唸博士真的有用嗎？｜AMD CEO 蘇姿丰給年輕人的職涯建議｜如何從 A- 學生變成半導體女王｜中英對照 聽力訓練

Um, first of all, President Chen, thank you so much for inviting me. Actually,

you invited me several times to come to NTEU and I was honored every time uh to come here, but sometimes when I come to Taiwan, it's very very busy and

especially around Computex timing, it's always very busy. So, this is actually a great time that uh I can visit and you know, thank you for the warm um indivmentmentation and professor Woo as

well. Thank you so much. It's really my

well. Thank you so much. It's really my pleasure to be here to meet uh all of you very talented students and you make

me feel very old but hopefully I can show I can um give you a few uh stories to um for your

career. So maybe I should say uh my

career. So maybe I should say uh my background was in electrical engineering and computer science. So that's where I went to school for um uh MIT for

bachelor's, masters and PhD and my specialty was semiconductor devices. So

when I was a a young student that is what I um I studied now for AMD, I've been in the semiconductor industry for

over 30 years and what you find is that the technology moves so fast. So for um AMD, I've been with the company since uh

2012 and CEO for uh since 2014. And

probably the most important thing with um any company is to decide uh what do you want to be when you grow up. And for

AMD uh my uh belief is that our specialty is in high performance computing. And high performance

computing. And high performance computing is really at the center of um all of our lives today. I'd like to say that AMD's technology touches billions

of people every day. Whether you're

talking about uh cloud data centers or you're talking about um anywhere in the edge of the network or you're talking about PCs or automotive or industrial,

you probably can touch um some of AMD's technology. But from a technology

technology. But from a technology company, the most important thing is you have to have a long-term strategy and roadmap. And so our uh focus was really

roadmap. And so our uh focus was really on three things. Um first is high performance computing. Um the second is

performance computing. Um the second is to have uh really excellent partnerships. So uh uh TSMC is at the

partnerships. So uh uh TSMC is at the center of um our partnership um overall.

And then the third p piece is that the semiconductor industry is changing and um many of you have heard that Moore's

law is slowing down and changing and uh we made a big bet on triplet technology as a very key area going forward. So

these are the some of the things that um were important for AMD and today um I'm very happy to say that when you look at

the top supercomputers in the world uh we have uh five out of the top 10 supercomputers are based on AMD's technology and from that standpoint um I

really feel that high performance computing is at the center of what's most important uh going forward. So I

heard that many um many students still view that chip design is the most exciting thing. I would say that there's

exciting thing. I would say that there's always a uh movement. Sometimes hardware

is more exciting. Sometimes software is more exciting. I think from my

more exciting. I think from my standpoint the most exciting part of design is actually holistic design. So

you have to really design uh hardware and software and systems together.

Especially when we're talking about AI going forward, you cannot design each piece separately. You actually have to

piece separately. You actually have to think about what is the overall system application especially thinking about um the uh the algorithms and the applications and then you can think

about how do you want to design your chips. Yeah. The one thing I can say

chips. Yeah. The one thing I can say about AI industry is uh every week there is uh new uh technology and innovation

that is coming out and I think deepseek was one of those areas when it came out in January. Uh I think there was some

in January. Uh I think there was some surprise. Yeah, there was some surprise

surprise. Yeah, there was some surprise and when you look at what they've done, it's actually, you know, pretty clever technology in terms of uh really gathering all of the different

innovations using the mixture of expert models and so on and so forth. But the

main thing that we saw with deepseek is when you can find a way to make AI more accessible and actually uh cheaper,

you actually will have more usage and application of AI. So the main thing that DeepS has done is it really has

accelerated the application of AI. Okay,

of course that's not a market statement.

That's a a technology statement. From

everything that we see, the key point about AI is that there's no one-sizefits-all for uh the the

computing uh ecosystem. Uh you have the very large frontier models where uh the few people who can invest at that

level, hundreds of billions of dollars, are going to go as fast as possible because they want to build the best model. But you also find that that is um

model. But you also find that that is um just one application of AI. You also see there will be many different um

medium-siz models, smaller models and um you will see that uh people will use different um algorithms to get there. So

my expectation is that there will continue to be a very significant investment including very large investments as well as more affordable investments going forward. So the most

important thing I can say about AI is that the market is huge. We expect that the market over the next three or four

years will be uh 500 billion. Yeah. Uh

area and in that type of market you need all different kinds of computing.

There's no one company that has all of the solutions. Actually uh my view is

the solutions. Actually uh my view is that you need every kind of computing.

you need CPUs, you need GPUs, you need AS6, you need FPGAAS. Uh and um in that framework, I think our AI strategy is

number one to have a very complete solution. So complete solution includes

solution. So complete solution includes hardware, software and um systems. And the second piece is to have a open

ecosystem. One of the most important

ecosystem. One of the most important things is to foster collaboration and to have an open ecosystem and uh that way people can learn from each other. That

was also another important uh part about deepseek is it was an open uh model and as a result many people uh innovated on top of that. So my answer to your

question is uh the uh the main thing to say is the market is huge and our um AI is throughout every aspect of our business including the largest data

centers to uh even at the PC level. Uh

we believe that there will be um a lot of AI. So there's a lot of market for

of AI. So there's a lot of market for all of us. So I think the Taiwan ecosystem is very special and when you look at the Taiwan ecosystem, you know,

starting from the TSMC uh ecosystem which uh really has built this uh incredible manufacturing capability and then including all of the um you know

ODM, OD uh OEM ecosystem, manufacturing ecosystem, design ecosystem and I think more investment on the software and system side uh will come uh from an AMD standpoint, you know,

we're very uh you know, proud of our team uh here in Taiwan is very important piece of our overall R&D. But the other thing I would say is in this world

today, the ecosystems are very interconnected and um from a United States standpoint, I think we uh we do rely on a lot of the the Taiwan

manufacturing ecosystem and trying to build resilience into that ecosystem is important. I think happy to say that you

important. I think happy to say that you know I was born in Tynan and my uh my parents uh immigrated to the United States and as you can imagine uh that

was a very big move uh for them. Yeah.

And when we went to the US as a family I think the the key part that my parents always instilled upon me is uh education is so important. So for all of you in

school, I remember when I was in school at um MIT, I very much wanted to finish school. So um after my bachelor's

school. So um after my bachelor's degree, I actually thought I should um finish school and go to work. And my

parents said, uh no, Lisa, you cannot do that. They uh strongly strongly

that. They uh strongly strongly suggested I get a PhD.

And at the time I was very impatient, you know, very impatient. I was like, "Oh, school is enough. It's

enough. Maybe some of you feel the same way."

way." And the truth is uh my parents were right because uh what you get from uh your university and your graduate school

is uh really an opportunity to learn many different things and to uh build the confidence to um to really solve the

future problems. So even though I was uh not happy with my parents at the time, I think they were absolutely right and um

as uh the uh experience to be CEO of AMD, I would say it was kind of my dream job to start from a semiconductor

engineer to become a business uh person and then become a um semiconductor leader. Uh but it was not easy

leader. Uh but it was not easy especially in those early years. Yes.

And uh my you know parents and family have always been uh very uh very supportive of that and um I'm very lucky in that way. I see. My thinking is that

uh there are many different kind of leaders. Many different kind of leaders

leaders. Many different kind of leaders and we have a lot of respect for all kinds of leaders. But I truly believe that the university education is so

helpful. Like these years that you're in

helpful. Like these years that you're in the university is really the best time uh to learn and to decide what you enjoy the most so that uh you can decide your

next step in your career. Okay. First of

all, I have a tremendous respect for Jensen. Okay. Tremendous respect. I

Jensen. Okay. Tremendous respect. I

think as a founder of Nvidia, he has um you know really built an amazing company. Um I would say that we run

company. Um I would say that we run independent companies though. So uh

probably um as a semiconductor industry uh we do have a semiconductor industry association in the US and for where all semiconductor companies are there and

you know together in that forum we would exchange some ideas. So I can say for um in my experience I had a one piece of

probably the best advice that I received from a mentor was to tell me you know Lisa whatever you do uh I would suggest that

you pick the hardest problems. So uh the advice they gave me is to run towards problems and actually I think that's the best advice okay that I can give uh

students because uh you know there's so many interesting things to do in the world choosing a project or a uh a problem or a company where you can make

a big impact uh you will learn so much and be able to contribute so much and I can say for sure that I learned the most from the biggest mistakes I made. So

mistakes are not a bad thing. Mistakes

are actually a good thing because you can learn from those lessons and uh be uh you know better in the future.

Political science is not not the easiest question. You know technology is

question. You know technology is actually a little easier.

Uh uh I think I would maybe not talk about the individual countries but I would talk to you a little bit about my thinking on um on AI. uh because AI is

such a new technology when you know if you think about the last few years uh we're all learning together and countries are learning companies are

learning individuals are learning uh the initial thought process is AI could be um quite dangerous if it's not fully uh

if they're not good guard rails. Um

however you also have on the other side you don't want to stop innovation and from my standpoint I think AI is the type of technology where we have to

innovate and go as fast as possible and understand that there does have to be regulation but not too much regulation so that you slow down the progress. So

uh in general that's um that's my philosophy and I think when you look across the world I think people are recognizing that people are recognizing

uh it's a very dynamic time so every you know I said every week sometimes every day something changes but generally speaking uh I think countries that

embrace technology are going to really benefit from it and uh many countries around the world are all building their AI policies. I think everybody should be

AI policies. I think everybody should be educated in AI. It doesn't matter uh which department that you're in. Um when

you say you're in the literature and linguistics department, I think AI can also be very uh you know beneficial to understand how to apply AI in your um in

your field. And most of what we see

your field. And most of what we see today is uh I think as I said people are still learning very much what you can use um AI for in in each field. So we

talk a lot about the development of AI but there's also a whole field on the application of AI you know going forward. So I would certainly encourage

forward. So I would certainly encourage um all students uh to do that. I'm not

so sure that I can say that but I I will say that I think Taiwanese um Americans and and Taiwanese in general are um you

know extremely uh hardworking. Yeah. and

uh really thinking about how to contribute to uh their fields and their professions. And I think that's

professions. And I think that's something that we can be very proud of.

Yeah. Although some people say that I'm more American than Taiwanese, but I'd like to think I'm Taiwanese American.

So probably the most difficult time for AMD was when we had to uh really um when

I started it was in 2014 time frame and at that time you know uh the company had have a road map that was not

competitive. So if you're in a

competitive. So if you're in a technology company you must have your product is the most important thing. I

would say that's the number one thing and so our goal was to really uh reframe our product roadmap and ensure that you know I had um three priorities. Uh the

first one was to build great products.

Uh the second was to have very very deep customer relationships because I think uh together in this world uh there's as I said earlier there's no one company that has all of the good ideas. Uh

collaboration is very important. And the

third is to be uh very agile and move uh very quickly. So in that standpoint um

very quickly. So in that standpoint um our biggest job was to give ourselves time to build this uh great road map and

we are um now I would say our uh CPU road map is based on our Zen technology.

We're now on the fifth generation of Zen technology. I can say we went from let's

technology. I can say we went from let's call it uh not the best to today I feel that we have the best uh CPU technology

um in the world and uh we have um also um a lot of a broader road map um going forward so I think the key is in those times to be very clear about what you're

trying to achieve and for us it was always in a technology company is to build great products that's the number one foundation so in terms of the semiconductor trends Um I think there

are a few key trends. Uh I would say we start with the idea that we have to have a very holistic design environment. So

hardware software um uh systems all need to be optimized together. Um probably one of the most

together. Um probably one of the most exciting parts about AI is we're still in the very early innings. So I would say you know we still have probably 10

years of very significant innovation in AI and on the application space we're going to go from uh today we talk about

AI making our uh businesses more productive or making our people more productive as we go forward over the next many years I think we can see um

much larger changes in terms of what AI can do for us. uh probably one of my most exciting areas or most interesting areas is in uh health care and life

sciences. So I know that there's a

sciences. So I know that there's a medical school here and um the application of AI together with medicine I think can help us uh solve things like

uh cure diseases and find new uh drug therapeutics um also help diagnose uh the you know future um health problems much sooner. So that's just one example

much sooner. So that's just one example of what AI I think can do uh going forward. I would say the following. I

forward. I would say the following. I

would say that I would encourage you to be uh very ambitious in what you would like to achieve. So the most important

thing for um you know either young students or early in your career is you know you can do anything. Okay. So have

a very big dreams and um if you tell somebody or tell people tell your um uh your colleagues or your managers or your

you know your future uh mentors uh people will help you to achieve those um those opportunities. So that's one of

those opportunities. So that's one of the things I would say because sometimes uh I would say for Taiwanese or Asian uh people we are sometimes a little shy.

Well, I would say that um when you have the right people, the right people are very important because not everyone has

the same motivation. Uh some people are motivated um you know for their project.

Some people are motivated uh to start a company. Some people may be motivated

company. Some people may be motivated for other reasons. Um at AMD we try to have the right people to build that team

and for our engineers um you know back to this question when we started uh we wanted to build the best that was our um

ambition. Uh but we also thought we

ambition. Uh but we also thought we should be very uh realistic that you cannot go overnight from um uh not

competitive to be the best. So you have to build a plan to to do that and um I think that's what uh we were able to do.

But engineers basically love to solve problems. Okay, we have to give them good problems to solve. I think we're all motivated by seeing success and

success comes in different ways. Success

comes in recognition of the product.

Success comes in uh you know great people wanting to join the company and yes success comes in uh revenue growth which leads to uh market share growth

which leads to stock price growth. But

my personal opinion is uh a PhD is actually pretty useful. That's my

personal opinion because you get to work on some problem you know during your PhD thesis you work on some problem that

should nobody else is really working on in in the world and that gives you a lot of opportunity to explore much broadly

than sometimes in a company you become much more focused so I actually think PhD is pretty useful but it's different for every every person the top you said

soft skills. So I would say um a few

soft skills. So I would say um a few things. Uh one is you know you really

things. Uh one is you know you really have to think that the the world changes very fast. So it's

important to be actually quite agile in your thinking. Maybe today you think

your thinking. Maybe today you think this is the right path but you will learn something every day. And um I have a big philosophy that says uh you need to continuously learn all the time. Like

every day you're going to learn something new and you can decide how you might change your path. That's um that's something that uh I use very much and I

would suggest that that's an important um you call it soft skill. And then to your second question about um how do you deal with setbacks? I said earlier, you

actually learn the most when you have a setback, when you have a failure, when you have something like I can tell you so many examples of my uh my setbacks,

but my advice to you is uh you can be sad for just a short time. Of course, we're all, you know,

time. Of course, we're all, you know, human beings, right? So if you have something that you're working on that didn't go the way you wanted to, you might be a little bit unhappy, uh but

only for a very short time and then after that uh you learn and you get uh uh really focused on your next thing.

And I think that's the the key piece I can say for sure I'm just a okay student. Uh, okay.

student. Uh, okay.

Meaning pretty good, but not A++, you know, maybe like, you know, sometimes A minus, you know, that kind of I I actually think everyone

is different. Everyone is different and

is different. Everyone is different and the most important thing is uh what take whatever you learn to apply

it in the real world. So I would say that's what I learned the most even uh as my device physics was very very hard class. I don't know if any of you are

class. I don't know if any of you are taking device physics class. So I

thought the problem the written problem is very hard but working in the lab on a project I was really good and so

everyone has their specialty. Uh let me say one thing I'm a big believer in it's not the number of people that you have for anything for you know, if you think

about a company or if you think about a project, it's not the number of people that you have, it's do you have the right people uh to do the the things.

So, when um when I first started at AMD, we had 8,000 people. Uh people used to say, "Oh my goodness, you're so small compared to Intel." Actually, at that

time, I think Intel maybe had 10 times the people Yeah. of AMD. And they said, "Lisa, you're crazy. Why are you doing that? you cannot possibly win. And I

that? you cannot possibly win. And I

said, really? Why is why do you say that? I I actually don't feel that way.

that? I I actually don't feel that way.

I don't feel that it's a number of people. I feel that it's the right

people. I feel that it's the right people with the right project with the right focus and the right vision. And to

the to the point about how do you optimize? I think everyone has, you

optimize? I think everyone has, you know, in your mind uh what you can see as your path and uh every company's path is different. So I think from u my

is different. So I think from u my standpoint uh and you know maybe since this is the last question I will make it a little bit broader. We are incredibly

lucky to be in this technology field. I

hope all of you feel this way because you know we're working on things. What I

really enjoy about um what we're doing in technology is the technology that we're working on is touching billions of people every day. It is having an impact

on people's lives. It will make companies better. It will make your life

companies better. It will make your life better. It will make our uh our future

better. It will make our uh our future um health care better. It will make our future um sustainability better. from

all of those aspects we're very lucky to work on this type of technology and so we have to find you know what is it that we're going to do best and for us at AMD

it's around high performance computing I don't believe that um we can do everything so but I do believe that in high performance

computing uh it can have a a very uh wonderful impact on people's lives and you know that's what we uh we choose to is