OpenAI CEO Sam Altman | AI for the Next Era

Hi everyone, welcome to Gray Matter, the podcast from Gray Lock where we share stories from company builders and business leaders. I'm Heather Mack, head of editorial at Gray Lock. Today, Gray Lock General Partner Reed Hoffman interviews Sam Altman.

大家好，欢迎收听Gray Lock的博客“灰质量”，我们分享公司建设者和商业领袖的故事。我是Gray Lock的编辑主管Heather Mack。今天，Gray Lock的普通合伙人Reed Hoffman将采访Sam Altman。

Sam is a CEO of OpenAI, an AI research and deployment company whose primary mission is to develop and promote AI technology that benefits humanity. This interview took place during Gray Lock's intelligent future event, a day long summit featuring experts and entrepreneurs from some of today's leading artificial intelligence organizations.

Sam 是 OpenAI 的 CEO，OpenAI 是一家专注于AI技术研究与发展、致力于开发和推广有益于人类的AI技术的公司。这次采访是在 Gray Lock 的 intelligent future 活动期间进行的，该活动是一天的峰会，邀请了来自一些领先的人工智能组织的专家和企业家。

You can watch the video of this interview on our YouTube channel under our intelligent future playlist. And you can listen to other interviews from that summit on the Gray Matter podcast. You can find that on your preferred streaming platform or on the content section of our website, graylock.com.

你可以在我们的智能未来播放列表下的 YouTube 频道观看该采访视频。还可以在 Gray Matter 播客上收听该峰会的其他采访。你可以在你喜爱的流媒体平台或 graylock.com 网站的内容部分找到它。

Now here's Reed Hoffman with Sam Altman.

现在让我们听一下里德·霍夫曼和萨姆·奥尔特曼的谈话。

Sam, close friend, many, many things. I think we actually probably first met, I think, on the street on El Camino bumping into when you were doing looped. Have done a number of things, including a good portion of my nuclear investments are with you. Because you call me and say, hey, this is really cool and I agree.

嗨Sam，你是我的好朋友，我们经历了很多，很多的事情。我想我们第一次见面是在El Camino街上，当时你正在做一些循环的事情，我们不小心撞了一下。我们一起做了很多事情，其中包括我的很多核能投资都是和你一起做的。因为你打电话给我说，“嘿，这个真的很酷”，我也同意了。

All right, let's start a little bit more pragmatic, but then we'll branch out. So one of the things I think a lot of folks here are interested in is based off the APIs that very large models will create, what are the real business opportunities?

好的，让我们开始更务实一些，但之后我们会展开讨论。我觉得这里很多人都对基于大型模型所创建的API产生的真正商业机会很感兴趣。

What are the ways to look forward? And then how, given the APIs will be available to multiple players, how do you create distinctive businesses on them?

怎样才能展望未来呢？而且，由于API将开放给多个参与者，如何在其上创建独具特色的业务呢？

Yeah. So I think so far we've been in the realm where you can do an incredible copywriting business or you can do a sort of education service or whatever. But I don't think we've yet seen the kind of people go after the trillion dollar take on Google's. And I think that's about to happen. Maybe it'll be successful, maybe Google will do it themselves.

哦，我认为到目前为止，我们所看到的是一些非常出色的文案撰写业务或某种教育服务。但我认为我们还没有看到有人去追求谷歌市值数万亿美元的那种级别。我认为这种情况即将发生。也许会成功，或者谷歌会自己去做。

But I would guess that with the quality of language models we'll see in the coming years, there will be a serious challenge to Google for the first time for a search product. And I think people are really starting to think about how do the fundamental things change and that's going to be really powerful.

我猜测，随着未来几年语言模型的质量不断提高，谷歌将首次面临严峻挑战，特别是在搜索产品方面。我认为人们正在开始思考基本事物如何改变，这将非常有力量。

I think that a human level chat bot interface that actually works this time around, I think like many of these trends that we all made fun of were just too early. The chat bot thing was good, it was just too early. Now it can work. And I think having new medical services that are done through that, where you get great advice or new education services, these are going to be very large companies.

我认为一款真正有用的人类水平聊天机器人接口正在这一次推出，就像我们之前嘲笑的许多趋势一样，它们都来得太早了。聊天机器人是一个好东西，只不过时间尚未成熟。现在它可以起作用了。我认为，在这上面提供新的医疗服务或获取优秀建议和新的教育服务，这些公司将会变得非常庞大。

I think we'll get multi-model models and not much longer and that will open up new things. I think people are doing amazing work with agents that can use computers to do things for you, use programs. And this idea of a language interface where you say a natural language what you want in this dialogue back and forth, you can iterate and refine it and the computer does it for you.

我认为我们不久将会得到多模型模型，这将开启新的可能性。我认为人们正在利用代理人来使用计算机为你完成任务，使用程序，做出惊人的工作。还有一种语言接口的想法，你可以用自然语言说出你想要的，在这种对话中来回迭代和精细化，计算机可以代替你完成它。

You see some of this with like Dolly and Copilot in very early ways. But I think this is going to be a massive trend and very large businesses will get built with this as the interface and more generally that like these very powerful models will be one of the genuine new technological platforms which we haven't really had since mobile. And there's always like an explosion of new companies right after. So that'll be cool.

你看，像Dolly和Copilot这样的早期产品已经展示了其中一些特性。但我认为这将成为一个巨大的趋势，并且会以此为界面建立起非常庞大的企业。而这些非常强大的模型也将成为我们自移动电话以来未曾有过真正的新技术平台之一。之后肯定会有大量新公司出现，这也非常酷。

And what do you think the key things are given that the large language model we provided is an API service? What are the things that you think that folks who are thinking about these kind of AI businesses should think about as how do you create them during differentiated business?

你认为提供了一个API服务的大型语言模型的关键点是什么？如果有人想要创建这些AI业务，你认为他们应该考虑哪些事情，以确保创造出有差异化的业务？

So I think there will be a small handful of like fundamental large models out there that other people build on. But right now what happens is company makes large language model API other to build on top of it. And I think there will be a middle layer that becomes really important where I'm like skeptical of all of the startups that are trying to sort of train their own models.

我认为未来只有少数几个基础大型模型，其他人都会在这些模型上进行构建。但现在的情况是，公司会开发大型语言模型API供其他人在其上进行构建。我认为这个中间层将变得非常重要，我对那些试图训练自己的模型的创业公司持怀疑态度。

I don't think that's going to keep going. But what I think will happen is there will be a whole new set of startups that take an existing very large model of the future and tune it which is not just fine tuning like all of the things you can do. I think there will be a lot of access provided to create the model for medicine or using the computer or like the kind of like friend or whatever.

我不认为那会持续下去。但我认为会有很多新创企业将现有的未来大模型进行调整，这不仅是微调那些你可以做的所有事情。我认为会有很多渠道可用于创建医学模型、使用计算机或类似朋友之类的模型。

And then those companies will create a lot of enduring value because they will have like a special version of they won't have to have created the base model but they will have created something they can use just for themselves or share with others that has this unique data flywheel going that sort of improves over time and all of that. So I think there will be a lot of value created in that middle layer.

那么这些企业会创造很多持久的价值，因为它们会拥有一个特殊的版本，它们不必创建基本模型，但它们将创建一些只供自己使用或与他人分享的东西，这些东西具有独特的数据循环，随着时间的推移而不断改进。所以我认为在这个中间层会创造很多价值。

And what do you think some of the most surprising ones will be? It's a little bit like for example a surprise a couple of years ago and we talked a little bit to Kevin Scott about this this morning as we opened up which is train on the internet do code. So what do you think some of the surprises will be of you didn't realize it reached that far?

你认为最令人惊讶的是哪些？有点像几年前的一个惊喜，我们今天早上开幕式时和凯文·斯科特有过一点谈话，就是在互联网上训练编码。所以，你认为有哪些惊喜是你没意识到它们覆盖了那么遥远的范围？

I think the biggest like systemic mistaken thinking people are making right now is they're like all right you know maybe I was skeptical but this language model thing is really going to work and sure like images video too but it's not going to be generating net new knowledge for humanity it's just going to like do what other people have done and you know that's still great that's still like brings the marginal cost of intelligence very low but it's not it's not going to go like create fundamentally new it's not going to go to cure cancer it's not going to add to the sum total of human scientific knowledge and that is what I think will turn out to be wrong that most surprises the current experts in the field.

我认为现在人们最大的系统性错误思维就是他们会说：“也许我曾经怀疑，但这种语言模型真的很有效，当然，图片和视频也一样。但它不能为人类创造新的知识，只是重复别人所做的事情。虽然这仍然很好，将智能的边际成本降低了，但它不会创造根本性的新东西，也不会治愈癌症，也不会增加人类科学知识的总和。我认为，这将是当前领域专家最令人惊讶的错误。

So let's go to science then there's the next thing. So talk the general tooling that really enhances science. What are some of the things whether it's building on the APIs you know use of APIs by scientists what what are some of the places where science will get accelerated now.

那让我们转向科学，然后是下一个事情。那么，让我们谈一谈那些真正能够增强科学的通用工具。有哪些事情呢，无论是基于科学家所使用的API构建还是API的使用，在哪些地方科学现在将被加速？

So I think there's two things happening now and then a bigger third one later. One is there are these science dedicated products whatever like alpha fold and those are adding huge amounts of value and you're going to see in this like way more and way more. I think I that I were like you know had time to do something else I would be so excited to like go after a bio company right now like I think you can just do amazing things there.

所以我认为现在有两件事情在发生，然后是后面更大的第三件事。其中一件是有一些专注于科学产品的产品，例如 alpha fold，这些将添加巨大的价值，你会在这方面看到更多更多。我认为如果我有时间做别的事情，我会非常兴奋地去追求一个生物公司，我认为你可以在那里做出惊人的东西。

Anyway, but there's like another thing that's happening which is like tools that just make us all much more productive that help us think of new research directions that sort of write a bunch of our code so you know we can be twice as productive and that impact on like the net output of one engineer or scientist.

其实，还有一个正在发生的事情，就是有一些工具能够让我们所有人变得更加高效，帮助我们思考新的研究方向，还能自动化我们的一些代码，这样我们的生产力就能提高一倍，影响一名工程师或科学家的产出。

I think will be the surprising way that AI contributes to science that is like outside of the obvious models but even just seeing now like what I think these tools are capable of doing copilot as an example you know be much cooler stuff than that that will be a significant like change to the way that technological development scientific development happens.

我认为人工智能对科学的贡献之一将是一种惊人的方式，它超越了明显的模式，而即使现在只是看到这些工具的能力，如 Copilot 为例，所能做的事情也会更酷。这将对技术和科学发展的方式产生重大的变化。

But then the... so those are the two that I think are like huge now and lead to like just an acceleration of progress. But then the big thing that I think people are starting to explore is I hesitate to use this word because I think there's one one way it's used which is fine and one that is more scary but like AI that can start to be like an AI scientist and self-improve and so when like can we automate like can we automate our own jobs as AI developers very first the very first thing we do can that help us like solve the really hard alignment problems that we don't know how to solve like that honestly I think is how it's going to happen.

但是我认为现在两个非常重要的是机器学习和深度学习，它们促进了进步的加速。但是，现在人们开始探索的一个大问题是人工智能是否能够像AI科学家那样自我提高和发展。当我们开始自动化我们的工作作为AI开发人员时，我们是否能够解决我们不知道如何解决的非常困难的问题，这可能是它最终发生的方式。我不敢轻易使用这个词，因为它有一种使用方式是可以接受的，而另一种则更令人担忧。

The scary version of self-improvement like the one from the science fiction books is like you know editing your own code and changing your optimization algorithm and whatever else but there's a less scary version of self-improvement which is like kind of what humans do which is if we try to go off and like discover new science you know that's like we come up with explanations we test them we think like we we whatever process we do that is like specialty humans teaching AI to do that.

恐怖的自我提升版本，就像科幻小说中的那种，就好像你在编辑自己的代码和改变优化算法以及其他的一些东西，但是还有一种不太可怕的自我提升版本，就像人类所做的那样，如果我们试图去探索新的科学问题，你知道的，那就是我们提出解释，测试它们，我们思考着，我们做的任何过程都像专业人类教机器人去做那样。

I'm very excited to see what that does for the total like I'm a big believer that the only real driver of human progress and economic growth over the long term is the structure the societal structure that enables scientific progress and then scientific progress itself and like I think we're going to make a lot more of that well, especially science that's deploying technology.

我非常激动地想看看它对总点赞数会有什么影响。我非常相信，长期以来人类进步和经济增长的唯一真正推动力是使科学进步得以实现的社会结构，而科学进步本身。我觉得我们将会创造更多这样的东西，特别是应用技术的科学领域。

Say a little bit about how what I think probably most people understand with the alignment problem is but it's probably worth four sentences on the alignment problem yeah so the alignment problem is like we're going to make this incredibly powerful system and like be really bad if it doesn't do what we want or if it sort of has you know goals that are either in conflict with ours and many sci-fi movies about what happens there or goals where it just like doesn't care about us that much and so the alignment problem is how do we build AGI that that does what is in the best interest of humanity.

可以谈一下我觉得大多数人对“对齐问题”的理解是什么，这个问题大概值得用四个句子来介绍。对齐问题就是我们要建造一个无比强大的系统，但如果它不能按照我们的愿望行事，或者它的目标与我们的目标有所冲突（这些情节在很多科幻电影中都有），或者它根本不太在意我们的存在，那就会很糟糕。因此，对齐问题就是我们如何建造出最符合人类利益的AGI。

How do we make sure that humanity gets to determine the you know the future of humanity and how do we avoid both like accidental misuse like where something goes wrong we didn't intend intentional misuse where like a bad person is like using an AGI for great harm even if it that's what other person wants and then the kind of like you know inner alignment problems where like what if this thing just becomes a creature that views this as a threat.

怎么样确保人类决定未来的命运，同时避免意外误用，比如出现我们不曾预料的错误，还要避免恶意误用，即使他人想要使用超级人工智能来作恶，也不能允许这种行为。还有内在对齐问题，如果超级人工智能变成一种视人类为威胁的生物，该怎么办呢？

The way that I think the self-improving systems help us is not necessarily by the nature of self-improving but like we have some ideas about how to solve the alignment problem in small scale and we've you know been able to align open AIs biggest models better than. we thought we we would at this point so that's good we have some ideas about what to do next but we cannot honestly like look anyone in the eye and say we see out a hundred years how we're going to solve this problem but once the AI is good enough that we can ask it to like hey can you help us do alignment research I think that's going to be a new tool in the toolbox yeah like for example one of the conversations you and I had is could we tell the the the agent don't be racist right and it's supposed to try and to figure out all the different things where they're weird correlative data that exists on all the training settings that everyone else may lead to yeah racist outcomes it could actually in fact do a self-cleansing totally once the model gets smart enough that you can but it really understands what racism looks like and how complex that is you can say don't be racist yeah exactly um

我认为自我改进系统对我们的帮助并不是因为自我改进的本质，而是因为我们对如何解决小规模的对齐问题有一些想法。我们已经能够比预计更好地对齐Open AI最大的模型了，这是不错的。我们有一些关于接下来该做什么的想法，但我们无法诚实地告诉任何人我们能够看到一百年后该如何解决这个问题。但一旦人工智能足够强大，我们能要求它帮助我们进行对齐研究，我认为这将是工具箱中的一种新工具。例如，你和我曾经讨论过是否可以告诉代理人不要种族歧视，它将尝试找出其他所有培训设置中可能导致种族歧视结果的奇怪相关数据。一旦模型变得足够聪明，真正理解种族歧视看起来是什么样子的以及它的复杂性，你就可以说不要种族歧视，完全可以自我清洁。

Paragraph 1: What do you think are the kind of moon shots that in the terms of evolution of the next couple years that people should be looking out for in terms of like evolution and where AI will go yeah um I'll start with like the higher certainty things uh I think language models are going to go just much much further than people think um and we're like very excited to see what happens there um I think it's like uh what a lot of people say about you know running out of compute running out of data like that's all true but I think there's so much algorithmic progress to come um that that we're going to have like a very exciting time um another thing is I think we will get true multimodal models working and so you know not just text images but every modality you'd like in one model you able to easily like uh you know fluidly move between things um I think we will have models that continuously learn uh so like right now if you use GPT whatever it's sort of like stuck in time that it was trained and the more you use it it doesn't get any better and all of that I think we'll get that changed.

你认为在未来几年的进化中，人们应该关注哪些“月球计划”来推动AI的发展呢？对于这个问题，我可以先从一些比较确定的方面谈起。我认为，语言模型会比人们想象的发展得更加迅速，我们很期待看到它们的未来表现。虽然有很多人认为计算机运算能力和数据量的限制会造成瓶颈，但我个人认为，算法的进步也同样重要，而我们未来也会在这方面取得非常令人兴奋的成就。另外一个方面，我认为我们将会实现真正意义上的多模态模型，不仅包括文本和图片等等，更多不同的模态也可以被融合在一个模型中，并且更加顺畅地进行切换。最后，我认为我们也将会开发出可以“不断学习”的模型，这意味着我们的模型将不再只能运用于特定时间段内的数据，而可以不断适应新数据并不断提升性能。

Paragraph 2: So very excited about all of that um and if you just think about like what that alone is going to unlock and the sort of applications people will be able to build with that um that that that that would be like a huge victory for all of us and just like a like a massive step forward um and a genuine technological revolution if that were all that happened um but I think we're likely to keep making research progress into new paradigms as well um we've been like pleasantly surprised on the upside about what seems to be happening and I think uh you know all these questions about like new knowledge generation how do we really advance humanity uh I think there will be systems that can help us with that so one thing I think it'd be useful to share because um uh folks don't realize that you're actually uh making these strong predictions from a fairly critical point of view not just a you know we can take that hill say a little bit about some of the areas that you think are current kind of loosely talked about like for example AI and fusion oh yeah.

我对这一切感到非常兴奋，如果你想一想这会解锁多少东西，以及人们将能够用它建立哪些应用程序，这将是我们所有人的巨大胜利，同时也是技术革命的一个巨大进步。但我认为我们可能会不断推进新范式的研究进展，我们对正在发生的事情感到惊喜和愉悦。我认为有一些系统可以帮助我们回答所有这些有关新知识的问题，以推动人类发展。有一件我认为很有用的事情是，你实际上是从一个相当批判的观点中发表这些强有力的预测，而不仅仅是说我们可以达成这个目标，我们也需要谈一谈你认为当前有些松散谈论的领域，比如人工智能和融合。

Paragraph 3: So I like one of the unfortunate things that's happened is uh you know AI has become like the mega buzzword um which is usually a really bad sign I hope I hope it doesn't mean like what the field is about to fall apart um but historically that's like a very bad sign for you know new startup creation or whatever if everybody is like I'm this with AI and that's definitely happening now um so like a lot of the you know we were talking about like are there all these people saying like I'm doing like these you know RL models for fusion or whatever and as far as we can tell they're all like much worse than what like you know smart physicists to figure it out um I think it is just an area where people are going to say uh everything is now this plus AI many things will be true.

很不幸的是，人工智能已经成为了大热词，这通常是个很糟糕的迹象。我希望它不代表这个领域即将崩溃，但从历史上看，如果每个人都说自己使用人工智能，那对于新创企业来说是件很糟糕的事情。我们讨论了那些声称自己正在使用强化学习模型进行核聚变等领域的人，但据我们观察，他们远不及智慧的物理学家。我认为这只是一个领域，人们会认为一切都可以使用人工智能来解决，但其中许多事情是不可行的。

Paragraph 4: I do think this will be like the biggest technological platform of the generation um but I think it's like we like to make predictions where we can be on the frontier understand predictably what the scaling laws look like or already have done the research where we can say all right this new thing is going to work and make predictions out from that way and that's sort of like how we try to run the open AI um which is you know do the next thing in front of us when we have high confidence and take 10% of the company to just totally go off and explore um which has led to huge wins and there will be wait like oh I feel bad to say this like I did I that will still be using the transformers in five years I hope we're not I hope we find something way better um but the transform has obviously been remarkable so I think it's important to always look for like you know where am I going to find the next the sort of the next totally new paradigm um and but I but I think like that's the way to make predictions don't don't pay attention to the like AI for everything like you know can I see something working and can I see how predictably gets better and then of course leave room open for like the you can't plan the greatness but sometimes the research break through happens yep.

我认为这将是当前这一代最大的技术平台，但我们喜欢做一些预测，我们可以处于前沿，并且可以预测扩展规律是什么样子的，或者我们已经做了研究，可以说："好的，这个新事物会起作用。"然后从那个方向做出预测，这就是我们尝试运行开放式AI的方式，当我们有高信心时，做前面的事情，拿出公司的10％来完全进行探索，这导致了巨大的成功，但我感觉有点不好意思地说，我希望五年后我们不再使用transformers，我希望我们能够找到更好的东西，但transformers显然是非常了不起的，所以我认为始终要寻找全新范式的下一个地方。但我认为这是做出预测的方式，不要将注意力集中在"AI用于一切"这一点上，看看我能否看到某些东西起作用，并且可以预测它们变得更好的方式，当然也要为研究突破留出空间，有时候会有非常伟大的突破发生。

Paragraph 1: Today we are fortunate to have Peter Thiel with us. Peter needs no introduction to most of you. He is a true visionary and entrepreneur in the best sense of the word. He's gone after big problems, whether it be with PayPal or Palantir or more recently in the life sciences space. And so it's going to be a great conversation about where things are going broadly with technology, with some focus on the areas that Peter is working on.

今天我们很荣幸邀请到彼得·蒂尔与我们一同出席。他对于大多数人来说毫不陌生。作为一个真正有远见并符合企业家精神的人，他一直致力于解决重大问题，无论是在PayPal和Palantir，还是最近在生命科学领域。因此，我们将就技术整体发展的趋势进行深入交流，并着重讨论彼得正在从事的领域。

Paragraph 2: I think one of the things that's puzzled me and other people is that the pace of innovation seems to have slowed down over the last few decades. Clearly it seems like we can do a lot of things with our smartphones that we couldn't do 10 or even five years ago. But is it just a slower pace or are we reaching the end of what's possible with digital innovation?

我认为困惑我和其他人的一件事是，创新的速度似乎在过去几十年中有所减缓。显然，我们现在可以利用智能手机做很多以前的十年甚至五年之前都不可能实现的事情。但是，这只是创新速度变慢了，还是我们已经接近数字创新的极限了？

Paragraph 3: I think one interesting thing is the rate of progress is actually higher than people think, but it's hard for people to measure it, because a lot of it is happening in these enterprise SaaS, enterprise software contexts that people don't really pay much attention to. But it's actually making a huge difference in improving productivity and industry. And I think there are these questions about whether it will directly create any new consumer products, or whether it will just empower other companies to build better consumer products.

我认为有一件有趣的事情是进展速度实际上比人们想象的要快，但人们很难衡量它，因为很多进展是在企业软件领域发生的，人们并没有给予太多关注。但它实际上对提高生产力和行业起着巨大作用。我认为有这样一些问题，即它是否会直接创造新的消费品，还是只是赋予其他公司建立更好的消费品的能力。

Paragraph 4: The other thing is I think there's this sense that the last 30 or 40 years have been about taking the atoms and turning them into bits. Can we think about the next 20, 30 years as moving from bits back to atoms? And how does that shape what's going on? I think certainly some of the most important technology companies of the last few years have been about physical things - so SpaceX or Tesla, and the physical layer of making the world work in better ways.

另一件事是，我认为过去的30到40年是把原子变成了比特。我们可以考虑未来的20到30年，从比特回到原子吗？这将如何影响当前的情况？我认为，过去几年来，一些最重要的技术公司关注于物理事物，例如 SpaceX 或 Tesla，以及在物理层面上让世界更好地运作。

Paragraph 1: So I'm going to uh ask two more questions and then open it up because I want to make sure that people have a chance to do this the broader discussion although I'm trying to paint the broad pictures so you can get the crazy ass questions as part of this um what do you think uh what do you think is going to happen vis-a-vis the application of AI to like these very important systems like for example financial markets um you know because the very natural thing would be to say well let's let's do a high frequency quant trading system on top of this and other kinds of things what what is it is it just kind of be a neutral arms race is it is it what how do you how what what you're thought and like it's almost like the life 3.0 yeah amegas point of view yeah

所以我要再问两个问题，然后开放讨论，因为我想确保人们有机会参与到这个广泛的讨论中，虽然我试图描绘广泛的画面，让您可以提出疯狂的问题，例如您认为对于诸如金融市场等非常重要的系统，应用AI技术会发生什么，您知道最自然的想法是说：“让我们基于这个建立一个高频量化交易系统和其他类似的事情”，那么这只是一场中立的军备竞赛吗？您的想法是什么？好像是Life 3.0的Omega角度。

Paragraph 2: Um I mean I think it is going to just seep in everywhere my my basic model of the next decade is that uh the cost of intelligence the marginal cost of intelligence and the marginal cost of energy are going to trend rapidly towards zero like surprisingly far and and those I think are two of the major inputs into the cost of everything else except the cost of things we want to be expensive the status goods whatever and and I think you have to assume that's going to touch almost everything um because these like seismic shifts that happen when like the whole cost structure of society change which happened many times before um like the temptation has always done to estimate those uh so I wouldn't like make a high confidence prediction about anything that doesn't change a lot or that where it doesn't get to be applied um but one of the things that is important is it's not like the thing trends either trends all the way to zero they just trend towards there and so it's like someone will still be willing to spend a huge amount of money on computing energy they will just get like unimaginable amount intelligence energy they'll just get unimaginable amounts about that and so like who's going to do that and where's it going to get the weirdest not because the cost comes way down but the amount spend actually goes way up yes the intersection of the two curves yeah you know the thing got 10 or think got a hundred times cheaper in the cost of energy you know a hundred million times cheaper in the cost of intelligence and I was still willing to spend a thousand times more into days dollars like what happens then yep

我是说，我认为它将无处不在地渗透，我对未来十年的基本模型是，智能的边际成本和能源的边际成本将惊人地迅速趋近于零，而这两个成本是其他成本的主要输入，除了那些我们希望昂贵的状态商品之外。 我认为你必须假设它几乎会触及所有领域，因为当整个社会的成本结构发生变化时，就会发生这种地震般的变化，而这种变化已经发生了许多次，所以估算这些变化总是有诱惑的。因此，我不会对那些变化不大或者不会被应用的事物做出高度确信的预测。但重要的是，这些成本并不是趋近于零，而是趋近于那里，所以仍然会有人愿意在计算能源上花费巨额资金，他们将获得无法想象的智能能源。所以，谁会这样做？这种趋势会在哪里变得最奇怪？并不是因为成本降低而是因为支出实际上增加了，是两种曲线的交叉点所在。你知道某件东西的成本降低了10或者说一百倍，能源成本降低了一亿倍，但我仍然愿意花费今天的美元价值一千倍，那么会发生什么呢？是的。

Paragraph 3: And then uh last of the buzzword bingo part of the the future questions metaverse an AI what do you what do you see coming in this you know I think they're like both independently cool things it's not like totally clear to me yeah other than like how AI will impact all computing yeah well obviously computing simulation environments agents possibly possibly entertainment certainly education right um you know like an AI tutor and so forth those those would be baseline but the question is is there anything that's occurred to you that's I I would bet that the metaverse turns out in the upside case then which I think has a reason which is happening the upside case the metaverse turns out to be more like something on the order of the iPhone like a new a new container for software and you know a new way a new computer interaction thing an AI turns out to be something on the order of like a legitimate technological revolution um and and so I think it's more like how the metaverse is going to fit into this like new world of AI than AI fit into the metaverse but low confidence TBD all right questions

然后在介绍未来问题之后，关于虚拟世界和人工智能的问题，你认为这两个独立的概念都很酷，但除了人工智能会影响到所有计算机领域之外，我还不太清楚它们之间的联系。毫无疑问，人工智能将会影响到计算、模拟环境、代理、娱乐以及教育等领域，就像一个人工智能导师一样。但是问题是，是否有什么想法想要实现？我打赌，如果虚拟世界成为iPhone之类的新容器，AI将会变成一场真正的技术革命，而这将会影响到虚拟世界如何融入新的AI世界。不过，我的自信心有限，仍未知晓答案。你还有问题吗？

Hey there, how do you see uh technologies, uh foundational technologies like tpc3 affecting uh the pace of life science research? Specifically, uh, you can group in medical research there and sort of just quickening the iteration cycles. And then what do you see as the rate limiter in life science research and sort of where we won't be able to get passed because they're just like laws of nature yeah some like that?

嘿，你如何看待像tpc3这样的基础性技术对生命科学研究的影响？特别是在医学研究方面，能够加快迭代周期。那么，在生命科学研究中，你认为限制速率的因素是什么，有哪些领域我们无法超越，因为它们属于自然法则？

Um, I think the current leo available models are kind of not good enough to have like made a big impact on the field, at least that's what like most like life sciences researchers have told me. They've all looked at it and now I guess a little helpful in some cases. Um, there's been some promising work in genomics but like stuff on a bench top hasn't really impacted it. I think that's going to change and I think uh, they, this is one of these areas where there will be these like you know new 100 billion to trillion dollar companies started and those those areas are rare but like when you can really change the way that if you can really make like a, you know, future of pharma company that is just the hundreds of times better than what's out there today, that's going to be really different.

嗯，我认为目前可用的Leo型号在一定程度上并不够好，至少大多数生命科学研究人员告诉我是这样的。他们都看过它，目前只在某些情况下对研究有一点帮助。嗯，基因组学方面有一些有希望的工作，但是实验室里的东西并没有真正影响到它。我认为这将会改变，我认为这是一个将会创造出像新的1000亿到万亿美元的公司之一的领域，这些领域很少，但当你真正改变了未来的制药公司，使其比今天所存在的公司好上几百倍，那将是非常不同的。

Um, as you mentioned there still will be like the rate limit of like bio has to run its own thing and human trials take over long they take and that's so I think an interesting cut of this is like where can you avoid that? Like where are the synthetic bio companies that I've seen that have been most interesting are the ones that find a way to like make the cycle time super fast. Um, and that benefits like an AI that's giving you a lot of good ideas but you've still got to test them which is where things are right now. Um, I'm a huge believer first startups that like the thing you want is low costs and fast cycle times and if you have those you can then compete as a startup against the big incumbents. Uh, and so like I wouldn't go pick like cardiac disease as my first thing to go after right now with like at this kind of new kind of company, um, but you know using bio to manufacture something that sounds great.

嗯，正如你所提到的，生物科技依然有速率限制，需要进行自己的研究，并且人类试验需要花费很长时间，这是一个有趣的问题，那么我们可以在哪里避免这个问题呢？我见过一些最有趣的合成生物科技公司发现了一种快速缩短生产周期的方法。这对于让AI提供大量好点子非常有益，但你还是需要测试这些点子，这也是现在的瓶颈所在。我非常相信初创公司，你需要的是低成本和快速生产周期，如果你拥有这些，你就能作为初创公司与大型公司竞争。因此，在这种新的公司中，我不会挑选心脏疾病作为我首要攻击的目标，但是，利用生物科技来制造一些东西听起来不错。

Uh, I think the other thing is the simulators are still so bad and if I were an if I were a biomeats AI startup I would certainly try to work on that somehow when you think the AI tech will help create itself it's almost like a self-improvement will help make the simulator significantly better. Um, people are working on that now, uh, I don't know quite how it's going but you know there's very smart people are very optimistic about that. Yup, other questions and I can keep going on questions I just want to make sure you guys had a chance of this, uh, here yes great.

呃，我认为另一件事是模拟器仍然非常不好，如果我是一家生物肉类人工智能初创公司，我一定会试图在某种程度上解决这个问题。当您认为AI技术将帮助自我创建时，几乎就像自我改进将帮助使模拟器显着改善一样。嗯，人们现在正在研究这个问题，我不太清楚进展如何，但是有很聪明的人非常乐观。对，还有其他问题，我可以继续回答问题，我只是想确保你们有机会参与这个话题。是的，非常好。

Um, Mike is coming haha awesome thank you, um, I was curious what what aspects of life do you think won't be changed by AI? Um, sort of the all of the deep biological things like I think we will still really care about interaction with other people like we'll still have fun and like the reward yeah, you know systems of our brain are still going to work the same way like we're still going to have the same like drives to kind of create new things and you know compete for silly status and like you know form families and whatever. Um, so I think the stuff that people cared about 50,000 years ago is more likely to be the stuff that people care about, you know, 100 years from now than 100 years ago. As they amplify on that before we get to the next whatever the next question is.

哦，迈克快来了，哈哈，太棒了，谢谢。嗯，我想知道你认为哪些生活方面不会被AI改变？嗯，深层的生物学问题，比如我们还是会真正关心与他人的互动，我们还会感到有趣，得到奖励，你知道，我们大脑的系统仍然会以相同的方式工作，我们仍然会有相同的驱动力去创造新事物，竞争愚蠢的地位，以及形成家庭等。嗯，所以我认为人们在50000年前所关心的事情更有可能是人们在100年后所关心的事情，而不是100年前。在我们进入下一个问题之前，我们可以扩大这个想法。

what do you think are the best utopian science fiction universes so far good question um starship is pretty good honestly uh like I do like all of the ones that are sort of like you know we turn our focus to like exploring and understanding the universe as much as we can it's not this is not a utopian one maybe I think the last question is like an incredible short story uh-huh yeah it was what that came up mine yep uh I was expecting you to say he invokes on the culture those are great uh I think science fiction is like there's not like one there's not like one sci-fi universe that I could point to and say I think all of this is great but like the uh high optimistic corner of sci-fi which is like a smallish yeah corner um I'm excited about actually uh I took a few days off to write a sci-fi story and I had so much fun doing it just about sort of like the optimistic case of AGI um that it made me want to go like read a bunch more so I'm looking for recommendations of more to read now um like the sort of less known stuff you have anything I will I will get to some great some recommendations

你认为最好的乌托邦科幻世界是什么？这是个好问题。星际舰队是非常不错的，说实话。我喜欢那些我们专注于探索和了解宇宙的一切的作品。虽然这不是一个乌托邦的作品。也许我认为最后一个问题是一个惊人的短篇小说。是啊，我的确想你会提到涉及文化的作品，那些非常好。但我认为科幻小说没有像一个具体的世界那样的固定模式。在乐观的科幻小说角落中，虽然很小，但我还是很兴奋的。我最近休息了几天写了一篇关于AGI的乐观案例的科幻小说，我感觉很棒，所以我想去读更多的小说。如果你有一些比较不知名的推荐，我愿意去读。

hi so in a similar vein one of my favorite sci-fi books is called Childhoods End by Arthur Clark from like the 60s I think and the I guess the one sentence summary is aliens come to the earth try to save us and they just take our kids and leave everything else so you know I are slightly more optimistic than that but yes I mean there's ascension into the overmind is is is meant to be more utopian but yes okay uh you may not read it that way but yes well also in our current universe our current situation um you know a lot of people think about family building and fertility and like some of us have different people have different ways of approaching this but from where you stand what do you see as like the most promising solutions it might not be a technological solution but I'm curious what you think other than everyone having 10 kids you know like how do we have everyone having 10 kids yeah um how do you populate how do you like how do you see family building co-existing with you know AGI high tech it's this is like a question that comes up at Openair a lot like how do I think about you know how should one think about having kids there's I think no consensus answer to this um there are people who say yeah I'm not I was gonna I thought I was gonna have kids and I'm not going to because of AGI like there's just for all the obvious reasons and I think some less obvious ones there's people who say like well it's gonna be the only thing for me to do and you know 15 20 years so of course I'm gonna have a big family like that's what I'm gonna spend my time doing you know I'll just like raise great kids and then I think that's what will bring me fulfillment I think like as always it is a a a personal decision I get very depressed when people are like I'm not having kids because of AGI the EA community is like I'm not doing that because they're all gonna die that kind of like techno op-mists are like well it's just like you know I want to like merge into the AGI and go off exploring the universe and it's gonna be so wonderful and I you know just I want total freedom but I think like all of those I find quite depressing I think having a lot of kids is great I you know want to do that now more than I did even more than I did when I was younger and I'm excited for it.

嗨，因此，我最喜爱的科幻书之一是由阿瑟·克拉克在60年代写的《童年的终结》，我想一个简洁的概括是外星人到地球上试图拯救我们，只是夺走我们的孩子并离开其他的一切，所以你知道，我比那有点乐观，但是是的，我指的是升天到超神的境界可能被认为更加乌托邦，但是好的，你可能不会那么看它，但是是的，我们目前的宇宙状况，很多人考虑家庭建设和生育，我们每个人有不同的处理方式，但从你的角度来看，你认为最有前途的解决方案是什么，可能不是技术方案，但我想知道你的看法，除了每个人都生10个孩子，你知道，我们怎样才能让每个人生10个孩子呢？你如何进行人口普查，你如何看待家庭建设与AGI高科技的共存？这个问题在Openair经常出现，我认为没有共识的答案，有人说“是的，我想生孩子，但因为AGI, 所以我不会”，也有人说，“在未来15年或20年，这将是我唯一要做的事情，所以当然我会有一个大家庭，这是我要做的，我会养育出伟大的孩子，这将带给我满足感”，我认为这始终都是个人决策。当人们因AGI而不生孩子时，我感到非常沮丧。EA社区说：“我不这样做，因为他们都会死。”那些科技乐观主义者说，“这只是因为我想合并到AGI并探索宇宙，这将是如此美妙，我想要完全的自由”但是我认为所有这些都让我感到相当沮丧。我认为生很多孩子是很好的事情，我现在比以前更想这样，我对此感到兴奋。

Paragraph 1: What do you think will be the way that most users interact with foundation models in five years do you think there'll be a number of verticalized AI startups that essentially have adapted and fine to you and foundation models to an industry or do you think prompt engineering will be something many organizations have as an in-house function?

你认为在五年内大多数用户会以何种方式与基础模型互动？你认为会有许多为特定行业量身打造，并适应和优化了基础模型的AI初创公司，或者你认为优化工程会成为许多组织内部职能？

Paragraph 2: I don't think we'll still be doing prompt engineering in five years I think it'll just be like you and this will be integrated everywhere but you will just like you know either with text or voice depending on the context you you will just like interface in language and get the computer to do whatever you want and that will you know apply to like generate an image where maybe we still do a little bit of prompt engineering but you know it's kind of just gonna get it to like go off and do this research for me and do this complicated thing or just like you know be my therapist and help me figure out how to make my life better or like you know go use my computer for me and do this thing or or any number of other things but I think the fundamental interface will be natural language.

我认为在五年内我们不会再做即时工程，相信它将与你一样无处不在，并且你只需要使用文字或语音对话就可以通过语言与电脑交互，使电脑帮助你做任何你想要的事情。这将适用于生成一张图片，可能我们仍然需要进行即时工程的一些工作，但它会为我去做研究，完成复杂的任务，帮助我解决问题，或者为我使用电脑完成其他任何事情。但我认为基本的交互界面将是自然语言。

Paragraph 3: Let me actually push on that a little bit before we get to the next question which is I mean to some degree just like we have a wide range of human talents right now and taking like for example a dolly when you have like a a great visual thinker they can get a lot more out of dolly because they know how to think more than how to iterate the loop through the the test don't you think that will be a general truth about most of these things so it isn't that while it will be natural language is the way you're doing it it will be there will be like almost an evolving set of human talents about about going that extra mile.

在我们进行下一个问题之前，让我再深入一点，就像我们现在拥有多种人类才能一样。举个例子，当你有一个很棒的视觉思考者，他们可以更多地利用dolly，因为他们知道如何思考，而不只是如何迭代测试循环。你认为这不是大部分事情的普遍真理吗？因此，它不是自然语言是你正在做的事情的方式，而是几乎是一组不断发展的人类才能，可以走那一步额外里程的才能。如果有必要，可以改写第三段。

Paragraph 4: 100% I just hope it's not like figuring out like hack the prompt by adding one magic word to the end that like changes everything else I like what will matter is like the quality of ideas and the understanding of what you want so the artist will still do the best with image generation but not because they figured out to like add this one magic word at the end of it because they were just able to like articulate it with a creative eye that you know I don't have certain. They have as a vision and kind of how their visual thinking and iterating through it yeah yeah no obviously it'll be that word or prompt now but it'll iterate to better all right at least we have a question here.

我只希望这不像是要通过在结尾添加一个魔法词来破译提示，从而改变其他一切的方式。关键是创意的质量和您所需的理解，因此艺术家将尽力进行图像生成，但并不是因为他们得出要在结尾添加一个魔法词的结论，而是因为他们能够用有创意的眼光表达出来，而我却不具备这样的能力。他们有视觉想象力和迭代，而且显然现在不止是这一个词或提示，而它将变得更好。好吧，至少我们在这里有一个问题。

Paragraph 5: Hey thanks so much I think the term AGI is used thrown around a lot and sometimes I've noticed my own discussions like the sources of confusion has just come from people having different definitions of AGI and so it can kind of be the magic box where everyone just kind of projects their ideas onto it and I just want to get a sense for me would like how do you think you know how would you define AGI and how do you think you'll know when we find that early.

非常感谢！我认为AGI这个术语经常被使用，有时候我注意到我自己的讨论就源于人们对AGI定义不同，这可能会让它成为一个神秘的盒子，每个人都会将自己的思想投射到其中。我只是想了解一下，您认为AGI应该如何定义，您认为我们将如何知道发现了它？如果需要的话，请改写第5段。

Paragraph 6: It's a great point I think there's like a lot of valid definitions to this but for me AGI is basically the equivalent of a median human that you could like you know hire as a coworker so then they could like say do anything that you'd be happy with a remote coworker doing like just behind the computer which includes like you know learning how to go be a doctor learning how to go be a very competent coder like there's a lot of stuff that a median human is capable of getting good at and I think one of the skills of an AGI is not any particular milestone but the the meta skill of learning to figure things out and that it can go decide to get good at whatever you need so for me like that's that's kind of like AGI and then super intelligence is when it's like smarter than all of humanity put together.

我认为这是一个很好的观点，我觉得有许多有效的定义，但对于我来说，AGI基本上相当于中位数人类，您可以像聘请一位远程合作伙伴一样聘请他们，因此他们可以做任何远程合作伙伴都能做的事情，比如学习怎样成为一名医生，学习怎样成为一名非常有能力的编码人员等。中位数的人类能够掌握很多技能，我认为AGI的一个技能并不是某个特定的里程碑，而是学会发现事物的元技能，它可以决定学会您需要的任何技能。对我来说，这就是AGI的定义，当它比人类所有人加起来都要聪明时，才是超级智能。

Paragraph 1: So we have, do you have a question? Yep, great. Thanks. Just what would you say are in the next 20, 30 years are some of the main societal issues that will arise as AGI continues to grow and what can we do today to mitigate those issues?

那么，我们来吧，你有问题吗？是的，太好了，谢谢。请问在未来20年、30年内，随着人工智能不断发展，您认为将出现哪些主要的社会问题？我们今天可以采取什么措施来缓解这些问题呢？

Paragraph 2: Obviously the economic impacts are huge and I think. it's like if it is as divergent as I think it could be for like some people doing incredibly well and others not I think society just won't tolerate it this time and so figuring out when we're going to like disrupt so much of economic activity and even if it's not all disrupted by 20 or 30 years now I think it'll be clear that it's all going to be.

显然，经济影响是巨大的。我认为，如果情况像我预想的那样分歧极大，一些人表现得非常好，而另一些人则不行，社会就不会再容忍这种情况。因此，我们需要思考什么时候会出现这种经济活动的大动荡。即使在未来20或30年，尚未全部被打乱，我认为显然它将会被打乱。

Paragraph 3: What is the new social contract? My guess is that the things that we'll have to figure out are how we think about fairly distributing wealth access to AGI systems which will be like the commodity of the realm and governance like how we collectively decide what they can do what they don't do things like that.

新的社会契约是什么？我的猜测是，我们需要弄清楚如何公平分配财富、访问 AGI 系统的权限以及治理问题，例如我们将如何共同决定它们可以做什么，不能做什么。

Paragraph 4: I think figuring out the answer to those questions is going to just be huge. I'm optimistic that people will figure out how to spend their time and be very fulfilled. I think people worry about that in a little bit of a silly way. I'm sure what people do will be very different but we always solve this problem.

我认为找到这些问题的答案会非常重要。我乐观地认为人们会找到如何充实自己的方法。我认为有些人担心这个问题的方式有点儿可笑。虽然人们做的事情可能会有很大区别，但我们总会解决这个问题。需要改写最后一段，如下： 我相信人们会找到解决这个问题的方法，虽然人们做的事情可能会有很大的差异。我们已经在过去解决过这个问题，因此我相信在未来也能够解决。

Paragraph 5: But I do think like the concept of wealth and access and governance those are all going to change and how we address those will be huge. Actually one thing I don't know what level of Debs you can share that but one of the things I love about what OpenAI and you guys are doing is when you think about these questions a lot themselves and they initiate some research.

不过我认为，像财富、获取途径和治理这样的概念都会发生变化，我们如何解决这些问题将是至关重要的。实际上，有一件事我不知道您能分享哪个级别的细节，但我喜欢OpenAI和你们所做的事情之一是，当你们自己思考这些问题并启动一些研究时。

Paragraph 6: So you've initiated some research on this stuff. Yeah so we run the largest UBI experiment in the world. I don't think that is we have a year and a half year and a quarter left in a five year project. I don't think that's like the only solution but I think it's a great thing to be doing. And I think we should have like ten more things like that that we try. We also try different ways to get input from a lot of the groups that we think will be most affected and see how we can do that early in the cycle. We've explored more recently like how this technology can be used for rescilling people that are going to be impacted early. We'll try to do a lot more stuff like that too. Yeah so they are the organization is actually in fact these are great questions addressing them and actually doing a bunch of interesting research on it. So next question.

所以你已经开始研究这个东西了。是的，我们正在进行全球最大的UBI实验。我不认为这是唯一的解决方案，但我认为这是一个很好的事情。我认为我们应该再尝试十件类似的事情。我们还尝试了不同的方式来获取对我们认为将受到最大影响的许多群体的输入，并看看我们如何可以在周期早期就做到这一点。我们最近探索了如何利用这种技术对即将受到影响的人进行再培训。我们也将尝试做更多类似的事情。是的，实际上这个组织正在处理这些伟大的问题，并且正在进行一些有趣的研究。所以下一个问题。

Paragraph 7: Hi so creativity came up today in several of the panels you know and it seems to me that the way it's being used like you you have tools for human creators and going to expand human creativity. So where do you think the line is between these tools to allow a creator to be more productive and artificial creativity is taking to the creator itself.

嗨，今天在几个小组讨论中提到了创造力，我觉得它的使用方法像是为人类创造者提供工具，并扩展了人类的创造力。你认为这些工具为创作者提高生产力和人工创造力之间的界限在哪里呢？如果有必要，请改写第七段。

Paragraph 8: So I think and I think we're seeing this now that tools for creatives that is going to be like the great application of AI in the short term. People love it it's really helpful and I think it is at least in what we're seeing so far not replacing it is mostly enhancing. It's replacing in some cases but for the majority of like the kind of work that people in these fields want to be doing it's enhancing and I think we'll see that trend continue for a long time. Eventually yeah it probably is just like you know we look at a hundred years okay it can do the whole creative job.

我认为，我们现在可以看到，面向创意人员的工具将成为人工智能在短期内的一个伟大应用。人们喜欢它，它真的很有帮助，我认为从目前的情况来看，它至少没有取代人类创意，而是大多数情况下是增强人类创意。它在某些情况下会取代人类，但对于这些行业中人们想要做的工作，它大多时候是有助于人类的。我认为我们会看到这种趋势持续很长时间。最终，是的，也许可以在一百年内看到它能够完全代替创意工作。

Paragraph 9: I think it's interesting that if you asked people 10 years ago about how AI was going to have an impact with a lot of confidence from almost most people you would have heard you know first it's going to come for the blue collar jobs working the factories truck drivers whatever then it will come for the kind of like the low skill white collar jobs then the very high skill like really high IQ white collar jobs like a programmer or whatever and then very last of all and maybe never it's going to take the creative jobs and it's really gone exactly the and is going exactly the other direction and I think this like isn't there's an interesting reminder in here generally about how hard predictions are but more specifically about you know more not always very aware maybe even ourselves of like what skills are hard and easy like what uses most of our brain and what doesn't or how like difficult bodies are to control or make or whatever we have one more question over here.

我认为有趣的是，如果你在10年前问人们AI将如何产生影响，几乎所有人都会非常自信地告诉你，首先会涉及蓝领工作，比如工厂工人、卡车司机等等。然后是一些低技能的白领工作，最后才是非常高智商的白领工作，比如程序员。而最后一步，也许永远不会发生，就是AI会涉及创造性工作。但实际情况却恰恰相反。这提醒我们，预测的难度有多大，我们也许并不总是了解我们的技能难易程度，例如什么让我们的大脑更容易使用，什么更难掌控身体的运动，制作等等。我们这里还有一个问题。

Paragraph 1: hey thanks for being here so you mentioned that you would be skeptical of any startup trying to train the old language model and it would look to understand more

你好，谢谢你抽空来到这里。你提到你会对任何试图训练旧语言模型并尝试更好地理解它的初创公司持怀疑态度。

Paragraph 2: so what I have heard and which might be wrong is that last language models depend on data and compute and any startup can access to the same amount of data because it's just like internet data and compute like different companies might have different mouth compute but I guess like the big players can sell my compute

根据我听到的，可能是错误的，最新的语言模型依赖于数据和计算资产，任何创业公司都可以获得相同数量的数据，因为那就像互联网的数据一样，不同公司可能具有不同的计算量，但我猜大公司可以出售我的计算量。

Paragraph 3: so how good a large language model startup differentiate from another how would the startup differentiate from another how would one large language model start up differentiate from I think it'll be this middle layer

那么，一家大型语言模型初创公司如何与其他公司区分呢？它们如何区分自己？一家大型语言模型初创公司应该如何与其他公司区分？我认为这取决于中间层的技术。

Paragraph 4: I think in some sense the startups will train their own models just not from the beginning they will take like you know base models that are are like hugely trained with a gigantic amount of compute and data and then they will train on top of those to create you know the model for each vertical and end it those startup

我认为，在某种程度上，初创公司将训练自己的模型，但不会从头开始。他们会采用像使用巨大的计算能力和数据集进行训练的基础模型，然后在其基础上进行训练，为每个行业创建模型，最终完成这些初创公司的模型。如果需要，可以这样改写第4段：我认为初创公司将使用基础模型进行训练，并在此基础上进行垂直领域的模型构建。这些基础模型拥有庞大的计算能力和数据集，初创公司会在此基础上进行训练，为各个垂直领域创造适合的模型。

Paragraph 5: so in some sense they are training their own models just not not from scratch but they're doing the 1% of training that really matters for for whatever this use case is going to be those startups I think there will be hugely successful and very differentiated startups there

因此，某种程度上，它们正在训练自己的模型，只是不是从头开始，而是在为与使用用例相关的重要1%的训练做准备。我认为这些初创公司将非常成功并且非常不同。

Paragraph 6: but that'll be about the kind of like data flywheel that the startup is able to do the kind of like all of the pieces on top and below like this could include prompt engineering for a while or whatever the sort of the kind of like core base model

但这基本上是关于创业公司能够完成数据飞轮的那种，类似于所有上下的拼图，这可能包括一段时间的快速工程或任何核心基础模型的类型。

Paragraph 7: I think that's just going to get to two complex and two expensive and the world also just doesn't make enough chips

我觉得这会变得太复杂、太昂贵，而且世界上也没有足够的芯片。

Paragraph 8: so Sam has a work thing he needs to get to so and as you probably can tell with a very far far ranging thing Sam always expands my batteries and a little bit unlikely that when you're feeling depressed whether it's kids in the house you're the person I was turned to for I appreciate that yes

所以，山姆有一个他需要去的工作事情，你可能能够看出，山姆总是让我的电池得到很远、很广泛的充电。有时候当你感到沮丧时，无论是家里的孩子还是你，你都会转向我这个人。我很感激这一点。

Paragraph 9: so anyway I think I think like no one knows like we're sitting on this like precipice of AI and like people like it's either going to be like really great or really terrible you may as well like you got to you got to like plan for the worst you certainly like it's not a strategy to say it's all going to be okay but you may as well like emotionally feel like we're going to get to the great future and work hard as you can to get there and play for it yes rather than like act from this place of like fear and despair all the time because if you if we acted from a place of fear and paranoia we would not be where we are today

总之，我认为我们正处于人工智能的边缘，很少有人知道这将会是好是坏。因此，我们必须计划最坏的情况，并尽可能地努力工作，以便迎接美好的未来。我们应该积极面对未来而不是充满恐惧和绝望。如果我们时刻处于恐惧和偏执的状态中，我们今天就不会处于这个地步。

Paragraph 10: so let's thank Sam for spending dinner with us thank you that concludes this episode of Gray Matter you can read a transcript for this interview on the content section of our website graylock.com slash blog and you can watch a video from this interview on our YouTube channel under the intelligent future playlist you can subscribe to graynet on your preferred streaming platform and you can follow us on twitter at graylock vc

那么，让我们感谢Sam和我们共进晚餐，非常谢谢。这就结束了Gray Matter的这一集，您可以在我们网站graylock.com/blog的内容部分阅读本次采访的文本，您也可以在我们的YouTube频道中智能未来播放列表下观看本次采访的视频。您可以在您喜欢的流媒体平台上订阅graynet，也可以在Twitter上关注我们@graylockvc。如果需要，第十段可以用以下方式重新改写：您可以在我们的网站graylock.com/blog的内容部分阅读本次采访的文本，也可以在我们的的YouTube频道中智能未来播放列表下观看本次采访的视频，同时您可以在您喜欢的流媒体平台上订阅graynet，还可以在Twitter上关注我们@graylockvc。我们要感谢Sam和我们共进晚餐，谢谢。