。我确实认为，在某个时候，对于她来说，公开支持OpenAI的行为可能会显得尴尬，尤其是考虑到OpenAI的目标直指苹果的核心。而我认为现在她决定这样做并不尴尬，这很有意思。至于原因，我也有些理论，但可能不好在这里猜测。
But I viewed this event as sort of the coming out of her allegiances in some way. Not that her investment was a secret. It wasn't not that her relationship with Joy. But Tim is her best friend or a very, very close friend. So to do something this publicly was notable. I think if you look at the overarching themes of the events, they present a very humanist worldview in opposition to kind of a highly chaotic political world, social world, you know, etc. And so a lot of what I think she was trying to pull out of them was this very human, more humanist view of what computing might be and that the ability to be, you know, calm, contextual, proactive without being intrusive, you know, they talked a lot about joy and simplicity and being human centered.
我将这件事视为她某种程度上的立场公开。并不是说她的投资是个秘密，她与Joy的关系也不是什么秘密。但Tim是她最好的朋友或者说非常非常亲密的朋友。所以她如此公开地参与是很值得注意的。我认为，如果你从整体主题来看，这些事件展现了一种人文主义的世界观，与那种高度混乱的政治和社会世界形成对比。我觉得她试图从中展现的是对于计算机更多的人性化、人文主义视角，强调能在不具侵入性的前提下保持冷静、贴合情境、积极主动。他们谈论了很多关于快乐、简单以及以人为本的理念。
But all of that's marketing, Dave, right? Like, I don't think that it is. It's a design philosophy. It is a design philosophy and it's designed to like, which is marketing. No, it's design philosophy. Yeah, that's products you are not marketing. No, but it is like, and I admire that design philosophy. I think it's a cool design philosophy. But it's also like, she's tugging on the heartstrings of the people who believe it while most people are going to decide if it's just a better product or not.
但是，这些都是营销手段，对吧，Dave？我觉得不是的。这是一种设计理念。这确实是一种设计理念，而且它是被设计成让人喜欢的，而这也是一种营销。不，这就是设计理念。对，你的产品不是营销。但是，我确实很欣赏这种设计理念，我觉得这是一种很酷的设计理念。但这也像是，她在拨动那些相信这种理念的人的心弦，而大多数人还是会通过产品本身好不好来做出决定。
Well, just, just be clear, I want to go back in the history of Apple and the whole relationship between design and marketing. Do you really say I'm because look at who's on the other side of the podcast here? No, no, no, no, I'm not, I'm not, I'm actually wanting to hear these guys talk about this. One of the interesting things about Apple, we talk about is it marketing or is it product? Is Apple is the rare tech company where marketing is super freaking powerful and way more powerful than like the product managers on teams, right? Like Apple is driven by the marketing department.
好的，我只是想澄清一下。我想回顾一下苹果的历史以及设计和营销之间的关系。你真的认为我是这么说的吗？因为看看谁在播客的另一边？不不不，我不是，我实际上是想听这些人谈论这个话题。关于苹果，有一件很有趣的事情，那就是我们总是讨论到底是营销还是产品。苹果是一个非常特别的科技公司，因为它的营销非常强大，甚至比团队里的产品经理还要有影响力，对吧？就像苹果是由营销部门驱动的。
No, I think that's a, that's a misunderstanding of how Apple works. Well, explain it, explain how it works because the way it's always been explained to me is that Apple is the rare company where marketing is super powerful and like actually in Google or Facebook or places like that. But the marketing are very unpowerful. I've worked at Google and Apple. So explain it, Brett. They were equally as powerful. Go ahead, Brett. I would say at Google, engineers are king and queen like for sure. It's about the code. It's about how fast it is. It's about the A, B test of like conversion rate percentages at like 100th of a percentage one. The analytical. This is all 10 years in the in the rear of your mirror, but that's how it was. For sure. Fine. I think at Apple, it is like product design, which inherently becomes marketing and engineering. But it's about what is the form factor? How are people going to use it?
不，我认为这对苹果运作方式的理解是个误解。解释一下，苹果的运作方式，因为据我所知，苹果是一个很少见的公司，在那里市场营销非常有力，而在像谷歌或Facebook这样的公司，市场营销却没有那么大的影响。我曾在谷歌和苹果工作，所以我可以解释一下。布雷特，你来解释吧。我会说在谷歌，工程师是绝对的核心，因为一切都与代码、速度以及极小百分率的转化率A/B测试有关。这在十年前就是这样了，完全没错。而在苹果，则是以产品设计为主，设计本身就包含了营销和工程的功能。关键在于产品的外观设计，以及人们如何使用它。
It's about the customer and the customer's delight in this thing. It's about Steve saying like we want to build the bicycle for the mind. We want to unleash human capabilities through the devices we're creating. And of course, that becomes marketing too, but it starts at the design and the product. Beyond that, Apple is completely human centric in its worldview. And the marketing, Sam, to your point is in service of a business model that requires literally selling a physical object to someone. Like since the beginning of Apple, they had to go to conferences and trade shows and make a better designed product through and through. Design is not how it looks. It's not the aesthetic. Design is how it works.
这关乎顾客以及顾客对这一事物的喜悦。正如史蒂夫所说，我们想要打造"心灵的自行车"——通过我们创造的设备来释放人类的潜力。虽然这也涉及到市场营销，但一切始于设计和产品。更进一步讲，苹果公司始终以人为本，他们的世界观完全围绕人展开。正如Sam所说，市场营销是为了服务于一种商业模式，这种模式需要将实体产品直接出售给消费者。自苹果公司成立以来，他们就必须参展和参与交易会，并通过彻底改进的设计来打造更好的产品。设计不仅仅是它的外观，也不是简单的美学，设计是产品的实际运作方式。
Right. I'm just saying, I'm not saying I just think Apple starts from, I think in my understanding, you know, I never worth there. But like the way it was always explained to me is that Apple was the tech company. We think about the religion, the product and so on. Whatever. It's just empowered. Right. And the most powerful as a relative one, it plays like Facebook or MetaBach. And today, marketing was not powerful at all. Right. Like it had no role. Right. They were like the fifth person you call maybe. Well, I think you're lumping together marketing. You're lumping together a bunch of things. Right. So there's like depends on what you mean by that.
好的。我只是想说，我不是在下结论，只是根据我的理解，虽然我从未在苹果工作过，但它一直被解释为一家科技公司。我们常常把它与宗教、产品等联系在一起。不管怎样，总之它是充满活力的。相对来说，比起像Facebook或MetaBach这样的公司，它更加强大。而且今天的营销其实一点也不具影响力。就像是没有任何作用的感觉。可能在重要程度上，他们甚至是你会第五个才想到要联系的人。不过，我觉得你把营销和很多不同的东西混在一起了。因此，这取决于你指的具体是什么。
Product marketing is different than marketing. Okay. Fine. Fine. We'll talk about product marketing as opposed to at Facebook, we had product marketing and product management. Product marketing was not important. Right. Like that was like an afterthought on a relative basis. Right. Whereas that Apple was very important. So when you talk about its product centric, I think it's worth noting, look at the history of errors in collective and how these things fit together. It comes from a legacy with Steve Jobs that really put marketing and storytelling first. Right. In a way, and I'm not saying that negatively. I'm just saying that as a strategy. So that when you then look at how immersed in collective or how the narrative comes, maybe that's coming through Sam Altman, etc.
产品营销与一般营销不同。好吧，没问题。我们来谈谈产品营销，而不是像在Facebook那样，我们有产品营销和产品管理。在Facebook，产品营销不那么重要，对吧？相对来说，那更像是事后才考虑的。然而，在苹果，产品营销非常重要。因此，当你谈论以产品为中心时，我认为值得注意的是，看看过去出现的错误以及这些事情如何结合在一起。这源于史蒂夫·乔布斯的遗产，他真正将营销和讲故事放在首位。这并不是说这是消极的，而是一种战略。因此，当你看协同工作或叙事的方式时，也许这些是通过山姆·奥特曼等人传达出来的。
It's not unsurprising to be like we're a storytelling company first from that perspective. I agree with everything everyone's saying. I'm just not sure it matters because I think like people are, well, I guess I'm wrestling with why it matters. Right. Well, why it matters is when you look at something like this story with Johnny and Sam and the narrative of human centric computing and like what's that? That's all a narrative, right. And then like you fit products into that narrative. I'm not saying that's a bad way to do it. I'm saying it's different, right? Yeah. You have companies have cultures. They have, you know, and they have them for a lot of reason. I think it's also the business model, the business model of OpenAI and the business model of Apple are the same.
从这个角度来看，说我们首先是一家讲故事的公司并不令人感到意外。我同意大家所说的一切。我只是不确定这是否重要，因为我觉得人们在想这个问题时，我也在思考它为什么重要。它之所以重要，是因为当你看像这则关于约翰尼和山姆的故事以及以人为中心计算的叙述时，一切都是一个叙事，然后你把产品融入那个叙事。我并不是说这是一个糟糕的方法，只是说它不同，对吗？是的，公司有自己的文化，这有很多原因。我认为这也是商业模式的问题，OpenAI和苹果的商业模式是一样的。
And I've been making this argument on the podcast for a long time. They're clearly not the same. They're exactly the opposite. Just give me a second. Like the, my point a long time ago was that this is a, you know, OpenAI is selling computer, quote unquote, right? Like people buy OpenAI, they buy Chatchy Pt Pro, like they buy an iPhone. You use it, you pay for it. It is not an advertising driven model today. It might be in the future. I don't know. But they have to do marketing like Apple does marketing because they're selling a, literally selling a product that does something both utilitarian and gives people a feeling of joy when they use it, which is the identical thing that Apple does.
我在播客上已经为这个观点辩护很久了。它们显然不一样，实际上完全相反。稍等一下。我的观点是，早在很久以前，我就认为OpenAI是在“卖电脑”。就像人们购买iPhone一样，他们购买OpenAI的产品，如ChatGPT Pro。你使用它，你为此付费。今天，这并不是一个依靠广告收入的模式，未来可能会是，但我不知道。但他们必须像苹果那样进行市场推广，因为他们在销售一种产品，这种产品既有实用价值，又能给人们使用时带来快乐的感觉，这和苹果做的事情是一样的。
This will be interesting because if that is the thing that you believe, which I'm not criticizing, I'm saying if you believe that as their philosophy, what we're actually going to see over AI on the consumer side is a literal business model war, right? Because that is not what Google is writing us. Certainly, what matters right now. I know. I think that is what's going on.
这会很有趣，因为如果你相信那种理念（我不是在批评，只是说如果你相信那是他们的理念），那么我们在消费者领域真正要看到的将是一场实实在在的商业模式之战，对吧？因为这并不是谷歌所采取的方式。当然，目前重要的是这一点。我明白。我觉得这就是正在发生的事情。
So, Dave, here's what you should do, Dave. Dave, you need to go sit down with Kate Roush, CMO of OpenAI, who's wonderful and convince her of that because that is not the playbook OpenAI is running right now. OpenAI wants us to think their meta, like explicitly. Everything that they're saying public and privately is to compete with meta. They're talking about time because there's a lot of meta people working there, including Kate. Yes. Literally, everyone there is from meta. Fiji. Still, it's insane. A lot of our old friends from meta work there.
所以，戴夫，你应该这样做：去找OpenAI的首席营销官凯特·鲁什聊聊，她很出色。你需要说服她，因为这不是OpenAI现在正在执行的计划。OpenAI希望我们明确地认为他们在与Meta竞争。他们公开和私下所说的一切都是为了与Meta竞争。他们谈论时间，因为那里有很多来自Meta的人，包括凯特。是的，几乎所有人都来自Meta。菲吉也是。真是不可思议，我们很多以前在Meta的朋友现在都在那里工作。
By the way, can we just like put a pin in the fact that meta, the meta mafia is actually the mafia in this town and nobody talks about it? We could circle back to it, but as meta mafia shows, I'm not sure you can weigh it on that. You can't talk about it. Yeah. You're too biased. Sure we can. PayPal talks about itself. It's worth thinking about this because Dave, you have been very consistent about your vision of OpenAI as a computer.
顺便说一下，我们能不能暂时搁置一下meta（形而上学集团）其实就是这个城市里的黑手党这一事实，而没人谈论这个问题？我们可以回头再谈，但正如meta黑手党所表现出来的那样，我不确定你对这个问题能发表什么客观意见。你不能谈论这个问题。是的，你太有偏见了。当然，我们可以。PayPal也会谈论自己。思考这个问题是有意义的，因为Dave，你对OpenAI作为一台计算机的愿景一直都很一致。
I thought it was a smart point when you made it six months ago and every time since. That is not how they think about themselves right now. Their CFO is on the call with investors talking about time spent and changing content moderation. Their CEO of apps is sub stacking about group chat and to build your community, pull your network into this. Their own advertising is sort of going head-to-head with Google search with like recipes and how to do pull-ups based on ads I saw that I was confused by.
我觉得你六个月前提到的观点很聪明，而且之后每次提到时我也都觉得有道理。但这并不是他们现在对自己的看法。他们的首席财务官正在与投资者通话，讨论用户使用时间和内容审核的变化。他们的应用程序首席执行官则忙于在博客上谈论群聊，以及如何通过将你的社群和网络拉入其中来建立你的社区。他们自己的广告有点像是和谷歌搜索正面交锋，提供类似食谱和健身教程这样的内容，这些广告让我有些困惑。
But that's not how they're making money today. That's also not how their CEO is talking about it on stage two days ago. Well, but that's a huge disconnect because that is a huge problem. But we started this by talking about the interview that was going on on stage. Yeah. I'm glad you watched the interview. Also relative to this point, we published a memo we pushed so ago, or details from a memo, quotes from a memo in which Sam Altman ahead of the Gemini launch, but clearly when the team had been testing it, needed to rally his troops.
但这并不是他们现在赚钱的方式。他们的CEO在两天前的演讲中也没有这么说。不过，这存在一个巨大的脱节，因为这是一个大问题。但我们一开始是谈论台上正在进行的采访的。是的，我很高兴你看了那次采访。关于这一点，我们刚刚发布了一份备忘录，或者是备忘录中的一些细节和引述。在Gemini发布之前，当团队显然已经在测试时，Sam Altman需要鼓舞他的团队士气。
And he said, the vibes are going to be tough out there for a while. He also said, it sucks that we have to do so many hard things at one time and many other things that you can read on the information. But I was struck by it because, I mean, it was fine. It's sort of like a tune out the noise and execute, but there was definitely a nervousness to it. They're trying to do everything. They're literally trying to be Google, Apple, and Meta all at once.
他说，接下来一段时间的气氛会比较紧张。他还表示，我们不得不同时做很多困难的事情，这真让人沮丧，还有很多其他的信息你可以自己去看。但这句话让我印象深刻，因为，我的意思是，尽管这些话听起来无所谓，有点像不去理会外界干扰然后执行下去，但其中确实透露出一种紧张感。他们试图同时做到一切，几乎是在努力成为谷歌、苹果和Meta。
And it's a point of pride for him. He uses that as a rallying cry. I get it. And I think he's sticking to his long-term vision. That's what he should do as a CEO. I don't actually think anything is wrong with the way he framed that. And I do think there will be short-term a lot of misses because Gemini and others are executing really well and have more resources. But they have a long-term vision.
这对他来说是一种自豪感。他以此作为号召。我理解他的想法，我认为他在坚持自己的长期愿景。作为一名CEO，这正是他应该做的。我认为他这样表达并没有问题。我确实认为短期内会有很多失误，因为Gemini等公司执行得非常好，而且资源更充足。但他们有一个长期愿景。
And better business models. I think this comes down to this business model question, which is everything that we're talking about, all of these random activities that we're listing out over and over again, are them trying to find a bigger business model than the one that currently exists. That's the real truth. Because they have to justify a half a trillion dollar valuation and figure out how to get more money to buy the things to fight all the wars at once.
更好的商业模式。我认为这归结为商业模式的问题，我们所讨论的一切，这些我们一再列举的各种随机活动，都是他们在尝试寻找一个比现有更大的商业模式。事实就是这样。因为他们必须证明自己值五千亿美元，并想办法赚更多的钱，以便同时购买各种资源来应对所有挑战。
Yes. Right now, they do two things. They sell a computer to people and they sell a computer to business. Well, to be clear, they sell very little of either. I mean, it's a lot of revenue, but not to not compare to the old competing with. Yes, I understand. Guys, what they're selling as a computer is pretty much available for free. We can quibble on the margins, but the free version of Gemini is the number of people who are going to find the difference.
好的。现在，他们主要做两件事情。他们向个人出售电脑，也向企业出售电脑。不过，说实话，他们卖的其实不多。我的意思是，收入虽然不少，但与以前的竞争相比，还是显得微不足道。我明白，他们所谓的电脑基本上可以免费获得。我们可以在小细节上讨论，但能分辨出与免费版Gemini的区别的人并不多。
So that is a big problem. That's a little problem. That's a big problem. Yes, which is why we're talking about this dissonance, right? It makes sense that the troops are talking about literally the meta business model, right? Because the underlying hope, not current reality, is that they're going to be able to generate a ton of engagement and apps like group chat and many, many others, probably productivity, probably health, probably like everything that will then convert into a attention based business model just like meta, right?
这真是个大问题。这是个小问题。这是个大问题。是的，这就是我们为什么要讨论这种不一致的原因，对吧？军队在谈论的其实是元公司的商业模式，这很合理。因为希望（而非当前的现实）是，他们将能够产生大量的用户参与度和各种应用，比如群聊、生产力工具、健康应用等等，然后将这些转变为一种基于注意力的商业模式，就像元公司一样，对吧？
Well, that's just to push you. That is the opposite of the Apple business model. I understand. And it's also the opposite of what Sam said on stage. Yeah. I think Dave is arguing that they should change their strategy and hark and align it more with the spirit of Sam's on stage interview. I'm not sure, Jess. I'm not like, this is a very complex problem. I don't, I'm trying to correctly communicate what Sam was talking about and what was going on at this Emerson collective thing as well as these Apple questions. I don't know how to solve the business model problem, which comes back to Matt, the macro, right?
好吧，那只是为了给你施加压力。这与苹果的商业模式完全相反。我明白。这也和 Sam 在舞台上所说的相反。是的，我认为 Dave 认为他们应该改变策略，使其更符合 Sam 在舞台采访中表达的精神。我不确定，Jess。这个问题很复杂，我努力想准确传达 Sam 所谈论的内容，以及在 Emerson Collective 活动上发生的事情，还有这些关于苹果的问题。我不知道如何解决这个涉及到 Matt 和宏观层面的商业模式问题。
Sam knows his, Sam knows his audience. We talked last week about if chat GPT is the clean ex of a category, but it turns out that that isn't the category per se, right? And that the distribution model is going to be more distributed. It's going to be through these other products and, and I think that kind of sums up the dilemma. Do you have data on how much like, like, how do we think about this in terms of who's using which product and at what scale? Like, who has the best data on that?
山姆了解他自己，也了解他的观众。我们上周讨论了ChatGPT是否是某个类别中的典型代表，但结果证明，这并不是那个类别本身。它的分发模式将更加分散，会通过其他产品进行分发。我觉得这很好地概括了这个难题。你有没有数据说明这些产品的用户使用情况及其规模？谁拥有关于这个问题的最佳数据？
Like, how many DAU of chat GPT is there versus Gemini versus Grock versus Anthropic versus? I haven't looked at the latest, but Gemini and chat GPT both have huge numbers without a doubt. Like, how is it, is it like one billion on chat GPT versus 400 million on Gemini? Like, that was what I remember last seeing somewhere, but I didn't know how accurate the data's are. I would have to compare, but my sense is very healthy engaged, hundreds of millions, very active. But that's like the difference between meta or Facebook and Twitter and Snapchat, right?
就像，我想知道ChatGPT、Gemini、Grock和Anthropic的每日活跃用户（DAU）分别有多少。我最近没查看最新数据，但可以肯定的是，Gemini和ChatGPT的用户数都很庞大。比如，ChatGPT是不是有10亿，而Gemini有4亿？我记得在某处看到过这样的数据，但不知道这些数据有多准确。我需要比较一下，不过我的感觉是，它们的用户参与度都很高，活跃用户有数亿。这就像是Meta或者Facebook与Twitter、Snapchat之间的区别，对吧？
Like meta was always north of a billion DAUs, Twitter and Snapchat were multi-hundreds of millions. But I think the trajectory matters more. It's like because you don't, you haven't really been like comparable products. And again, look, chat GPT could, if we wake up and open AI has, you know, some new AI breakthrough that gets us closer to that San Francisco consensus, that could change the game again. I think it would change it for a shorter window, but it could again, right? So. Well, this is the point is there's no lock in, right? It's like, if you only think it changes it, which is, it was the original pitch, was that we're so close to this like game over a nuclear style outcome that sound will get to first and then there's things locked, everyone else locked out, right? And like, that's not going to happen.
Meta 的每日活跃用户（DAUs）一直都高于10亿，而 Twitter 和 Snapchat 则有数亿用户。但我认为发展轨迹更重要。这是因为它们没有真正作为可比产品存在。而且，如果有一天我们醒来发现 OpenAI 有了新的 AI 突破，让我们更接近所谓的旧金山共识，那可能会再次改变游戏。我觉得这种变化可能会持续更短时间，但仍然有可能发生。关键是，市场上没有一家是绝对稳固的。最初的想法是我们非常接近一种“游戏结束”的结果，即某家公司会先到达这一点，从而将其他公司锁在外面。但这不会发生。
Back to the thing that we were somewhat talking about, each other about, I don't think you can claim that there's zero percent lock in. I think that memory, not preferences, memory, provides some level of lock in. Yeah, it's not zero. Yeah. Do you think wearables are going to add more of a lock in? Like if, if we are measuring all of our daily, quantified self, no, I think we've got to get back to the every man here. Like, I think we have to, I don't know what analogy we're in, but it's like we've been through this like hardcore power user phase of. And story telling things. But just that's why I'm asking you the DAU number because I think, I think it was Sam that said, or you guys may be quoted in an article said something like chat GPT is AI to most people. And I don't know how true that is, right?
回到我们刚才有些讨论的话题，我不认为你可以声称完全没有任何锁定效应。我认为记忆，而不是偏好，确实带来了某种程度的锁定效应。是的，这种锁定效应不是零。那么，你认为可穿戴设备会增加更多的锁定效应吗？比如说，如果我们记录了所有日常的量化自我数据，我觉得我们应该回到普通人的角度来考虑。我不太确定我们目前处于什么样的情境，不过感觉就像是经历了一个硬核资深用户的阶段。这就是为什么我问你的日活跃用户数量，因为我记得，好像是Sam说过的，或者你们可能在一篇文章中被引用，说什么ChatGPT对大多数人来说就是AI。我也不太确定这是不是事实，对吧？
Like if it's true that most of the DAU like in the world right now is in chat GPT every day, it's not. I don't think it is. I would just like to know what the number is. No, it will get back to you. There was an era where most of, to most of the world, AOL was the internet, right? And like, eh, they had a lot of lock in in a lot of different ways. And you know what? It just doesn't matter long term. I'm also not sure. I'm not sure you want to be AI to the world to our conversation last week. I think, I think that is an open question. Don't you just want to be like free shipping, cool stuff faster answers to the world and free? Isn't that what you want to be?
如果说现在世界上大多数的日活跃用户（DAU）每天都在使用ChatGPT，那是不准确的。我不这么认为。我只是想知道具体的数字是多少。不会有人告诉你。在某个时代，对世界上大多数人来说，AOL就等同于互联网，对吗？他们通过各种方式将用户锁定。但从长远来看，这并不重要。我也不确定。上周的对话中，我不确定你是否想成为世界的AI。我认为这是一个有待讨论的问题。你难道不想成为提供免费送货、有趣商品、更快答案的服务吗？这不正是你想要的吗？
And then the idea that you like, you know what actually even wants to buy a computer, right? They want the outcomes from a computer. I think this is like the key thing. It turns out, it might be that in chat GPT is clean X, but like clean X is like the plague, right? And like you're like, I actively don't want. Okay, that's a bit harsh. I will also say as we wrap this segment and perhaps this pod. Yeah, I want to go to the pool. I also want to. I'm sick of the surfshad. I want to get the sauna to the pool. I also am like not ready to like short open AI, despite everything I said.
翻译成中文并尽量易读：
然后，你知道，人们其实并不是真的想买电脑，对吧？他们想要的是电脑带来的结果。我觉得这就是关键点。事实证明，可能在聊天机器人GPT方面，像清洁剂X这样，但清洁剂X就像瘟疫一样，对吧？像是你根本不想要。好吧，这有点过于苛刻了。我也想说，当我们结束这个部分，甚至是这个播客的时候，我想去游泳池。我也想去。我已经厌倦了冲浪，想去桑拿或者游泳池。同时，尽管我说了那么多，我也还没有准备好做空OpenAI。
Like, yeah, I guess that's why I'm trying to get to this question, Jess, which is like, well, I guess I'd short it. I'd short it beyond like to some degree, you know, below its current valuation. Yes. But I would not it's not going to zero. It's not going to zero. And so I know, well, someone will buy it. So I think that's really interesting. And there's a real puzzle in front of them. I just there are certain queries I prefer them for to Gemini. They have a much more straightforward style Gemini so long winded. I want like Gemini short Gemini TLDR.
好的，我想这就是为什么我想问这个问题，Jess。就是说，我大概会做空它，但只是在某种程度上，低于它目前的估值。是的，但我不认为它会归零。我知道会有人买下它。我觉得这很有趣，他们面前确实有一个谜题。不过在某些问题上，我更倾向于用Gemini。Gemini的风格更简洁明了，而不像其他那样冗长。我更喜欢Gemini的简短总结。
And you know, they've got a lot to play with to do some cool stuff. But the ball is in their court again. And these things swing. These things are not going to zero because if nothing else, they're taking dollars and turning into 20 cents worth of infrastructure. Right? So the zero is not a number, right? Like if I sell you, if I sell you 20 cents of infrastructure for a dollar and I take the difference, like fine, it's, and I also say shorting in anything where an incinity is possible. And even I believe there's some crazy case where there is incinity possible. You can't short that.
你知道，他们有很多资源可以用来做一些很酷的东西。但现在球在他们那边。这种情况下会有起伏，这些事情不会变成一文不值。至少，他们在把美元转换成价值20美分的基础设施，对吧？所以零不是一个实际的数字。比如说，如果我以一美元卖给你价值20美分的基础设施，我获取了差额，那也没问题。另外，我还想说，在任何存在无限增长可能性的情况下做空股票是不明智的。即使我相信有些极端情况可能会导致无限增长，你也不能在这种情况下做空。
That's the only way to lose incinity money. I don't know. Like is there an AI put in the market? Like how far out would you play that right now? I am too much afraid to try things, even things I don't. And I wish, this is going to be my next evolution is I'm going to learn how to be an adult trader of the public markets. You're going to become a Cirque de Soleil gymnast. I saw Josh Wolf tweeting a bunch about buying an AI put, you know, basically just like literally shorting QQQ, buying a put like six months out because there's no possible way.
这是失去不正当收入的唯一方法。我不知道。比如说，市场上是否有AI期权？如果真的有，你现在会投资多久？我对尝试新事物感到非常害怕，即使是我不了解的事情。我希望我的下一个进化阶段是学习如何成为一名成熟的公众市场交易者。你将成为一个像太阳剧团的体操运动员。我看到Josh Wolf在推特上发了一堆关于购买AI期权的内容，基本上就是像做空QQQ，买一个六个月后的期权，因为没有可能。
Well, my thing is I want, I was going to buy a QQQ short and to pair with a Google long, right? Because I'm scared of the down, I'm scared of market go down, but I think Google on relative basis go up even now. And I'm just annoyed I didn't do it too. ECO. Now I'm going to go to the pool to drown my sorrows. You're going to be here in one week beating your brow again on this. So should I just buy a bunch of Google stock then? I can't, I can't give you an opinion on that. Again, just by the stock Sam, you've just talked about it for months now. I know, and I just, I did so much poorer because I didn't. I don't offer investment advice.
好吧，我的想法是我想买QQQ的空头，然后配合做多 Google，对吧？因为我担心市场下跌，但我认为相对而言，Google 现在依然会上涨。而且我很懊恼没这么做。唉，我要去泳池排解一下忧愁。你可能一周后还在为这事懊恼。所以我是不是应该干脆买一堆 Google 的股票呢？我不能给你任何建议。是啊，Sam，你都讨论了好几个月了。我知道，因为没有买我表现太差了。我不提供投资建议。
Dave, you wanted a conspiracy corner. I know. Oh, conspiracy corner, speaking of Department of Energy. Well, there's an amazing new documentary on Amazon called the Age of Disclosure. It's not that much new information, but it is a consolidation of all of the closures. Yes, of all the narrative around the UFO, UAP thing. It's worth, you know, if you're looking for something to watch over Thanksgiving, just check it out and draw your own conclusions. Am I going to believe in UFOs after I actually don't know that I have opinion actually? I might be. It's really fascinating to watch.
戴夫，你想要一个关于阴谋的讨论角，我知道。啊，说到阴谋，特别是能源部，有一个很棒的纪录片在亚马逊上，叫做《揭秘时代》。它的信息不算太新颖，但把关于UFO和UAP所有的说法都整合在一起了。如果你在感恩节期间找东西看，可以去看看，然后自己做出判断。我看完后会相信UFO吗？我其实不知道我有什么看法，也许会吧。这部纪录片确实很吸引人。
I mean, there's a lot of really senior people that worked at, I mean, we're talking Marco Rubio is on this documentary and that are talking about a lot of the various details related to this that we have craft, that we have none, or what do they call them, nonhuman biologics, that we have been interacting with these things for over 50 years that most of it is hidden in the Department of Energy that we're seeing an increasing number of reports. Anyway, it's like really interesting to watch.
我意思是，有很多非常资深的人参与了这个纪录片，比如马可·鲁比奥也在其中，他们讨论了许多相关的细节，比如我们所拥有的，这些被称为“非人类生物”的事物。我们与这些东西互动已有超过50年的历史，而大部分信息被隐藏在能源部。与此同时，我们看到相关报告越来越多。总之，这个纪录片真的很有趣。
Well, Dave, you should download and have you used enigma? Yeah, of course. Yeah, Alice Lloyd George. What is enigma? Yeah. Oh, that's the UFO spotty nap. Yeah, that's great. Are you an investor? I don't want to talk about it. I don't want to talk about it right now. Do you brought it up? I brought it up. I brought it up in one very specific sense. All right, I'm going to the pool. Bye.
好吧，Dave，你应该下载一下，你用过Enigma吗？当然用过。Alice Lloyd George，什么是Enigma？哦，就是那个UFO观测应用。对，挺不错的。你是投资者吗？我现在不想谈这个。是你提到的？我提到的。但我是从一个非常具体的角度来提的。好吧，我要去游泳池了，再见。
He leaves us for meetings. He leaves us. It's just like we're over time. No one's listening anymore. We could say goodbye to the people, Sam. You got to sign off. Okay, friends, listeners, colleagues, random people who stumbled into this podcast via Sam's advocate school. They can't possibly still be high. They'll launch the podcast at a kid school. And Sam needs to sign up to be a student.
他离开我们去开会。他离开我们。就好像时间到了，没有人在听了。我们可以向大家道别，山姆。你得结束节目了。好的，朋友们、听众、同事、那些在山姆的辩护学校意外发现这个播客的人们。你们不可能还在兴奋。他们会在一个孩子的学校启动这个播客，而山姆需要报名成为一名学生。
Guys, we need a way to compete that doesn't involve doing more things because then we have no time for anything. I'm just saying. The key is to just do them all less well. We hope that by the time you listen to this, you had a nice Thanksgiving, we appreciate you listening. And we will see you back here again next week for another episode of more or less. Bye. Bye. Bye.
大家，我们需要一种竞争方式，不是通过做更多的事情，因为那样我们就没有时间做其他事情。我只是随便说说。关键是把每件事都做得没那么好。希望当你听到这段话的时候，你度过了一个愉快的感恩节，非常感谢你们的收听。我们下周再见，准备好迎接更多或更少的新一期节目。再见，再见，再见。
If you enjoyed this show, please leave us a virtual high five by rating it and reviewing it on Apple Podcasts, Spotify, YouTube, or wherever you get your podcast. Find more information about each episode in the show notes and follow us on social media by searching for at more or less at Dave Moran, at lesson, at J lesson. And as for me, I'm at Brit. See you guys next time.
如果你喜欢这个节目，请通过在 Apple Podcasts、Spotify、YouTube 或其他你收听播客的平台上进行评分和评论，给我们一个虚拟的击掌支持。在节目的备注中可以找到关于每集的更多信息，并通过搜索以下账号在社交媒体上关注我们：@more or less、@Dave Moran、@lesson 和 @J lesson。而我则是 @Brit。下次再见！