Written by: Techub News Compilation

Introduction

In early 2025, NVIDIA became the highest-valued company in the world, with its founder and CEO Jensen Huang appearing on the cover of Time magazine as "Person of the Year" alongside AI architects. In a deep interview at NVIDIA's headquarters in the Bay Area, Huang addressed the ongoing global AI boom. NVIDIA's GPUs have almost monopolized the core computing power driving this revolution, but debates around market bubbles, geopolitical risks, and technological ethics are also intensifying. This dialogue takes place against the backdrop of Donald Trump's return to the White House and a new phase in US-China tech competition; Huang’s insights pertain not only to NVIDIA's future but also reflect the trajectory of the entire tech industry and the global landscape.

Summary

• The AI investment boom is not a bubble; it is fundamentally rooted in a paradigm shift from general computing to accelerated computing, and the current level of investment is "quite moderate" compared to the overall size of the computing market.
• The US and China are deeply interdependent in technology and economy; the notion of "decoupling" is flawed, and the wise approach is to manage this century-defining relationship well, avoiding conflict through negotiation, communication, and compromise.
• If the US wants to maintain its lead in the AI era, it must be determined to win at every layer of the technology stack (energy, chips, infrastructure, models, applications) and not allow the overall level to be constrained by its weakest link.
• Sales of advanced chips to markets like China can help ease tensions, prevent adversaries from feeling cornered, and enable the US technology stack to become the global standard, ultimately benefiting US national security and prosperity in the long run.
• AI will change all jobs; some positions will disappear, but it will create a large number of new jobs and significantly expand the global economy; individuals must learn to use AI, or they risk being eliminated.

The AI Boom: Bubble or Fundamental Change?

When asked if he agrees with Google CEO Sundar Pichai's view that there is an "irrational" element in the current AI stock market, Jensen Huang acknowledged that markets always have irrational aspects, but he believes people generally overlook a deeper reality.

He pointed out that the foundation of chatbot and intelligent agent AI is a comprehensive transformation of computation from general computing to accelerated computing. This is a fundamental reshaping of the entire computing industry. When evaluating incremental investments made for generative AI or AGI (artificial general intelligence) on this basis, it becomes apparent that these investments are "actually quite moderate" compared to the multitrillion-dollar overall computing market. While the scale of investment is significant, it is far from excessive.

In response to concerns that NVIDIA's investments in numerous AI startups (such as OpenAI) might bring "cyclical" risks and "vendor financing" issues, Huang firmly rebutted this. He explained that every deal and order at NVIDIA is examined through first principles: Is there a real demand from customers? If not, the chips will go to other customers who do have demand.

He further revealed that NVIDIA’s investments are primarily driven by a few reasons: firstly, to expand the ecosystem; secondly, the investment amounts are "only a minuscule single-digit percentage" of the company’s revenue, such a small proportion cannot create systemic cyclical risks; thirdly, and most importantly, the companies being invested in are some of the most influential in today's era, establishing front-row partnerships with them and investing based on a deep understanding of their potential for success will ultimately prove to be great investments.

Working with Trump: Listener, Workaholic, and Pragmatist

In 2025, as Trump returned to the White House, Huang established a close working relationship with the new president. Trump referred to him as a "close friend" and a "great leader." When asked about his views on Trump after observing him up close, Huang shared some traits that are not easily noticeable on television or in the news.

First, Trump is an "exceptionally great listener." Huang noted that almost everything he has said is remembered by Trump. Second, he possesses "incredible diligence." Trump has a strong work ethic, starting work early, often discussing work late at night when Huang is already ready to rest, and Trump can still work for a few more hours. Huang remarked that he has never seen anyone work so desperately; it seems Trump works all seven days a week with almost no leisure time. Additionally, Huang believes that Trump is "super smart." These qualities left a profound impression on him.

US-China Relations: Deep Interdependence and the Fallacy of "Decoupling"

During his travels with Trump around the world, Huang deeply felt a reality that is not fully understood by the outside world: The US and China are deeply interdependent in economics and technology. He believes the notion of US-China "decoupling" is flawed and that the level of interdependence is much deeper than people imagine; every time pressure is applied, new points of dependency are discovered.

Huang asserts that US-China relations will be "the most decisive relationship of the next century." Therefore, the most important thing is not to create conflict but to manage this relationship well. He praises the wisdom of the Trump administration in continually addressing this relationship through negotiation, communication, and compromise, believing it brings a more positive impact than isolation and avoidance.

He particularly emphasized that the AI industry is "deeply reliant" on outstanding students and scientists from China. About 50% of the world’s AI researchers are from China. Chinese companies and people want China to win, which is good; Americans also want America to win. Both sides can cooperate while competing fairly. Over-emotionalizing competitive sentiments will only make the relationship harder to manage.

Chip Game: Technological Leadership, Market Access, and National Security

The discussion touched on the most sensitive issue of chip exports. Huang had successfully lobbied the Trump administration to ease restrictions on the export of H20 chips to China, but subsequent comments from US Commerce Secretary Howard Lutnick (stating the goal was to make China "addicted" to US chips) and China’s reactions led to NVIDIA’s revenues in China dropping to zero.

In this regard, Huang first clarified the concept of the complete tech stack for AI: it is not just about the models but rather a full-stack reconstruction from energy, chips, infrastructure, cloud services, AI models to applications. If the US wants to seize the opportunity of this "largest industrial revolution in history," it must be determined to win at every layer. If the overall level is constrained by the weakest link, the US will surely fall behind.

At the chip level, he believes the key is to ensure that every AI model globally runs on American technology stacks, making the US technology stack the global standard. The idea that selling chips to China will enhance its military power is "absurd" because China already possesses a strong chip industry and manufacturing capability.

Huang proposed a seemingly contradictory but pragmatic viewpoint: allowing American companies like NVIDIA to compete freely in the Chinese market and win business is beneficial for the American people and the US market. This can generate income for America, and "the safest country with the strongest military is often also the wealthiest country." By enhancing US wealth through business success, this supports its military strength, which is his contribution to national security.

He acknowledged this is walking a tightrope. The US wants to limit technology exports to China while also wanting to participate in the Chinese market; China wants access to the best technologies to develop its AI stack while dominating the domestic market. These two demands exist in both countries. He believes that "the best answer is to let the system operate on its own." A refined policy is needed to ensure that American companies can acquire the most advanced American technology while allowing them to engage in the world’s second-largest market.

Supply Chain Resilience vs. Decoupling: Rethinking Taiwan's Role

As NVIDIA's first Blackwell chips roll off the production line at its factory in Arizona, and with China vigorously developing its semiconductor industry, semiconductor manufacturing seems to be undergoing a "decoupling." This raises concerns about whether geopolitical risks in Taiwan will increase as a result.

Huang clearly stated that any conflict should be reduced and that industry should not play a role in exacerbating conflicts. He believes Trump is a "president of peace" who seeks peace. In terms of manufacturing, he emphasized that the focus should not be on "decoupling" but rather on "enhancing resilience." Manufacturing the most critical products in multiple locations, across multiple continents, due to various reasons such as power, weather, and earthquakes is a wise move, as it concerns supply chain diversity and redundancy.

He simultaneously affirmed Taiwan's irreplaceable position: "For the next several decades, we will still rely on Taiwan to manufacture chips and electronic products." The talent, culture, ecosystem network, and manufacturing efficiency there are unparalleled, and duplicating all of this would take decades.

When asked whether selling chips to China would change the related risks of China preparing for "military action against Taiwan" by 2027, Huang's answer was: "The likelihood decreases." His reasoning is based on an ancient wisdom: never corner your adversary. Taking extreme (all or nothing) strategies can lead to unpredictable and unintended consequences. A refined strategy—ensuring the US possesses the best, most abundant technology while providing technology for global competition and expanding America's technological footprint—is the wise approach. He believes this aligns with the interests of both sides and helps maintain low tensions and a balanced relationship.

Middle Eastern Chess Game: Global Competition in Technology Stack

A similar logic applies in the Middle East. Huang traveled with Trump to Saudi Arabia and promoted a major deal for chip exports to Saudi AI companies (previously blocked by the Biden administration due to concerns about chips flowing into China via the Gulf).

His argument to persuade the Trump administration was that the Middle East will develop its own AI capabilities. The core question is: "Do we want the Middle East to build its capabilities on the US technology stack or someone else's?" Surrendering technological leadership and a massive market does not make sense to Trump. Huang believes that, with the world's best technology, it makes no sense to relinquish hundreds of billions of dollars in market opportunities without reason.

Recently, in a meeting with Saudi Crown Prince Mohammed bin Salman (MBS), the latter expressed great pleasure at receiving GPU licenses. Huang revealed that the Crown Prince himself is a loyal gamer and long-term customer of NVIDIA GeForce graphics cards, with a gaming room set up at home equipped with dozens of top gaming PCs, and he is pleased with the partnership between both sides in the AI field.

The Social Impact of AI: Job Transformation and Economic Growth

When discussing the social impact of AI automation on human labor and his responsibilities, Huang maintained an optimistic attitude. He believes AI will make tasks in jobs more efficient, while human work itself is more meaningful than operating spreadsheets or typing. He is confident that AI will drive productivity increases and income growth, thus creating more employment.

He challenges the notion that there is a ceiling on global GDP, predicting that AI could grow the global GDP from the current approximate 100 trillion to 200 trillion, 300 trillion, or even 500 trillion dollars. "There is no fundamental limit to GDP scale." The proportion of the population currently engaged in global economic activity is still small, and if more of the global population is empowered by AI to participate in the economy, it is very possible that GDP could multiply several times.

But he also confidently pointed out: "Everyone’s job will change because of AI. Some jobs will disappear." This has been the case in every industrial revolution, but it will also create a large number of new jobs. He offered a straightforward suggestion: "Everyone must use AI because if you don't, your job will be replaced by someone using AI."

Personal Reflections: At the Top of the World, What Is There to Envy?

At the end of the interview, the reporter mentioned that "NVIDIA" means "envy" in Latin and asked Huang, standing at the top of the world, if there is still anything to envy. His answer was filled with warmth and satisfaction.

He reflected on his "very beautiful" life: a partner he met at 17, two outstanding children (both now work at NVIDIA and collaborate with him), a great company building the most influential technology in history, 40,000 employees he has worked with for decades, and two healthy dogs who just had ultrasonic checks. He also enjoys the pleasure of traveling the world, visiting night markets, and tasting Korean fried chicken.

"For an ordinary person, this is already a dream come true," Huang summed up.

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。