AI技能差距是真实的：谷歌副总裁谈人们为何落后及当下该怎么做 | 约西·马蒂亚斯

2026年发生了一些变化，但大多数人还没真正反应过来。

Something shifted in 2026 that most people haven't really sat with yet.

现在有三成六的新公司，都是一个人单干起来的。

36% of new companies are now solo founded.

就一个人。

One person.

没有联合创始人。

No co founder.

没有团队。

No team.

可这样的公司，居然在做以前得十二个人才干得动的事？

And this company is building things that used to require 12 people?

甚至一百个人才干得动的事？

A 100 people?

五年前，这个比例才百分之二十三。

Five years ago, that number was 23%.

过去两年里，对分析和技术能力的需求上涨了20%。

Demand for analytical and technical skills is up 20% in two years.

真正动手搞AI的人，而不是光会用的人，薪资和适应这个世界的速度都明显领先了。

And people who actually work with AI, not just use it, are pulling ahead visibly in terms of salary, in terms of how they're catching up with the world.

我想搞清楚这背后到底发生了什么，好让你能抓住这些趋势，要么换个新职业，要么创业，至少也能把生活优化一下。

And I wanted to understand what's happening so you can jump on these trends and maybe build a new career or build a new business or at least optimize your life.

如果你生活中有什么事让你觉得‘要是能用AI改一改就好了’，那你真得好好看看这变化了。

So if you have things in your life that you wish, AI is changing how we do things, and you need to understand this.

这期节目里，我还会采访一位我很佩服的人——谷歌研究负责人约西·马蒂亚斯。

In this episode, I will also talk to someone who represents a company I admire, Yossi Matias, head of Google Research.

他开发了谷歌趋势，我天天都在用。

He built Google Trends, something that I use all the time.

他还做了搜索自动补全功能。

He built Autocomplete.

他在谷歌干了二十多年，而谷歌正是这场变革的领头羊之一。

He's been at Google for more than twenty years, and Google is one of the top companies leading this revolution.

他很少接受访谈，所以我很高兴我们能做成这次对话。

He doesn't do many interviews, so I'm really glad we did this one.

咱们先说第一个趋势：AI代理。

Let's start with trend number one, AI agents.

大多数人一提到AI，就想的是在聊天机器人里问问题。

When most people think about AI, they think about asking for information in a chatbot.

你问一句，它答一句。

You ask something and answers.

但现在真正发生的事，其实不一样。

What's actually happening right now is different.

现在的代理能替你做事，你把任务交给它，它会自动处理你邮箱、日历、研究、客户管理系统里的一堆事，你根本不用动手。

Agents execute for you and you hand off tasks across your entire system, your email, your calendar, your research, your CRM, and you don't have to touch it.

比如，我刚 setup 了一个代理。

For example, I just set up an agent.

用 Perplexity Computer 做这个其实特别简单。

It's really easy to do that with Perplexity Computer, for example.

它把所有事情都自动化得干干净净。

It just automates the hell out of everything.

我平时关注了几个Instagram账号，想看看他们那儿啥内容火了。

So I'm keeping track of a couple of people on Instagram and I wanna know what's going viral for them.

Perplexity就自动去分析哪些内容在爆，每天早上我都会收到一封邮件，告诉我他们账号的动态、网友的反应，还给我一个类似主题的视频脚本建议。

So Perplexity just goes, analyzes what's going viral, and every morning I get an email, what's going on in their accounts, how people are reacting, and I get a personalized script on a similar topic so I can make a video.

这简直太离谱了。

And it's crazy.

斯坦福的研究显示，目前35%的效率提升，都来自这种能感知上下文的智能代理。

Stanford tracked that 35% of productivity gains is happening right now from using context aware agents.

这直接决定了你是过四天班还是六天班。

That's the difference between a four day week and a six day week.

对你来说，这意味着，会 setup 和管理智能代理的人，不只是快一点。

What this means for your job is that the person who learns to set up and run an agent isn't just faster.

他们一个人干的活，以前得两个人甚至三个人才能完成。

They are doing work that used to require two or three people.

第二个趋势： vibe 编程。

Trend number two, vibe coding.

这个我 personally 觉得最让人兴奋，说实话有点疯狂。

This one I personally find the most exciting and honestly a little wild.

Vibe 编程就是用大白话描述你想做什么，然后让 AI 直接给你写代码。

Vibe coding means describing in plain language what you wanna build and having AI write the actual code.

我团队里有个人根本不会编程。

I have a person on my team who doesn't know how to code.

她负责我的领英账号，上周二她有个想法。

She runs my LinkedIn and she had an idea on Tuesday.

到周四，东西就跑起来了。

On Thursday, it was a working product.

而且她真的完全不会写代码。

And again, she doesn't know how to code at all.

谷歌的约西团队在搞一个叫生成式界面的东西，你只要说清楚想要什么，几分钟内就能生成一个带按钮、逻辑和界面的完整交互应用。

Yossi's team at Google has been working on something they call generative UI where you describe what you want and in about a minute you get a fully interactive application button, logic, interface all in a matter of a few minutes.

所以如果你一直想做个东西——不管是工具、产品还是工作流程，或者你有个想法，别光跟同事嘴上说说，直接动手做出来就行。

Again, what this means for you if you've been waiting to build something a tool, a product, or a workflow, or if you have an idea, instead of just talking about this idea to your teammates, just go and build it.

这个障碍已经没了。

This barrier is gone.

你跟应用说一声，就能把原型给‘聊’出来。

You talk to the app and you pitch a prototype.

我觉得这些技术被严重低估了。

I think they're underhyped.

是吗？你也这么觉得？

Oh, you think so?

对啊。

Yeah.

明白了。

Okay.

这可是我的看法。

And here's mine.

首先，我们得明白，今天看到的这些并不是未来。

So first, we should always keep in mind that what we're seeing today is not the future.

我们眼下看到的，只是现在。

We're just seeing today.

我干搜索这行十多年了，以前我们总是琢磨用户到底想干嘛，然后尽量给最准的信息。

And, you know, having been working for over a decade on search, in search, we suddenly take the user intent and we try to give the best information.

现在有了AI，我们能更懂用户真正想要什么，还能让他们直接说出自己的需求。

Now with AI systems, we can actually understand much better the user intent and let them actually express what we do.

比如 vibe coding，现在大家能直接口述想开发啥，一个人就能做出过去得靠一整个团队才能搞定的应用。

Now, vibe coding is a great example where people can now actually say what they want to be developed, and you can already everybody can now develop an application that previously required a team.

我们最近还搞了个叫生成式界面的东西——你随便提个需求，它立马给你出个交互界面。在 Gemini 里有个实验功能叫 Animate View，你丢个问题进去，一分钟内就能生成一个完整可操作的界面，就像有人在替你写程序一样。

Even we recently shared something that we call generative UI that enables you to actually, for any prompt, get a result out of we have an experiment within Gemini app called Animate View that gives the full interaction full interactive user interface in about a minute for any prompt that you ask, as if somebody was developing an application on your behalf.

你只要说清楚你想学啥，它甚至能自动把按钮、模拟操作都给你配齐。

So only you need to say what you'd like to learn about, and it will sometime even give all the buttons and the simulation.

而且现在，这技术也已经用到搜索的 AI 模式里了。

And in fact, the technology is now available also in Search AI mode.

第三个趋势：真正重要的技能正在发生变化。

Trend number three: The skills that actually matter are shifting.

大多数人第一反应都是：AI要取代工作了，我得变得更懂技术才能活下去，赶紧学编程，拿个数据科学证书。

The reflex most people have is AI is replacing jobs, so I need to become more technical to survive, learn to code, get a data science certificate.

这没错，但不全面。

That's not wrong, but it's incomplete.

约西一直在谷歌研究部招人，那是现在最让人兴奋的地方之一，我就直接问他：你现在最想找什么样的人？

Yossi hires constantly at Google Research, one of the most exciting places to be right now, and asked him directly: What are you actually looking for right now?

答案不是某个具体技能，而是能比技术变化更快地思考、适应和学习的能力。

And the answer wasn't a specific skill, it was the ability to think, adapt, and learn faster than technology changes.

他说，哪怕到了资深级别，人们也得重新学怎么工作，因为工具每个月都在变。

He said people need to relearn how to work even at senior levels because the tools shift every single month.

但说白了，真正跑在前面的，不是懂得多的人，而是知道该把什么交给AI、当AI反馈回来时又该问什么问题的人。

But what that actually means in practice, the people pulling ahead aren't the ones who know more, they are the ones who know what to give to AI and what questions to ask when it gives back.

所以，如果你真得挑一件事来重点提升，那就是你的判断力。

So if there is one thing you should be working on, it's actually your judgment.

你怎么知道自己工作里什么好、什么不好？

How do you know what is good and what is bad in your job?

这种判断力是可以学的。

That judgment is learnable.

你可以找一个品味超棒的上司，从他身上学，但几乎没人有意识地在练这个，而恰恰就是这个，能让你在找工作时脱颖而出。

You can work for somebody who has perfect taste and learn it from them, but almost nobody is deliberately practicing it, but that is exactly what's gonna give you a premium when you're looking for a job.

我招人的时候，特别看重有没有品位，能不能做战略判断，能不能判断出什么内容才叫牛，技术能力反倒是其次——因为Claude什么都知道，但它做不出人那样精准的判断。

I try to hire people who have taste, who can make strategic decisions, who know what's great for my content, and I care less about the technical skills because again, Claude is someone who knows everything about what's going on in the world, but it can't make a judgment as good as a person.

对。

Yeah.

说起来，我记得有人问过我：你觉得五年后会变成什么样？

So in a way, I remember I was asked, so how do you think things are going to change five years?

我认真想了好一阵。

And and I I thought about it.

一方面，一切都会不一样。

In one hand, everything is going to be different.

但另一方面，其实啥都没变。

On the other hand, nothing is going to be different.

我确实在达沃斯听过这种说法。

I've heard that at Davos, specifically.

说到底，我们毕竟是人，对吧？

In in a way, I mean, we're humans, right?

那我们做这些事的动力到底是什么？

And what's motivation for what we're doing?

并不是为了完成某个预先设定好的任务。

It's not really to fulfill a particular predetermined task.

我们的目标和任务，都是自己定的。

We define our own ambitions, our own tasks.

学习也是同样的道理。

Same with learning.

我还记得刚那会儿，谷歌突然对所有人开放了，大家都能查到信息，那时候我就想：这对孩子来说到底意味着什么？

I remember the early days when suddenly Google was made possible for everybody to get facts and to decide, wait a minute, what's going to do for kids?

因为我们以前让他们做作业、去图书馆查资料，现在这些都太容易了。

Because we ask them to do homework and collect facts in library, and now it's easy.

他们会因此变懒吗？

Are they going to be lazy?

不会的。

Well, no.

现在这些工具已经是理所当然的了，就跟铅笔一样。

Because now this is a given, that's a tool.

所以我们现在期待他们能更进一步。

So now we expect them actually to go to the next level.

我们希望他们能整合信息、融会贯通。

We're expecting them to synthesize.

有了AI之后，自然又有了新问题：它到底会带来什么改变？

Now with the AI, of course, there's another conversation, what is it going to do?

我的预测是，我们只会不断提高对他们的期待。

And my prediction is that, in fact, we're just going to uplevel what we expect.

所以我才觉得，AI是人类创造力的放大器。

That's why I'm thinking about AI as an amplifier for human ingenuity.

那我们接下来该怎么用它呢？

Now, what are we going to do with that?

这又回到了动机的问题上。

That goes back to the motivation.

我们到底为什么要这么做？

Why are we doing what we're doing?

这么说，我们的生活方式真要变了？

So in a way, our lifestyle is going to change?

肯定的。

Definitely.

但那些根本的东西，估计根本不会变。

But the basics are probably not going to change at all.

对。

Yeah.

其实就是人和人一起合作，解决那些让他们热血沸腾的重要问题。

Which is about people working with other people, about solving important problems that they're excited about.

对。

Yeah.

作为一直都在招人的你，能聊聊你最看重哪些技能吗？不管是技术的还是非技术的？

And as a person who's constantly hiring, can you talk about the skills that you're looking for, both technical and nontechnical?

我一直觉得最关键的是思考能力、适应能力、进化能力——能真正去分析问题，然后想办法解决它。

One thing that I always thought is critical is the ability to think, the ability to adapt, the ability to evolve, the ability to actually think about problems and then try to solve them.

现在这些能力比以往任何时候都更重要了，毕竟技术发展得太快了。

Now, today, these are more important than ever, of course, because, you know, technology is moving fast.

不管经验多丰富，人都得不断调整自己的学习方式。

People need to adjust their learning even no matter how experienced they are.

新技术层出不穷。

There are new technologies.

工程师得重新学怎么用AI来提升效率，而很多人已经在这么做了。

Engineers need to relearn, you know, how to use AI in order to be more productive, and people are doing it.

所以，适应能力、学习能力，还有扎实的基础，现在比以往任何时候都更重要。

So the ability to adapt, to learn, to have a strong foundation is more important than ever.

比如我说AI是人类创造力的放大器，能让每个研究者都用它来提出自己的问题，但你也得学会怎么用、怎么适应。

So for example, when I refer to AI as an amplifier of human ingenuity and the opportunity for every researcher to be able to use AI in order to actually ask your questions, it's important to also adapt and learn how to do that.

对。

Mhmm.

你懂的吧？

You know?

如果我们真想让普通人也能胜任那些原本只有资深人士才能干的活儿，那他们就得学得更快。

If we really want people to be able to do the kind of roles that today we expect many more senior people who are much more senior to do, obviously, we need them to learn how to do that much quickly.

好消息是，我觉得AI也能帮上忙，比如当导师、给反馈。

The good news is that I believe that AI can help us with that as well, with kind of tutoring, with feedback.

就拿写论文来说，AI现在能给你以前非得靠导师盯着才能得到的反馈。

The fact that to write a paper, for example, AI can give you very the kind of feedback that in the past you actually needed the attention of your adviser perhaps to do.

哦。

Oh,

是的

yeah.

所以我觉得，这些方面都会慢慢发生改变。

So so I think we are we things are going to adapt in all of these dimensions.

而且这么多年下来，我们人类证明了自己特别能适应。

And I think humans are what we proved over the years is that we're extremely adaptive.

适应力强。

Adaptive.

第四个趋势：AI正在变得隐形。

Trend number four: AI is becoming invisible.

约西管这叫无处不在的智能。

Yossi called it ambient intelligence.

意思是，技术越强大，你越感觉不到它的存在。

The idea is that technology becomes powerful precisely when you stop noticing it.

想想谷歌翻译。

Think about Google Translate.

你上一次想到它是怎么运作的，是什么时候？

When did you last think about how it works?

你就是直接用呗。

You just use it.

还有自动补全，Yossi 就是做这个的。

Or Autocomplete, Yossi built that.

现在没人会去想自动补全这回事了。

Nobody thinks about Autocomplete anymore.

大家觉得这本来就是理所当然的。

It's just expected.

AI 工具现在就是这个趋势。

That's what's happening with AI tools right now.

这对你的意义是，比如作为雇主，我已经默认你能做出精美的演示文稿了。

What this means for you is that as an employer, for example, I already expect beautiful presentations.

我也默认你会对业务现状做非常深入的分析。

I already expect that you do a very, very deep dive into what's happening in the business.

这些精美的文档都是必须的，因为用AI生成它们也就几分钟的事。

Like all of these beautiful documents, they are a must because it takes a few minutes to generate them with AI.

项目里当然有深度分析，但说到底，现在大家觉得漂亮的报告、深入的数据分析都理所当然，毕竟做起来太简单了。

There's deep intelligence on the project, but again, it brings me back to this idea that beautiful reports, deep analytics and everything is already expected just because it's so easy to do that.

我们真正愿意付高价的，是那些创意决策，对吧？

What we're paying premium for is creative decisions, right?

比如开电话会，跟团队讲清楚我们接下来要往哪个方向走。

Being on a call and explaining to the team the trajectory we're moving into.

这种事，AI可没法随便生成出来。

And that is something you can't just generate with AI.

我一直以来特别关注的一个概念，叫‘无处不在的智能’——就是那些你天天用、根本不用想的技术。

So one notion that I've been quite passionate about for quite some time is what I would call ambient intelligence, which is that you have technologies that you just use, you don't think about them.

它们变得太普及、太顺手了，你压根就默认它们能正常工作，对吧？

They're becoming so available and so intuitive that you actually don't you just assume they work, right?

就像自动补全，这些年我和团队在搜索领域有幸参与了它的开发。

Think about Autocomplete, which I also had the privilege to develop with my team over the years in search.

人们一打字，就觉得系统会自动给你推荐内容。

People just assume that you start typing and it will just suggest to you.

我记得早期有人还惊叹：这玩意儿怎么这么神奇？

And I remember that people were in the early days, oh, what about this magic?

哇，它怎么猜得这么准？

Wow, how does it make a guess?

可现在呢？大家觉得这本来就是理所当然的。

And today you just expect it to be the case, right?

对。

Mhmm.

类似的，你还能想到好多其他技术。

Similarly, you can think about so many other technologies.

比如语音技术，现在你张嘴说，就默认系统能听懂你说啥。

Think about voice technology, the fact that now you can speak and, you know, you expect what you're speaking to be understood.

你可以把文字转成语音听，也能把语音转成文字看。

You can take text, you can listen to it.

我还记得早年亲自参与这些技术开发的时候，那会儿这可是我们的终极目标。

I remember actually working on these technologies in the early days, this was the aspiration.

对。

Yeah.

可现在呢，你理所当然觉得就该是这样。

And now you just assume this is the case.

想想多语言翻译吧。

Think about multiple languages.

我依然记得有一天，我孩子从学校回来，跟我说：‘爸，你那个翻译怎么搞的？’

I still remember the day that one of my kids came from school and say, hey, dad, what's going on with your translate?

那句话翻得真不怎么样。

That line, that sentence was not translated very well.

简直翻得太烂了。

It's actually quite, bad translation.

我当时就想，几年前，能自动翻译整页内容还跟科幻片似的，可他却觉得这玩意儿本来就应该有。

And I was thinking to myself, you know, just a few years ago, having automatic translation of a page was a science fiction, and he assumes this is just available.

第五个趋势：AI正在重塑教育。

Trend number five: AI is rebuilding education.

如果你还没用过Google的Notebook LM，我真强烈推荐，特别适合教育场景。

Oh, I absolutely love Google's Notebook LM if you haven't used it for education yet.

你只要上传一堆文件，然后让它帮你改写——比如，给一个爱踢足球的10岁孩子重新讲一遍，用自由球来解释重力，或者做成播客，或者生成一张能发在LinkedIn上的信息图，这些图真的特别火。

It's basically you upload a bunch of files and then you ask it And now imagine it re leveled for a 10 year old who loves soccer, or gravity is explained using a free kick, or turned into a podcast, or an infographic that you can then post on LinkedIn, and those infographics do really well.

你甚至能直接做成视频，发到YouTube上。

Or you can make a whole video out of it and post it on YouTube.

这不光是解释知识，还能顺便产出大量社交媒体内容。

Like that's explanation for you but also a lot of social media content.

那种‘一本教材、一个难度、所有人一样’的老模式，都两百多年了。

The model where you had one textbook, one level, same for everyone is two hundred years old.

AI正在打破它。

AI is breaking it.

说到这个，我特别会想到我两个女儿。

And I think a lot about my daughters on this one.

从小就开始用个性化AI家教的孩子，到18岁的时候，基础会跟没用过的人完全不同。

Kids who grow up with personalized AI tutors from age five are going to arrive at 18 with a completely different foundation than kids who didn't.

这可是十年的领先优势。

That's a ten year advantage.

所以你现在要转行或学新技能，千万别错过这些高级工具。

So if you are reskilling now, make sure you use all of the advanced tools.

教育啊，这可是他们特别重视、我觉得也特别关键的一个领域，正等着被彻底重塑。

You know, education, which is one of theirs and I think is highly important, is a way that is really poised to be transformed.

我们最近就在想，能不能重新设计教科书？

And and we have recent experiment of asking ourselves, can we reimagine the textbook?

与其原封不动地用课本，不如用AI把它变成个性化的、贴合场景的学习体验？

Instead, can we take a textbook and use AI in order to actually give it in different experiences that are gonna be personalized and contextualized?

比如，能不能把文字变成沉浸式的体验？

For example, can I have an immersive Can I take the text and make it immersive?

想想《哈利·波特》那种感觉。

Think Harry Potter.

我能把它变成对话式的吗？

Can I make it conversational?

能配上一个速写本吗？

Can I have a sketchbook with that?

还有，能不能根据受众水平来调整内容？

A part of And by the way, can I have it in a level that is suitable for the audience?

比如，能不能用足球的例子，让一个十岁喜欢踢球的孩子理解重力？让教材的语言更贴近他们的表达方式？

So, for example, can I explain gravity to a 10 year old who likes soccer so that the textbook can actually be releveled to 10 year old kind of language and give examples from soccer?

答案是肯定的。

The answer is yes.

我们其实已经做了一些实验。

We actually have some experiment.

现在还只是初期，但我相信未来会变得特别自然流畅。

These are early days, so I expect that in the future, it's going to be seamless.

到时候，这些功能会像空气一样随手可用。

It's going to be kind of just available to us.

作为妈妈，我现在看到的趋势是，孩子一上学就得会认字了，对吧？

The trend that I'm seeing now as a mom, though, is that now kids are expected to know how to read when they start school, right?

以前可不是这样，孩子去上学，连字母和数字都不认识。

Because back in the day, you're like, your kid goes to school, no letters, no numbers.

现在呢，五岁的孩子上课都在读书了，你简直惊呆了，好吧。

Now they're like, oh, I actually have the classes already reading and they're five years old and you're woah, okay.

我觉得现在的孩子，比我们那一代聪明多了。

You know, kids are already smarter, I think, than the older generation.

我觉得下一代会更聪明，因为他们有AI可以随时用。

And I think that the next generation is going to be even smarter because they're going to have AI in their disposal.

是啊。

Yeah.

顺便说一句，以前我们大多数人得专心学一门课，就盯着这一个领域。

So And by the way, you know, typically we had to Most people had to focus and learn certain subject and focus mostly on that one.

想跨学科合作的话，还得找别人，想办法一块儿干。

And when they wanted to work across disciplines, had to meet with other folks and try to somehow do that together.

以后每个人口袋里都会有个通才。

We're going to have everybody's going to have a polymath in their pocket.

第六个趋势：这真让我特别乐观。

Trend number six: that makes me actually really very optimistic.

以前觉得不可能解决的问题，现在五年内就解决了。

Problems that were impossible are being solved in five years.

我想用这个收尾，因为我觉得它彻底改变了我刚才说的那些话的意义。

I want to end on this one because I think it really changes how you hold everything else I just said.

约西的团队在开发洪水预测系统。

Yossi's team builds flood prediction systems.

七年前，他们问过的每个专家都说这根本不可能。

Seven years ago, every expert they talked to said it was impossible.

变量太多，数据不干净，根本没法预测七天后的情况。

Too many variables, no clean data, no way to get to seven days.

现在，这个系统已经覆盖了150个国家，二十亿人口，从开始到完成，只用了不到五年。

Today, that system covers 150 countries and 2,000,000,000 people under five years, start to finish.

而且我们变得越来越厉害了。

And we're getting better and better.

斯坦福的经济学家把2025年称为AI的收获期。

Stanford economists called 2025 the AI harvest period.

实验阶段已经结束了。

The experiments are done.

现在关键是分清哪些有用，哪些没用。

What works is separating from what doesn't.

目前AI渗透率最高的那些行业，劳动生产率增长速度是全球平均水平的4.8倍。

The industries with the highest AI exposure right now are seeing labor productivity grow 4.8 times faster than the global average.

这都已经有数据了。

And that's already measured.

我们的效率提升快得简直离谱。

Like, how productive we're getting is just crazy.

所以当你听说AI还做不到那事的时候，也许现在确实不行，但你等两周看看会发生什么。

So when you're hearing AI can't do that yet maybe now, but see what happens in two weeks.

对AI能做什么、不能做什么，保持好奇心。

Stay curious about what AI can and can't do.

我们都听过那个男人为患癌的爱犬研发疫苗的事。

We all heard about a guy creating a vaccine for his dying dog to treat cancer.

愿意每隔几个月就更新自己的看法，这可能是你现在能为职业生涯做的最被低估的事。

And being willing to update your answer every few months, that might be the best underrated thing you can do for your career right now.

我说AI是人类创造力的放大器，这不只是预测，更是我们的设计目标。

When I say AI is an amplifier of human ingenuity, this is not only a prediction, this is a design goal.

我希望我们构建系统的方式是这样，希望我们能看清自己如何帮助社会实现这一切，如何影响教育，让我们的孩子能顺利迎接这个未来。

This is how I'd like us to build our systems, how I'd like us to see how we are helping, you know, the society to actually do that, how we're influencing education so that our kids can actually grow into this future.

我对下一代如何解决世界上更多问题，非常乐观。

And I'm quite optimistic about what how we can actually have those next generation solve many other problems in the world.

是的。

Yeah.

因为他们手里的工具更多了。

Because they have more to work with.

我最后还想问你一个问题。

I just wanted to ask you one last question.

如果有人只想从这次对话里记住一件事，那为了在2026年保持积极、不被时代淘汰，他们该抱有什么样的心态？

If somebody wants to remember one thing from this conversation, what is the mindset that they should adopt for 2026 to stay as positive as you are and also relevant in the job market?

是的，科技和研究特别让人兴奋的一点是，有些看似不可能解决的问题，其实并没有那么绝。

Yeah, so one exciting thing is about using technology and research is that certain problems that seem impossible are not necessarily impossible.

说实话，我还没见过真正没法解决的事。

In fact, I've yet to see something that is impossible to tackle.

举个例子吧。

Here's an example.

我们现在用AI用得比较多的一个领域，是应对气候变化、提升抗灾能力。

One area that we're using AI quite a bit is on climate resilience, how to address natural disasters.

那时候我了解到，有一个特别没被重视的领域，就是洪水预测。

One thing I learned at the time is that one area that were not very helpful is an area of natural disaster of flood flood prediction.

每年因为洪水死的人，成千上万。

Floods are, you know, causing thousands of deaths every year.

现在呢，我们已经开发出一个系统，能在150个国家提供洪水预警，覆盖20亿人口，提前七天就能预测。

Fast forward, we now have a system that provides flood predictions in 150 countries, covering 2,000,000,000 people, predictions up to seven days in advance.

七年前还觉得根本不可能的事，现在我们真做到了。

And from what seemed to be impossible just seven years ago, we have a system doing that.

这正是创新创造价值的绝佳机会。

It's a really opportunity for innovation to create value.

顺便说一句，在医疗领域，我们推出了一个叫MedJemma的模型，到现在已经下载了两百多万次，帮开发者们搭建自己的医疗应用。

By the way, on health care, we put out a model called MedJemma, which now have over 2,000,000 downloads by which enables developers to develop their own applications on medical capabilities.

所以在这段视频里，我分享了六个趋势，全是真实数据，来自一个做了二十年这行的人的亲身经历。

So in this video, six trends, real data, real examples from someone who has been building this for twenty years.

我想让你记住一点：趋势就在那儿，不管你搭不搭车，它都在发生。

Here's what I want you to take away from this: Trends are happening regardless if you're with them or not.

现在跑在前面的人，不一定是最懂技术、最拼命的，而是最有好奇心的。

The people who are pulling ahead right now aren't necessarily the most technical, most hardworking people in the room they are the most curious.

他们是那些愿意去试新工具、玩点新东西、刷刷Next、看优质YouTube视频、主动提问、调整自己的工作流，然后立马冲向下一个目标的人。

They're the ones who are trying the tool, who are playing with things, reading something on Next, watching some good YouTube videos, asked the question, updated their workflow and moved on to the next thing.

你今天已经做过了。

You already did that today.

这可不是小事。

That's not nothing.

非常感谢观看，我们很快再见。

Thank you so much for watching this and I'll see you soon

在

on

这个频道。

this channel.