I'm really annoyed with giving speeches. Jensen Huang reveals that NVIDIA has 61 CEOs and never fires employees who make mistakes: CEOs are the most vulnerable group.

"There won't be a second CEO like me in the world." Recently, Jensen Huang, the co-founder and CEO of NVIDIA, said this in a private interview.

It is reported that this in - depth conversation has been in the works for thirty years, revealing the little - known side of Jensen Huang to the public. The host, Jodi Shelton, first crossed paths with Huang in her professional life over thirty years ago, when the graphics processing unit (GPU) had not yet sparked the global AI revolution. From the origin of accelerated computing to the prospects of generative AI, this conversation is like a masterclass in foresight.

During the interview, Huang said that in a sense, NVIDIA actually has 61 "CEOs". Over the years, many people, including himself, have made serious mistakes, but at NVIDIA, no one has ever been fired for making a mistake. "We've created a safe enough environment." He also revealed that the position of CEO is far more fragile than people think. "In fact, we might be the most vulnerable group in the company. But for me, it's not difficult to admit this vulnerability."

Interestingly, he mentioned that in many ways, he is a "reluctant CEO". "Public speaking scares me to death. I'd rather stay within the company than be in the spotlight outside; I'd rather work quietly than give speeches; I even don't like giving keynote speeches at all, but I have to do these things for the company."

In addition, Huang said that NVIDIA's success is not achieved by high production volume. "Although NVIDIA invented the GPU, in terms of production volume, we are actually the smallest GPU manufacturer in the world. Many unknown manufacturers produce more GPUs than we do." And the fact that "there is no ultimate goal" has really played a crucial role in NVIDIA's development.

Regarding the world five years from now, Huang asserted that NVIDIA's and the entire industry's investment in the AI field will surely completely change the operation mode of computers. Future computers will evolve from being "programmed by humans" to "learning to program autonomously under human guidance". Moreover, 100% of job positions will change, but 50% of the jobs will not disappear. The future trend will not be a decrease in job opportunities; instead, people will become busier than they are now. And those who are currently unemployed may well find a means of livelihood because of AI.

Netizens have commented on Huang after this interview, saying, "I've never seen him so candid and straightforward. It's incredible."

The following is the detailed content of the conversation. We have translated and edited it without changing the original meaning for our readers.

"It took a full 33 years to see the results"

Jodi Shelton: The public is really curious about people like you. After all, you are defining the future of technology, and the future of technology is the future of the whole world. So what we want to do is to explore your personal experiences behind the halo of success and the values that have supported you to get to where you are today. What do you think of this positioning?

Jensen Huang: To be honest, I don't like it.

Jodi Shelton: Really? But you're a celebrity now, and everyone wants to know the stories of celebrities.

Jensen Huang: I never think of myself as a celebrity, and I'm not one at all. I just happen to be in charge of a very important company, the CEO of one of the most successful technology companies in history. A long time ago, we made some right decisions. Back in 1993, we were determined to reshape the computing industry, and we had our own unique insights into computer architecture. For a long time, this view was not well - received and was even quite controversial. You know, at that time, the entire industry's focus was on microprocessors and CPUs. By the way, that's when I met you. We met at the end of 1993 or in 1994, right? Since then, NVIDIA has been doing what we're still doing now: reshaping computing.

Jodi Shelton: Yes, I remember it clearly. At that time, in Silicon Valley, it was the era when CPUs reigned supreme, Moore's Law was in full swing, and the personal computer revolution was in full swing.

Jensen Huang: Yes. And all of our early customers were startups in the PC chipset field. These companies were the founders of the glorious semiconductor industry, like Cirrus Logic, S3 Graphics, Western Digital, Trident Microsystems. Do you still remember these names?

Jodi Shelton: Of course, I do.

Jensen Huang: These companies can be regarded as NVIDIA's "predecessors". And now, we're still on this path, committed to creating a new computing model. It took us a full 33 years to see the results. I just happened to become the CEO of this company, that's all.

Jodi Shelton: Maybe for you, all this came naturally, but for the whole world, NVIDIA's rise was like a bolt from the blue. Since around November 2023, the global technology landscape has changed because of you. How do you view this transformation?

Jensen Huang: You know, if you want to create the future, you have to be in the future before it arrives. To be honest, from the moment we invented CUDA technology and launched related products, we've been on the road to the future. What I'm most proud of about NVIDIA is that we're not only good at technological invention but also at turning technology into products and bringing them to the market. There are so many companies, researchers, and inventors in the world. They do create advanced technologies, but in the end, they often can only sigh, "I invented this technology a long time ago" or "I had this idea a long time ago". Every time I hear such words, I feel sorry for them. These excellent inventors unfortunately didn't meet equally excellent product innovators.

A product innovator is someone who can turn a technological invention into a mature product that can be brought to the market. And that's not enough. You also have to formulate a precise market strategy for the product and even cultivate a brand - new market yourself so that the market can accept the products you've developed and the strategies you've formulated. NVIDIA is such a company. We have the full - chain capabilities of technological invention, product innovation, strategy formulation, ecosystem building, and even market cultivation, and we've successfully achieved this many times. So for me, this state of "being in the future" has lasted for a long time.

Jodi Shelton: That's true.

Jensen Huang: A long time ago, we had a strategy that we don't talk about much now, called "CUDA everywhere". Many people have heard the story of me promoting CUDA everywhere back then, visiting universities, startups, and established companies. Sometimes, there were only three people in the audience, but I would still take out my laptop and demonstrate CUDA to them, telling them how this technology would change the world. I visited countless research institutions and laboratories and attended countless industry conferences. I've probably promoted CUDA more times than anyone else in the world. For a long time, I've been immersed in this "future scenario". After telling so many stories, I even had an illusion that "the future has already arrived".

Jodi Shelton: I really have that feeling.

Jensen Huang: So now, seeing all this become a reality still makes me very happy. And in my opinion, all this is actually not a surprise because what supports NVIDIA's development is the most fundamental underlying logic in the field of computer science, not just a momentary intuition or subjective preference. In many ways, today's results are an inevitability. But I want to say that when you increase the speed of something by a thousand times, or expand its scale by a thousand times, or reduce its volume by a thousand times, no matter what that thing is originally, it will undergo a qualitative leap. And the result of this qualitative change is often beyond imagination.

We've long foreseen that deep - learning technology has great expansion potential, which is why we've bet the whole company on this field. We knew that AlexNet (a deep convolutional neural network) would not be the end of deep learning. This technical architecture is inherently highly scalable, and with the world's massive data resources, the explosion of deep learning was a natural thing. But I also knew at that time that there was a technology that would become an obstacle on our way forward, that is, unsupervised learning, or self - supervised learning, which means allowing computers to get rid of the shackles of manually labeled data and achieve autonomous learning. Because the efficiency of manually labeling data would sooner or later become a bottleneck for technological development. And when the unsupervised learning technology made a breakthrough, I knew that our era had come.

Not long ago, at an investor roadshow, someone told me that I had clearly mentioned this "qualitative change" at that time. If you go back and look at the earnings conference calls at that time, you'll find that whenever I talked about technology topics that were crucial to the world, I would explain this point very clearly. At every investor roadshow and every occasion where I gave a speech, I would emphasize this view. Now, the unsupervised learning technology has indeed made a major breakthrough, and the scale effect of deep learning has been completely released. Only then have we really entered the fast lane of development. But even so, the problems that this technology can solve today still surprise me. We've long anticipated that there would be a qualitative change in technology and a transformation in the computing platform, but we didn't expect that the results of the transformation would be so fruitful.

Now we can interpret the "language" of proteins, the "language" of cells, the "language" of quantum, and understand all kinds of manifestations of everything in the world. The way we used to describe information is now being completely reshaped. From geometric shapes and textures to today's 3D Gaussians and 3D point clouds, the presentation forms of information are changing with each passing day. It's like humans have suddenly become extremely intelligent, and even the English language system has changed accordingly. We no longer use the old words, grammar, and sentence patterns because our wisdom has evolved to a new dimension, and we can communicate in a brand - new way. Maybe in the future, human communication will become simple "beeping" signals. This reminds me of the scene in the movie Arrival, where humans suddenly start communicating with abstract graphics and can transmit a huge amount of information just through graphics, achieving deeper and more efficient communication.

The most incredible thing is that many of the problems we're solving now were completely unimaginable in the past, and the speed at which we solve problems is also much faster than before. In the past, we often talked about Moore's Law, and now NVIDIA's development speed can be described by the "NVIDIA Law", which is a thousand times faster than before. The next ten years will surely be a magnificent decade, just thinking about it makes me extremely excited.

Jodi Shelton: To do what you've done, to be able to foresee the future and firmly believe that it will come, you need such strong self - confidence. As you said before, we met in 1994, and over the years, you've always been like this.

Jensen Huang: Yes, I remember it clearly.

Jodi Shelton: I was in my twenties at that time. You were probably a bit older than me, right?

Jensen Huang: I was about 29 years old at that time, almost 30.

"NVIDIA has 61 'CEOs', and no one has ever been fired for making a mistake"

Jodi Shelton: I still remember the first time we met. I was interviewing you for a magazine. I asked you, "Jensen Huang, there's a high - turnover rate of talent in Silicon Valley. Many people come and go. Are you worried about this?" After all, many CEOs were complaining about this at that time. And you were only 29 or 30 years old then. You answered me like this: "NVIDIA is neither a church nor a prison. Those who want to come can come, and those who want to leave can leave." I was really shocked at that time and thought, "Who on earth is this person?" He's so young but has such self - confidence and wisdom. I also heard a similar story. When Morris Chang first met you, you said right away, "I'll be your biggest customer, or at least one of the biggest." His reaction was, "Wow, this young man is really bold." So I'm really curious, where did you get this self - confidence when you were so young?

Jensen Huang: Haha, you know, it's actually quite painful to know everything. I'm just joking. By the way, Morris Chang would be very happy if he knew that NVIDIA is now TSMC's biggest customer.

Jodi Shelton: Definitely. He would be proud of you.

Jensen Huang: I'm also proud of him. You know, during the personal computer revolution, NVIDIA was TSMC's biggest customer. Now, we've become their biggest customer again, and I'm very pleased about that. Back to the point, I think a person must firmly believe in what they believe. And this belief can't be based on hearsay. You can't just believe something because someone else said it. You have to think carefully, sort out the logic of why you believe in this thing, and break down this logic into reliable underlying principles. Then, you also need to regularly test these principles to ensure that the beliefs you hold and the actions you take are based on a solid foundation.

If this foundation is not stable or changes for some reason, it means that it may not be a real underlying principle. Maybe it's not anchored in physical laws or objective facts. Once this happens, you have to re - evaluate and then adjust your direction in time. I've always done it this way. Moreover, if you really believe in something, you should take action to achieve it. I've believed in what we've been doing since 1993, and this belief hasn't changed until today. Because of my firm belief, I keep reasoning and sorting out the logic in my mind. I keep reviewing past decisions and predicting future trends.

Just like yesterday, we had so many meetings. At each meeting, I would re - sort out the logic of our journey. You'll find that some of the past assumptions were correct, but some were wrong. It's because we're flexible enough to adjust our direction in time according to the actual situation that we've finally come this far. So, looking back and reviewing past decisions regularly is a very meaningful thing. It can help you better exercise your ability to reason forward. Because I've always insisted on doing this, I've always lived in the truth I've recognized. Even now, I still think of myself as just an employee of NVIDIA. I care very much about this company, and there are many people in the company who, like me, have deep feelings for it.

In a well - managed company, the role of the CEO is clearly defined. The CEO needs to report to the board of directors, and the board of directors is responsible to the shareholders. If the CEO's performance doesn't meet the board's expectations, no matter if there are 12, 13, or 15 members on the board, he will be fired. So, in fact, the CEO is also an employee in the company's organization. That's why I said that NVIDIA is neither a church where anyone can come in nor a prison where no one can leave. This mindset can keep you down - to - earth, humble, and enterprising because you have to work hard every day to be worthy of your job.

Sometimes, someone will ask me, "Jensen Huang, do you love your job?" I'll tell them that I don't love it every day, but I'll do my best to do it well every day. I think this attitude comes from two aspects. First, I firmly believe that I'm the best person for this job. Second, I have to work hard every day to be worthy of this identity.

Jodi Shelton: In everyone's eyes, you're the synonym for NVIDIA, and NVIDIA is you. Over the years, you've been deeply bound to this company.

Jensen Huang: I'm probably the person who has been photographed the most within NVIDIA.

Jodi Shelton: That's right. But if there's a new CEO in the future, can this person really take over from you?

Jensen Huang: There won't be a second CEO like me in the world. The reason is simple. I was cultivated by this company step by step. When we first founded NVIDIA, I knew nothing about how to be a CEO, how to make strategic plans, how to develop products, or how to start a brand - new industry. I only knew how to raise funds but didn't know how to communicate with shareholders, understand the thoughts of shareholders, policymakers, leaders of various countries, and corporate managers, or how to understand the mindset of employees and build corporate culture. I couldn't even accurately define what the term "corporate culture" meant, let alone formulate the company's strategy. That was my real state on the first day I took over the job. And in the past 33 years, I've become proficient in these fields step by step.

If there's anyone in the world who can be called a "master of corporate strategy" or an "industry pioneer", it's probably a short guy like me. I've devoted my entire career to learning these abilities, and I'm a good student. In addition, my level of commitment to this job and my deep feelings for it are hard to replicate through recruitment. In my heart, NVIDIA is like my child, and I've poured all my efforts into it. My family has also accompanied me and contributed to