本集简介
双语字幕
仅展示文本字幕，不包含中文音频；想边听边看，请使用 Bayt 播客 App。
我认为我们已经实现了通用人工智能。
I think we have AGI.
我认为我们已经拥有了人工通用智能。
I think we have artificial general intelligence.
我们真的已经拥有了。
We really have.
你经常听说95%的项目都会失败，但你知道吗，这其实正是你想要的。
You you hear these 95% of projects fail, but, like, you know, like, that's that's that's actually what you want.
我认为大语言模型已经是一种商品。
I I think the LLM is a commodity.
人们还没有这样说，但它确实是一种商品。
People are not saying that, but it is a commodity.
你可以从这个加油站加到汽油。
Like, you can get gas from this gas station.
你可以从那个加油站加到汽油。
You can get gas from that gas station.
这不重要。
It doesn't matter.
只需比较价格。
Just compare price.
AI处于泡沫中吗？
Is AI in a bubble?
AI确实存在泡沫。
There is an AI bubble.
好吧。
Okay.
所以格伦也在泡沫中。
So then Gleen is also in the bubble.
人人都在泡沫里。
Everybody's in the bubble.
不。
No.
我会说存在一个泡沫。
I would I would say there is a bubble.
我会说那三个阵营。
I I would say those three camps.
嗯。
Yeah.
有一个超级智能追求派别。
There is a super intelligence quest camp.
嗯。
Mhmm.
我对这一点非常担忧。
I would be very worried there.
第二个是那些研究人员，你知道的，他们肯定不在泡沫里。
There's a second, the researchers doing the you know, that's definitely not in a bubble.
他们就像
They're like the
他们很清醒。
They're sober.
嗯。
Yeah.
他们非常清醒，但没人关心他们。
They're they're super sober and nobody cares about them.
然后呢？
Then there's right?
而他们可能是对的，不幸的是。
And they're probably the ones that are right, unfortunately.
然后是第三个阵营，也就是我们努力使其产生价值。
And then there's the third camp, which is us trying to make this valuable.
从某种意义上说，我们并不在泡沫中，因为我们没有在所做的事情上投入大量资本。
We're not in a bubble in a sense that we're not spending huge amounts of capital on what we are doing.
我们只是试图在这些组织内部实现真正的经济价值。
We're just trying to get actual economic value inside of these organizations.
两位传奇的建设者，阿里和埃尔文德。
Two legendary builders, Ali, Ervind.
我非常兴奋能和你们一起探讨这个话题，因为你们俩见证了我经历过的每一个超级周期——互联网、移动、云计算、数据、人工智能，不仅经历了这些周期的高潮，也经历了泡沫与幻灭的低谷，而这一次真的不同了。
I'm so thrilled to get into this with you because both of you have seen every super cycle I've lived through, Internet, mobile, cloud, data, AI, not just through the super cycles, but also through the hype, the trough of disillusionment, and this time it's different.
今天，我们将深入讨论人工智能的现状。
Today, we're gonna chop it up on the state of AI.
让我们先从一个宏观的视角开始。
You know, let's let's start with a 20,000 feet view.
评估一下我们目前所处的位置。
Take stock of where we are.
在消费级AI领域，我们已经看到了数十亿用户，ChatGPT，三年前枪声已经打响，云计算、Perplexity、ChatGPT，人们在房间里都在使用它。
AI, we've seen consumer AI, billions of users, chat GPT, said the guns went off three years ago, cloud, perplexity, chat GPT, people use it in the room.
在中小企业和开发者方面，我们有数亿用户在使用Cursor、Codex、Cloud Code等等。
On the SMB and developer side, we've got hundreds of millions of users with Cursor and Codex and Cloud Code and and and so on.
而在企业领域，却存在巨大的分歧。
Enterprise, on the other hand, there's a lot of divide.
很难看清战争的迷雾。
It's hard to see a lot of fog of war.
一方面，有些模型在数学、科学和工程基准上表现出色。
On one side, you've got models that are earning math benchmarks and science benchmarks and engineering benchmarks.
但另一方面，麻省理工学院的报告称，95%的AI部署都失败了。
But on the other side, you've got the MIT report that's saying 95% of AI deployments don't work.
现实究竟是什么？
What's the reality?
为我们弥合这一差距。
Bridge that gap for us.
从你的视角，全面地剖析一下。
Lay it out as you see it, view from the top.
所以我认为，首先我们应该意识到，人们在个人生活和工作中都在使用AI。
So I think, first of all, I think we should we should know that people use AI in their personal and work lives both.
因此，这种分隔并没有那么明显。
So there's not so much of a divide.
比如，你公司里的每个人可能每天都在使用ChatGPT、Claude和其他工具。
Like, you know, everybody in your company is probably using ChatGPT and Claude and other tools on a daily basis.
我觉得在企业中正在发生的是，你听到这些95%的项目都失败了，但事实上，这正是你想要的。
Thing that I feel is happening in enterprises, you hear these 95% of projects fail, but, like, you know, like, that's that's that's actually what you want.
当你真正尝试新技术时，如果所有的项目都失败了，那说明你还不够努力，你知道吗？
Like, you you like, when you are actually experimenting with new technology, if if if all of all of your projects are failing, that means you didn't just not trying enough, you know, at the moment.
所以当我做这项研究时，这对我来说并不意外。
So so I think when when I did the study, like, it was not a surprise for me.
你知道，我们希望明年也能看到类似的统计数据，因为我们希望整个行业都能充满热情地去实验，真正弄清楚如何融合、如何真正从这项技术中获益。
You know, we can actually see, hopefully, like, you know, similar stats next year too because we want everybody in the industry to be to be really eager and experiment and actually figure out, like, you know, how to mix how how how to actually make, you know, get better you know, get benefits from this technology.
这会让你们自然而然地成为那5%成功应用AI的团队，没错。
This would make you guys by default the 5% of AI that is working, which is Yeah.
二十分之一。
One in 20.
也许你们能进入那5%。
Maybe maybe you go to the 5%.
什么用例是真正有效的？
What what is the use case that is working?
不仅仅是节省时间，而是真正改变了你的公司。
And not just working like it's like saving me time, but like it's working and it's transforming my company.
一个你可以拿去给CFO看、让CFO注意到、而法律部门不会叫停的用例。
Something that you can take to the bank, to to the CFO while while the CFO will notice it, the legal won't shut it down.
是的。
Yeah.
没错。
Right.
是的。
Yeah.
我的意思是，我们看到了很多正在见效的用例。
Well, I mean, look, we're seeing a lot of use cases that are working.
但你得明白，你不能只是放任智能体自己运行就指望它自动生效。
It's just that you know you just have to, it's not just you can just unleash the agents and it just works.
这是一门工程艺术。
It's an engineering art.
如果你要打造一家真正具有差异化优势的公司——无论是我的公司、你的公司还是任何公司的公司，想要超越竞争对手，就不能随便拼凑一个东西，就以为竞争对手不会做同样的事。
Like if you're gonna have a company that's gonna be really differentiated, my company or your company or anyone's company and you wanna beat the competition, you can't just like quickly put something together and think that your competition is not gonna do the same thing.
因此，这需要评估，需要你将其投入生产，需要付出努力。
So that's gonna be something that needs evaluations, it needs something that you're gonna productionize, it's gonna take effort.
你需要一支优秀的团队来支持它。
You need a great team around it.
但我们已经看到了很多这样的案例。
But we're seeing a lot of them.
我来给你举几个例子。
Like I'll give you some examples.
加拿大皇家银行与我们合作构建了智能代理，这些代理在财报发布后，立即自动处理——股权研究分析师的工作就是撰写报告，判断哪些股票是买入、持有等。
Royal Bank of Canada built agents with us that basically take, as soon as an earnings report comes out, so equity research analyst their job is to put together these reports that say like this is a buy, this is a hold and so on.
这些智能代理会获取最新的财报、所有过往财报、竞争对手的财报，以及市场上的所有相关信息，进行全面分析，包括新闻动态等。
The agent goes, gets the earnings report, gets all the previous earnings reports, gets all the competitors earnings reports, gets everything that's going on in the market, does the full analysis, the news, everything.
它将所有信息整合在一起，能够在财报电话会结束后十五分钟内生成股权报告。
Puts it all together and it can get the equity report out in fifteen minutes from the earnings call.
行业标准是两小时。
Industry standard is two hours.
当然，这种技术最终会变得商品化，其他人也会跟进。
Of course it's gonna get commoditized and others are gonna do that as well.
但这是我们看到的金融领域一个非常重要的应用场景。
But that's actually really important use case that we're seeing in finance.
这就是金融领域的例子，还有许多类似的应用，比如筛选成千上万份文件、SEC报告等等。
So that's like finance, example in finance, And there's lots of examples like this, sifting through hundreds of thousands of documents, SEC reports, so on.
这就是金融领域。
That's finance.
让我们换个话题，谈谈医疗健康。
Let's switch gears, let's go to healthcare.
医疗健康领域则完全不同。
Healthcare is completely different.
在医疗领域，我们有一位客户默克公司，在生命科学领域创建了一个名为TEDDI的模型。
In healthcare we have a customer Merck that in the life science space created a model called TEDDI.
TEDDI代表Transformer驱动的药物发现。
TEDDI stands for Transformer Enabled Drug Discovery.
这是一个Transformer模型，类似于大型语言模型，但它不是预测下一个词，而是能判断如果你移除某个基因组，哪个基因组缺失了。
And this is a transformer model, kinda just like large language models that can predict But the next it instead can figure out which genome is missing if you remove a genome.
因此，它真正理解基因调控网络，能够准确告诉你基因表达等方面发生了什么。
So it really understands the gene regulatory network and can really start telling you what's happening with gene expression and so on.
这对药物发现至关重要。
So this is really important for drug discovery.
这还处于起步阶段，但将真正帮助我们实现以前无法做到的事情。
It's the beginnings, but this is gonna actually help us do things that we couldn't do before.
我们再来看看零售业。
Let's pick retail also.
所以我选择的是不同的领域。
So I'm picking different.
医疗是一个例子。
Healthcare is one.
我跟你提过金融，就是加拿大皇家银行的那个。
I gave you finance, right, the RBC one.
我们来看看零售，7-Eleven。
Let's go to retail, seven Eleven.
完全自动化营销栈的智能代理。
Agents that completely automate the marketing stack.
实际上，营销栈将受到严重颠覆。
Actually the marketing stack is gonna get disrupted pretty heavily.
这些代理可以基本上进行准备，它们可以细分受众，比如这个群体想听这个，还能生成直接针对你们的全部营销材料。
So these agents can basically prepare, they can segment audience like this segment wants to hear this, and they can prepare all the marketing material that's like directly targeting you guys.
它还能整合活动并完成这些工作。
And it can put the campaigns together and do that.
7-Eleven之前也在做这件事。
Seven Eleven was doing this before as well.
但这一点，我们在Databricks也看到了，越来越多的事情由代理自动完成，让你能更快地实现。
But this, and we're seeing this at Databricks as well, more and more is being done by agents and being automated so you can just do it faster.
而且你可以进行更精细的细分。
And you can segment more fine grain.
因为以前你必须为各个群体创建内容，这需要大量人工内容创作。
Cause before you had to create the content for the groups that was a heavy content creation with something that was human manual labor.
现在你实际上可以做得多得多。
Now you can actually do that much much more.
你可以让你所有的网络内容完全针对目标群体进行定制。
You can have all your web materials completely customized for a target group.
这些都是已经取得成效的例子。
So these are examples where it is working.
但也有很多例子表明它并未成功，即使在Databricks，我们也不只是那5%。
There are also lots of examples where it's not Even with Databricks, we're not just the 5%.
也要看看那95%的情况。
Have some of that 95% too.
但有一些我们看到成功的例子。
But some examples where we're seeing success.
阿里，接着说一下这个。
Ali, follow-up on that.
这些都是很好的例子。
These are great examples.
谢谢。
Thank you.
也许如果你往上再深入一层，这些用例、这些组织或这些首席信息官之间有什么共同点，使得这些用例能够成功？
Maybe if you if you were to book take it a layer up, what is common across these use cases or these organizations or these CIOs that's making these use cases work?
我们能否找出某种模式？
Is there something that we can pattern match?
是的。
Yeah.
看。
Look.
我认为大语言模型已经成为一种大宗商品。
I I think the LLM is a commodity.
人们还没有这么说，但它确实是一种大宗商品。
People are not saying that, but it is a commodity.
就像我上经济学课时学到的，大宗商品就是可以互相替代的东西。
Like, and you know, when I took econ classes, commodity was when it's interchangeable.
你可以从这个加油站加油，也可以从那个加油站加油。
Like you can get gas from this gas station or you can get gas from that gas station.
这并不重要，只需比较价格即可。
It doesn't matter, just compare price.
大语言模型已经变成了这样。
LLMs have become that way.
现在这个模型更好，但下周另一个可能更好，你根本跟不上了，对吧？
Like it doesn't really matter, this one is better right now, next week that one is better, you can't even keep up anymore, right?
到底发生了什么？
What's happening?
所以它们是商品。
So they're a commodity.
所以这并不是重点。
So it's not about that.
真正重要的是你的公司。
It really comes down to your company.
你的公司拥有哪些竞争对手没有的特殊数据？
What data does your company have that's special that your competitors don't have?
你能利用这些数据并构建出真正理解这些数据的AI吗？
Can you leverage that and can you build AI that really understands that data?
因为那不是商品。
Because that's not a commodity.
目前没有哪款AI能理解你公司所有的业务流程、你的核心秘诀和你的数据。
There's not an AI out there that understands all your business processes in your company, your secret sauce and your data.
那不是商品。
That's not a commodity.
事实上，这更接近95%。
In fact that's closer to the 95%.
归根结底就是这一点。
It really comes down to that.
或者如果你公司有某种复杂的流程，只有你们公司才拥有，这就是你们提供产品和服务的方式。
Or if you have a complicated process that just your company has, this is how you deliver your product and services in your company.
而这一部分可以通过AI以某种方式实现突破。
And it's, that portion can be disrupted with AI somehow.
如果你能做到这一点，就能超越竞争对手。
If you can do that, now you can get ahead of your competition.
但这又回到了什么让你们公司与众不同。
But it comes back to what makes your company special.
不幸的是，许多公司只是在构建普通的产品。
Unfortunately a lot of companies are just building commodity stuff.
你不应该做这种事，因为这是每个公司都能做到的。
You should not be building that because it's a thing that every company can do.
这对你公司来说并不独特，我认为这是整个行业的问题。
It's not special to your company or to to your that's that's I think the problem in a lot of the industry.
行业中的另一个问题是大量所谓的演示软件。
Another problem in the industry is that a lot of demo ware.
用生成式AI制作酷炫的演示非常容易。
It's really easy to make cool demos with GenAI.
因此，我们看到了很多酷炫的演示，但它们就只是演示而已。
And, you know, therefore we're seeing a lot of cool demos, but that's all they are.
是的。
Yeah.
我们在Altimeter经常提到的一点是：你的AI战略应从你的数据战略开始。
Well, you know, something we say around at Altimeter quite a bit is your AI strategy start as your data strategy.
是的。
Yeah.
所以你必须先把数据基础整理好。
And so you gotta get the data house in order first.
而且，你知道，有很多原因导致一些用例我们尝试过，但并没有成功。
And, you know, there's a lot of reasons for for for use cases that are, you know, we were trying, we're not working.
能否给我们举个例子，说明你们在Databricks或Glean时，95%的AI投入最终没有成功，以及失败的原因是什么？
Maybe give us an example of the 95% of an AI bet that either of you had at Databricks, at Glean that did not work out and why it didn't work out.
实际上，如今工程领域有个有趣的现象：你构建系统时，前所未有地经常遇到这种情况——你一开始有个很棒的想法，但短短两周内，它就不再像当初那样靠谱了，因为我们总能看到新的进展。
It's actually an interesting thing, you know, with engineering today is you build systems and never before have you been in this mode where you start with a great idea and it doesn't seem like a good idea anymore, like within two weeks, you know, because we see a new development that happens.
因此，我们在工程方面遇到了大量类似的失败。
So there are we have, like, numerous failures in engineering on that front.
比如，我们的一些微调工作，为产品中的特定用例构建模型，实际上并没有取得预期效果。
Like, you know, for example, some of our fine tuning work, building models for specific use case within our product, like, know, didn't really pan out for us.
最终，我们的选择是：直接采用已有的模型，无论是托管在Databricks上的小型开源模型，还是一些大型基础模型。
And ultimately, the choice was that, you know, we can go with already built models, whether they are small open source models hosted on Databricks or one of the large, you know, foundation models.
但就企业内部的使用场景而言，我们实际上也处于这样一种状态：很多工作虽然不能说完全失败，但确实需要比预期长得多的时间才能看到成效。
The but internally, like, you know, from a corporate, you know, use cases perspective, actually, like, we are also, like, in in many ways in this mode where a lot of our work actually like, I would not say, like, fail, but it actually takes much longer than, you know, to actually generate success.
你知道，我们实际上正在努力自动化公司内部的许多业务流程。
You know, there are you know, we're actually trying to automate a lot of our business processes internally.
比如，我希望我们公司里的每个人都能清楚知道自己这一周的首要任务是什么，该做什么。
And, like, example, like, you know, one thing that I want is, in our company, I want everybody to actually know exactly what their top priority for the week is, what they want to work on.
也许我们希望一个AI代理先告诉他们应该优先做什么，所有内容都要有文档记录，并且我们需要一个系统能将这些信息汇总起来，让我每周都能快速看到公司里每个人都在做什么，以及这些工作是否与我希望他们做的保持一致？
And and maybe, you know, we want an AI agent to actually first tell them what their priority should be, and we want it all to be documented, and we want a system which actually then, you know, rolls it rolls it all up, and I get a view every week where I can actually quickly see, you know, what are all the different people working on in the company, and are they aligned is that aligned with, you know, what I want them to work on?
嗯。
Mhmm.
这其实是个很简单的事情，公司一直都在尝试实现这一点，作为CEO，你总希望做到，但一直很难实现。我们原本以为AI能轻易地完成这一切，因为它掌握了公司内部的所有上下文信息，但到目前为止我依然没有实现。
And and this this is a simple thing, like, you know, companies are always tried to actually have this, you know, as CEOs, you always want it, and it's always hard to make happen, and we thought that AI would simply just, like, you know, magically do all of this work because, you know, like it has all the context, the billing has all the context inside the company to make it happen, but I still don't have it.
所以，事情确实需要时间，就像Dolly说的，AI只是你工具箱里的又一个工具。
So things do take time to actually, you know, Dolly's point, like, you know, there is AI is just one more tool that you have in the toolkit.
它并不能突然让构建复杂的企业系统变得轻而易举，不可能一天之内就搞定。
It does not suddenly make building complex enterprise systems, you know, you know, like it doesn't make it like that you can, you know, build it up like in in one day.
不是吗？
Doesn't it?
是的。
Yeah.
你知道吗，上一次企业对某种工具如此兴奋的时候，那叫RPA。
You know, the last time enterprises got this excited about a tool was called RPA.
嗯。
Mhmm.
而我们知道它最终结果如何。
And we know how that ended.
不幸的是，它逐渐消退了。
It unfortunately fizzled out.
昨天有位观众说：‘这次和RPA有什么不同？’
And, you know, somebody in the audience yesterday is like, hey, how is this time different from RPA?
看起来像是同一部电影，只是预算更高、演员更好。
It seems like the same movie, bigger budgets, better actors.
这次有什么不同呢？
What what's different this time?
这次的技术本质或架构，与之前的自动化浪潮有何不同？
How how is the nature or the architecture of the the the the technology different from the previous automation cycle?
你们俩任何一个。
Either of you.
嗯。
Yeah.
我的意思是，首先，RPA 根本没引起我的注意，所以我真的没在意过，没错。
I mean, I I first of all, RPA, like, it didn't take you know, it didn't capture my attention at all, so I no so I actually can't, Right.
你知道 RPA 是什么吗？
You know What is RPA?
嗯。
Yeah.
所以我觉得，我根本不会把这两种技术相提并论。
So so so I think I I I think, like, I I would not compare these two technologies at all.
你知道，我们现在看到的 AI 是如此根本性的变革，当你第一次看到它时，简直像魔法一样。
Like, you know, you know, what we're what we're seeing now with AI is so fundamental, you know, it's it's, you know, it's it's the, you know, when when we saw it first, it was basically magic.
我们简直不敢相信，这居然是机器在完成这些工作。
And and we couldn't believe that this is a machine that is doing this work.
机器根本不可能做到我们看到的这些事情。
Machines just simply cannot do these kind of things, you know, that we saw them do.
比如自己骑行、拥有情感、理解情感。
Like riding on their own, having emotion, understanding emotion.
所以这是一项根本性的变革。
So it's a it's, you know, it's it's fundamental.
它是不同的。
It's different.
而且，是的。
And and the yeah.
正因如此，我认为这项技术不会消退。
And and that's why, like, you know, I don't think, you know, we this this this technology is going to fizzle out.
你不需要是金融专家，或者对商业有深刻见解。
And it's not like, you know, you don't have to be like a financial expert or, you know, a bit, you know, like sort of a deep thinker on business.
这是显而易见的事情，我们所有人都能感受到，都能看到这项技术的能力，我们知道它很特别，而且会持续存在。
This is obvious stuff, like, you know, of us know, all of us feel it, all can of see the capability of this technology, and we know it's special and it's going to be around.
是的。
Yeah.
你想听听我的RPA吗？
Do you wanna hear my RPA?
请说。
Please.
你知道，它基于规则，但问题在于，尤其是当你想自动化桌面操作和正在进行的工作时，总会遇到太多意外情况，设置起来既困难又脆弱。
You know, mean, was rule based and the problem with it, especially if you're, you know, you want something that automates what's going on on your desktop and automate the work that's happening, It's just that there's too much unexpected things that happen and it's just hard and brittle to set it up.
它从来不会学习。
It wasn't learning ever.
所以根本没有任何学习能力。
So there was like zero learning.
你得明确告诉它具体的规则是什么。
It was like you tell it exactly here are the rules.
如果它出错了，你就得回去重新扩展规则。
And if it got something wrong you need to go and go back and expand the rules.
这里你有一个能够学习的系统，它可以不断改进、泛化，并理解模式，进行模式识别。
Here you have something that's learning, So it can improve and it can generalize and it can understand the patterns and do pattern recognition.
因此，这是这两者之间的根本区别。现在有许多在生成式AI领域失败的初创公司。
So that's the fundamental difference between these Now there has been many startups that have failed in the generative AI.
我们将用生成式AI模型取代RPA。
We're gonna replace RPA with generative AI models.
实际上，我知道有不少失败的初创公司，其中一些还很有名。
There's many startups that failed actually that I know of, pretty some high profile ones.
这是因为我们今天所处的AI范式仍然存在问题。
It's because the paradigm we live in today with AI is there's still problems.
最大的问题是，你训练好一个模型，让它学会所有需要学习的内容，然后就把它冻结了。
The biggest problem is that you bake a model and that's where it's learned everything it needs to learn and then you freeze it.
然后你把它发布出去。
And then you launch it.
之后你或许会给它一些上下文，但仅此而已，它已经被冻结了。
And then maybe you give it some context, but that's it, it's frozen.
因此，问题在于我们需要一种AI，它能够在使用桌面、点击操作的同时持续学习。
So therein lies the problem that we need an AI that really can sort of continue learning while it's using desktop and clicking around.
我认为这个问题很难解决，但我认为阿尔文说得对，这就像僵化的基于规则的系统与学习型遗传系统之间的对比。
So I do think this problem is hard to nail, but I think Arvind is right that it's like there's no comparison at It's like brittle, rule based stuff versus learning a genetic system.
我认为它会完美地实现这一点。
I think it's gonna nail it perfectly.
但我们还没有真正攻克计算机操作这一难题。
But we haven't really nailed computer use yet.
是的。
Yeah.
正在努力解决。
Working on it.
最重要的转变是从如果-那么-否则语句转向一种能够自行找出解决方案的生成式方法。
The number one shift is this move from if then else statements to a more generative solution that figures out the solution.
是的。
Yeah.
所以你是在用不确定性换取确定性。
And so you're trading breath for for maybe determinism.
这似乎是关键区别。
That seems to be the difference.
在座的有很多CIO，他们手头都有即将到来的预算规划。
And, you know, there's a lot of CIOs in the room we've got here, and they've got budgets coming up to plan.
如果你要根据你对客户群的了解，给他们一两个建议，告诉他们必须弄清楚并协调好激励机制，这可能涉及可靠性问题或组织架构。
If you were giving advice to them of, like, hey, based on everything I know from my customer base, here's one thing or two things that you've gotta figure out and and align incentives on, or it could be a reliability problem or org design.
对于正在考虑AI预算的CIO们，你有什么建议？
What advice would you have for CIOs who are thinking about their AI budgets right now?
那就多花钱。
Well, spend more.
把钱花在Glean上。
Put it on Glean.
花在Glean上。
It on Glean.
多花钱，没错，把钱花在Glean上。
Spend more, yeah, put it on Glean.
但我认为，如今AI市场中非常重要的一点是，它非常新，而且有很多参与者。
But the I I think the like, one thing, you know, which which is important in AI market today is that it's very new and there are many players.
事实上，每一家软件公司现在也都是一家AI公司。
In fact, every software company is also an AI company now.
你可以去查看他们的网站。
You can go and check their websites.
所以，我认为很难真正确定如何分配这些预算。
So the I think it's just hard to actually figure out where to allocate those budgets.
我们告诉人们的是，赢家尚未出现，因此要多尝试几家供应商，签订短期合同。
And what we tell people is that I think the winners are yet to be identified and so experiment with more vendors, do shorter term contracts.
虽然说起来容易，但实际实施起来很难，因为你尝试的每一个产品都需要付出成本，才能让它真正得到测试。
And and, you know, while that's easy to say, it's hard to actually implement because every product that you try has, you know, it's a cost that you have to pay to make it make it sort of even tested.
所以，你还得选择那些容易测试的产品。
So so you had to also pick products that are are easy to test.
我的意思是，那些不需要你花接下来六个月去尝试实施、却完全不知道最终会得到什么的产品。
I mean, those are the the ones that don't require you to, you know, spend the next six months trying to implement something and you have no idea what's gonna come out after that.
比如，如今那些用正确AI技术构建的产品，应该能很快为你发挥作用。
Like, you know, the products of today, the the products that are built with the right AI, they should work, you know, very, very quickly for you.
卡尔，走一圈。
Karl, walk around.
我们要窥探一下未来。
We're gonna take a peek into the future.
换个话题，让我这样的投资者夜不能寐的一件事，就是有2500亿美元被花在了英伟达和半导体领域。
Shifting gears, you know, one of the things that keeps investors like me up in the nights is a quarter trillion being spent on NVIDIA on on the semi side of things.
假设这只是资本支出的50%，那么你们在资本支出上就花了大约5000亿美元，而要让这笔支出物有所值，你们必须赚取约一万亿美元的AI收入。
Assuming that is just 50% of the CapEx, you're spending about half 1,000,000,000,000 on CapEx, and then you've gotta earn about a trillion dollars of AI revenue for all of this CapEx to be worth it.
为了提供背景参考，整个软件行业目前的收入大约为4000亿美元。
This and just to put this in context, the entirety of the software industry earns about $400,000,000,000 of revenue.
到目前为止，这看起来更像是一个物理难题。
This seems like a physics problem at this point.
你怎么能怎么能够
How do you how do you
你觉得这会如何发展？
think this this plays out?
你知道，你必须赚到大约一万亿美元的收入，才能证明当前正在进行的这笔支出是合理的。
You know, you've got you've got to make about a trillion dollars of revenue to justify this present spend that's already happening.
你觉得这会如何收场？
How do you think this shakes out?
也许我们先从你开始，阿尔文。
Maybe we start with you, Arvind.
找错人了，但你知道，我是个工程师，我其实并不太关心谁在花多少钱，你知道，我们能打造出产品并创造价值。
Wrong person to start with, but, you know, I'm an engineer, and I actually don't really, you know, think too much about who's spending what money, like, you know, we're able to build our product and add value.
所以某种程度上，你知道，我确实没怎么思考过这个问题。
So that's so in some sense, you know, I'm I've not really thought too much about this problem.
但如果你想想人工智能，你知道，人工智能实际上并不是在边际上扩展软件。
But but if you think about AI, the, you know, AI is not actually, you know, extending software in a marginal way.
这是一个不同的产品，事实上，它将攫取大量目前属于服务行业的收入，而服务行业的规模是软件行业的25倍。
It's a different product, and in fact, you know, it's actually going to grab a lot of revenue that actually today is in services industry, which is 25 times larger than software industry.
因此，会有大量的支出发生转移。
So there's a lot of spend that is gonna move.
我的意思是，你看到的在AI上的支出，实际上就是那些从服务费用或软件费用转化而来的资金。
I mean, the spend that you see happen on AI is actually sort of, you know, those service dollars that are converting into AI or software dollars.
不过话虽如此，也许你对这个问题有更深入的看法。
And I think the, but with that said, know, maybe you have a more informed view on this.
顺便问一下，你是否认为，正如你所说，我是个工程师，只想打造一些酷炫的东西？
By the way, do think that's just to build on, you said, you know, I'm an engineering, I wanna just build something that's cool.
我认为这并不是非黑即白的，对吧？
I do think it's not binary, right?
并不是说物理原理行不通，整个体系就会崩溃。
It's not like okay so the physics doesn't work out, so the whole thing will collapse.
不，总会有一些东西是能成功的，因此继续专注于那些明显已经奏效的领域，并进一步拓展，是个好主意。
No, there's gonna be things that work, and so it is a good idea to continue focusing on the stuff that is obviously already working, continue expanding on that.
但我觉得如果你放宽视角，我认为有三种范式或三个阵营。
But I think if you zoom out, I think there's like three paradigms or three kind of camps.
我把阿文归入第三个阵营。
And I put Arvind in the third camp.
实际上，我也把自己归入第三个阵营。
I actually put myself also in the third camp.
但让我们先从第一个阵营开始。
But let's start with the first camp.
我认为第一个阵营是追求超级智能的阵营。
I think the first camp is this quest for super intelligence camp.
我认为所有前沿实验室都在做这件事。
And it's, I think all the frontier labs are doing this.
不管是三个、四个还是五个，不管你如何计数。
Like all three, four, or five of them, however you wanna count them.
我认为这仍然深受规模定律思维的影响，即拥有最多GPU和最多数据的人将赢得超级智能的竞赛，这种智能近乎神级，能实现AI的自我递归改进，一旦达成，就能治愈癌症并解决所有经济问题。
And I think it's really still being, a lot of it comes from the scaling laws mentality, which is whoever has the most GPUs and the most data is gonna win the quest for super intelligence, which is kind of intelligence that's like an almost godlike, it leads to recursive self improvement of the AI, which then once you have that, it can cure cancer and solve all economical problems.
我们可以在几年内将GDP提升十倍。
And we can probably 10x GDP over a few years period of time.
那你所说的物理限制问题到底是什么意思？
So what the hell are you talking about that there's a physics problem?
任何你的成本方程与这件事所带来的经济价值相比都会相形见绌。
Like anything, any of your cost equations are gonna pale in comparison to the economic value that this thing is gonna provide.
所以这是一种流派，他们的发展方式是构建越来越大的集群，消耗越来越多的能源，这就是他们的做法。
So that's like one camp and the way they're developing it is bigger and bigger clusters, more and more energy and that's how they're going about it.
大部分资本都流向了这个领域，对吧？
And that's where most of the capital is going, right?
你花的钱或者我花的钱，都属于这一派。
The kind of capital you're spending or I'm spending, but that's that camp.
那他们怎么知道自己成功了呢？
And then how do they know that they're succeeding?
他们可不是简单地说：‘相信我们就好。’
They're not just like, let's trust us.
这些人都非常聪明，正在从事这项工作。
They're very smart people working on this.
所以他们的方式是，把我们面临的最困难的问题交给现有的任何人工智能来解决。
So the way they're approaching it, they're saying, we'll throw the hardest questions we have at whatever AI we have now.
如果人工智能能完美解决这些问题，而且我们正在取得飞速进展，那你的问题到底出在哪里？
And if it nails them, and we're making really rapid progress, so what's your problem?
看看数学奥林匹克竞赛吧。
Like look at Math Olympiad.
我们正在轻松解决这些数学奥林匹克和物理奥林匹克的问题。
We're like nailing these Math Olympiad problems, Physics Olympiad.
编程竞赛的表现已经超越了任何人。
And programming contest is like better than any human being.
所以他们把所有最具智力挑战性的问题都丢给了人工智能。
So like that's what they're throwing all the most intellectually challenging.
还有第二个阵营，那就是最初技术的创造者、开发出这项技术的科学家们，他们因此获得了计算机科学界的诺贝尔奖——图灵奖。
There's a second camp which are the people that created the original technology, scientists who created the technology, got them the computer science Nobel Prize for it, it's called Turing Award.
这正是里奇·萨顿，他创造了强化学习，而许多这些技术都是建立在其基础上的。
And that's Rich Sutton who created reinforcement learning, which a lot of this stuff is built on.
你还有杨·勒昆，他是三位奠基人之一，还有许多其他人。
You have Jan Lakun who was one of the three founding fathers, and many others.
多年来，实际上我问了他们很多年，他们一直认为第一派的观点是错误的，这不是正确的方向。
They have for many years, actually I've been, I asked them for years, they've been saying that that first camp is not gonna, that's like not even the right approach, is their view.
他们说，这只不过是自回归的下一个词元预测。
They're like no, that's just like autoregressive next token prediction.
这只是概率性地预测下一个词元。
It's just probabilistically predicting the next token.
这并不是人类学习的方式，他们通常会这么说。
That's not how, and usually they will say that's not how humans learn.
这也不是动物学习的方式。
That's not how animals learn.
我们的运作方式是不同的。
We operate in a different way.
你的大脑并不是那样的。
Your brain is not that way.
一个例子是，即使孩子也能非常迅速地学会走路、说话和做事情，所需的数据量极少，绝对没有哪个孩子是在读遍互联网数据四遍之后才学会说话的。
And one example is that even a child learns very quickly to walk and talk and do things with very little data compared to, certainly no child is reading all of the internet's data four times over before they learn to speak.
所以这是第二派。
So that's like camp number two.
顺便说一下，这些人说还需要二十年。
Those guys by the way, they say it's twenty years out.
他们认为，这是一个物理学问题，需要二十年才能实现。
So they're saying, hey, it's a physics problem and it's gonna take twenty years to get there.
这让我觉得，别管我了。
Which to me it's like, don't know, like leave me alone.
让我做研究吧。
Let me do research.
第三派，我认为我们正属于这一派，即我不认为我们需要超级智能。
Third camp, which is I think what we are in is, I don't think we need super intelligence.
我觉得我们现在并不需要那种超级智能。
Like I don't think we need that super intelligence right now.
也许他们最终能做到，如果真做到了那就太棒了。
Maybe they'll get there, that's awesome if they do.
但我认为我们已经有了通用人工智能。
But I think we have AGI.
我认为我们已经有了人工通用智能。
I think we have artificial general intelligence.
我们确实已经有了，我们绝对已经有了。
We really have it, we absolutely have it.
任何说我们需要实现AGI的人，都是从一个错误的前提开始的。
It's like anyone who says we need to get to AGI, that's like it's false premise to start with.
我们已经拥有AGI了。
We already have AGI.
我于2009年来到美国，在加州大学伯克利分校，离这里不远。
I came to United States in 2009 at UC Berkeley, not far away from here.
我当时在一个名为AMP实验室的AI实验室工作。
And I was in an AI lab, was called AMP Lab.
A代表算法和人工智能，机器与人。
The A was for algorithms and AI, machines and people.
这些都是AI领域的人。
And these are all AI people.
当时我们对AGI的定义，我们已经满足了。
And back then the definition of AGI we had, we already have satisfied that.
我记得我们当时的讨论。
I know the discussions we had.
我实际上回去联系了那些人，想看看是我一个人这么想，还是2009年时大家都是这种看法？
And I actually went back to some of those folks to see like is it just me or what was the sentiment back in 2009?
我问过的每个人都说，是的，按我们当时的标准，我们确实已经拥有AGI了，但现在我们改变了定义。
And everybody that I talked to said, yeah, that's by those standards we had AGI, but we've changed the definition now.
我们有那些定义，你知道的，广告。
We have those definitions, you know ads.
所以三十到四十年来，我们一直有一个AGI的定义，而我们已经达到了。
So for thirty, forty years we had a definition of AGI, we've already hit that.
现在我们正在改变它，移动目标。
Now we're changing it and moving the goalpost.
但很明显，我们已经拥有AGI了。
But very obviously we already have AGI.
随便使用任何一个这些大语言模型，让它做一些推理，它肯定比你很多朋友都聪明，对吧？
Just use any of these LLMs and have it do some reasoning, and certainly it's smarter than a lot of friends that you have, right?
别提那些同事或别的什么人了，对吧？
Let's not them, or coworkers or whatever, right?
所以你已经拥有AGI了。
So you already have AGI.
现在我们不再纠结它到底有多聪明。
Now we're not haggling over exactly how smart is it.
你有没有朋友比它更聪明？
Do you have a friend that's smarter or not?
既然我们已经有了AGI，那么现在只需要让它在企业中变得有用。
So if we already have AGI we just need to make it useful inside the enterprise.
我们需要把这5%扩大到10%、20%、30%。
We need to just expand that 5% to be 10%, 20%, 30%.
所以我认为阿尔敏的回答其实是个很好的回答。
So that's why I think Armin's answer is actually a good answer.
我们已经有了所需的AGI，现在只需专注于解决组织内部的实际问题。
We have the AGI we need, let us just focus on solving the actual problems inside the organizations.
我认为我们现在已经足够强大，能够自动化许多任务，并从中获得巨大的经济价值。
And I think we can already, that's enough to automate a lot of the tasks and get huge economic value out of it.
我们其实并不需要超级智能来实现这一点。
We don't actually need super intelligence for that.
如果那些研究超级智能的人成功了，太棒了，那样我们就治愈了癌症。
That's a good idea if the super intelligence guys nail it, amazing, then we've cured cancer.
如果他们没成功，希望第二阵营在未来二十年内能提出新的突破，那也同样很棒。
If they don't hopefully the second camp comes up with a new thing in the next twenty years, that's also awesome.
展开剩余字幕（还有 361 条）
我们已经拥有了所需的一切。
We already have whatever we need.
是的，让我们专心做好工程工作吧。
Yeah, let us just do our engineering.
对。
Right.
是的。
Yeah.
是的。
Yeah.
这个表述非常好。
That's really good framing.
这种理念在现实中的体现就是存在一个数据层。
And the way this manifests in the in the you know, in the world is there's a data layer.
嗯。
Mhmm.
还有一个智能层，第一阵营 presumably 正在这一层产出大量优秀的模型。
There's the intelligence layer, which is where camp one is presumably producing a lot of great models.
然后是用户交互的软件层。
And then there's the software layer where the users engage with.
如果你要在这三个层——数据层、智能层和软件或应用层——分配100单位的价值，你认为价值会集中在哪儿？
Where do you think value will accrue if you were to design a 100 units of value across these three layers, the data layer, the intelligence layer, and the software or the application layer?
你认为在未来五年里，价值会集中在哪儿？
Where do you think value accrues in the next five years?
好吧。
All right.
这是一个很难的问题。
This is a this is a tough question.
我的意思是，我认为这三个层实际上都非常重要。
I mean, I think the all all those three layers actually are very fundamental.
是的。
Yeah.
我以为你会再加几个，但其实并没有。
I thought you're gonna add a few more which are not.
是的，我觉得正如阿里所说，这些模型将会对我们所有人开放，它们会变成一种商品，很难看出更多的投入会流向模型本身，而不是这些上层的架构。
Yeah, I think, I feel like the, like as Ali was saying that the models are going to be available to all of us, you know, they are going to be commodity, and it's going to be hard to sort of see that it it the more spend goes to them versus, you know, these layers on top.
但你怎么看呢？说实话，很难判断哪里会创造最大的价值。
The but how do you how do you like, you know, I it's hard to sort of come up with, you know, where the most value will be.
而且我也不知道，这是否真的会改变当今的技术架构——比如，在人工智能出现之前，任何企业应用和数据系统都是如此。
And and I also don't know if if actually it changes from today's technology architecture where, again, like, you know, when you think about in a pre AI world, any any sort of, like, you know, enterprise, you know, application and data systems.
你知道，你有数据系统。
You know, you have you have data systems.
我想你目前确实缺乏足够的智能层，然后你还有应用层。
You you do have I guess, you you don't have enough of that intelligent layer today, and then you you have the application layer.
所以，我想会有一些资金转向这个领域。
So so so I so I I guess, you know, some some dollars will shift into it.
我知道我们确实认为，智能层将会是一个非常重要的部分。
I know we do think that the intelligence layer is actually gonna be a pretty thick one.
也许，它会捕获一半的企业价值。
Maybe, you know, it'll capture half of the enterprise value.
阿里，你还有什么要补充的吗？
Anything to add, Ali?
嗯。
Yeah.
没有。
No.
我认为，取决于你如何去做，栈中会有更多的层次。
I I I think that, you know, yeah, there are more layers in the stack depending on how you wanna do it.
但正如我所说，大语言模型已经成为一种商品，就像阿文说的，你可以获得它们，但这并不意味着这些公司不会有价值。
But I think as I said, LLMs is a commodity as Arvind You can get them, but that doesn't mean those companies are not gonna be valuable.
它们可以非常有价值，比如台积电就非常有价值。
They can be very, I mean TSMC is very valuable.
它们将会像这些晶圆厂一样的公司。
And they're gonna be kind of like these fab like companies.
但它们是可以互换的，我们从未见过这样的情况。
But they're interchangeable, and we've never seen something like that ever.
在这整个过程中，我从未见过人们像在一天之内就切换大语言模型那样。
I have not during all these, people just switch LLMs like in one day.
这和你的iPhone与Android、Windows与Mac，或者其他任何东西之间的区别都不一样。
That's not the case with your iPhone versus Android, or your Windows versus your Mac, or your anything versus anything.
比如，Google Sheets和Excel之间的差异，就像我们公司内部一场巨大的宗教战争。
Like know, Google Sheets versus Excel is like huge religious battle inside our company.
大语言模型就像我所说的那样，是一种商品。
LLMs it's like, it's a commodity as I said.
它只会说英语或者你喜欢的任何语言，并且每次给出的答案都不同，不如直接试试更便宜的那个，更便宜的商品或者稍微聪明一点的商品。
It just speaks English or any language you like, and it gives you different answers every time, might as well just try the cheaper one, the cheaper commodity or the slightly smarter commodity.
你根本都看不出区别，对吧？
You can't even really tell the difference, can you?
所以真正特别的是你拥有的数据。
So then what is special is the data that you have.
再说一遍，如果你的公司拥有其实际收集到的、竞争对手没有的数据。
Again, if your company has data that it has actually collected that your competitors do not have.
比如Glean非常棒，但如果你把Glean的所有数据都移除，它就毫无用处了，对吧？
Like Glean is amazing, but if you remove all the data from Glean there's no use to it, right?
所以关键在于你拥有的数据。
So it's all about the data that you have.
而且你能否保护好这些数据？
And can you secure the data also?
如果我们让智能代理四处访问这些数据，比如，哦，那是他的HR数据。
So if we're gonna have agents running around accessing this data, like oh that's his HR data.
哦，这是Purva的薪资信息。
Oh here is the Purva's salary information.
哎呀，我不小心把它们说出来了。
Oops I blurped it out to all of you.
那么，你该如何加以限制呢？
So how do you lock it down?
你如何确保有良好的治理？
How do you make sure that there's governance?
人们还非常担心，如果使用的是中国模型，如果它访问了这些信息，如果它将这些信息分享给竞争对手，如果它进行了交互怎么办？
There's also a lot of worry around what if it's using a Chinese model, what if it's accessing this information, what if it's sharing this information with a competitor, what if it's interacting.
因此，治理和安全层将变得极其重要。
So the governance security layer is gonna be super super important.
但我认为大部分价值将体现在应用程序上。
But I do think most of the value will accrue to the apps.
所以这有点像，我认为这是常识，只是我不确定具体是哪些应用。
So it's kinda, and I think that's common sense, I just don't know which apps.
我认为Glean确实很棒。
I do think Glean is amazing.
你把它看作是一个应用吗？或者我不知道，但是
Do you think of it as an app or I don't know but
现在我们把自己视为既是应用也是平台。
Now we see us as both app and a platform.
是的。
Yeah.
所以我认为这可以称为一个应用平台，我觉得它非常棒，因为它有潜力自动化组织内部的大量繁琐工作。
So I think it's a, let's call it an app platform, I do think it's amazing because it has the potential to automate so much of the overhead inside of an organization.
比如，想想为什么一些组织会有成千上万的员工，有的甚至五万、两万。
Like if you think about why do organizations have hundreds of thousands of employees, some organizations, or 50,000, 20,000.
其中很大一部分原因是协调成本——这么多人需要彼此沟通，比如‘发生了什么？你这话到底是什么意思？’
A lot of it is the coordination overhead of so many people have to communicate with each other, hey what happened, what did you exactly mean by this?
我们得开个会，让你向我解释一下，或者我问一些问题，再邀请其他人加入，然后把内容记录下来。
Let's do a meeting where you explain to me I asked some questions, or let's invite these other guys also and then write it down.
组织的协调成本巨大，对吧？
Then just the coordination overhead of organizations is massive, right?
这就像是一个N平方的问题——每个人都需要和每个人沟通，但他们只在各自孤立的组织架构内交流，那怎么跨越界限呢？
It's like this N squared problem that everybody needs to communicate with everybody and they're communicating inside their siloed org chart, but how do we get it across?
所以，我们通过文档、Excel表格、PPT和会议来推动公司和组织向前发展。
So this, through Docs and Excel sheets and PowerPoints and meetings is how we like move companies and organizations forward.
这些工作中的很大一部分都可以通过Glean得到增强并变得更加高效。
So much of that can be augmented and be made more efficient with Glean.
这就是我认为Glean很了不起的原因。
That's why I think Glean is amazing.
但这有点像是2000年的时候。
But this is kinda like 2,000.
你问互联网上的杀手级应用是什么。
And you ask what are the killer apps on the internet.
顺便说一下，那时候我们认为会是思科路由器，或者是拥有成千上万个链接的门户网站。
By the way, back then we thought it's like Cisco routers, portals maybe with thousands of links on them.
实际上，我当时刚上大学。
Actually I was like in, I just started college.
我们知道互联网的未来将是门户网站。
And we knew that the future of internet would be portals.
也就是那些包含一百个链接的网页，你只需点击正确的链接即可。
Are Which these web pages with a 100 links on it, and you just click on the right link.
这发生在谷歌搜索之前。
This was before Google search.
但互联网的未来实际上并没有朝那个方向发展。
But the future of internet actually didn't look that way.
最终变成了一些像Facebook这样的社交平台，像Airbnb这样的租房服务，像Uber这样的打车行业，以及Twitter等等。
It ended up being things like Facebook for friends, and things like Airbnb for rentals, and Uber for your cab industry, and Twitter and so on.
这些都成为了伟大的公司。
Those became great companies.
所以我不知道未来的那些应用会是什么。
So I don't know what those are for the future.
它们会突然出现。
They will pop up.
是的。
Yeah.
而且它们会变得极其有价值。
And they will be extremely valuable.
但好吧，这是否意味着Databricks和Glean最终会消亡，然后会出现一批新的公司？
But okay, does that mean that Databricks and Glean then basically will die, and there'll be a new set of companies?
不，早在1998年就已经有amazon.com了。
No, back then there was actually an amazon.com already in 'ninety eight.
实际上在1998年，谷歌就已经存在了，等等。
There was already a Google actually existed already in 'ninety eight and so on.
顺便说一句，思科至今仍然存在，它是一家价值约3000亿美元的公司，对吧？
Cisco by the way is still around and it's only a $300,000,000,000 company or something like that, right?
所以这不是非黑即白的。
So it's not binary.
我们拭目以待会发生什么。
We'll see what happens.
但我确实认为，未来涌现的应用程序将创造巨大的价值。
But I do think there's gonna be really a lot of value will go to the future apps that will emerge.
让我们深入探讨一下这一点。
Let's double click into that.
今天在这个层级，像Salesforce、ServiceNow这样的3000亿美元公司，很多人说软件已经死了。
The $300,000,000,000 companies of today at that layer, software, apps, Salesforce, ServiceNow, lot of talk about softwares being dead.
萨提亚称它们为CRUD应用。
Satya calls them the CRUD apps.
这个如今被称为软件的层级，未来会如何发展？它似乎正朝着变成数据库的方向前进。
What is the future of this layer that today is called software that seems to be heading towards becoming a database?
那么，价值会如何积累到这一层级上呢？
And what do you see the the the value accrue to to to those to these this part of the layer?
也许先从你开始，阿拉文德。
Maybe start with you, Aravind.
嗯。
Yeah.
我觉得这是一种过度简化。
I I I think that's the oversimplification.
比如，阿拉文德说Salesforce只是一个数据库，但实际上它是一个完整的生态系统，包含许多构建在该基础设施之上的工作流和其他应用。
Like, for example, Aravind to say that Salesforce is, you know, is just a database, you know, it's a full sort of ecosystem of workflows and other applications, you know, that are sort of built on top of that infrastructure.
所以，我有点不太理解这个概念，就是说，你有一个数据库，里面存着所有企业数据，然后人们可以自己在这个数据之上创建动态的用户界面体验，比如每个企业都可以自己完全独立地创建所有界面。
So I sort of, like, you know, haven't really understood this concept of that, you know, you have this, like, a, you know, database, you know, where all your enterprise data is, and then and then people can just go and create dynamic UI experiences on their own, on top of that data, on it, like, you know, every business can, for example, just create all the UI by themselves on this.
我不觉得事情会像那样发展，没错，AI 确实让你更容易构建，有了数据库，你只要跟 AI 说话，就能创建出完全符合你需求的界面和体验，但大多数时候，你其实并不知道自己真正想要什么。
I don't think, you know, it's gonna be happening like that, because, yes, you know, AI makes it easy for you to build, can have a database and you can build, you can just talk to AI and create a UI and experience that is exactly what you want it to be, but most times you actually won't know what you want.
我觉得，软件公司的真正优势在于，它们会认真思考如何将这些数据加以利用，然后以一种自然、易于交互或修改的方式呈现出来，从而真正提升人的工作效率。
Like, you know, I think a lot of like, you know, good thing about software companies is that they actually think about how to actually take that data, but then present it in a way, let you know, make people interact with it or modify it in a way which sort of is natural and which, you know, drives, you know, like more productivity from a human.
所以，我认为，从本质上讲，软件是一个端到端的完整体系，这些公司我觉得都不会消失。
So so I think ultimately, like, a software is an end to end stack in my opinion, and all of these companies, you know, I don't think they're going away.
我不觉得它们会被简化成仅仅成为一个数据库。
I don't think, you know, they're gonna relegate it to becoming a database.
在过去二十年里，人类已经对这些屏幕上瘾了。
Humans, you know, over the last twenty years, we got addicted to these screens.
我们低头盯着屏幕，用键盘和下拉菜单输入信息，嘿。
We scrunch over the screens, and we would input this information with our with our keys with the drop down and, hey.
我今天见了阿文，这就是我学到的东西。
I met Arvin today, and this is what I learned.
它应该真的是，嘿，聊天。
It should really be, hey, chat.
见了阿尔文。
Met with Arvin.
这是我学到的。
This is what I learned.
提醒我两天后跟进
Remind me in two days to catch up with
他们。
them.
对吧？
Right?
这会发生。
That will happen.
我明白了。
I understand.
我觉得这件事会发生。
Think it's gonna happen.
是的。
Yeah.
这将在未来几年内发生。
That will happen in the next couple years.
甚至，格莱恩，你也不再想打字了。
And even, Glyn, you'll you won't be wanting to type.
你想跟它说话。
You wanna talk to it.
但我认为关键是数据录入。
But I think the big thing is data entry.
数据是如何进入那个数据库的？
How does the data appear in that database?
而今天这还远未完全自动化。
And that's today not completely automated.
所以，你知道，我只是觉得，一家非常适合做这件事的公司实际上是Zoom。
So, you know, just just to like I I think a company that would be well positioned to do that would actually kinda be Zoom.
但很多人并没有这样想过。
And a lot of people don't think about it that way.
但Zoom其实应该是完美的数据录入应用，对吧？
But Zoom is really, should be the perfect data entry application, right?
因为所有对话都在那里进行，所有信息也都从那里产生。
Because that's where you're having all the conversations and that's where all the information's coming out.
如果它能与Glean协同工作，提取出最重要的信息，并将这些信息存储在系统记录中，而不是结构化的数据表里。
And if it could work with Wein and extract the most important information, store it all, not in like a structured date table, but like store that information in system of record.
如果你拥有这样的功能，那将彻底颠覆SaaS行业。
If you had that, that would be the full disruption of the SaaS
我们已经有了，Raj。
We have that, Raj.
这实际上是现在最常见的智能代理之一，你知道的，使用Glean，你可以获取这些会议录音，分析出你和客户讨论了什么、有哪些行动项，然后智能代理会自动更新Salesforce中的备注。
That's actually, is one of the most common agents these days, you know, with Glean, which is you take these meeting recordings, you figure out, like, you know, what you talked to the customer, what were the action items, and then the agent goes updates the notes in Salesforce with that.
这些事情其实已经在发生了。
Like, these these kind of things are happening already.
会议啊，是这样的，我们在Glean有一个政策，会记录每一次会议，无论是内部会议还是外部会议（如果客户允许），因为里面包含了太多太多的信息。
Meeting meetings is, yeah, is you know, like, we in in clean, we have this policy where we record every single meeting, internal meeting, external meeting if our customers allow, because there's there's so much so much information, you know, in there.
我上周参加了一个会议。
I I joined a meeting last week.
那是四个真人和六个AI记事员。
It was four humans and six AI notetakers.
是的。
Yeah.
我听说了，我觉得昨天我们还在讨论，有一个会议里有17个记事员。
I heard about I think yesterday, we were talking about 17 notetakers in one of the, you know, discussions.
这
This it
感觉就像一部电影的开场场景，AI开始接管了。
felt like the first, you know, it's like the first scene of a movie where where the AI takes over.
显然，这里存在很多分散现象。
Clearly, there's a lot of sprawl.
工具太多了，几乎多到过头了，整合正在到来，是的。
There's there's, like, only almost too many tools and and consolidation coming Yeah.
在某个时刻。
At some point.
但也许你们个人的工作流程，你们是AI时代的CEO。
But but maybe your guys' personal workflow, you know, you guys are CEOs in the age of AI.
在座的很多CIO们，他们手头有更多工作和空闲时间。
A lot of CIOs in the room, they've got more jobs and time on their hands.
你们如何将AI用于个人生活？
How are you using AI for both your personal self?
你们又如何推动你们的组织，或大型组织，采用AI并从中受益？
And how are you driving your organizations or both large organizations to to adopt AI and and benefit from it?
也许给我们看看你们在AI时代的领导风格。
Maybe give us a glimpse of your leadership in in the age of AI.
也许这次我让你先开始。
Maybe I'll leave you start with you this time.
好的。
Yeah.
我的意思是，我们为各种事情都使用了智能代理。
I mean, we have agents for all kinds of stuff that we use.
我们有一些非常擅长理解客户的代理。
Everything from, we have agents that are really good at understanding our customers.
我们有一个叫拉菲的代理，拉菲就是它的名字。
We have an agent, Rafi, Rafi's is the name which Rafi.
是的。
Yeah.
如果我想了解任何关于客户的故事，比如最棒的客户案例，我就可以直接问它。
If I wanna understand anything about any, tell me best customer story on this.
我跟你说过RBC，加拿大皇家银行，但我现在可以直接问它。
I told you about RBC, Royal Bank, Canada, but I can just ask it.
我需要一个用例，我要上台演讲，得谈谈金融行业。
I need a use case, I'm gonna get on stage, I need to talk about finance sector.
给我一个符合这些条件的用例，它会自动为你找到所有收集到的信息。
Give me a use case that has these, it'll just find you all the information collected.
所以当我像这样上台时，这类功能对我真的非常有帮助。
So it's really really helpful for me for these kind of things when I get on stage like this.
但同时，如果你去见客户，我想向客户X介绍他们的最大竞争对手Y是如何使用Databricks的。
But also customer, if you go into a customer meeting, I wanna tell customer X about their biggest competitor Y, how they're using Databricks.
也许Y并没有使用Databricks，那我就不能用Y，而应该用Z，因为Z确实正在使用Databricks。
Now maybe Y is not using Databricks, so then I shouldn't use, I should use Z, which actually is using Databricks.
也许Z是他们的第二大竞争对手。
Maybe that's like the number two competitor.
我该怎么快速获取这些信息？
How do I get this information super quickly?
所有这些在Databricks上都已准备好了。
All of those are prepared at Databricks.
所以在市场推广方面，很多工作都已完全自动化，我们正在使用它。
So on the go to market side a lot of this is being completely automated and we're using this.
我之前提到的营销工具链已经高度自动化了。
The marketing stack I already mentioned is heavily automated already.
比如营销中发生的许多任务都是如此。
Like a lot of the tasks that's happening in marketing.
所以我们正在看到这种工具链的实施。
So we're seeing that stack, that happening.
然后是工程领域。
Then there's engineering.
那是一个庞大的领域。
That's like a whole big thing.
这就是我们如何应对的，我认为这涉及整个变革管理以及如何正确实施。
That's how we sort of, and I think there's a whole change management and how to do it right.
Databricks 最初尝试自动化大量软件工程工作时，某种程度上失败了。
Initial attempts at automate a lot of the software engineering at Databricks kinda failed.
AI本身并没有问题。
Even there's nothing wrong with AI.
问题在于人类以及我们的组织方式。
The problem is the humans and how we were organized.
但这些就是两个主要的部门。
But that's, so those are like the two big orgs.
Databricks 是一个拥有六千人的市场部门和三到四千人的研发部门。
Databricks is a big 6,000 person go to market org and three, four thousand person R and D org.
然后还有一些后台工作。
Then there's some back office stuff.
对于这两个部门，我们已经看到大量使用智能代理来自动化各种任务。
Those two already we're seeing heavy automation using agents for all kinds of the tasks.
然后是后台工作。
Then there's back office.
比如财务和其他职能。
So there's finance and these functions.
财务部门全部使用Databricks，所有的预测工作都转向了基于机器学习的模式。
Finance is all on Databricks and it's all the forecasting, all the sort of, it's all moved to machine learning based.
但这个过程花了很长时间，因为他们有自己的Excel模型，而且对此非常自豪，不愿意轻易改变。
But it took them a long time because they had their Excel models and they're very proud of them and they didn't wanna, you know.
但这里同样存在变革管理的问题。
But again, there's a change management there.
我们实际上请了一个外部的数据科学团队来构建AI模型，最终这些模型变得足够优秀。
We actually had an external data science team build the AI models, and then eventually they became good enough.
现在财务部门已经接手了这些模型，基本上在Databricks内部，财务团队已经从Excel转向了Python。
And now finance has taken those over and like now finance have kind of moved from Excel to Python largely at Databricks.
但这是一段漫长的旅程，因为大多数人只会用Excel。
But it was a journey because most of us speak with Excel.
类似的情况现在也正在人力资源和其他部门发生。
Similar thing is now happening to HR and other departments as well.
但我认为，一般来说，人力资源部门并不太贴近这种用Excel进行分析的工作。
But I think they're like, I think in general HR departments are like, they're not the closest to doing this kind of analytical work with Excel and so on.
所以也许这方面进展没那么快，但没错，我们在各个地方都看到了这种趋势。
So maybe that's not quite as far along, but yes, we're seeing it everywhere.
是的，Arvind，你还有什么要补充的吗？
Yeah, anything to add Arvind?
我们这边也一样。
Same for us.
我觉得我可以分享一些我个人的看法，我们有一个每日准备代理，我非常喜欢。
And I think I can share some of my own personal views with with it, so I I so one of our agents is daily prep agent, which I really love Yeah.
因为每天早上它都会告诉我，我的一天会怎样，我需要阅读什么、准备什么。
Because, you know, every morning it tells me, like, you know, my day is gonna be, what I need to read, what I need to prepare.
大多数会议，你都知道，我都没有背景信息。
Like most of the meetings, you know, I will not have context.
它实际上会为我整理好这些会议的计划。
It actually brings, you know, like the plan for those meetings for me.
所以这是我最喜欢的代理之一，它让我感觉更有信心，知道我当天的会议该怎么进行。
So that's that's one of my favorite agents, you know, that helps me feel more confident, like, you know, for, like, how I'm gonna do my meetings in the day.
另一个，我昨天也提到过，我已经改变了我的本能，我认为改变本能需要很长时间。
The other one, which I which I shared yesterday also, the like, you know, I've I've changed my instinct, and I think, you know, changing changing instincts, you know, take take take take long time.
是的。
Yeah.
当你担任首席执行官时，你是老板，大家都听你的，你可以随时带着一个小问题，好奇地去问别人。
And, you know, when when you're the CEO, like, you're the boss and everybody listens to you and you can just, like, you know, whenever you have, you know, a small question, curiously, just go and ask somebody.
他们会立刻安排三十个人来帮你找到那个答案。
And they're gonna, you know, put 30 people on the task to actually get that get that answer from me.
他们会
This is They're gonna
在准备会议之前先开个准备会议。
have a prep meeting before the prep meeting.
是的。
Yeah.
所有这些都包括在内。
So it's all of that.
所以，这对我来说其实挺简单的。
So so that so that's sort of like, you know, and and so but it's sort of like, for me, it was easy.
我只是去问别人，而我知道，我之前的行为实际上造成了很多浪费，成本非常高。
I just get to ask somebody and and that, you know, I changed that because I knew I was actually causing, like, you know, a lot of those that was very expensive.
所以现在，每当我产生好奇、有疑问、需要做数据分析或写东西时——比如每月写给公司的那封信——所有这些事情，我从根本上都会使用AI，当然这里指的是Glean，来帮助我完成任务。
So the so today, like, you know, my my instinct is to whenever have curiosity, whenever I have questions, when I need to do data analysis, when I need to write something, you know, my my letter to the company every month, all of those things, you know, like, fundamentally, I use, you know, AI, of course, know, Glean in this case, but to actually help me do my tasks.
是的。
Yeah.
越来越多地，你必须持有这种信念。
More and more, I think the, you have to, you have to sort of have that belief.
很多人并不这么做。
A lot of people don't do it.
你必须相信，AI是一个优秀的合作者。
You you have to have that belief that AI is a good collaborator.
它不会替你完成工作，但如果你善用它，最终你产出的结果会更好。
It's not gonna do the work for you, but if you use it, you're gonna actually produce better output eventually.
即使你前几个月并没有节省时间，但你的输出质量实际上会提高。
Even if you don't save time, you know, for the first, you know, first few months, but you're actually gonna improve the quality of your output.
很有趣。
Fascinating.
这让我想到了我最喜欢的部分——快速问答。
Well, this brings me to my favorite part of this conversation, which is rapid fire.
简短的回答就可以。
Short answers are fine.
长回答也可以。
Long answers are welcome.
从现在起十二个月后，我们今天熟知的大型AI公司是上涨还是下跌？
Start with twelve months from now, are the big AI companies that we know of today up or down?
我们先从OpenAI开始，十二个月后，它的股价是涨还是跌？
We'll start with OpenAI 12 from now, stocks up or down.
然后是阿里和阿文德。
Ali and then Arvind.
上涨。
Up.
我会说收入也会上涨。
And I'll say revenue will be up.
我不太明白股票是怎么运作的。
I don't really understand how stocks work.
混乱的。
Entropic.
阿里，阿尔文。
Ali, Arvind.
上涨。
Up.
一样。
Same.
是的。
Yeah.
好的。
Okay.
我们让我
We me let
让我拉
me pull
我给你拿点课程吗？
can I get you a Of little bit course?
因为ChatGPT会继续增长，它正火得不得了，人人都在用。
Because ChatGPT is gonna continue growing and it's on fire and it's what everybody uses.
嗯。
Mhmm.
顺便说一下，Gemini也是。
So is Gemini, by the way.
还有Anthropic，因为越来越多地，你知道，编程领域，我们才刚刚触及到这个市场的一小部分。
And then Anthropic because more and more, you know, coding, we've only, like, eaten into a small portion of that market.
他们刚起步。
They just started.
所以
So
是的。
Yeah.
AI处于泡沫中吗？
Is AI in a bubble?
是或不是？
Yes or no?
确实存在AI泡沫。
There is an AI bubble.
就像说好吧。
Like saying like okay.
那么精益也处于泡沫中。
So then Lean is also in the bubble.
每个人都身处泡沫中。
Everybody's in the bubble.
不。
No.
我会说确实存在一个泡沫。
I I would say there is a bubble.
我会说那三个派别。
I I would say those three camps.
嗯。
Yeah.
有一个超级智能追求派别。
There is a super intelligence quest camp.
嗯。
Mhmm.
我会对此非常担忧。
I would be very worried there.
第二个是研究人员，他们做的是你知道的，那肯定不在泡沫里。
There's a second, the researcher's doing the you know, that's definitely not in a bubble.
他们就像是
They're like the
他们很清醒。
They're sober.
是的。
Yeah.
他们非常清醒，但没人关注他们。
They're they're super sober and nobody cares about them.
然后呢？
Then there's right?
而他们可能是对的，不幸的是。
And they're probably the ones that are right, unfortunately.
然后是第三个群体，就是我们努力让这件事变得有价值。
And then there's the third camp, which is us trying to make this valuable.
我们并不处于一个泡沫中，因为我们没有在我们的工作中投入巨额资本。
We're not in a bubble in a sense that we're not spending huge amounts of capital on what we are doing.
是的。
Yeah.
我们只是试图在这个组织内部创造真正的经济价值。
We're just trying to get actual economic value inside of this organization.
所以我不认为这是非黑即白的，但确实存在一个泡沫。
So I don't think it's binary, but there is a bubble.
我的意思是，有些初创公司没有任何收入，估值却高达一千亿、两千亿、三千亿美元。
I mean, there are startups with zero revenue worth, you know, $10.20, 30,000,000,000.
这就是泡沫。
That's a bubble.
是的。
Yeah.
一样。
Same.
我的意思是，有不少公司存在过度乐观的情况，它们的估值远远超过了实际业务水平。
I mean, I think the, there are quite a few companies where there's poor optimism and valuations which are well ahead of the business that those companies have.
而且，我想可以说，与非AI公司相比，AI公司的估值倍数确实更高更高，但我认为这背后是有合理原因的，因为这些AI公司肯定比非AI公司增长得更快。
And like, I guess you can say, like, you know, compared to non AI companies, like, of course, AI companies do have higher higher higher multiples, but I think, you know, this sort of comes from that, you know, that there's a good reason for it, you know, because, you know, these are, these AI companies are gonna grow more than non AI companies for sure.
是的。
Yeah.
我在‘终极我们问CEO’节目中最喜欢的游戏是多空游戏：如果你要选一家公司、一个产品或一个想法，认为它未来会比现在更重要，那会是什么？
My favorite game at Ultimate We Ask Our CEOs is a long short game, is if you were to pick a company, a product, an idea that you're long, that you think is gonna be a bigger deal than it is today, what is that?
然后是空方，也就是那些表面光鲜但实质不足、炒作多于现实的公司。
And then short, which is, you you know, there's more sizzle than their steak, more more hype than reality.
选一个你非常看好的方向。
Pick along, something that you're very optimistic on.
顺序一样，先阿里，然后是阿尔文。
Same order, Ali and then Arvind.
我非常看好智能代理。
I am very long on agents.
是的。
Yeah.
你知道，我非常看好语音交互。
You know, I think I'm very long on speech Speech.
作为一种交互方式。
As an interaction.
我觉得键盘基本上会完全消失。
Like I think keyboards are kind of basically gonna disappear completely.
我们实际上还没有真正攻克语音技术。
We haven't actually nailed speech.
我知道感觉上好像已经做到了，但其实还没有，因为你还在用键盘。
I know it feels like we have, but we haven't because you're still using your keyboard.
所以只要你还在用键盘，我们就还没真正攻克语音技术。
So as long as you're using your keyboard, we haven't nailed speech.
但我认为我们离完全淘汰键盘已经近在咫尺了。
But I think we're this close to completely eliminating keyboards.
我觉得这是其中一个大问题。
I think that's that's a big one.
我该说什么呢？
What's what would I say?
我觉得编程有点被高估了。
It's like, you know, I do think coding is a little bit overhyped.
我不知道我是否会看空它。
I don't know if I would short it.
我的意思是，我认为它仍然是未来。
It's I mean, I think it's still the future.
所以我觉得这是其中之一。
So I think that's that's one of them.
我觉得自动化客户服务和支持有点被高估了。
I think automating our customer service and support is a little bit overhyped.
所以，你知道，我认为业界认为那些了不起的事情，我们确实取得了很大进展。
So, you know, I basically I think the things that the industry thinks are like amazing and we've made great progress.
我们在这方面的进展可能没那么多。
We probably haven't done as much progress on it.
然后还有很多其他被忽视的事情，你知道，我们会在那些领域取得突破。
And then a lot of the other things that are being ignored, you know, we're gonna have breakthroughs in those.
很有趣。
Fascinating.
是的，
Yeah,
对我来说，我认为会改变格局的产品是这样的：不再是你开发一个产品然后期待用户来找你，而是如果你能深刻理解你的用户和客户，并真正将AI带给他们，这才是我感兴趣的领域。
and for me, I think the products that are going to change the paradigm where instead of you building a product and expecting people to come to you, if you understand your user, your customer very deeply and actually bring AI to them, that's the category that I'm excited about.
我希望明年能看到更多主动型的AI产品进入市场。
I want to see more proactive proactive AI products coming coming into the market next year.
是的。
Yeah.
那才是真正能让用户从5%的高级用户扩展到100%的关键。
That that's when that's that is what is going to actually take it from a 5% of the users being power users to a 100%.
嗯。
Yeah.
嗯。
Yeah.
嗯。
Yeah.
你生活中最喜欢的AI工具是什么？
Your favorite AI tool that you use in your lives?
我觉得Dune太棒了。
I think Dune is awesome.
我的意思是，如果这不公平的话
I mean, if that was not fair
我们开始吧。
Let's Let's go.
所以我经常用它。
So I use it all the time.
实际上，我向团队提出的很多问题，关于你提到的那些改动，我首先会问Glyn，然后看看它是否准确解决了问题。
I actually a lot of the questions I would ask from the team, the the thing you said you changed, I I I first ask Glyn and then see, you know, if it nails it or not.
如果没解决好，我就会组建一个30人的团队，花一周时间开预备会议，等等，只为让我理解某个简单概念的解释。
Then if it doesn't, then I'll spin up a 30 person team to go spend a week and have pre meetings and all that to get, you know, the explanation of some simple concept for me.
但通常来说，它都能准确解决。
But usually, nails it.
是的。
Yeah.
是的。
Yeah.
对我来说，我对笔记工具感到兴奋。
For for me, I'm excited about note takers.
我自己用过Granolah、Fathom和其他一些工具。
I've I've used Granolah myself and Fathom and a few others.
是的。
Yeah.
但做笔记实际上非常有趣。
But note taking is actually fascinating.
我的意思是，我觉得如果你做了这些笔记，然后以正确的方式加以利用，比如像阿里所说的，这将成为真正创造知识、将数据保存在你们系统中的来源，这将改变公司的运作方式。
I mean, I think the I I feel like, you know, if you if you take those notes and then if you utilize it the right way, like for example, what Ali was saying, like, you know, that becomes a source of what then actually creates knowledge, saves data in your systems, that's gonna change, you know, how companies work.
是的。
Yeah.
最后，我很想听听你们对公司未来的愿景。
You know, in closing, I'd love to get your vision for your companies.
我们先从阿里最喜爱的工具Glean说起。
We'll start with Ali's favorite tool, Glean.
恭喜你。
Congrats.
你们刚刚宣布达到了一个重要里程碑，年收入达到2亿美元。
You just announced crossing a big milestone, $200,000,000 in revenue run rate.
你们正在签下大单，单笔交易高达1000万美元。
You've you're signing big deals, $10,000,000 deals.
你拥有超级用户。
You've got super users.
我看到你也在接触普通用户。
I'm seeing you're seeing casual users.
描绘一下Glean从现在到实现十亿美元收入的愿景吧。
Paint us the vision for for Glean from here to a billion in revenue.
我认为他们仍在进行年度规划，一些AI公司告诉我这太老派了。
I think they're still doing annual planning, which also some, you know, AI companies are telling me that's that's so old school.
但我们还是在做。
But we're doing it regardless.
就是在做。
Doing it.
这是因为早期的初创公司。
That's just because there's early startups.
当你刚开始做Glean的时候，你做过年度规划吗？
Like, did you do annual planning when you started clean?
没有。
No.
没有。
No.
但对我来说，最让我兴奋的是，我们非常关注人工智能素养，以及如何让每个人都能跟上这趟旅程。
So the, but I think for us, thing that I'm most excited about, again, is, so we think a lot about AI literacy and how do you get everybody along on this journey.
但目前我们还没有看到这种现象。
And we're not seeing it right now.
Glean 是一个被广泛使用的工具，但顶级用户和底层用户之间仍存在很大差异。
Like Glean is a heavily used product, but still there's a big variance between the top users and the ones, you know, at the bottom.
这正是我们想要改变的。
And that's what we want to change.
因此，我们未来的愿景是让 Glean 成为全球每个公司中每个人专属的个人伙伴。
So for the future for us is we want to be, we want Glean to be this very personal companion for every person in every company in the world.
这个伙伴与你有着非常私密的关系——无论你向它提出什么问题，或与它进行何种交流，都完全属于保密范畴。
This this companion with which, you know, is is is, you know, you have a very confidential relationship with this companion in the sense that whatever you ask this companion, you know, whatever communication you have with them, you know, it's it's fully privileged.
没有人能看到它。
Nobody else gets to see it.
但这个伙伴了解你和你工作生活的方方面面。
But this companion knows everything about you and your work life.
它了解你的一天。
It knows your day.
它了解你的一周。
It knows your week.
它知道你每天会见谁。
It knows who are you gonna meet, you know, in the day to day.
它了解你的周目标。
It knows your weekly goals.
它知道你擅长什么、不擅长什么，以及你的职业抱负。
It knows what, you know, what things you're not good at or what your career ambitions are.
有了这些信息，这个私人伙伴现在正帮助你更好地工作。
And with all of that, you know, this this personal companion is is sort of helping you now with your work.
它希望能自动处理你大部分任务，甚至在你开口之前就主动完成这些工作。
It, you know, hopefully takes majority of your tasks automatically, you know, works on them before you ask it to work on them.
这正是我们正在努力实现的愿景。
And that's sort of the vision that we are, you know, taking our part to.
我们已经基本搭建好了这一系统的底层基础。
We we have most of the found you know, foundation for this in place already.
如今，你必须亲自来到Glean才能完成大部分这些工作。
Today, you have to come to Glean to get most of that work done.
未来，我们希望Glean能主动找到你，并替你完成这些工作。
In the future, we want Glean to actually come to you and do that work.
太有趣了。
Fascinating.
我们可以再聊一会儿，但Andi已经准时到了。
Well, we can keep going for a bit, but Andi called on time.
非常感谢你抽出时间和我们交流。
Thank you so much for chopping it up with us.
你们这里有很多前沿内容和深刻见解。
You got a lot of alpha, a lot of insights here.
非常感谢。
Really appreciate it.
谢谢。
Thank you.
谢谢。
Thank you.
谢谢。
Thank you.
谢谢
Thank
谢谢。
Thank you.
各位。
Gentlemen.
非常感谢。
Thank you so much.
关于 Bayt 播客
Bayt 提供中文+原文双语音频和字幕，帮助你打破语言障碍，轻松听懂全球优质播客。