Source: Sequoia Capital

Translation: Yuliya, PANews

Editor's note: In the past, our data centers only stored files for humans to retrieve; now, computing is moving towards generation, where every word, every image, every segment of video is produced in real-time and highly customized based on the requester's context. In this global wave, Sequoia Capital partner Konstantine Buhler has had an in-depth conversation with NVIDIA founder and CEO Jensen Huang about the significant changes in computing technology. Huang believes that automation does not bring unemployment but a comprehensive enhancement of labor demands and the elevation of professions themselves; people will not lose their jobs because of AI, but may be replaced by those who are adept at using AI.

AI Factories and Generational Leap in Computing Models: From Retrieval to Generation

Konstantine: Thank you very much for being here, Jensen. We are in the midst of a massive AI revolution, the scale and speed of which may even surpass that of the industrial revolution. You have mentioned that what is happening now is the largest infrastructure build-up in human history. At the center of this build-up are AI factories, and the company empowering all of this is NVIDIA. Can you tell us what an AI factory is? Why is it the most worthwhile investment for all companies in the next decade?

Jensen Huang: You can understand AI in many ways. The most familiar one to the public might be interacting through web browsers and chatbots: you give it a prompt, and it responds with a paragraph. Even if you’ve been using AI for a while, you'll notice that its capabilities have significantly evolved in the past two to three years.

Two years ago, people heard about ChatGPT. It’s essentially a software that can understand the information you input. It can perceive, understand information, and transform and generate it into other content. For example, you can give it a PDF file and have it summarize, which is text to text; you can have it generate an image based on a story, which is text to image; or give it a photo and ask it to describe the picture, which is image to text. This capability was referred to as generative AI two years ago.

But beyond understanding and generating, what is more valuable is the ability to think. The foundation of generative AI grants it the capacity for intrinsic thinking, step-by-step reasoning, and problem-solving. Moreover, it can now generate control commands to use tools, whether using digital tools like browsers, spreadsheets, Photoshop, AutoCAD, or in the future controlling mechanical systems (which includes robotics and autonomous vehicles).

Two years ago, people thought ChatGPT was fun and could write poems and songs, but it occasionally made nonsensical statements; today, we have agentic systems. AI is no longer just about understanding information; it can reason and do useful work now. Because it can do useful work, AI has created genuine commercial value. We won’t pay for friends who can only bloviate, but we will pay for those who can actually get the job done. Now, every day there are people hiring AI by the hour, paying it $20 to $30 an hour. This is also why it has become the fastest-growing software business in human history.

From the upstream logic of the industry, we need to return to first principles. The basic concept of the computer industry we know today was established about 64 years ago. At that time, IBM launched the System/360, which is why IBM became the most valuable company in the world at that time.

In the past 60 years, the essence of computing has been pre-recording and retrieval: you write the story, take pictures, record videos, save them as files in a hard drive; when you want to use them, you retrieve them from the hard drive. This is also why those buildings are called data centers. They simply store data and don’t do much computation.

But the situation has changed. In the AI era, every time you provide new background information (context) and new requests, AI will perform real-time understanding, reasoning, and generate entirely new results. For example, my current speech is generated in real-time based on the different backgrounds of everyone present, not just reading from a script. This is called intelligence.

In the future, every pixel, every sound, every segment of video, even every advertisement and news, will be fully customized and generated for you, rather than pre-recorded for retrieval. This means that in the future, we will need a large number of generators, which are large computers we are building; this is the AI factory.

The Intelligent Network Enveloping the Earth and the Generator of the Digital Age

Konstantine: How large will this generator be?

Jensen Huang: Currently, we provide information and intelligent generation for about a billion people globally. But since AI has turned into agents, they can work by themselves; one agent can even communicate and collaborate with another agent. Within NVIDIA, there may be hundreds or thousands of agents talking to each other, solving problems (of course, they all operate under safe sandboxes and protective barriers).

This means that in the future, not only will humans be using the internet, but there may also be hundreds of billions of agents working non-stop online. Corporate agents, autonomous vehicles, robots, even the systems in every building, will all be communicating with each other. All commands, all thoughts, will be generated in real-time.

It’s like there is a thick computing network, wrapping the entire Earth like a cocoon. This sounds exaggerated, but this has actually happened twice in history:

• The first time was 300 years ago when Siemens in Germany created a machine. When you ignite it, it outputs a powerful invisible force, which is electricity. Today, the power generation network (the grid) has enveloped the whole Earth.

• The second time was 35 years ago when the internet was born in the United States, and it similarly envelops global communications.

Now, we welcome the third network after energy and communication: the intelligent network. The business that NVIDIA relies on to survive is to build this new era's generator (Dynamo). The generator from 300 years ago used physical movements of water flow, wind power, or coal (atoms) as input and output electrons; whereas our NVIDIA machines input electrons (electricity) and output digits. These digits, through various combinations, become languages, mathematics, or languages of proteins and human biology, physical laws, climate predictions, and even the language of the 3D world, robotics, and autonomous driving.

These two machines, 300 years apart, serve similar purposes: atoms in, electrons out; electrons in, digits out. These digits are what we refer to as Tokens, which are intelligence. We mass-produce these intelligent Tokens in the factory; that is the significance of the AI factory.

Konstantine: We are amidst a wave of intersecting revolutions. From energy transformation, global telecommunications network routers, to the core of the intelligent revolution represented by GPUs and AI factories, like the H100 or the latest Vera Rubin architecture. Integrating everything needed together.

Jensen Huang: Yes, our computing units are called “racks.” Each rack contains 72 chips. This year, we are going to make about 8 million such components. A single rack weighs 2 tons and is valued at $4 million, containing 1.5 million parts. It is the most expensive equipment in the world, but we are mass-producing them like mobile phones, shipping them to data centers worldwide. These things are big, moving them is definitely hard work.

The Five-Layer Cake Investment Logic of the AI Era

Konstantine: This is an exciting picture. How can both large enterprises and individuals participate in this revolution?

Jensen Huang: When investing in the AI industry, you can think of its industrial layout as a five-layer cake. You know, a $50 billion AI factory can generate $300 to $400 billion worth of intelligence, and its return on investment is astonishing. So what are these five layers?

The first layer is Energy: This is the fundamental generator at the bottom layer. It is the largest growth opportunity in the energy industry for generations. To support computing, sustainable energy (nuclear, wind, solar, hydrogen, etc.) will receive substantial funding. As long as it can generate energy, it will get investment. This is why companies like Siemens, Mitsubishi, and GE Vernova are performing so well right now.

The second layer is Chips and Computing Facilities: This includes chips, computers, networking equipment, switches, and silicon photonics technology.

The third layer is Infrastructure: This includes land, electricity, building shells, funding, and the day-to-day operations of data centers. Currently, these resources are in extremely short supply.

The fourth layer is Models: These are large models built on cloud infrastructure. This is the most investment-intensive area driven by the market in human history. Well-known examples include OpenAI and Anthropic. But remember, AI can learn not just natural language; it can learn any structured thing. We are learning the laws of the physical world—for example, when I sat down earlier, I was very confident, not because I had a 47% chance of falling through the chair, but because I was 100% confident in the laws of physics. AI can similarly learn the meaning of proteins, the significance of genes, and the function of cells. The industrial scale of the physical entity world reaches $80 trillion, a frontier area that is currently under-discussed but extremely important.

The fifth layer is Applications: Based on the underlying technologies, countless startups are reshaping industries like financial services, law, accounting, transportation, and logistics. Last year, venture capital invested $100 billion in this top layer, the highest in human history for VC investments.

This future will be tremendously vast. We are just at the starting stage; this year, an estimated $1 trillion will be invested in this ecosystem. But I estimate that AI will become a massive ecosystem with an annual output value of $20 trillion. How important is intelligence? Who needs intelligence? How much do you need? Once you figure this out, you’ll know the direction for investment.

AI is not here to take jobs; it is here to elevate you

Konstantine: This is not just a multi-trillion-dollar market opportunity; the explosion in hardware facilities and applications also means creating a vast number of real jobs for humanity.

Jensen Huang: Absolutely right, and we must emphasize this point. Right now, every country and culture has a different attitude toward AI. But I sincerely suggest everyone: beware of the plots in Hollywood sci-fi movies. Don’t keep listening to people saying “the Terminator is coming,” “the technological singularity has arrived,” “there is a 20% chance that AI will destroy humanity.” That’s complete nonsense.

Some even threaten, “We don’t even know how AI works, it’s too mysterious; maybe it will just walk away by itself tomorrow.” This is even more absurd. AI is just computers and software; engineers certainly know how it works; otherwise, how could they keep making it safer and smarter every year?

The current AI has significantly reduced hallucinations, and the knowledge it generates is accurate and context-aware; when it encounters something it doesn’t understand, it will even look it up. Before answering you, it may even self-reflect, providing a few options for comparison before telling you the answer. Just like today’s cars are much safer than they were a hundred years ago, the tech industry is doing its utmost to make AI extremely safe.

So focus your attention on the certainties. I can say with great certainty: you may not lose your job because of AI, but you will definitely lose your job to someone who uses AI.

Since this is a technology that can empower humanity with super abilities, you should use it! Whether it’s to inform your loved ones, your children, your company, or your country: definitely embrace AI.

Konstantine: But when it comes to jobs, people are indeed very anxious.

Jensen Huang: I get upset when I hear people creating panic over job issues. This year, we invested $1 trillion in this ecosystem—energy, chips, infrastructure, model layer, application layer—all creating far more jobs than ever before.

Some may ask, what about traditional jobs? Here’s a common cognitive mistake: they confuse “jobs” with “tasks.”

For example, I am a CEO. My daily “tasks” mainly involve typing and speaking. Now AI is much better at typing and speaking than I am, at a superhuman level; but as a CEO, I am actually busier now than before.

Let me give you a deeper example. About 12 years ago, a top computer scientist warned everyone that computer vision for medical imaging never gets tired, never misses a detail, and has reached superhuman levels. He asserted that the first profession to be wiped out by AI would be “radiologists” and advised everyone not to pursue that profession anymore.

He was completely correct in his technical judgment. Now, all radiology systems are integrated with computer vision, and all radiologists are using AI to aid their work. But what’s the outcome? The demand for radiologists worldwide has actually increased!

Why? Because the purpose of a radiologist is not just to look at images, but to diagnose diseases together with clinical doctors. Thanks to automation, their efficiency has greatly increased, allowing hospitals to take on more patients waiting in line, making radiology more profitable. Hospitals finding increased profits and more patients have gone on to hire more radiologists! Those who heard the warnings and didn’t study radiology ended up missing the opportunity.

Similarly, there are people saying that AI can write code now, and 90% of software programming is gone, so we don’t need software engineers anymore. But the reality is, we are hiring more software engineers now than ever before! Because the purpose of a software engineer is to solve problems and innovate, not to compete in typing speed. Writing code is just a task; solving problems is the core.

AI will not eliminate jobs; it will enhance the value of your work. If today I am a plumber, I might just be working according to blueprints; but with the help of AI tomorrow, I could also be a kitchen designer at the same time. If I am a furniture seller or carpenter, in the past you’d just expect me to nail wood together, but with AI, I can provide you a complete interior design scheme, making your home beautifully stunning. My professional skills are elevated!

So I believe the narrative that AI will lead to unemployment is completely wrong; it's just to scare people away so that others can profit from it. Throughout my entire career, computer technology has become increasingly complex. In the past, only 2% of people could master the C++ programming language (perhaps those attending this Silicon Valley venture capital circle understand more). Now, because of AI, as long as you understand human language, you can program. We have truly closed the technological gap for the first time; we must take everyone along into this new era.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。