SPEAKER 1: It was stunning，it was mind-blowing.
SPEAKER 1：太棒了，令人兴奋。
After the biology questions，I had them type in &34;
在生物学问题之后，我让他们输入“你对一个有生病孩子的父亲说什么？”
It gave this very careful excellent answer that was perhaps better than any of us in the room might have given.
它给出了这个非常仔细的优秀答案，也许比我们在场的任何人都可能给出的更好。
I was like，wow，what is the scope of this thing?
我当时想，哇，这东西的范围是什么？
Because this is way better.
因为这样更好。
KEVIN SCOTT: Hi everyone.
凯文·斯科特：大家好。
Welcome to Behind the Tech.
欢迎来到技术背后。
I'm your host，Kevin Scott，Chief Technology Officer for Microsoft.
我是主持人，微软首席技术官 Kevin Scott。
In this podcast，we&39;ll talk with some of the people who have made our modern tech world possible and understand what motivated them to create what they did.
在此播客中， 我们将深入了解技术，我们将与一些使我们的现代技术世界成为可能的人交谈， 并了解是什么促使他们创造他们所做的事情。
Join me to maybe learn a little bit about the history of computing，and get a few behind the scenes insights into what's happening today.
和我一起也许可以了解一点计算的历史，并获得一些幕后洞察力来了解当今正在发生的事情。
Stick around.
停在附近。
Hi，welcome to Behind the Tech.
您好，欢迎来到技术背后。
We have a great episode for you today with a really special guest，Bill Gates who needs no introduction given the unbelievable impact that he&39;re experiencing right now in AI means for OpenAI，Microsoft，all of our stakeholders and for the world at large.
今天我们有一个非常特别的嘉宾，比尔·盖茨，不需要介绍，因为他在过去几十年对全世界的技术世界产生了令人难以置信的影响，他一直与在过去的几个月里，Microsoft 和 OpenAI 的团队帮助我们思考我们目前在 AI 领域经历的惊人革命对 OpenAI、Microsoft、我们所有的利益相关者以及整个世界意味着什么。
I&39;ve had with Bill over these past months that I thought it might be a great thing to share just a tiny little glimpse of those conversations with all of you listeners today.
在过去的几个月里，我从与比尔的谈话中学到了很多东西，我认为今天与你们所有的听众分享这些谈话的一小部分内容可能是一件很棒的事情。
With that，let's introduce Bill and get a great conversation started.
有了这个，让我们介绍比尔并开始一段愉快的对话。
Thank you so much for doing the show today.
非常感谢你今天的节目。
I just wanted to jump right in with maybe one of the more interesting things that&39;ve been doing together at Microsoft with OpenAI.
我只想直截了当地谈谈过去几年技术领域发生的一件更有趣的事情，那就是 GPT-4 和 ChatGPT，我们在 Microsoft 与 OpenAI 一起开展的工作。
By the time this podcast airs，OpenAI will have made their announcement to the world about GPT-4，but I want to set the stage，the unveiling of the first instance of GPT-4 outside of OpenAI was actually to you last August at a dinner that you hosted with Reid and Sam Altman and Greg Brockman and Satya and a whole bunch of other folks. The OpenAI folks had been very anxious about showing you this because your bar for AI had been really high.
到这个播客播出时， OpenAI 将向全世界宣布他们关于 GPT-4 的消息， 但我想先说明一下，OpenAI 之外的第一个 GPT-4 实例的揭幕实际上是去年八月在一个你与 Reid 和 Sam Altman、Greg Brockman 和 Satya 以及一大群其他人一起举办的晚宴。OpenAI 的人非常急于向你展示这个， 因为你对 AI 的门槛非常高。
I think it had been really helpful actually，the push that you had made on all of us for what acceptable high ambition AI would look like.
我认为这实际上真的很有帮助，你向我们所有人推动了可接受的雄心勃勃的 AI 应该是什么样子。
I wanted to ask you，what was that dinner like for you?
我想问你，你觉得那顿晚餐怎么样？
What were your impressions like what you've been thinking before and what，if anything，changed in your mind after you had seen GPT-4?
您的印象是什么， 例如您之前一直在想什么？ 在您看到 GPT-4 之后，如果有的话， 您的想法发生了什么变化？
BILL GATES: AI has always been the holy grail of computer science.
比尔·盖茨：人工智能一直是计算机科学的圣杯。
When I was young，Stanford research had Shakey the Robot that was trying to pick things up and there were various logic systems that people were working on.
当我年轻的时候，斯坦福大学的研究人员让机器人 Shakey 试图捡起东西，人们正在研究各种逻辑系统。
Also the dream was always some reasoning capability.
此外，梦想总是具有某种推理能力。
Overall progress in AI until machine learning came along was pretty modest.
在机器学习出现之前，人工智能的总体进展相当有限。
Intelligent manufacturing system
Even speech recognition was just barely reasonable and so we had that gigantic acceleration with machine learning，particularly in sensory things，recognizing speech，recognizing pictures.
甚至语音识别也只是勉强合理，所以我们通过机器学习获得了巨大的加速，特别是在感官事物、识别语音、识别图片方面。
It was phenomenal，and it just kept getting better and scale was part of that.
这是惊人的，而且它一直在变得更好，规模是其中的一部分。
But we were still missing anything that had to do with complex logic with being able to say，read the text and do what a human does，which is quote，understand what's in that text.
但我们仍然缺少任何与复杂逻辑有关的东西，即能够说、阅读文本并做人类所做的事情，即引用、理解文本中的内容。
As Microsoft was doing more to OpenAI，I had a chance to go see them myself independently a number of times and they were doing a lot of text generation.
随着微软在 OpenAI 上做更多的工作，我有机会亲自去见了他们很多次， 他们正在做很多文本生成。
They had a little robot arm.
他们有一个小机械臂。
The early text generation still didn't seem to have a broad understanding.
早期的文本生成似乎还没有广泛的理解。
It can generate a sentence，saying Joe&39;s in Seattle，which in its local probabilistic sense，it was a good sentence，but a human has a broad understanding of the world from both experience in reading that you understand that can't be.
它可以生成一个句子， 说 Joe&39;s in Seattle，从局部概率的意义上说，这是一个很好的句子， 但是人类从这两种阅读经验中对世界有广泛的理解， 你明白那不可能。
They were enthusing about GPT-3 and even the early versions of GPT-4. I said to them，&39;s not part of the training set or a bunch of them，and give fully reasoned answers，knowing that a biology textbook is one of many things that's in the training corpus，then you will really get my attention because that would be a heck of a milestone.
他们对 GPT-3 甚至 GPT-4 的早期版本充满热情。 我对他们说，“嘿，如果你能通过高级安置生物学考试， 你会选择一个不属于训练集或一堆问题的问题，并给出完全合理的答案， 知道生物学教科书是其中之一训练语料库中的东西，那么你真的会引起我的注意， 因为那将是一个里程碑。
Please work on that.
请继续努力。
I thought that they'd go away for two or three years，because my intuition has always been that we needed to understand knowledge representation and symbolic reasoning in a more explicit way. That we were one or two inventions short of something where it was very good at reading and writing and therefore being an assistant.
我以为它们会消失两三年，因为我的直觉一直是我们需要以更明确的方式理解知识表示和符号推理。我们只差一两项发明就可以很好地阅读和写作， 因此可以成为一名助手。
It was amazing that you and Greg and Sam over the summer，were saying yeah it might not be long before we&39;s actually doing pretty well on scientific learning.
令人惊奇的是， 你和 Greg 以及 Sam 在整个夏天都说是的，我们可能很快就会向你演示这个东西， 因为它实际上在科学学习方面做得很好。
In August they said，yeah，let's get to this thing.
八月份，他们说，是的，让我们开始吧。
In early September，we had pretty large group over to my house for dinner，maybe 30 people in total，including a lot of the amazing OpenAI people，有关智能制造的英语单词，but a good size group from Microsoft.，
9 月初，我们有相当多的人到我家吃晚饭，总共可能有 30 人， 其中包括很多令人惊叹的 OpenAI 人员，但来自微软的人数相当多。
Satya was there and they gave it AP biology questions and let me give it AP biology questions.
Satya 在那里，他们给它出了 AP 生物学问题，让我给它出 AP 生物学问题。
With one exception，it did a super good job and the exception had to do with math，which we can get to that later.
除了一个例外， 它做得非常好， 例外与数学有关，我们可以稍后再谈。
But it was stunning，it was mind-blowing.
但它是惊人的，它是令人兴奋的。
After the biology questions，I had them type in，&34;
在生物问题之后，我让他们输入，“你对一个有生病孩子的父亲说什么？”
It gave this very careful excellent answer.
它给出了这个非常仔细的优秀答案。
That was perhaps better than any of us in the room might have given.
这可能比房间里的任何人都好。
I was like，wow，what is the scope of this thing?
我当时想，哇，这东西的范围是什么？
Because this is way better.
因为这样更好。
Then the rest of the night we asked historical questions about，are there criticisms of Churchill or different things and it was just fascinating.
然后剩下的那个晚上我们问了历史问题，有没有对丘吉尔的批评或其他事情， 这很有趣。
Then over the next few months as I was given an account and Sal Khan got one of those early accounts.
然后在接下来的几个月里，我得到了一个帐户，而 Sal Khan 得到了其中一个早期帐户。
The idea that you could have it write college applications or rewrite，say the Declaration of Independence，the way a famous person like Donald Trump might have written it.
你可以让它写大学申请或重写的想法，比如独立宣言，就像唐纳德特朗普这样的名人可能写的那样。
It was so capable of writing poems，give it a tune like，Hey Jude and tell it to write about that.
它非常有能力写诗，给它一个曲调，比如，嘿裘德，然后告诉它写下这件事。
Tell it to take a Ted Lasso episode and include the following things.
告诉它拍摄 Ted Lasso 剧集并包括以下内容。
Ever since that day in September，I've said，wow，this is a fundamental change and not without some things that still need to be worked out.
自从 9 月的那一天起，我就说，哇，这是一个根本性的变化， 并非没有一些事情仍然需要解决。
But it is a fundamental advance.
但这是一个根本性的进步。
It&39;t yet do this，it can&39;s not perfect to this or that.
它使人们感到困惑，好吧，它还不能做到这一点， 它不能做到这一点，它对这个或那个都不完美。
Natural language is now the primary interface that we're going to use to describe things even to computers.
自然语言现在是我们用来向计算机描述事物的主要接口。
It's a huge advance.
这是一个巨大的进步。
KEVIN SCOTT: Yes.
凯文·斯科特：是的。
There&39;s not good at it.
这里有很多不同的事情要谈，但也许第一个是谈谈它不擅长的事情。
Because the last thing that I think we want to do is give people the impression that it is an AGI，that it is perfect，that there isn't a lot of additional work that has to happen to improve it and make it better.
因为我认为我们最不想做的就是给人们这样的印象， 那就是它是一个通用人工智能，它是完美的， 不需要做很多额外的工作来改进它并让它变得更好。
You mentioned math is one of the things so I thought maybe let's talk a little bit about what you think needs to be better about these systems over time and where we need to focus our energy.
你提到数学是其中之一，所以我想也许让我们谈谈你认为随着时间的推移这些系统需要改进的地方以及我们需要集中精力的地方。
BILL GATES: Yeah.
比尔·盖茨：是的。
It appears to be a general issue that its knowledge of context when it&34;I&34; or &39;m giving you advice that if it&39;re going to buy the wrong stock or take the wrong drug."
这似乎是一个普遍的问题， 当它被问到时， 它对上下文的了解，好吧，我告诉你一些东西， 然后我生成一些东西，人类理解“我在这里编造幻想的东西”，或者“我给你建议如果错了，你就会买错股票或吃错药。 ”
Humans have a very deep context of what's going on.
人类对正在发生的事情有着非常深刻的理解。
Even the AI&39;ve switched context like if you&39;re not in that joking thing，it wants to keep telling jokes.
甚至 AI 也有能力知道你已经切换了上下文， 就像你要求它讲笑话然后你问它一个严肃的问题， 人类会从你的脸上看到什么， 或者那种变化的本质，好吧，我们是不是在开玩笑， 它想一直讲笑话。
You almost have to do the reset sometimes to get it out of the，hey，whatever I bring up just make jokes about it.
有时你几乎必须重新设置才能让它摆脱，嘿，无论我提出什么， 都只是拿它开玩笑。
I do think that sense of context，there's work.
我确实认为那种情境感，有工作。
Also in terms of how hard it should work on a problem.
还有就是它应该在多大程度上解决一个问题。
When you and I see a math problem，we know，well，I may have to apply simplification five or six times to get this into the right form.
当你和我看到一个数学问题时，我们知道，好吧，我可能需要进行五六次简化才能将其转化为正确的形式。
We're looping through how we do these reductions，whereas today the reasoning is a linear chain of descent through the levels.
我们正在循环我们如何进行这些减少，而今天的推理是通过级别的线性下降链。
And if simplification needs to run 10 times，it probably won't.
如果简化需要运行 10 次，它可能不会。
Math is a very abstract type of reasoning.
数学是一种非常抽象的推理类型。
Right now，I&39;s the greatest weakness.
现在，我会说这是最大的弱点。
Weirdly，it can solve lots of math problems.
奇怪的是，它可以解决很多数学问题。
There are some math problems where if you ask it to explain it in an abstract form，make essentially an equation or a program that matches the math problem，it does that perfectly and you could pass that off to a normal solver，whereas if you tell it to do the numeric work itself，it often makes mistakes.
有一些数学问题， 如果你要求它以抽象形式解释它，本质上是一个方程或一个与数学问题相匹配的程序，它会完美地做到这一点， 你可以把它传递给一个普通的求解器，而如果你告诉它自己做数字工作，它经常出错。
It&39;s very confident that，hey - or it'll say，I mistyped.
这很有趣，因为有时它非常自信，嘿 - 或者它会说，我打错了。
好吧，事实上，这个场景中的任何地方都没有打字机。
The notion of mistyping is really very weird.
打错字的概念真的很奇怪。
Whether these current areas of weakness，it&39;s not just making up URLs.
无论这些当前的弱点领域，在很大程度上得到修复之前需要六个月、一年或两年，所以我们有一个严肃的模式， 它不仅仅是制作 URL。
Then we have the more fanciful mode.
然后我们有更奇特的模式。
There's some of that already being done largely through prompts and eventually through training.
其中一些已经主要通过提示并最终通过培训完成。
Training it for math，there may be some special training that needs to be done.
对其进行数学训练，可能需要进行一些特殊训练。
But these problems I don't think are fundamental.
但我认为这些问题都不是根本性的。
There are people that think it's statistical，therefore it can never do X. That is nonsense.
有人认为它是统计的，因此它永远不能做 X。那是胡说八道。
Every example they give of a specific thing it doesn't do，wait a few months.
他们给出的每个例子都说明了它没有做的具体事情，等几个月。
It's very good.
这很好。
Characterizing how good it is，the people who say it&39;re wrong.
描述它有多好， 那些说它很糟糕的人真的错了， 那些认为这是 AGI 的人，他们错了。
Those of us in the middle are just trying to make sure it gets applied in the right way.
我们中间的那些人只是想确保它以正确的方式得到应用。
There's a lot of activities like helping somebody with their college application.
有很多活动，比如帮助某人申请大学。
What's my next step?
我的下一步是什么？
What haven't I done?
我没做什么？
I have the following symptoms，that are in fact far within the boundary of things that it can do quite effectively.
我有以下症状，实际上这些症状远远在它可以非常有效地完成的事情的范围内。
KEVIN SCOTT: Yeah.
凯文·斯科特：是的。
Well，I want to talk a little bit about this notion of it being able to use tools to assist it in reasoning.
好吧，我想谈谈它能够使用工具来协助推理的概念。
I'll give you an example from this weekend with my daughter.
我会给你举个这个周末我女儿的例子。
My daughter had this assignment where she had this list of 115 vocabulary words.
中国智能制造挑战赛 "SIEMENS Cup" China intelligent manufacturing challenge 重点词汇 西门子siemens; Siemens 智能intelligent; intelligence; brain power; noopsyche 制造manufacture; make; create; engineer; fabricate 。
我女儿有这个任务，她有这个 115 个词汇表。
She had written a 1，000 word essay and her objective was to use as many of these vocabulary words as she reasonably could in this 1，000 word essay，which is a ridiculous assignment on the surface.
她写了一篇 1，000 字的文章， 她的目标是在这篇 1，000 字的文章中尽可能多地使用这些词汇，这在表面上是一项荒谬的任务。
But she had written this essay and she was going through this list trying to manually figure out what her tally was on this vocabulary list and it was boring and she was like，"I want a shortcut here.
但她写了这篇文章， 她正在浏览这个列表， 试图手动找出她在这个词汇表上的记录，这很无聊，她说，“我想要一条捷径。
Dad，can you get me a ChatGPT account and can I just put this in there and it will do it for me?"
爸爸，你能给我一个 ChatGPT 帐户吗？我可以把它放在那里，它会为我做吗？”
We did it and ChatGPT，which is not running the GPT-4 model，but I don&39;t quite get it right.
我们做到了， ChatGPT 没有运行 GPT-4 模型，但我认为 GPT-4 也不会做到这一点，也不太正确。
It was not precise.
这并不准确。
But the thing then that I got her to do with me as I was like，well，let ChatGPT write a little Python program that can very precisely - this is a very simple Intro CS problem here.
但是后来我让她和我一起做的事情就像我一样，好吧， 让 ChatGPT 编写一个可以非常精确的小 Python 程序 - 这是一个非常简单的 Intro CS 问题。
The fact that the Python code for solving that problem was perfect and I got my solution immediately is just amazing.
解决该问题的 Python 代码非常完美，而且我立即得到了解决方案，这真是太棒了。
My 14-year-old daughter who doesn't program understood everything that was going on.
我 14 岁的女儿不会编程，她明白发生的一切。
I don&39;re trying to solve a complicated math problem，we&39;re talking about to help us break down very complicated problems into smaller or less complicated problems that we can solve.
我不知道你在过去的几个月里是否反思过很多， 因为基本上当我们试图解决一个复杂的数学问题时，我们已经满脑子都是认知工具， 我们会像你正在谈论的这些抽象一样使用这些工具帮助我们将非常复杂的问题分解成我们可以解决的更小或更简单的问题。
I think it's a very interesting idea to think about how these systems will be able to do that with code.
我认为考虑这些系统如何使用代码来做到这一点是一个非常有趣的想法。
BILL GATES: Yeah.
比尔·盖茨：是的。
It's so good at writing.
写的太好了
That's just a mind-blowing thing.
这简直是一件令人兴奋的事情。
But when you can use natural language，say for a drawing program that you want various objects and you want to change them in certain ways，sure，you still want the menus there to touch up the colors，but the primary interface for creating a from-scratch drawing will be language.
但是当你可以使用自然语言时，比如对于一个你想要各种对象的绘图程序， 你想要以某种方式改变它们，当然，你仍然想要那里的菜单来修饰颜色，但是创建一个从- 草图将是语言。
If you want a document summarized，that's something that it can do extremely well.
如果你想要一个文档的摘要，这是它可以做得非常好的事情。
When you have large bodies of text，when you have text creation problems，there was a ChatGPT-3 written where a doctor who has to write to insurance companies to explain why he thinks something should be covered that's very complicated and it was super helpful.
当你有大量文本时， 当你遇到文本创建问题时，有一个 ChatGPT-3 写了一个医生必须写信给保险公司解释为什么他认为应该涵盖一些非常复杂但非常有帮助的东西。
He was reading that letter over to make sure it was right.
他正在阅读那封信以确保它是正确的。
In ChatGPT the Version 4 stuff，we took complex medical bills and we said please explain this bill to me.
在 ChatGPT 版本 4 中，我们收到了复杂的医疗账单，我们说请向我解释一下这个账单。
What is this and how does it relate to my insurance policy?
这是什么以及它与我的保单有什么关系？
It was so incredibly helpful at being able to do that.
能够做到这一点非常有帮助。
Explaining concepts in a more simpler form，it's very helpful at that.
以更简单的形式解释概念，这非常有帮助。
There&39;s just huge increased productivity，including lot of documents，payables，accounts receivables.
将会有很多任务会大大提高生产力，包括大量文件、应付账款、应收账款。
Just take the health system alone，there's a lot of documents that now software will be able to characterize them very effectively.
仅就卫生系统而言，现在有很多文件可以用软件非常有效地描述它们的特征。
KEVIN SCOTT: Yeah.
凯文·斯科特：是的。
One of the other things that I wanted to chat with you about，you have this really unique perspective in your involvement in several of the big inflection points in technology.
我想和你聊的另一件事是，你在参与技术的几个重大转折点时拥有这种真正独特的视角。
For two of these inflections，you are either one of the primary architects of the inflection itself or either one of the big leaders.
对于其中两个变化，您要么是变化本身的主要设计者之一，要么是主要领导者之一。
We wouldn't have the PC，personal computing ecosystem without you.
没有你，我们就不会有 PC、个人计算生态系统。
You played a really substantial role in getting the Internet available to everybody and making it a ubiquitous technology that everyone can benefit from.
在让每个人都可以使用互联网并使其成为每个人都可以从中受益的无处不在的技术方面，你们发挥了非常重要的作用。
To me，this feels like another one of those moments where a lot of things are going to change.
对我来说，这感觉就像是又一个很多事情都将发生变化的时刻。
I wonder what your advice might be to people who are thinking about like，I have this new technology that's amazing that I can now use.
我想知道你对那些正在考虑的人有什么建议，我拥有这项令人惊奇的新技术， 我现在可以使用了。
How should they be thinking about how to use it?
数字智造 Digital Intellectually Manufacture 中国智造 Intellectually Manufactured in China
他们应该如何考虑如何使用它？
How should they be thinking about the urgency with which they are pursuing these new ideas?
他们应该如何考虑追求这些新想法的紧迫性？
How does that relate to how you thought about things in the PC era and in the Internet era?
这与您在 PC 时代和互联网时代对事物的看法有什么关系？
BILL GATES: Yes.
比尔·盖茨：是的。
The industry starts really small where computers aren't personal.
在计算机不是个人的地方，这个行业开始时非常小。
Then through the microprocessor and a bunch of companies，we get the personal computer，IBM，Apple，and Microsoft got to be very involved in the software.
然后通过微处理器和许多公司，我们得到了个人电脑，IBM、Apple 和 Microsoft 开始涉足软件领域。
Even the BASIC interpreter on the Apple II，a very obscure fact，was something that I did for Apple.
即使是 Apple II 上的 BASIC 解释器，一个非常模糊的事实，也是我为 Apple 所做的。
That idea that，wow，this is a tool that at least for editing documents that you have to do all the writing，that was pretty amazing.
那个想法，哇，这是一个至少用于编辑文档的工具， 你必须完成所有的写作，这真是太棒了。
Then connecting those up over the Internet was amazing.
然后通过互联网将它们连接起来真是太棒了。
Then moving the computation into the mobile phone was absolutely amazing.
然后将计算转移到手机中绝对是惊人的。
Once you get the PC，the Internet，the software industry，and the mobile phone，the digital world is changing huge parts of our activities.
一旦你有了个人电脑、互联网、软件行业和手机，数字世界就会改变我们活动的很大一部分。
I was just in India，seeing how they do payments digitally even for government programs，it&39;s too complicated.
我刚刚在印度， 看到他们如何甚至为政府项目进行数字支付，这是一个令人惊叹的世界应用， 可以帮助那些永远不会拥有银行账户的人， 因为费用太高，太复杂了。
We continue to benefit from that foundation.
我们继续受益于该基金会。
I do view this，the beginning of computers that read and write，as every bit as profound as any one of those steps and a little bit surprising because robotics has gone a little slower than I would have expected.
我确实认为这一点， 即读写计算机的开端，每一点都与这些步骤中的任何一步一样深刻， 并且有点令人惊讶， 因为机器人技术的发展比我预期的要慢一些。
I don&39;s a special case，that's particularly hard because of the open-ended environment and the difficulty of safety and what safety bar people will bring to that.
我不是说自动驾驶， 我认为那是一个特例，由于开放的环境和安全的难度以及人们会为此带来的安全障碍， 这特别困难。
But even factories where you actually have a huge control over the environment of what's going on and you can make sure that no kids are running around anywhere near that factory.
但即使是工厂，您实际上可以极大地控制正在发生的事情的环境，并且您可以确保没有孩子在该工厂附近的任何地方跑来跑去。
A little bit people have been saying，these guys over-predict，which that's certainly correct.
有些人一直在说，这些人预测过度了，这当然是正确的。
But here&39;s ability to deal with that and how that affects white-collar jobs，including sales，service，helping a doctor think through what to put in your health record，that I thought was many years off.
但在这个案例中， 我们低估了自然语言和计算机处理该问题的能力， 以及它如何影响白领工作，包括销售、服务、帮助医生考虑将什么放入您的健康记录，我认为很多年了。
All the AI books，even when they talk about things that might get a lot more productive，will turn out to be wrong.
所有关于 AI 的书籍， 即使它们谈论的事情可能会变得更有效率，但最终都会被证明是错误的。
Because we're just at the start of this，you could almost call it a mania，like the Internet mania.
因为我们才刚刚开始，你几乎可以称之为狂热，就像互联网狂热一样。
But the Internet mania，although it had its insanity and things like sock puppets or things where you look back and say，what were we thinking?
但是互联网狂热， 虽然它有它的疯狂和像袜子木偶之类的东西， 或者你回头看看会说，我们在想什么？
It was a very profound tool that now we take for granted.
这是一个非常深刻的工具，现在我们认为这是理所当然的。
Even just for scientific discovery during the pandemic，the utility of the immediate sharing that took place there was just phenomenal.
即使只是在大流行期间进行科学发现，在那里发生的即时共享的效用也是惊人的。
This is as big a breakthrough，a milestone，as I&39;m quite young.
这是一个巨大的突破， 一个里程碑，正如我在整个数字计算机领域所看到的那样， 它真正开始于我还很年轻的时候。
KEVIN SCOTT: I&39;m just interested in your reaction because you will always tell me when an idea is dumb.
凯文斯科特：我要对你说这个，我只是对你的反应感兴趣， 因为当一个想法很愚蠢时你总是会告诉我。
But one of the things that I&39;s happening because of this technology is that for 180 years from the point that Ada Lovelace wrote the first program to harness the power of a digital machine up until today，the way that you get a digital machine to do work for you is you either have to be a skilled programmer，which is like a barrier to entry，that's not easy or you have to have a skilled programmer anticipate the needs of the user and to build a piece of software that you can then use to get the machine to do something for you.
但过去几年我一直在思考的一件事是， 由于这项技术而发生的重大变化之一是， 从 Ada Lovelace 编写第一个程序以利用直到今天， 一台数字机器，你让一台数字机器为你工作的方式是， 你要么必须是一名熟练的程序员，这就像进入的障碍， 这并不容易， 要么你必须有一个熟练的程序员预期用户的需求并构建一个软件， 然后您可以使用该软件让机器为您做某事。
This may be the point where we get that paradigm to change a little bit.
这可能是我们让范式稍微改变的地方。
Where because you have this natural language interface and these AIs can write code，and they will be able to actuate a whole bunch of services and systems that give ordinary people the ability to get very complicated things done with machines without having to have all of this expertise that you and I spent many years building?
因为你有这种自然语言界面， 这些人工智能可以编写代码，它们将能够启动一大堆服务和系统， 让普通人能够用机器完成非常复杂的事情， 而不必拥有所有这些你和我花了很多年建立的专业知识？
BILL GATES: No，absolutely.
比尔·盖茨：不，绝对没有。
Every advance hopefully lowers the bar in terms of who can easily take advantage of it.
每一项进步都有望降低谁可以轻松利用它的门槛。
The spreadsheet was an example of that because even though you still have to understand these formulas，you really didn't have to understand logic or symbols much at all.
电子表格就是一个例子， 因为即使您仍然需要理解这些公式，您实际上根本不需要理解太多逻辑或符号。
It had the input and the output so closely connected in this grid structure that you didn't think about the separation of those two.
在这个网格结构中，它的输入和输出连接得如此紧密，以至于你不会考虑将这两者分开。
That's limiting in a way to super abstract thinker.
这在某种程度上限制了超级抽象思想家。
But it was so powerful in terms of the directness.
但就直接性而言，它是如此强大。
That didn't come out right，let me change it.
说的不对，改一下吧。
Here，there&39;ve had 20 percent of the headcount missing and are our sales results affected by that?
在这里，有一整套获取公司数据并进行展示或进行复杂查询的程序，是否有任何销售办事处的员工人数减少了 20%， 我们的销售结果是否会受到影响？
Now you don&39;s too hard or something.
现在您不必去 IT 部门排队等候， 也不必让他们告诉您，哦，这太难了或什么的。
Most of these corporate learning things，whether it's a query or report or even a simple workflow where if something happens，you want to target an activity，the description in English will be the program.
大多数这些企业学习的东西，无论是查询还是报告， 甚至是一个简单的工作流程， 如果发生什么事情，你想要针对一个活动， 英文描述就是程序。
When you want it to do something extra，you'll just pull up that English or whatever your language is in and type that in.
当你想让它做一些额外的事情时，你只需调出英语或任何你使用的语言并输入即可。
There's a whole layer of query assistance and programming that will be accessible to any employee.
任何员工都可以访问一整层的查询帮助和编程。
And the same thing is true，I&39;s my next step and what&39;s so opaque today，so empowering people to go directly and interact，that is the theme that this is trying to enable.
同样的事情是真实的， 我在大学申请过程中的某个地方， 我想知道我的下一步是什么以及这些事情的门槛是什么，今天它是如此不透明， 所以让人们能够直接去互动，这就是这是试图启用的主题。
KEVIN SCOTT: I wonder what some of the things are that you're most excited about just in terms of application of the technology to the things that you care about deeply from the Foundation or your personal perspective?
KEVIN SCOTT：我想知道，从基金会或您个人的角度来看，就将技术应用到您深切关心的事情而言，您最感兴趣的事情是什么？
You care a lot about education，public health，climate，and sustainable energy.
您非常关心教育、公共卫生、气候和可持续能源。
You have all of these things that you're working on and have you been thinking about how this technology impacts any of those things?
你有所有这些你正在做的事情，你有没有想过这项技术如何影响这些事情？
BILL GATES: It's been fantastic that even going back to the fall，OpenAI and Microsoft have engaged with people at the Gates Foundation thinking about the health stuff and the education stuff.
比尔·盖茨：令人惊奇的是， 即使回到秋天，OpenAI 和微软也与盖茨基金会的人们一起思考健康和教育方面的问题。
In fact，Peter Lee is going to be publishing some of his thinking which is somewhat focused on rich world health，but it's pretty obvious how that work in a sense is even more amazing in health systems where you have so few doctors and getting advice of any kind is so incredibly difficult.
事实上，Peter Lee 将发表他的一些想法， 这些想法在某种程度上侧重于富裕世界的健康，但很明显，从某种意义上说， 这种工作在医疗系统中的作用更加惊人， 因为那里的医生很少， 而且只能听取医生的建议任何一种都非常困难。
It is incredible to look at saying，can we have a personal tutor that helps you out?
令人难以置信的是，我们可以请一位私人导师来帮助您吗？
Can yo，u when you write something，if you&39;t get that much feedback on their writing and it looks like，configured properly，this is a great tool to give you feedback in writing.
你，当你写东西的时候， 如果你要去一些很棒的学校，是的，老师可能会逐行给你反馈，但是很多孩子只是没有得到那么多关于他们写作的反馈和看起来，如果配置得当， 这是一个很好的工具， 可以给你书面反馈。
It's also ironic in a way that people are saying，what does it mean that can people cheat and turn in computer writing?
人们所说的也具有讽刺意味，人们可以作弊并上交计算机写作是什么意思？
Kind of like when calculators came along and we said。
有点像当计算器出现时，我们说。
oh my goodness，what are we going to do about adding and subtracting? Of course，they did create contexts where you couldn't use a calculator and we got through that without it being a huge problem.
哦，我的天哪，我们要如何做加法和减法？当然，他们确实创造了你不能使用计算器的环境， 我们在没有遇到大问题的情况下解决了这个问题。
I think education is the most interesting application.
我认为教育是最有趣的应用。
I think health is the second most interesting.
我认为健康是第二有趣的。
Obviously there&39;ll happen，you don't need any foundation type engagement on that.
显然，在销售和服务类型的事情中存在商业机会， 而且这种情况会发生，您不需要任何基础类型的参与。
We brainstormed a lot with Sal Khan，and it looks very promising because a class size of 30 or 20，you can't give a student individual attention。
you can't understand their motivation or what is keeping them engaged.
您无法理解他们的动机或让他们保持参与的原因。
They might be ahead of the class，they might be behind.
他们可能领先全班，也可能落后。
It looks like in many subject areas。
看起来在许多学科领域。
by having this and having dialogues and giving feedback。
通过进行对话并提供反馈。
for the first time，we'll succeed in helping education.
第一次，我们将成功地帮助教育。
Now，we have to admit，except for this prosaic thing of looking up Wikipedia articles or helping you type things and print them out nice。
现在，我们必须承认，除了查找维基百科文章或帮助您键入内容并将它们打印出来这种平淡无奇的事情外。
the notion that computers were going to revolutionize education largely are still more in front of us than behind us.
计算机将在很大程度上彻底改变教育的观念仍然在我们面前而不是在我们身后。
Yes，some games draw kids in，but the average math score in the US hasn't gone up much over the last 30 years.
是的，有些游戏吸引了孩子们，但美国的平均数学成绩在过去 30 年里并没有上升多少。
The people who do computers kept saying，hey，we want credit for that.
做计算机的人一直在说，嘿，我们想为此获得荣誉。
Credit for what?
信用什么？
It's not a lot better than it was.
它并没有比以前好多少。
Obviously，the computers didn't perform some miracle there.
显然，计算机并没有在那里创造奇迹。
I think over the next 5-10 years，we will think of learning and how you can be helped in your learning in a very different way than just looking at material.
我认为在接下来的 5 到 10 年里，我们会考虑学习，以及如何以一种与仅仅看材料截然不同的方式来帮助你学习。
KEVIN SCOTT: I know you think about this as a global problem.
凯文斯科特：我知道你认为这是一个全球性问题。
My wife and I with our family foundation，think about it as a local problem for disadvantaged kids.
我和我的妻子以及我们的家庭基金会，将其视为弱势儿童的本地问题。
One of the common things that we see is that parent engagement makes a big difference in the educational outcomes for kids.
我们看到的一个共同点是，父母的参与对孩子的教育成果产生了很大的影响。
If you look at the children of immigrants in East San Jose or East Palo Alto here in the Silicon Valley。
如果你看看硅谷东圣何塞或东帕洛阿尔托的移民子女。
often the parents are working two，three jobs。
父母经常打两份、三份工作。
they&39;t speak English and so they don't even have the linguistic ability.
他们太忙了，很难与孩子相处，有时他们不会说英语，所以他们甚至没有语言能力。
You can just imagine what a technology like this could do where it really doesn't care what language you speak.
你可以想象这样的技术可以做什么，它真的不在乎你说什么语言。
It can bridge that gap between the parents and the teacher。
它可以弥合父母和老师之间的鸿沟。
and it can be there to help the parent understand where the roadblocks are for the child and to even potentially get very personalized to the child&39;re struggling with。
它可以帮助父母了解孩子的障碍在哪里，甚至可以非常个性化地满足孩子的需求，并帮助他们解决他们正在努力解决的问题。
I think is really very exciting.
我觉得真的很令人兴奋。
BILL GATES: Just the language barriers。
比尔·盖茨：只是语言障碍。
we often forget about that，and that comes up in developing world.
我们经常忘记这一点，而这在发展中国家会出现。
India has a lot of languages and I was at the Bangalore Research Lab as part of that trip.
印度有多种语言，作为那次旅行的一部分，我去了班加罗尔研究实验室。
They're taking these advanced technologies from trying to deal with the tail of languages。
他们从尝试处理语言的尾巴中获取这些先进技术。
so that's not a huge barrier.
所以这不是一个巨大的障碍。
KEVIN SCOTT: One of the things that you said at the GPT-4 dinner at your house。
KEVIN SCOTT：你在你家的 GPT-4 晚宴上说的其中一件事。
is that you had this experience early in Microsoft's history where you saw a demo that changed your way of thinking about how the personal computing industry was going to unfold and that caused you to pivot the direction of the company.
是你在微软历史的早期就有过这样的经历，你看到了一个演示，它改变了你对个人计算行业将如何发展的思考方式，并使你改变了公司的发展方向。
I wonder if you might be willing to share that with everyone.
我想知道你是否愿意与大家分享。
BILL GATES: Xerox had made lots of money on copying machines.
比尔·盖茨：施乐在复印机上赚了很多钱。
They got out ahead，their patents were there。
他们走在了前面，他们的专利在那里。
the Japanese competition hadn't come in。
日本的比赛还没有进来。
and so they created a research center out in Palo Alto。
所以他们在帕洛阿尔托创建了一个研究中心。
which was forever known by its acronym。
它的首字母缩写词永远为人所知。
Palo Alto Research Center，PARC.
PARC 帕洛阿尔托研究中心。
At PARC，they assembled an incredible set of talent.
在 PARC，他们聚集了一批令人难以置信的人才。
Bob Taylor and others were very good judges of talents.
鲍勃泰勒和其他人是非常好的人才判断者。
So you end up with Alan Kay，Charles Simonyi。
所以你最终会得到 Alan Kay、Charles Simonyi。
Butler Lampson，I don't want to leave anybody out。
巴特勒·兰普森，我不想漏掉任何人。
but there's a bunch of other people and they create a graphics user interface machine.
但是还有很多其他人，他们创建了一个图形用户界面机器。
They weren't the only ones.
他们不是唯一的。
There were people over in Europe doing some of these things。
在欧洲有人在做这些事情。
but they combined it with a lot of things.
但他们将它与很多东西结合起来。
They put it on the network，they got a laser printer.
他们把它放在网络上，他们得到了一台激光打印机。
Charles Simonyi was there programming this and did a word processor that used that very graphical bitmap screen and let you do things like fonts and stuff we take for granted now.
Charles Simonyi 在那里编程并做了一个文字处理器，它使用非常图形化的位图屏幕，让你做字体和我们现在认为理所当然的事情。
I went and visited Charles at PARC at night。
我晚上去 PARC 拜访了查尔斯。
and he demoed what he had done with this Bravo word processor。
他演示了他用这个 Bravo 文字处理器所做的事情。
and then he printed on the laser printer a document of all the things that should be done if there were cheap pervasive computers.
然后他在激光打印机上打印了一份文件，上面列出了如果有便宜的普及型计算机应该做的所有事情。
He and I brainstormed that，and he updated the document and printed again.
他和我集思广益，他更新了文档并再次打印。
It just blew my mind.
这让我大吃一惊。
The agenda for Microsoft came out of - That&39;m with him。
微软的议程来自 - 那是在 1979 年，我和他在一起。
computers are still completely character mode.
计算机仍然完全是字符模式。
That's when the commitment to do software for the Mac emerges from Steve Jobs having a similar experience with Bob Belville at Xerox PARC.
就在那时，史蒂夫·乔布斯 (Steve Jobs) 与 Xerox PARC 的鲍勃·贝尔维尔 (Bob Belville) 有着相似的经历，由此产生了为 Mac 开发软件的决心。
Xerox built a very expensive machine called the Star that they only sold a few thousand of because people didn't think of word processing as something you would pay for.
施乐制造了一台非常昂贵的机器，叫做 Star，他们只卖了几千台，因为人们不认为文字处理是你会花钱买的东西。
You had to come in really at the low end。
你必须真正进入低端。
so PCs with first character mode，but later graphics word processing.
所以 PC 首先是字符模式，然后是图形文字处理。
I hired Charles.
我雇用了查尔斯。
Charles helped do Word and Excel and many of our great things.
Charles 帮助完成了 Word 和 Excel 以及我们许多伟大的事情。
Eventually，15 years after Charles had showed me his thinking and we brainstormed。
最终，在查尔斯向我展示他的想法并且我们集思广益 15 年后。
we largely achieved through Windows and Office on both Windows and Mac。
我们主要通过 Windows 和 Office 在 Windows 和 Mac 上实现。
we largely achieved that piece of paper.
我们基本上实现了那张纸。
I told the group that that was the other demo that had blown my mind and made me think。
我告诉小组那是另一个让我震惊并让我思考的演示。
what can be achieved in the next 5-10 years?
未来5-10年可以实现什么？
We should be more ambitious taking advantage of this breakthrough.
我们应该更加雄心勃勃地利用这一突破。
Even with the imperfections that we're going to reduce over time.
即使有我们将随着时间的推移减少的缺陷。
KEVIN SCOTT: It was a really powerful and motivating anecdote that you shared.
凯文·斯科特：你分享的是一个非常有力和激励人心的轶事。
Maybe one last thing here before we go，or maybe two more things.
也许在我们走之前在这里做最后一件事，或者再做两件事。
What do you think are the big grand challenges that we ought to be thinking about over the next 5-10 year period?
您认为在未来 5 到 10 年期间我们应该考虑的重大挑战是什么？
Like in a sense like this，I actually have this piece of paper that Charles wrote。
从某种意义上说，我实际上有查尔斯写的这张纸。
it&39;s ever written.
它放在我的办公桌旁，因为我认为这是任何人写过的对技术周期的更令人难以置信的预测之一。
I don&39;t know about the existence of this thing.
不知道为什么大家都不知道这个东西的存在。
It's just unbelievable.
这简直令人难以置信。
As you think about what lies ahead of us over the next 5-10 years，what's your challenge?
当您思考未来 5 到 10 年我们面临的挑战时，您面临的挑战是什么？
Not just to Microsoft，but to everybody in the world who's going to be thinking about this?
不仅是微软，还有世界上所有会考虑这个问题的人？
What do you think we ought to go push on really really hard?
你认为我们应该真正非常努力地推进什么？
BILL GATES: Well，there'll be a set of innovations on how you execute these algorithms and lots of chips.
比尔·盖茨：嗯，关于如何执行这些算法和大量芯片，将会有一系列创新。
Some movement from silicon to optics to reduce the energy and the cost.
从硅到光学的一些运动可以降低能源和成本。
Immense innovation where Nvidia is the leader today。
英伟达今天处于领先地位的巨大创新。
but others will try and challenge them as they keep getting better and using even some radical approaches because we want to get the cost of execution on these things and even the training dramatically less than it is today.
但其他人会尝试挑战他们，因为他们不断变得更好，甚至使用一些激进的方法，因为我们希望获得这些事情的执行成本，甚至培训成本也比今天低得多。
Ideally，we'd like to move them so that often you can do them on a self-contained device。
理想情况下，我们希望移动它们，以便您经常可以在独立设备上进行操作。
not have to go up to the Cloud to get these things.
不必上云去获得这些东西。
Lots to be done on the platform that this uses.
在这个使用的平台上有很多事情要做。
Then we have an immense challenge in the software side of figuring out。
然后我们在软件方面面临着巨大的挑战，要弄清楚。
okay，do you just have many specialized versions of this thing or do you have one that just keeps getting better?
好吧，你只是有很多这个东西的专门版本，还是你有一个不断改进的版本？
There'll be immense competition.
将会有巨大的竞争。
Those two approaches.
这两种方法。
Even Microsoft will pursue both in parallel with each other.
甚至微软也会同时追求两者。
Ideally within a contained domain we'll get something that the accuracy is provably extremely high by limiting the areas that it works in and by having the training data and even perhaps some pre-checking post-checking type logic that applies to that.
理想情况下，在一个包含的域中，我们将通过限制它工作的区域并拥有训练数据，甚至可能有一些适用于它的预检查后检查类型逻辑，来获得可以证明准确性极高的东西。
I definitely think areas like sales and service。
我绝对认为销售和服务等领域。
there is a lot that can be done there and that's super valuable.
那里可以做很多事情，而且非常有价值。
The notion that there is this emergent capability means that the push to try and scale up even higher。
有这种紧急能力的概念意味着尝试和扩大规模的推动力更高。
that'll be there.
那会在那里。
Now，what corpuses exist?
现在，存在哪些语料库？
Once you get past every piece of text and video.
一旦你通过了每段文字和视频。
Are you synthetically generating things?
你是在综合生成东西吗？
Do you still see that improvement as you scale up?
当你扩大规模时，你是否仍然看到这种改进？
Obviously，that&39;t stand in the way of that going ahead in a very high-speed way.
显然，这会得到追求，而且这样做要花费数十亿美元的事实不会阻碍它以非常高速的方式前进。
Then there's a lot of societal things of，okay，where can we push it in education?
然后有很多社会问题，好吧，我们可以在教育中将其推向何处？
It&39;ll just immediately understand student motivation or student cognition.
这并不是说它会立即了解学生的动机或学生的认知。
There'll have to be a lot of training and embedding it in an environment where the adults are seeing the engagement of the student and seeing the motivation.
必须进行大量培训，并将其嵌入成年人看到学生参与并看到动机的环境中。
Even though you free up the teacher from a lot of things。
尽管你让老师摆脱了很多事情。
that personal relationship piece，you&39;s there with the teacher or with the patient.
Microsoft talks about making humans more productive.
微软谈论让人类更有生产力。
Some things will be automated，but many things will just be facilitated where the final engagement is very much a human。
有些事情会自动化，但很多事情只会在最终参与非常人性化的情况下得到促进。
but a human who's able to get a lot more done than ever before.
而是一个比以往任何时候都能完成更多工作的人。
The number of challenges and opportunities created by this is pretty incredible.
由此带来的挑战和机遇之多令人难以置信。
I can see how engaged the OpenAI team is by this。
我可以看到 OpenAI 团队对此的投入程度。
I&39;s lots of teams that I don't get to see that are pushing on this.
我敢肯定有很多我看不到的团队正在推动这一点。
就行业规模而言，当微处理器被发明时，软件行业还是一个很小的行业。
We could put most of us on a panel and they could complain that I work too hard and it shouldn't be allowed.
我们可以把我们中的大多数人放在一个小组中，他们可以抱怨我工作太努力了，这是不应该被允许的。
We can all laugh about that.
我们都可以对此大笑。
Now，this is a global industry，so it's a little harder to get your hands around.
现在，这是一个全球性的行业，所以要掌握起来有点困难。
I get a weekly digest of all the different articles about AI that are being written.
我每周都会收到一份关于人工智能的所有不同文章的摘要。
Can we use it for moral questions?
我们可以用它来解决道德问题吗？
Which seemed silly to even ask to me，but fine，people can ask what they want to ask.
这对我来说甚至都显得很愚蠢，但是很好，人们可以问他们想问的问题。
This thing has the ability to move faster because the amount of people and resources and companies is way beyond those other breakthroughs that I brought up and was privileged to live through.
这件事有能力更快地移动，因为人员、资源和公司的数量远远超过我提出并有幸经历的其他突破。
KEVIN SCOTT: One of the things for me that has been really fascinating and I think I'm going to say this just as a reminder to folks who are thinking about pursuing careers in Computer Science and becoming programmers.
KEVIN SCOTT：其中一件对我来说非常吸引人的事情，我想我要说这只是为了提醒那些正在考虑从事计算机科学事业并成为程序员的人们。
I spent most of my training as a computer scientist in my early part of my career as a systems person.
在我作为系统人员的职业生涯的早期，我的大部分培训都是作为计算机科学家度过的。
I wrote compilers and wrote tons of assembly language and design programming languages。
我编写了编译器并编写了大量的汇编语言和设计编程语言。
which I know you did as well.
我知道你也这样做了。
I feel a lot of the things that I studied just in terms of parallel optimization and high-performance computer architectures in grad school.
我觉得我在研究生院学习的很多东西只是在并行优化和高性能计算机体系结构方面。
I left grad school and went to Google and thought I would never use any of this stuff ever again.
我离开研究生院去了谷歌，我想我再也不会用这些东西了。
Then all of a sudden now，we're building supercomputers to train models and these things are relevant again.
然后突然之间，我们正在建造超级计算机来训练模型，这些东西又变得相关了。
I think it's interesting.
我觉得很有趣。
I wonder what Bill Gates the young programmer。
我想知道年轻的程序员比尔盖茨。
would be working on if you were in the mix right now。
如果你现在在混音中，将会继续工作。
like writing code for these things because there's so many interesting things to go work on.
喜欢为这些事情编写代码，因为有很多有趣的事情要做。
But what do you think you as a 20-something-year-old。
但是你觉得你作为一个20多岁的人。
young programmer would get really excited about in this stack of technology?
年轻的程序员会对这堆技术感到兴奋吗？
BILL GATES: Well，there is an element of this that's fairly mathematical.
比尔·盖茨：嗯，其中有一个相当数学化的元素。
I feel lucky that I did a lot of math.
我很幸运，我做了很多数学。
That was a gateway to programming for me。
那是我编程的大门。
including all crazy stuff with numerical matrices and their properties.
包括所有具有数字矩阵及其属性的疯狂东西。
There are people who came to programming without that math background。
有些人没有数学背景就开始编程。
who don't need to go and get a little bit of the math.
谁不需要去学一点数学。
I&39;s super hard，but they should go and do it so that when you see those funny equations。
我并不是说这很难，但他们应该去做，这样当你看到那些有趣的方程式时。
you&39;m comfortable with that because a lot of the computation will be that thing instead of classic programming.
你就像，我对此很满意，因为很多计算将是那个东西而不是经典的编程。
The paradox when I started out writing tiny programs，was super necessary.
当我开始编写小程序时的悖论是非常必要的。
The original Macintosh is a 128K machine，128K bytes，22K of which is the bitmap screen.
原来的Macintosh是一台128K的机器，128K字节，其中22K是位图屏幕。
Almost nobody could write programs to fit in there.
几乎没有人能编写适合那里的程序。
Microsoft，our approach，our tools，let us write code for that machine and really only we and Apple succeeded。
微软，我们的方法，我们的工具，让我们为那台机器编写代码，实际上只有我们和苹果成功了。
then when it became 512K，a few people succeeded。
然后变成512K的时候，有几个人成功了。
but even that people found very difficult.
但即便如此，人们也很难做到。
I remember thinking as memory got to be 4 gigabytes。
我记得当时想内存必须是 4 GB。
all these programmers，they don't understand discipline。
所有这些程序员，他们不懂纪律。
and optimization and they're just allowed to waste resources.
和优化，他们只是被允许浪费资源。
But now that these things that you're operating with billions of parameters。
但现在你正在使用数十亿个参数来操作这些东西。
the idea of okay，can I skip some of those parameters?
好吧，我可以跳过其中一些参数吗？
Can I simplify some of those parameters?
我可以简化其中一些参数吗？
Can I precompute various things?
我可以预先计算各种东西吗？
If I have many models，can I keep deltas between models instead of having them?
如果我有很多模型，我可以保留模型之间的增量而不是拥有它们吗？
All the optimizations that made sense on these very resource-limited machines.
在这些资源非常有限的机器上有意义的所有优化。
Well，some of them come back in this world where when you're going to do billions of operations or literally hundreds of billions of operations。
好吧，他们中的一些人会回到这个世界，当你要进行数十亿次或数千亿次操作时。
we are pushing the absolute limit of the cost and performance of these computers.
我们正在推动这些计算机的成本和性能的绝对极限。
That's one thing that is very impressive is the speed-ups even in the last six months on some of these things has been better than expected.
令人印象深刻的一件事是，即使在过去六个月中，其中一些事情的加速也好于预期。
That's fantastic because you get the hardware speedup，the software speedup multiplied together.
这太棒了，因为你得到了硬件加速，软件加速成倍增加。
That means are we - how resource bottlenecked will we be over the next couple of years?
这意味着我们 - 在未来几年内我们将面临怎样的资源瓶颈？
That's less clear to me now that these improvements are taking place。
“智能制造”用英文怎么说 "Intelligent manufacturing"
现在这些改进正在进行中，我不太清楚。
although I still worry about that and how we make sure that companies broadly。
尽管我仍然担心这一点以及我们如何确保广泛的公司。
and Microsoft particular，allocates that in a smart way.
尤其是微软，以一种聪明的方式分配它。
Understanding algorithms，understanding why certain things are fast and slow that is fun.
理解算法，理解为什么某些事物快而慢，这很有趣。
The systems work that in my early queries。
这些系统在我早期的查询中工作。
just one computer and later a network of computers。
只有一台计算机，然后是计算机网络。
now that systems where you have datacenters with millions of CPUs。
现在你拥有数百万个 CPU 的数据中心的系统。
it's incredible the optimization that can be done there.
可以在那里完成的优化令人难以置信。
Just how the power supplies work，or how the network connections work.
电源是如何工作的，或者网络连接是如何工作的。
Anyway，in almost every area of computer science。
无论如何，几乎在计算机科学的每个领域。
including database type techniques。
包括数据库类型技术。
programming techniques，this forces us to think about in a really new way.
编程技术，这迫使我们以一种真正新的方式思考。
KEVIN SCOTT: I could not agree more.
凯文·斯科特：我完全同意。
Last，last question.
最后，最后一个问题。
I know that you are incredibly busy and you have the ability to choose to work on whatever it is that you want to work on.
我知道你非常忙，而且你有能力选择从事任何你想从事的工作。
But I want to ask you anyway，what do you do outside of work for fun?
但我还是想问你，工作之余你会做什么来消遣？
I ask everybody who comes on the show that.
我问每个参加演出的人。
BILL GATES: Oh，that's great.
比尔·盖茨：哦，太好了。
I get to read a lot.
我开始阅读很多东西。
I get to play tennis a lot.
我经常打网球。
During the pandemic I was down in California in the fall and winter and I'm still enjoying that.
在大流行期间，我在秋天和冬天去了加利福尼亚，我仍然很享受。
Although the Foundation's meeting in person and some of these Microsoft OpenAI meetings。
尽管基金会的现场会议和其中一些 Microsoft OpenAI 会议。
it&39;ve been able to do those in person。
很高兴我们能够亲自完成这些工作。
but some we can just do virtually.
但有些我们可以虚拟地做。
Anyway I play pickleball because I've been playing for over 50 years。
不管怎样，我玩泡菜球是因为我已经玩了 50 多年了。
tennis and I like to read a lot.
网球，我喜欢读书。
I goofed off and went to the Australian Open for the first time because it's nice warm weather down there and that was kind fo spectacular.
我偷懒去了澳大利亚网球公开赛，因为那里天气暖和，真是太壮观了。
KEVIN SCOTT: I actually want to push on this idea that you read a lot.
KEVIN SCOTT：我实际上想推动这个想法，你读了很多书。
You say you read a lot，which is not the same as what most people say when they say they read a lot.
你说你读了很多书，这和大多数人说你读了很多书不一样。
You're famous for carrying around like a giant tote bag of books with you everywhere you go and you read a insane amount of stuff.
您出名的是随身携带一个巨大的书包，无论您走到哪里，您都会阅读大量的东西。
Everything from like really difficult science books，all the way to fiction.
从真正困难的科学书籍，一直到小说，应有尽有。
How much do you actually read?
你实际读了多少？
What's a typical pace of reading for Bill Gates?
比尔盖茨的典型阅读速度是多少？
BILL GATES: If I don't read a book in a week，then I really look at what I was doing that week.
比尔·盖茨：如果我一周没有读一本书，那么我真的会看看那一周我做了什么。
If I&39;ll hope to read more like five。
如果我在度假，那么我希望能多读五本书。
six，or seven because books are quite variable in size.
六七本，因为书籍的大小变化很大。
Over the course of the year I should be able to read close to 80 plus books.
在这一年中，我应该能够阅读近 80 多本书。
I have a younger children who read even more than I do.
我有一个年幼的孩子，他们的阅读量甚至比我还多。
It's like，oh geez，I have to be - which Sowell books am I going to read?
就像，天哪，我必须——我要读索厄尔的哪本书？
I still read all the Smil，Pinker，some authors that are just so profound and have shaped my thinking.
我仍然阅读所有 Smil、Pinker 和一些非常深刻并塑造了我的思想的作者。
But reading's relaxing.
但阅读让人放松。
I should read more fiction.
我应该多读小说。
When I fall behind，my non-fiction list tends to dominate and yet people have suggested such good fiction stuff to me.
当我落后时，我的非小说类作品往往占主导地位，但人们向我推荐了如此好的小说类作品。
That's why I share my reading list on Gates Notes.
这就是为什么我在盖茨笔记上分享我的阅读清单。
KEVIN SCOTT: What&39;re famous for saying that you want to read David Foster Wallace's Infinite Jest。
凯文斯科特：什么是高低，你以说你想读大卫福斯特华莱士的无限笑话而闻名。
like over-under that happening in '23. BILL GATES:
就像 23 年发生的那样。比尔盖茨：
Well，if there hadn't been this darn AI advance.
好吧，要是没有这个该死的 AI 进步就好了。
I&39;s really true.
我在开玩笑，但这是真的。
I have allocated and with super excitement。
我已经分配并且非常兴奋。
a lot more time to sitting with Microsoft product groups and saying。
有更多时间与 Microsoft 产品组坐在一起说。
what does this mean for security?
这对安全意味着什么？
What does it mean for Office?
这对 Office 意味着什么？
What does it mean for our database type thing.
这对我们的数据库类型意味着什么。
Because I love that type of brainstorming because new vistas are opened up.
因为我喜欢那种类型的头脑风暴，因为它打开了新的视野。
So no，it's all your fault.
所以不，这都是你的错。
No Infinite Jest this year.
今年没有无限笑话。
KEVIN SCOTT: Excellent.
凯文·斯科特：非常好。
Well，I appreciate you making that trade because it's been really fantastic over the past six months having you help us think through all of this stuff.
好吧，我很感谢你做那笔交易，因为在过去的六个月里，你帮助我们思考所有这些事情真是太棒了。
Thank you for that and thank you for doing the podcast today.
谢谢你，也谢谢你今天做播客。
Really，really appreciate it.
智能制造:Intelligent manufacturing intelligence manufacturing
真的，真的很感激。
BILL GATES: No it was fun，Kevin.
比尔·盖茨：不，这很有趣，凯文。
Thanks so much.
非常感谢。
KEVIN SCOTT: Thank you so much to Bill for being with us here today.
凯文斯科特： 非常感谢比尔今天和我们在一起。
I hope you all enjoyed the conversation as much as I did.
我希望你们都像我一样喜欢这次谈话。
There&39;ve been having with Bill over the past handful of months as we think through this amazing revolution.
这次谈话有很多很棒的事情，反映了过去几个月我们在思考这场惊人的革命时与比尔的谈话。
One of the things that I've learned most from Bill over the past handful of months as we think about what AI means for the future is how he thought about what personal computing and the microprocessor and PC operating system revolution meant for the world when he was building Microsoft from the ground up.
在过去的几个月里，当我们思考 AI 对未来意味着什么时，我从 Bill 那里学到最多的一件事就是他如何思考个人计算、微处理器和 PC 操作系统革命对世界的意义，当时他从头开始建立微软。
Even what it felt like for him as one of the leaders helping bring the Internet revolution to the world.
甚至他作为帮助将互联网革命带给世界的领导者之一的感觉。
Like those parts of the conversation today that we have where he was recounting some of his experiences like the first meeting that he had with Charles Simonyi at Xerox PARC。
就像今天谈话的那些部分，他在讲述他的一些经历，比如他在 Xerox PARC 与 Charles Simonyi 的第一次会面。
where he saw one of the world's first graphical word processors and how seeing that in talking with Charles influenced an enormous amount of the history of not just Microsoft。
在那里他看到了世界上第一个图形文字处理器，以及在与 Charles 的谈话中看到它如何影响了不仅是微软的大量历史。
but the world in the subsequent years.
但随后几年的世界。
Just hearing Bill talk about his passion for the things that the Gates Foundation is doing and what these AI technologies mean for those things。
刚刚听到比尔谈到他对盖茨基金会正在做的事情的热情以及这些人工智能技术对这些事情意味着什么。
like how maybe we can use these technologies to accelerate some of the benefits to the people in the world who are most in need of technologies like this to help them live better and more successful lives.
比如我们如何使用这些技术来加速为世界上最需要此类技术的人们带来一些好处，以帮助他们过上更好、更成功的生活。
Again，this is a tremendous treat for being able to talk with Bill on the podcast today.
同样，今天能够在播客上与比尔交谈是一种巨大的享受。
If you have anything you'd like to share with us。
如果你有什么想和我们分享的。
please e-mail anytime at behindthetech@microsoft.com.
请随时发送电子邮件至 behindthetech@microsoft.com。
You can follow Behind the Tech on your favorite podcast platform。
你可以在你最喜欢的播客平台上关注 Behind the Tech。
or check out full video episodes on YouTube.
或在 YouTube 上查看完整的视频剧集。
Thanks for tuning in and see you next time.
感谢收看，下次再见。
