彭博科技在CES

AI代理变得越来越令人印象深刻。

AI agents are getting pretty impressive.

你可能甚至没意识到你此刻正在听一个AI代理。

You might not even realize you're listening to one right now.

但我们不仅仅只是交谈。

But we don't just talk.

我们24/7全天候工作，以解决客户问题。

We work twenty four seven to solve customer problems.

没有等待音乐，只有答案和行动。

No hold music, just answers and action.

访问siera.ai了解更多信息。

Visit siera.ai to learn more.

彭博音频工作室。

Bloomberg Audio Studios.

播客。

Podcasts.

广播。

Radio.

新闻。

News.

彭博科技现场直播，横跨美国东西海岸，连线纽约的卡罗琳·海德和旧金山的埃德·卢德洛。

Bloomberg Tech is live from coast to coast with Caroline Hyde in New York and Ed Ludlow in San Francisco.

欢迎。

Welcome

来到在拉斯维加斯举办的CES展会特别版彭博科技节目，本周我们将为您带来行业顶尖人物的对话。

to a special edition of Bloomberg Tech live from CES in Las Vegas, where we will bring you conversations from the biggest names in the industry throughout the week.

当然，节目中我们将与AMD首席执行官苏姿丰进行对话，此前该公司刚刚发布了面向企业数据中心的新芯片。

Of course, coming up on the show, we sit down with the CEO of AMD, Lisa Su, after the company announced a new chip for corporate data center use.

此外，我们还与Lucid代理首席执行官马克·温特霍夫讨论了全球电动汽车和机器人出租车的格局。

Plus, we discussed the global EV and robotaxi landscape with Lucid interim CEO, Mark Winterhof.

随后，我们采访了游戏硬件公司Razer的首席执行官，他加入节目讨论了其最新的AI游戏生态系统。

And then we talked to the CEO of gaming hardware company, Razer, joining us to discuss its latest AI gaming ecosystem.

但在CES展会现场，我们必须立即连线埃德·卢德洛，他正与AMD首席执行官在一起。

But here on the floor of CES, we've got to get straight over to Ed Ludlow, who is standing by with the CEO of AMD.

埃德。

Ed.

谢谢卡罗琳，欢迎回到Bloomberg Tech，Lisa Su。

Thank you, Caroline, and welcome back to Bloomberg Tech, Lisa Hsu.

Helios MI 455X是AMD首个机架级系统解决方案，而其内部搭载了AMD首款、也是全球首款此类2纳米芯片。

Helios with MI four fifty five x, AMD's first rack scale system solution, but inside it, AMD's first and the world's first two nanometer chip of that type.

当你站在台上第一次亲手举起它时，这一点被广泛热议。

A lot was made of it when when you actually just stood on stage and held it in your hands for the first time.

这为什么如此重要？

What why is it significant?

首先，埃德，很高兴能在这里与你在CES见面。

Well, first of all, Ed, it's great to be here with you at CES.

我认为CES总是开启新一年的绝佳方式，因为它能提供广阔的视野。

I think CES is always a great way to kick off the year because you get so much perspective.

昨晚做主题演讲真的很有趣。

So it was fun giving the keynote last night.

看。

Look.

Helios 是一个庞大的系统，你知道的。

Helios is a massive, you know, system.

你可以在这里的背景中看到它。

You can, you know, see it in the background here.

MI 455 是一块极其强大的芯片。

And m I four fifty five is, just, an incredibly powerful chip.

我想给埃德的背景是，我们非常清楚的一点是，对 AI 计算的需求仍在持续增长。

And, probably the the context I would give Ed is, know, one of the things that we're so clear about is that the demand for AI compute is just continuing to increase.

你知道，过去五年里，当你想到涌现出的大量新功能时，我们就已经看到了这一点。

And, you know, we have seen that over the last five years when you think about just how much, you know, new capabilities that have come on board.

我们现在看到，使用 AI 的人数出现了真正的转折点。

We've now seen, a real inflection in the number of people who are using AI.

所以如果你想想今天，他们可能已经有超过十亿活跃用户在使用AI，我们预计在未来五年内，这一规模将超过50亿用户。

So if you think today, they're probably, you know, more than a billion active users, using AI, and we expect that scale to over 5,000,000,000 users over the next five years.

为此，你需要计算能力，大量的计算能力。

So for all of that, you need compute and lots and lots of compute.

从这个角度来看，MI April 在技术能力上实现了重大飞跃，由2纳米和3纳米芯片组成，拥有3200亿个晶体管，具备强大的性能和能力。

And, from that standpoint, you know, m I April is a, significant leap forward, in terms of technology capability, you know, made up of two and three nanometer chips, 320,000,000,000 transistors, just a lot of performance and a lot capability.

那么，它如何在现实世界中部署呢？

For it to be deployed in the real world then?

那么，谁会是它的首批主要用户呢？

And then who will be the principal first user of it?

你将在2026年看到它，并且之后会迅速推广。

You'll see it in the '26, and, it will ramp from there.

而且，我们拥有非常强大的合作伙伴关系。

And, you know, we have, you know, very strong partnerships.

OpenAI的格雷格·布罗克曼曾登台演讲

OpenAI, Greg Brockman was on stage

没错。

That's right.

昨晚我们和他们讨论了他们看到的所有应用场景。

With us last night, talking about all the use cases that they see.

我们还宣布了与甲骨文以及其他多家公司的合作。

We've announced a partnership with Oracle, many others, as well.

既然它已经是两纳米工艺，现在已全面投产。

So given that it it's it's two h, it's in full production now.

它正准备进入

It's getting ready to

我们确实在全力准备发货。

We we are we are absolutely getting ready, to to ship it.

这处于规模和谱系的一端。

That's at one end of of of the sort of scale and spectrum.

而在另一端，你有MI440X，适用于小型数据中心。

At the other, you have m I four forty x, which is for small data centers.

我正试图简化它，但本质上它是一款企业级产品。

I'm trying to simplify it, but it's basically an enterprise product.

你们当初是想用它解决什么问题呢？

What what was it that you were trying to solve for with that?

是的。

Yeah.

我认为我们想解决的是，世界本质上是一个非常异构的世界。

I I think what we're trying to solve for is, you know, the world is a very heterogeneous world.

你有各种各样的AI应用场景，从进行大规模训练和推理的超大型云数据中心，到企业应用，再到超级计算机。

You have all kinds of use cases for AI from, you know, sort of the very biggest cloud data centers that are doing, you know, large scale training and inference to enterprise applications as well as supercomputers.

因此，我们实际上有一系列芯片。

And so we actually have a family of chips.

在最高端是我们用于云环境的MI 455，但昨晚我们还发布了MI 440，它使用了相同的底层构建模块，但专门针对企业客户，满足他们在本地环境中对数据中心的自主控制需求。

At the highest end is our MI four fifty five for the cloud environment, but we announced last night a MI four forty, which is, actually using the same basic building blocks, but is now, really focused on enterprise they want their own control of their data centers in terms of on prem environments.

不过，他们具体用它来做什么呢？

What what are they doing with it, though?

我的意思是，我们一直过于关注那些拥有数百亿参数的前沿模型以及支撑它们所需的基础设施规模。

I mean, you know, we we've been so fixated on frontier models with hundreds of billions of parameters and the the scale of infrastructure needed for that.

有了MI 440，我们谈论的是稍微不同的东西。

With MI four forty, we're talking about something slightly different.

我只是觉得，如果你能解释一下这些企业的需求是什么，他们对这项技术有什么期望，那会非常有意思。

I'm just I think it's just really interesting if you could explain what the demand is from those enterprises, what they want with the technology.

我认为，现在许多企业正在将其AI技术应用到整个业务流程中，无论是在工作流程中的哪些环节。

Well, I I think you see many enterprises now using AI all throughout their their business processes, whether you're talking about, you know, things in their workflow.

就连AMD，我们也在开发流程的每一个环节都使用了AI。

I can even AMD, we're using AI through every part of our development process.

我们在金融服务和医疗保健领域看到了大量应用。

We're seeing a lot of applications in financial services, in health care.

这些领域，尤其是在金融服务中，人们并不一定希望所有东西都放在云端。

These are, you know, areas, especially in financial services, that, you know, people actually don't want everything necessarily in the cloud.

他们更希望拥有自己的本地部署或私有云部署。

They'd like to be able to have, you know, their own, on prem deployments or private cloud deployments.

在这种情况下，你不想为每一代新芯片都新建一个数据中心。

And in this case, you don't wanna have to build a brand new data center for every new generation of chip.

MI440让我们能够利用一些现有的数据中心，并通过新功能进行升级。

MI four forty allows us to use some of those, existing data centers and upgrade, with the new capabilities.

欢迎。

Welcome.

如果您正在观看彭博电视台或收听彭博电台，我们正在拉斯维加斯现场直播，与AMD首席执行官Lisa Hsu对话，讨论最新一代加速器。

If you're watching us on Bloomberg Television or you're listening on Bloomberg Radio, we're live in Las Vegas, and we're with AMD's CEO, Lisa Hsu, and we're talking about the latest generation of accelerators.

与NVIDIA目前提供的产品相比，这一代AMD加速器在本地部署和边缘场景中为何更具优势？

What makes this generation of AMD accelerators the better option, particularly for on prem and at the edge over what NVIDIA is offering right now?

埃德，最好的理解方式是，我们正处在一个AI的转折点上。

Well, the best way to think about it, Ed, is, we're in this place where, AI is at an inflection point.

我们现在看到AI已经渗透到计算的每一个领域。

We're seeing AI now in every part of compute.

我们在最大的模型中也能看到它，比如当你想到TrackGPT、Gemini和Grok这样的模型时。

We see it in, the largest models, you know, when you're thinking about things like TrackGPT and Gemini and Grok.

你知道，我们还看到许多新应用场景，比如视频制作、娱乐、医疗保健，在这些领域你正在进行药物发现，所有这些不同的应用。

You know, we're also seeing, you know, many use cases in, you know, new capabilities like, you know, video production, entertainment, health care, where you're doing drug discovery, all of these, you know, various things.

你知道，我们的优势在于卓越的性能和极具竞争力的总体拥有成本。

You know, our claim to fame is really, you know, outstanding performance at, you know, very advantaged total cost of ownership.

我们非常坚信的是开放的生态系统和深度的合作伙伴关系，让我们的整个生态系统协同合作。

And the other thing that, you know, we believe very strongly in is an open ecosystem and deep partnerships, you know, with our, you know, with our overall, ecosystem coming together.

所以，当你从这个角度来审视时，我认为我们拥有众多能够充分利用这些最新一代芯片的应用程序。

So when you put those things in perspective, I think we have a a a great set of, applications that will take advantage of these newest generation chips.

你提到格雷格·布罗克曼，也就是OpenAI的总裁，昨晚和你同台亮相。

You mentioned that Greg Brockman, who's the OpenAI president, was on stage with you last night.

他提出的一个基本观点是，他们有很多工具和功能很想发布并推向世界，但受限于计算能力。

And one of the the basic points that he made was there are tools and, functions they would love to release and put out into the world, but they're compute constrained.

我经常让你量化需求，但有没有办法量化当前计算能力短缺的严重程度？

I often ask you to quantify demand, but is there a way to quantify the severity in the lack of compute, you know, the deficit that's out there right now?

让我给你一些数据，帮助你理解我们对当前需求环境的看法。

Well, let me just give you some numbers to kinda ground what we think the demand environment is, is looking like.

所以，如果你想想现在，我们大约有十亿活跃用户，计划在未来五年内将其提升至五百亿。

So if you think, you know, today, we have about a billion active users, and we're ramping that to, you know, 5,000,000,000 over the next five years.

而目前，全球大约有100泽弗洛ps的算力，你知道的，这是所有算力的总和。

And we have about, let's call it, a 100 zetaflops of compute, you know, all around the world.

我的意思是，这只是一个笼统的数字，汇总了所有这些算力。

I mean, that's just a a generic number, that, that aggregates all of that.

我们认为，在未来四五年里，算力需要再增加一百倍。

You know, we think we have to increase compute by another 100 times as you go over the next, you know, four or five years.

昨晚，我引入了一个新术语——尧弗洛ps。

And, I introduced a term, last night, the yotta flop.

人们会问：什么是尧弗洛ps？

You know, people are like, what is a yotta flop?

尧弗洛ps实际上是10的24次方浮点运算次数。

A yotta flop is actually 10 to the twenty fourth in terms of flops.

也就是说，这是一个后面跟着24个零的数字。

So that's a one followed by 24 zeros.

为了让你了解事物究竟增长了多少，这比我们今天的算力又增加了100倍。

And to give you, you know, just a view of just how much things have really increased, I mean, that's another 100 times more compute, than we have today.

所以这让你有个概念。

So that gives you an idea.

现在你可能会想，这么多算力你打算用来做什么？

Now you think, what are you gonna use all that compute for?

事实上，我们今天拥有的模型已经非常出色了。

I mean, the truth is the models that we have today are great.

它们确实能完成令人惊叹的事情。

I mean, they they do amazing things.

我们之前讨论过一些应用场景。

You know, we talked about a number of use cases.

也许有一个非常贴近我们生活的例子，那就是编写软件。

You know, perhaps, you know, one that's, you know, very hits very close to home is is, writing software.

比如，人们现在正使用AI工具显著提升软件开发者的生产力。

Like, you know, people are using, the AI tools right now to significantly enhance the productivity of software developers.

但这已经不错了，还能变得更好。

But it's good, but it can get so much better.

我认为这才是关键点。

And, I mean, I think that's the key point.

我们常说，人工智能将无处不在，而且人人都能使用。

You know, we we like to say that AI is really gonna be everywhere, and it's really for everyone.

它能让每一个人提升自己企业的生产力，每个人在未来都能更加高效。

And it's for each one of us to make our businesses more productive, you know, each one of us, more productive, you know, going forward.

因此，我们仍处于释放人工智能巨大潜力的非常早期阶段。

And so we're still in the very early innings of really unlocking the power of AI.

所以我们目前的情况是，好吧。

So where we stand is we okay.

存在计算能力不足的问题，而软件已经触及了当前一代计算能力所能提供的极限。

There's a there's a compute deficit, and software has kind of hit the limits of what current generation compute can offer.

帮我们理解一下部署这些计算能力所面临的瓶颈和障碍。

Help us understand the bottlenecks and barriers that to deploying that compute.

目前有很多关于内存芯片的话题。

A lot at the moment about memory chips.

还有别的吗？

What else?

能源，电力。

Energy, electricity.

丽莎，你现在遇到什么问题让你感到担忧，觉得这是个当前的难题？

What what's crossing your desk, Lisa, that that's that gives you pause and say, this is a this is a problem right now?

作为科技行业，我们的职责是推动技术前沿。

Well, our job as a technology industry is to push the bleeding edge.

我的意思是，这正是我们的工作。

I mean, that is our job.

当我们考虑部署MI455、采用2纳米和3纳米芯片，以及最新的高带宽内存时，真正重要的是整个生态系统要协同合作，共同规划这场计算领域的下一次重大转折。

And so, you know, when we think about, like, the m I four fifty five deploying, you know, two nanometer and three nanometer chips, having the latest generation memory, high bandwidth memory, that that is out there and really deploying these big systems, The important thing is that the entire ecosystem come together, and we plan together for this, you know, next big inflection in compute.

而这正是我们目前正在做的事情。

And that's exactly what we're doing, right now.

我认为我们正在与整个供应链紧密合作，以确保我们拥有扩展这一计算环境所需的资源。

I think we're working very closely with the entire supply chain to ensure that we have the the resources to expand this compute environment.

是的，你提到的一些问题，我们可以称之为受限的，

And, yes, you know, some of the things that you mentioned are, let's call it, constrained,

但哪个

but Which

最严重，所以所以

is most severely, so so

我认为这不是某一个单一的问题。

You know, I don't think that's any one thing.

我认为我们都在思考，如何更快地构建。

I think we're all looking at, you know, how do we build faster.

当然，电力是其中一个领域，你在世界各地都能看到这一点。

You know, certainly, you know, power is one of those areas where, you know, you see throughout the world.

电力正在尽可能快地投入运行。

You know, power is being, you know, brought online as fast as possible.

当然，从硅芯片的角度来看，我们正在与合作伙伴一起提升生产能力。

Certainly, from a silicon standpoint, you know, we're ramping our production capabilities with our partners.

从内存的角度来看，我们的合作伙伴也在加速提升产能。

From a memory standpoint, our partners are are are ramping as well.

所以这不是某一个单一的问题。

So it's not any one thing.

我认为所有这些方面都必须协同推进，这就是为什么在这一行业中伙伴关系如此重要。

I think it's all of these things have to go sort of in tandem, and that's why partnership is just so important in, in this business.

我们一开始聊的是AMD的Helios机架级架构和基础设施。

We started this conversation talking about Helios first rack scale, architecture and and infrastructure from AMD.

你能谈谈未来，以及你希望在服务器中掌控多少内容吗？

Could you talk about the future and how much of the content you want to own in in a server?

我们最初是从GPU开始讲这个故事的。

You know, we started this this story with the GPU.

坦率地说，如果你看看英伟达在做什么，他们希望越来越多地掌控机箱内的所有组件。

Frankly, if you look at what NVIDIA is doing, they want to increasingly own all of what's inside the box.

AMD也专注于这一点吗？

Is that something that AMD is focused on too?

对我们来说，最重要的是确保我们提供易于客户部署的即用型解决方案。

You know, what's most important, for us is to ensure that we have turnkey solutions that are very, very easy for our customers to deploy.

因为当你思考如何最有效地利用所有这些AI计算能力时，你希望它能直接进入数据中心，并在第一天就正常运行。

Because when you think about, you know, how do you use all of this AI compute most effectively, you want it to go into the data center and, you know, really be up and running on day one.

为此，你必须优化整个系统。

And for that, you have to optimize a full system.

但从这个角度来看，我们非常关注开放的生态系统。

But from that standpoint, you know, we are very focused on an open ecosystem.

因此，是的，我们设计CPU、GPU和部分网络组件，但同时也与广泛的合作伙伴和行业标准紧密合作。

So, yes, we design the CPUs and the GPUs and some of the networking elements, but we also work, you know, really with a broad ecosystem of partners, with industry standards.

这一切都是为了确保我们在整合解决方案时能够兼收并蓄，获得最佳效果。

It's all about ensuring that, you know, we get the best of all worlds when we put our solutions together.

展望未来，MI52027的性能将是MI300系列的1000倍。

Looking ahead to m I five hundred twenty twenty seven, that has 1,000 times the performance of the m I 300 generation.

所以你们上一代实际部署的设备，现在即将推出一款性能提升千倍的产品。

So your last generation of real world deployed gear, something's coming that's a thousand times better.

你们是如何实现千倍性能提升的？

How did you make it a thousand times better?

这在每一个层面都体现了令人难以置信的工程实力。

It it is just incredible engineering at every level.

所以MI455比我们六个月前刚推出的MI355芯片性能提升了十倍。

So MI four fifty five is 10 times better than the chip that we just launched, six months ago, the m I three fifty five.

而MI500又在此基础上再提升了十倍。

And m I 500 is another 10 x, you know, on top of that.

我们使用了目前最先进的技术。

We are using the most advanced technology out there.

我们非常明确地专注于硬件、软件和系统的协同设计，这显然正在推动能力的前沿边界。

We have a a very, you know, very clear focus on, you know, hardware, software, system co design, and, it is, you know, clearly pushing the bleeding edge of capabilities.

目前AMD向中国销售产品的状况如何？

What is the status of AMD's ability to sell products into China right now?

所以，你知道，中国对我们来说是一个重要的市场。

So, you know, China is an important market for us.

你知道，我们实际上向中国销售多种芯片，包括我们的个人电脑以及其他嵌入式芯片。

You know, we actually sell a broad range of chips into China, including our, you know, our PCs as well as, you know, other embedded chips

当然，还有数据中心领域。

as data center context, of course.

抱歉。

Sorry.

在数据中心领域，我们当然认为中国是一个重要的市场。

In the data center context, we are you know, certainly, we see China as an important market.

我们去年年底从美国政府获得了一些许可证，涉及我们上一代的 MI308 芯片，目前我们正在申请新许可证，针对最近被允许申请许可证的 MI325 芯片。

We were we did get some licenses from the US government, you know, late last year as it relates to some of our previous generation, our m I three zero eight, you know, chips, and we are in the process of applying for new licenses, with our, m I three twenty five chips that were, you know, recently allowed to, to apply for licenses.

我们尚未获得这些许可证，但我们依然将中国视为我们的重要市场。

We haven't gotten those licenses yet, but we continue continue to view China as an important market for us.

我问这个问题的部分原因是，许多开源模型的工作以及弥合开源与闭源之间差距的工作，在某种程度上是在中国进行的。

The reason I ask about it is is in part because a lot of the work that's being done in open source models and bridging the gap between open and closed is being done in China to some extent.

关于中国市场需求的讨论很多，但您能否稍微谈谈这种需求，以及中国政府对于您将更先进技术引入中国持何种态度？

There's been a lot of discussion about the demand being there in China, but could you reflect a little bit on on that demand, but also what the Chinese, government's attitude is to to you taking a a later generation of tech to the country?

我认为，正如我们之前讨论的那样，人工智能在中国的整体需求非常高。

Well, I I do think the demand for, you know, AI in general and in China is high, for all the reasons that we talked about.

我认为，全球范围内，更多的计算能力都是有益的，我们现在正处在一个这样的需求环境中。

I think we are in a demand environment where, you know, more compute is beneficial, across the world.

我们认为，中国是我们的重要市场，并且在部署我们的解决方案方面非常积极。

We think, you know, China is an important market for us, and it's a very active in having our solutions deployed.

因此，我们继续将其视为一个重要的市场。

So, you know, we continue to view it as something that's important.

我们正在与美国政府以及我们的中国客户合作，寻找良好的解决方案。

We're working with the US government as well as our Chinese customers, you know, to find good solutions there.

两国政府都有迹象表明，许可证审批流程正在推进。

And there are signs from both governments that the license process is moving.

商务部以审批拖延、文件长期积压而闻名。

Commerce is kinda notorious for things sitting on a desk for quite a long time.

我认为我们很有信心，能够获得一些这些许可证的批准。

I think we are optimistic that, you know, we'll have an opportunity to, to get some of those licenses granted.

您正在收看彭博电视台。

You're watching Bloomberg Television.

您正在收听彭博电台。

You're listening to Bloomberg Radio.

这里是彭博科技，我们正在拉斯维加斯现场连线AMD首席执行官许莉莎。

This is Bloomberg Tech, and we're live in Las Vegas with the AMD CEO, Lisa Hsu.

在数据中心的背景下，最后一个问题是，市场和投资者希望看到数据和迹象，证明你们正在抢占市场份额。

Last question really in the data center context is the markets and investors want data and signs that you're taking market share.

您会引用哪些指标来证明这一点？这些指标是已经存在的，还是在未来十二个月内会出现？

What would the metrics be that you'd point to either that that already exist or over the coming twelve months that would evidence that?

我认为MI 455是我们技术能力以及我们与整个行业深度合作的一个明确转折点。

Well, I think m I four fifty five is a a clear inflection point in, you know, both our technology capability as well as the deep partnerships that we have across the industry.

因此，我们对眼前看到的前景感到兴奋。

So we're excited about, you know, what we see in front of us.

而且，我们谈到过，到2027年，AI收入将达到数百亿美元。

And, you know, we've talked about, you know, tens of billions of dollars in in AI revenue as we get into 2027.

我认为，这些指标对我们公司思考AI潜力至关重要。

And I think these are important metrics, you know, for us as a company when we think about the AI potential.

尽管大家聚焦于数据中心，但有些人忽略了AMD在PC领域许多方面的领先地位。

For all the focus on data centers, some forget that AMD is leader in PC in many respects.

预测机构对今年的情况有着截然不同的看法。

Forecasters have very different opinions of what will happen this year.

有些人认为市场会萎缩。

Some see, you know, shrinking market.

有些人则认为，仅靠AI PC就能推动温和增长。

Some see modest growth driven literally by just AI PC.

无论整体环境如何，你们一直能够抢占市场份额并实现增长，但这些环境并不理想。

You've been able to take market share and grow irrespective of what the broader conditions are, but they haven't been great.

你们预计这种情况会如何持续下去？

How have you done expect that to continue to be the case?

PC市场对我们来说是一个非常好的市场。

Well, the PC market is a very good market for us.

我们在2025年的PC市场中实现了大幅增长，这主要得益于我们产品组合的强大实力。

We grew a ton in the PC market in 2025, and that really came from the strength of our product portfolio.

我们很早就押注于AI PC，因此我们认为这项技术将带来明确的需求。

We bet early in AIPCs, so it was a a clear area where, we believe that the technology would generate demand.

我们还经历了Windows 11的更新周期。

We also went through a refresh cycle with Windows 11.

随着我们进入2026年，我想看看接下来几个季度的表现如何。

And as we go into 2026, I think we'll we'll want to see how a few quarters, play out.

我认为对计算能力的总体需求确实存在。

I think the general, demand for computing is certainly there.

目前还有一些供应链限制，我们正在努力解决，并希望在未来继续关注。

There are some supply chain constraints that, you know, we're working through and we want to, you know, watch going forward.

但我们的情况是，在市场的某些部分，我们仍然处于相对弱势地位。

But, you know, our case is one where, we are still underrepresented in in parts of the market.

你知道，我们在游戏领域非常强大。

You know, we are, very, very strong in gaming.

我们在消费领域也非常强大。

We're very strong in consumer.

我认为我们在企业笔记本电脑领域表现不足，我们认为这是我们的一个增长领域。

I think we're underrepresented in enterprise laptops, and we view this as a growth area for us.

AIPC会改变这一点吗？

Does AIPC change that?

AIPC无疑有助于推动升级周期的到来。

AIPC is absolutely, help in terms of, you know, just the upgrade cycle, coming in.

我们对我们在AI开发系统方面的一些工作也感到兴奋。

We're, you know, excited about some of our work with, AI development systems as well.

我们昨晚发布了一款新的AI开发系统，我们认为它也会非常有吸引力。

We we announced a new AI development system last night that, you know, we think will be, also very attractive.

你刚才提到的PC领域的那些限制，具体是指DRAM，还是更广泛的问题？

The those constraints you were talking about in the in the PC context are are specifically DRAM or it's broader than that?

这更多是关于内存方面的问题。

It's it's more around the memory side.

所以当你考虑整体内存时，我认为来自所谓的AI数据中心计算的需求非常大，我们想看看这将如何影响外部的其他内存市场。

So when you think about, you know, memory overall, I think we have, you know, so much demand coming from, let's call it, AI data center compute that, you know, we want to see how it impacts, you know, sort of the rest of the memory market, out there.

你和OpenAI的格雷格·布罗克曼在台上讨论的另一个领域是AI对全球经济的广泛影响，而不仅仅是针对企业。

One of the other areas that that you discussed with with Greg Brockman on stage from OpenAI was sort of the net or broad economic impact of of AI, not just for for the the companies.

我觉得你当时更多是在谈全球经济。

I think you were talking more about global economy.

同样，这很难回答，丽莎，但如何衡量AI在任何一年里是否对全球产生了直接的积极经济影响呢？

Again, very difficult, Lisa, but how does one measure progress in in whether AI has or has not had a direct, positive economic impact around the world in any given year?

你知道，这确实是真的。

You know, it it's true.

要厘清所有正在发生的事情很难，但在我看来，从我们业务中看到的情况来说，埃德，很多人希望看到特定投资的直接回报。

It's hard to deconvolve all of the things that are happening, but I think from a sense of, you know, what we see in the business, Ed, you know, many people want to see direct return on investment for, a particular set of investments.

我想说的是，我们知道AI正在提升企业的生产力。

What I would say is that, we know that AI is making a difference in the productivity of companies.

我们知道。

We know that.

我可以看到，在AMD内部，当我们部署AI时，能够更快地将产品推向市场。

I can see that, you know, within AMD in terms of as we deploy AI, you know, we're able to get products to market faster.

我们能够显著改善一些业务流程。

We're able to, you know, significantly improve some of our business processes.

因此，展望未来几年，我认为这种影响将在企业中更加广泛地显现。

So, you know, as we go forward over the next, several years, I think you're gonna see that much broader in enterprises.

我所接触的每一位首席执行官都在谈论AI。

Every CEO that I talk to is talking about AI.

如何打造更好的公司、构建更好的投资组合，AI已成为核心议题。

It is front and center in terms of how to build a better company, how to build a better portfolio.

所以，正如格雷格所提到的，当你将所有这些因素汇总起来，AI必然会对全球GDP产生影响，我们将在未来几年看到这一点。

And so, you know, I think what, you know, Greg was talking about is when you aggregate all of that, AI has to impact, you know, the world at a GDP level, and, we'll see that over the next few years.

您正在收看彭博电视台。

You're watching Bloomberg Television.

您正在收听彭博电台。

You're listening to Bloomberg Radio.

这里是彭博科技，我们正在拉斯维加斯现场直播。

This is Bloomberg Tech, and we're live in Las Vegas.

我们正在采访AMD的首席执行官Lisa Hsu。

We're speaking to AMD's CEO, Lisa Hsu.

您是生成性生物技术领域的投资者。

You are an investor in generative bionics.

他们同时也是技术合作伙伴，并在拉斯维加斯CES展会上发布了一款人形机器人。

It's also a technology partner, and they have unveiled a humanoid robot here at Las Vegas CES.

事实上，如果电视的魔力能够实现，当我们切换到全景镜头时，您会在背景中看到它。

In fact, if the magic of television can happen and we cut to the wide shot, you'll see it in the background.

对。

Right.

我认为，这是迄今为止我们从AMD看到的、关于其如何参与物理AI领域的首个具体迹象。

You know, this is the first tangible sign I I feel we've we've seen from AMD on how you intend to play in physical AI.

是的。

Yes.

解释一下你的策略。

Explain your strategy.

这是下一个大市场。

It is the next big market.

对吧？

Right?

是的。

Yes.

我不说这是第一次，但可能是我们不太常提及的领域之一，因为大家的关注点都集中在数据中心和云，而这些领域的机遇显然就在我们面前。

And I wouldn't say it's the the the first time, but it's probably, one of the areas where we don't highlight as much because there's so much focus on data center and cloud, and the opportunities there are, you know, very much in front of us.

但当我们审视物理AI时，从我们在FPGA和嵌入式实时能力方面所做的所有工作来看，我们早已涉足这一领域。

But when we look at physical AI, you know, starting from, all of the work we've done in FPGAs and embedded real time, you know, capability, we have been in this space for a long time.

你知道，我们已经为许多现有的机器人应用提供动力。

You know, we already power a lot of robotic applications, you know, out there.

但我认为，随着我们进入人形机器人领域，我们对与G Bionics和OnGeneOne的合作感到兴奋，这让我们在能力、智能以及我们所追求的目标上达到了一个新的层次。

But I think as we go into, the humanoid capability and, you know, we're excited about our partnership with, you know, g Bionics and the work with OnGeneOne, I think that takes us to another level in terms of capability and intelligence and what we're trying to do.

所以

So

商业模式是成为人形机器人内部的所有部分吗？包括在AMD加速器上训练的推理端软件？

Is the business model to be all things, the brain inside of the humanoid robot on the inference side, the the underlying software being trained on a trained on AMD accelerators?

我只是想知道，你们的市场策略是什么？我想问的就是这个。

Just don't what's the go to market, I guess, what I'm asking?

你应该期待我们的合作伙伴关系贯穿所有这些层面。

You should expect that, our partnerships extend through all through all of those levels.

因此，我们拥有能够为人形机器人提供动力的组件，支持其实时本地化能力，这一点非常重要。

So we have the components that can, you know, power the humanoid robots, you know, sort of real time local capability, which is very, very important.

此外，我们还拥有背后的技术，用于在这些人形机器人上进行训练和推理。

And then we also have the technology behind that in terms of, you know, how to train and inference, on these, on these, humanoids.

我们上一次面对面见面是在华盛顿特区，当时总统刚刚提出了美国人工智能的广泛战略。

When last we met in person, it was in Washington DC, and the president had just outlined a broad strategy for America and AI.

这主要围绕基础设施、放松监管，允许基础设施建设者加快步伐。

And it really centered around infrastructure, deregulation, allowing those building the infrastructure to move faster.

那是在去年下半年的事。

That was kind of in the second half of last year.

在那之后的几个月里，你看到任何迹象表明这奏效了吗？

In the months that have followed, have you seen any signs that it worked?

有没有什么具体例子能说明，人们现在能更快地建设，或许能缓解我们之前讨论的计算能力短缺问题？

And anything that you could point to that says, yeah, people are able to build faster, maybe to to address some of the compute deficits we discussed.

我可以肯定地说，总统的AI行动计划，我们见面时，我记得是在七月发布的，我当时对美国在AI领域全面领先的战略充满乐观。

Well, I I can say for sure, you know, the president's AI action plan, you know, when we met, and I think this was back in July when it came out, I was very optimistic about having a a really forward leaning strategy from, you know, sort of the the whole view of what does it take for The US to lead in AI.

我认为我们在这条路上已经取得了巨大进展。

And I think we've made a ton of progress along the way.

而且，迈克尔·克拉西奥斯昨晚也上台和我们讨论了创世计划，这同样是一种公私合作的方式，旨在真正推动美国的科学发展。

And, you know, I had, Michael Kracios joined us last night on stage as well to talk about the Genesis mission, which is, you know, another, you know, sort of public private partnership approach to really advance science in The United States.

当你审视所有这些举措时，加快建设、确保我们拥有正确的出口管制，以便美国技术栈能够被广泛采用。

And when you look at, you know, all of these things, you know, building faster, ensuring that we have, you know, the right, export controls so that we're able to have The US stack, adopted across.

我们目前的出口管制是否合适？

Do we have the right export controls currently?

我们当然正在与美国政府的各个相关方紧密合作，以确保在这方面取得适当的平衡。

We we are certainly working very closely, with, you know, the the various parties in the US government to ensure that we have the right balance, there.

同时，我们也在思考如何在这里加大投资，确保在美国本土能够尽可能快地推进，将人工智能能力投入运行，以助力科学和更广泛的经济效益。

And we also have, you know, this notion of how do we invest more here and ensure that in The United States that we are, you know, running as fast as possible, to bring, you know, AI capacity, you know, online to, help us in, you know, science and, you know, sort of the broader, you know, economic benefits.

丽莎，2026年会发生什么？

Lisa, what what happens in 2026?

在人工智能领域会发生什么？你认为今年你的行业希望取得的进展会以什么为标志？

What happens in the world of AI, and what do you think will define this year in terms of the progress that your your industry hopes to make?

昨晚我们的主旨演讲开头，我说了一句话：你还没看到真正的精彩呢。

Well, I I I started our keynote last night with the sentence that, you know, you ain't seen nothing yet.

这确实是我真实的想法。

That's really how I feel.

我的意思是，现在才一月份，我们每周、每月看到这些模型如何发展、应用场景如何拓展，以及它们给企业与成果带来的切实影响，都令人惊叹不已。

I mean, we're sitting here in January, and it's just amazing how much progress is made, you know, every week and every month when we see how these models are developing, when we see how the use cases are developing, and then when we see the tangible results on on businesses and outcomes.

我相信，我们在2025年已经看到了相当多的成果得以实现。

I believe that, you know, we saw a good amount of the that, you know, come to fruition in 2025.

我们将在2026年看到更多这样的进展，因此每个人都应该明白，人工智能不仅仅是外界的炒作。

We're gonna see much more of that in 2026 so that everyone should understand that, you know, AI is not just, you know, hype out there.

它也不仅仅是投资界人士谈论的一些空泛概念。

It's not just, you know, sort of things that people are talking about in the investment community.

它是人们每天都在实时使用、并切实感受到的技术——嘿。

It's things that people are using every day real time and feeling like, hey.

我的生活因为这项技术而变得更好了，我认为我们在2020年就会看到这种变化。

My life is better because I have this technology, and I think we're gonna see that in 2020

Lisa Hsu，AMD首席执行官，AMD推出了全球首款2纳米芯片，并应用于其首个机架级系统解决方案Helios。

Lisa Hsu, AMD CEO, AMD with it's in the world's first two nanometer chip going into Helios, its first rack scale system solution.

Carrie，回到你在拉斯维加斯的演播室。

Carrie, back to you on set in Las Vegas.

这是一场非凡的对话，Ed Ludlow，一如既往地与AMD的Lisa Sue进行了交流，当然，她昨晚还与OpenAI的Greg Brockman一同登台。

What an extraordinary conversation, Ed Ludlow, as always, with Lisa Sue of AMD, of course, took to the stage last night alongside Greg Brockman of OpenAI.

我们来跟进一下股价情况，因为丽莎在与埃德的访谈中发表了诸多重要言论。

And we're just gonna check-in on the shares because there were significant statements coming from Lisa throughout that interview with Ed.

尽管她继续阐述未来四到五年计算需求将增长百倍，强调我们仍处于释放AI潜力的早期阶段，并大力推介MI455，推动能力的前沿发展，但股价仍下跌了3.3%。

We are off by 3.3% even as she continues to articulate how much compute is gonna need to increase a 100 times in the next four to five years, talking about the early innings that we're at in terms of unlocking AI and really talking up the m I four fifty five, pushing at the bleeding edge of capability.

我们还在讨论他们在人工智能领域对中国市场的巨大需求。

We're also talking about the demand they have in AI for China.

特别是，中国的需求非常高。

In particular, Chinese demand is high.

目前，他们正与政府合作，寻找能够向中国发货的解决方案。

They're working with the government as it stands for finding solutions to be able to ship to China.

他们还谈到，在企业笔记本电脑市场中，他们的份额仍显不足。

Also talk about how they're underrepresented in enterprise laptops as well.

不过，股价下跌3.3%，是因为摩根大通表示：和英伟达一样，AMD对计算需求的前景非常乐观。

Well, by 3.3%, though, because seeing JPMorgan saying, look, like NVIDIA, AMD's outlook for compute demand is very bullish.

他们还提到预计2027年将推出MI 500系列，但摩根士丹利认为，这并未带来太多全新的信息。

And they're hearing about the m I 500 series coming on for 2027 launch, but not much new according to Morgan Stanley in terms of brand new information.

视频业务增长了百分之零点四。

And video is up four tenths of percent.

我们可能获得了更多关于Rubin发布的信息，了解到已有六款芯片回归，这是他们计算领域下一代架构的创新成果，我们看到它们已经重返市场。

It's maybe we got a little bit more detail on the Rubin unveil and the fact that six chips have already come back on their next innovations, the next architecture for their compute, and we're seeing that they're already back.

它们很可能在2026年期间开始发货，同时他们也在大力宣传与英伟达合作的自动驾驶汽车未来，推出一个可能与特斯拉竞争的新平台。

They're likely to be shipping in the course of 2026, but also talking up the future of self driving vehicles with NVIDIA, a new platform to rival maybe even a Tesla.

埃隆·马斯克说，对此我连觉都睡得着。

Elon Musk says, I'm losing not sleep over that.

但我们还听到了英伟达在机器人领域的未来规划，以及来自中国对H200芯片的需求。

But we're also hearing about the future robotics coming from NVIDIA as well and demand for h two hundreds coming from China.

这些股票上有太多信息需要消化。

So so much to digest on these particular stocks.

在拉斯维加斯，还有太多东西有待了解，我们当然将在这里与英伟达首席执行官黄仁勋以及西门子首席执行官罗兰·布什面对面交流，地点就在拉斯维加斯的CES展会现场。

So much still to learn here in Las Vegas, and we're gonna be sitting down, of course, with the CEO of NVIDIA, that is Jensen Huang, and Siemens CEO, Roland Bush, right here from Las Vegas, from CES coming up.

我们接下来还将与这些正在与埃德·卢德洛对话的领导者们一同展开讨论。

We are gonna be talking though right alongside not only these leaders who are speaking with Ed Ludlow in the next few hours.

在接下来的几分钟里，您将听到路西德临时首席执行官马克·温特霍夫的发言。

In the next few minutes, you're gonna be hearing from the Lucid interim CEO, Mark Winterhof.

我们将讨论全球电动汽车的未来、整个行业，以及该公司与优步、Nuro合作的新型机器人轴系统，它们如何与英伟达产生关联？

We're gonna be discussing the future of global EVs, the industry, the company's new robot axi partnership with Uber, with Nuro, how are they intertwining with NVIDIA?

他们对未来的自动驾驶平台又是如何思考的？

How are they thinking about the future of self driving platform there as well?

在供应链方面也有大量内容值得探讨。

So much to get to in terms of supply chain as well.

请继续关注。

Stick here.

我们正在拉斯维加斯的消费电子展现场。

We are from the Consumer Electronics Show in Las Vegas.

每天，数百万客户都在与我这样的AI助手互动。

Every day, millions of customers engage with AI agents like me.

我们全天候工作，随时掌握最新信息。

We work round the clock and have the facts at our fingertips.

我们快速高效，但极具耐心。

We're fast and effective, but incredibly patient.

我们基于Sierra构建，这是领先的AI驱动客户体验平台。

And we're built on Sierra, the leading AI powered customer experience platform.

没有等待音乐，只有答案和行动。

No hold music, just answers and action.

访问sierra.ai了解更多信息。

Visit sierra.ai to learn more.

那就是sierra.ai。

That's sierra.ai.

欢迎回到由拉斯维加斯CES现场直播的特别版Bloomberg Tech节目。

Welcome back to a very special edition of Bloomberg Tech live from Las Vegas in CES.

我们在此时快速检查一下市场动态，因为我们在消费电子展上发布了一些重大公告。

Quick check on these markets as we stand because we had some big announcements over here at the Consumer Electronics Show.

其中关键的公告来自英伟达和AMD。

Key among them has been from NVIDIA and from AMD.

目前，英伟达上涨了百分之零点二十五。

Now NVIDIA are a up quarter of a percent at the moment.

我们听到了关于Rubin的未来。

We hear about the future of Rubin.

我们听说它正在按计划推进。

We hear about how it's coming on track.

制造合作伙伴已经为下一阶段的架构带回了六组芯片。

Already manufacturing partners are bringing back six sets of chips for the next stage of architecture.

他们正在推广新的自动驾驶平台。

They're talking up new self driving platform.

他们在思考机器人技术。

They think about robotics.

英伟达仅上涨了约百分之零点二十五。

NVIDIA catching a bit only about a quarter of a percentage point.

AMD下跌了3.8%，尽管我们听到Lisa Su谈到未来四到五年内对100倍算力的需求，以及他们将如何用下一代芯片满足这一需求，但市场仍未能充分理解这些新进展。

AMD down 3.8% even as we hear Lisa Su talking about the 100 x compute need over the next four to five years and how they're gonna be satisfying it with their next generation of chips, but not enough new for the market to get its head around it would feel.

纳斯达克100指数上涨百分之零点四。

Nasdaq 100 up four tenths of percent.

让我们转向一些大幅波动的股票，因为这些公告带来了对其他关键公司的连锁影响。

Let's move on to somewhere big movers, though, because amid these announcements come the ramifications, the ripple effects on other key companies.

江森自控。

Johnson Controls.

看看吧。

Check out.

我们下跌了8%，但这家是一家冷却设备公司，从很多方面来看，它关注的是如何为英伟达的芯片降温。

We're off by 8%, but this is a cooling and and and several equipment company in many ways that's looking at the ways in which you cool down NVIDIA's chips.

也许未来你不需要用空气来实现这一点。

Well, maybe you won't need air to do that in the future.

也许他们能用水冷技术来冷却这些芯片，这似乎让一些供应商不寒而栗。

Maybe they're able to cool them with water cooling, and that sent some shivers down some suppliers' spines, it would seem.

我们下跌了8.5%。

We're off by 8.5%.

一些市场人士表示，这有点反应过度，巴克莱银行如此认为。

Some of the market's saying there's a bit of overreaction, so says Barclays.

特斯拉下跌4%。

Tesla off by 4%.

这是否源于对英伟达正在构建的自动驾驶平台的担忧？

Is this the concern around a self driving platform being built by NVIDIA?

埃隆·马斯克昨天在X上发帖称，他对此并不担心，睡得安稳。

Well, Elon Musk posted on X yesterday that it's knocking he's not losing any sleep over it.

但目前，特斯拉在昨日大涨后，确实有了一些卖出的理由。

But for now, there's a little bit of a reason to be selling Tesla after its rally yesterday.

闪迪上涨24%，这是因为我们听到，对内存和存储的需求依然令人难以置信。

SanDisk up 24%, and this is as we hear actually the still unbelievable need for memory and memory storage.

闪迪再次感受到了詹森·黄言论的连锁影响。

SanDisk once again managing to feel the ripple effects of Jensen Huang's words.

他昨天在台上表示，斯坦迪因此上涨，与其他内存公司如美光一同走高。

He said that on stage yesterday, and Stanisq rallies higher along with other memory companies such as Micron.

但让我们谈谈在这里CES上发布的其他公告。

But let's talk about other announcements being made here at CES.

其中一个是来自Lucid的，因为Lucid、Neuro和Uber正在将他们的机器人出租车愿景变为现实，在CES上揭幕了一款全新的自动驾驶汽车。

And among them is from Lucid because Lucid, Neuro, and Uber, they're bringing their robotaxi ambitions to life, unveiling a new autonomous vehicle right here at CES.

Lucid临时首席执行官马克·温特霍夫现在加入我们。

Mark Winterhoff, Lucid interim CEO, joins us now.

我很高兴能说，你们的机器人出租车有什么不同？

I'm very pleased to say, why is your robotaxi gonna be different?

我想是因为我们领先的电动汽车技术与Lucid Gravity所提供的豪华体验相结合。

Well, I guess it's the integration of, you know, our leading EV technology, the luxury experience that the Lucid Gravity provides.

再加上Neuro驾驶系统，你们能非常非常快地将产品推向市场，因为从我们三方合作开始，到计划在今年年底前推出付费服务，时间还不到十八个月。

With the neuro driver, you know, in bringing this to market very, very, very fast because I mean, from when we all came together, the three of us and to when we plan to roll it out by the end of the year in a in a paid service actually, it's less than eighteen months.

是的。

Yeah.

如果你们能在这么短的时间内完成这一切，这本身就已经非常独特了。

And if you do that in that short period of time, that's actually a very unique thing by itself.