Written by: Techub News Compilation

In this public dialogue surrounding "U.S. leadership in the field of artificial intelligence," Jensen Huang, founder and CEO of NVIDIA, U.S. Congressman Ro Khanna, and host H.R. McMaster discussed more than just chips, models, and export controls; they focused on a larger question: as artificial intelligence becomes a new general-purpose technology, what must a country rely on to maintain its lead? The answer is not only the technology itself but also the comprehensive abilities in talent, energy, manufacturing, university systems, policy design, social trust, and national narratives.

Content-wise, this dialogue encompasses at least three main threads: First, AI is not a singular technology but a multi-layered industrial system; second, if the U.S. wants to remain ahead, it cannot solely emphasize frontier innovations but must also rebuild manufacturing capabilities, expand technology diffusion, and benefit more ordinary workers; third, in the face of global competition, especially the complex industrial ties with China, the U.S. cannot simply "de-risk" to the point of suffocating innovation nor allow chaotic globalization to continue eroding domestic industries and social cohesion.

What is even more noteworthy is that this discussion did not remain stuck in the binary opposition of "tech optimism" or "AI panic." Jensen Huang emphasized that while AI will reshape industries, "task automation" does not equate to "jobs disappearing"; Ro Khanna reminded that even if productivity boosts may create more jobs in the long run, the transition period of technology diffusion could still be accompanied by significant unemployment, income divergence, and regional imbalances. Therefore, what truly matters is not whether to develop AI but how to develop AI in a more socially inclusive manner.

AI is not a model but an entire industrial infrastructure

In the dialogue, Jensen Huang repeatedly emphasized that one of the biggest misconceptions society has about AI is understanding it as a singular model or product. According to him, AI is essentially an industrial system with a "five-layer stack structure": at the bottom is energy, above that are chips, then come cloud and AI factories as infrastructure, followed by models, and at the top are applications.

This judgment is crucial because it expands "AI competition" from a competition of model capabilities to a competition of national-level infrastructure. In other words, whether a country can maintain its lead in the age of AI cannot be judged solely by the existence of several star model companies but must also consider whether power is sufficient, whether chip supplies are sustainable, whether data centers and cloud infrastructures are robust enough, whether the model ecosystem is thriving, and most importantly—whether AI applications have truly entered industries and society, forming large-scale usage.

Jensen Huang particularly emphasized that if the U.S. is strong in the first four layers but cannot diffuse at the application layer, the whole industrial flywheel will not turn, and the technology will struggle to truly amplify its value. He is not worried that the technology is not advanced enough, but rather that society, out of fear, excessively rejects AI, even "regulating it out of the industry, regulating it out of society." If the diffusion of applications is artificially suppressed, then even if the U.S. invents this round of industrial revolution first, it may not be able to fully reap the benefits it brings.

From this perspective, the core of AI policy is not just about "controlling risks," but also about "reducing barriers to effective use." These barriers can be institutional, psychological, or public opinion-related. If a country shapes AI purely as a threat on a social level, rather than as a tool to be learned and mastered, then it may lose its window for technology diffusion in self-doubt.

America's advantages lie not just in enterprises but also in an open talent and university system

Ro Khanna’s answer to "Why does America still have a chance to maintain AI leadership" complements Huang's industrial perspective. He believes that America's greatest comparative advantage is primarily its ability to attract global talent to the U.S. for study, research, entrepreneurship, and collaboration; secondly, it has a strong research university system; and thirdly, a culture of academic freedom and questioning authority, as well as a relatively mature mechanism for technology transfer among universities, government, and the private sector.

In this discussion, Ro Khanna specifically mentioned that many AI startups are founded by immigrants, and that a large number of AI researchers did not complete their undergraduate education in the U.S. but eventually came to the U.S. for innovation. This mechanism of "absorbing talent globally and forming high-density collaboration locally" is one of the fundamental sources of America's technological leadership.

He also pointed out that the importance of research universities cannot be underestimated. The long-term accumulation of America in basic research, talent cultivation, and technology spillover is not something that has happened by chance, but is closely related to sustained public investment. In other words, when discussing America's AI advantages today, it is necessary to acknowledge not only capital markets and leading enterprises but also the foundational role of national research investment and university systems.

This explains why, although this dialogue is led by a star entrepreneur and a congressman, its underlying logic is not about "the omnipotence of enterprises" or "the omnipotence of government," but rather the synergy among three parties: government provides long-term guidance and institutional environment, universities provide talent and basic research, and enterprises drive industrialization and large-scale applications.

Reindustrialization becomes a new keyword in AI competition

If discussions about AI in previous years focused more on computing power, models, and capital, the most distinctive reality of this dialogue is that it clearly included "reindustrialization" in the AI agenda. Ro Khanna bluntly stated that one of the major mistakes the U.S. has made over the past few decades is the fantasy of being able to be just a financial and innovation center without retaining a strong industrial base. This not only harms national security but also weakens social cohesion and leaves a long-term sense of deprivation in many regions.

He mentioned that the decline of traditional manufacturing is not an abstract macro trend but has specifically destroyed the dignity, employment, and intergenerational identity of many communities. Those cities and families that once relied on factories, steel, and supply chains are forced to face the reality that "if you can't enter the finance or technology industry, you can only be eliminated." This rupture ultimately also reflects in the anger, division, and distrust within American politics.

Therefore, Ro Khanna advocates for a form of "21st-century Marshall Plan"-style new economic patriotism: the U.S. should not just use tariffs as a gesture but should genuinely rebuild critical industries, form new industrial investment capabilities around rare earths, critical minerals, raw pharmaceuticals, robotics, advanced materials, and others, and reorganize the government, businesses, technology sector, and workers toward the same direction.

Jensen Huang provided a practical supplement echoing this point. He believes that the AI industry itself is becoming the engine for America’s reindustrialization. As AI factories, chip factories, and computing infrastructure are established in the U.S., the related construction has driven the demand for a large number of manufacturing, construction, electrical, plumbing, precision tool jobs, and has increased the wage levels of related roles. He also stated that companies are planning to invest massively in U.S. manufacturing, which is predicated on the necessity for the U.S. to maintain a sufficiently vibrant, profitable, and investment-friendly industrial environment.

This means that AI is not just a technology for "replacing labor" but may also become an opportunity to rebuild the real economy and regional employment. However, whether it can become an opportunity depends on whether policies guide capital into long-term construction rather than just revolving around short-term arbitrage.

"Will AI take jobs?" is not a question that can be simply answered

Regarding the impact of AI on employment, the most widely spread part of the dialogue was undoubtedly Jensen Huang’s direct rebuttal of the narrative that "AI destroys jobs." He believes that describing AI as a force that massively destroys jobs is not only inaccurate but also harms American society's acceptance of technology.

He provided a famous example: years ago, some prominent scholars in the field of AI predicted that as AI fully penetrates image reading, radiologists would become "irrelevant" within ten years. Jensen Huang acknowledges that the first part is correct—AI has indeed infiltrated almost every aspect of radiology; but the second part is wrong—radiologists have not decreased; in fact, they have increased.

Why is that? His explanation is: the "purpose" of a profession is not the same as the "specific tasks" within that profession. AI can automate certain tasks but does not necessarily eliminate the profession itself. On the contrary, when AI enhances efficiency, organizations can serve more patients, handle more demands, generate higher income, and thus need more professionals to participate in higher-level judgments, collaboration, and services.

He extends the same logic to software engineering. Within NVIDIA, software engineers have widely used agent-based AI tools, and the result has not been the replacement of engineers but rather that "engineers who are better at using AI" become more popular and successful, while the entire team can advance more projects faster. In other words, AI primarily changes the way work is organized and the boundaries of productivity, rather than simply reducing positions by headcount.

However, Ro Khanna provided a necessary correction. He does not deny that in the long run, technological progress will create new demands and new jobs, but history has also shown: from the industrial revolution to subsequent technological waves, productivity growth will not automatically and equitably be distributed to everyone. The process of technology diffusion often accompanies significant unemployment, widening income gaps, and certain groups being unable to share in the benefits for a long time.

Therefore, truly responsible policies should not simply repeat the slogan "technology will ultimately create more jobs" but should consider during the phase of technology adoption: do workers have bargaining power? Can they share in the benefits of productivity increases? How can young people and those in entry-level positions gain new access? Can those professions most likely to be impacted receive training, protection, and re-employment support during the transition period?

This also explains why Ro Khanna positions himself as an "AI democratist" rather than an "AI doomer" or "AI accelerationist." His core proposition is not against AI but against the concentration of AI benefits solely towards capital while ordinary workers bear the costs.

The real danger is not AI itself, but that only a few people will learn to use AI

Jensen Huang provided a very representative judgment on employment issues: most people may not "lose to AI," but they are likely to lose to those who know how to use AI. The emphasis of this statement is not to create anxiety but to point out the direction of technology diffusion—instead of fearing AI, it is better to quickly enable more people to learn how to use AI.

In his view, the reason AI has become one of the fastest adopted technologies in history is that its entry threshold is lower than many foundational technologies of the past. Ordinary people do not need to become chip engineers or algorithm researchers to use AI as a tool that enhances their capabilities in their professional scenarios. He even provided an example where a person who was originally just a carpenter could use AI to better complete design expressions and then elevate themselves to provide services closer to architectural or interior design.

The logic behind this is that the most important social value of AI is not to confine professional knowledge within a few institutions but to overflow the previously high-threshold cognitive and expressive abilities to more people. When more ordinary workers, entrepreneurs, and students can use AI to perform more complex work, the technological dividends have the potential to truly diffuse.

Ro Khanna further elevates this "diffusion" to the level of social contracts. He believes that one of the reasons American society is highly skeptical of AI today is not only that people do not understand the technology but also because many ordinary people no longer trust elite institutions and no longer believe that a new wave of technological revolution will automatically bring opportunities for themselves. To repair this distrust, it cannot rely solely on propaganda; it also requires visible employment plans, skills training, regional investments, and public commitments.

Between China, globalization, and regulation, what America needs is a "middle path"

Another highly sensitive topic in this dialogue is how America should handle its relationship with China and the global supply chain. Jensen Huang's attitude is very clear: the world is interdependent, and the AI and related industrial chains are not a system that can be completely closed off by one country; any approach of "closing everything and cutting off everyone else" could lead to serious unintended consequences.

He emphasized multiple times that AI is not a singular product but a complex industry deeply embedded in the global supply system. From energy, minerals, and equipment to manufacturing processes, the U.S. has extensive interdependencies with other countries, including China. For this reason, policy-making cannot approach technological competition in a simplistically emotional manner but must carefully assess long-term consequences, chain reactions, and the overall balance of the industrial system.

Ro Khanna shares consensus and adds to this. He agrees that the U.S. cannot completely "decouple" but also believes that the previous form of unrestrained globalization has also proven unsustainable. What the U.S. needs to do is not move toward closure but to rebuild some form of "more boundary-aware openness": acknowledging the monopolistic risks China poses in certain key resources and areas, and needing to push for rebalancing and local capability building; also acknowledging that the U.S. should not fall into a political sentiment of rejecting cooperation and being anti-China.

Jensen Huang also raised a very important reminder here: opposing China's national competitive rival should not slip into an anti-Chinese, anti-immigrant, or anti-international talent social mentality. Because one of America's core assets is precisely that "the best people in the world want to come here." Once the U.S. treats the narrative of competition as an identity-based hostility, what it damages is not others but its own most precious talent attraction and the "American Dream" brand.

On regulatory issues, their differences are not as significant as the outside world imagines. Ro Khanna advocates establishing moderate, refined rules that allow American AI to maintain competitiveness while also becoming "high-quality AI" trusted by the global market; Jensen Huang advocates focusing regulation on applications and usage scenarios, while being cautious of prematurely, excessively, and rigidly regulating foundational technologies that are still rapidly evolving.

In summary, they do not support either extreme: one end is unbounded laissez-faire, and the other end is suppressing innovation under the name of security. The truly feasible route is to maintain a dynamic balance among risk governance, industrial development, and global competition.

What this dialogue truly discusses is a new national narrative

If we only understand this discussion as an "AI policy seminar," we underestimate its significance. More deeply, it discusses whether, at the moment AI is reconstructing economic and social orders, the U.S. can still rebuild a national narrative that allows the majority to believe they have a share in participation.

Ro Khanna repeatedly mentions that one of the most serious problems in the U.S. today is the loss of confidence. People no longer believe they can share in growth, they no longer believe that national systems will prioritize ordinary workers, and they no longer believe that the "American Dream" still holds for the next generation. Therefore, he advocates using AI as an opportunity to rethink what should be the North Star of technological development—it should not just be technological breakthroughs themselves but rather the construction of a more cohesive, diverse society that provides more people with a sense of security and upward mobility.

Jensen Huang, from an entrepreneurial perspective, provided another inspiring response. He believes that now is one of the best times for young people to enter society, use AI, engage in entrepreneurship, and reshape industries. Because this round of technological revolution is not about mending the edges of the old world but about resetting the entire computing industry and thereby resetting almost all industries built on computing foundations. For students and young professionals, this means unprecedented equal starting opportunities.

In this sense, the most important consensus of this dialogue is not "America will definitely win" but rather "If America wants to win, it must allow more people to win together." Leadership in the AI era is not just about having the strongest chips, the most capital, and the most advanced models but whether it can weave technology, industry, education, manufacturing, governance, and social trust into a community project.

This may be the most worthwhile takeaway from this dialogue for outsiders: in the future AI competition, superficially, it looks like a technological race between enterprises and countries, but fundamentally, it is a competition about who can establish a more complete, open, and resilient national capability system. And what truly determines victory or defeat often is not the loudest slogans but whether it can simultaneously answer three questions: who will innovate, who will manufacture, and who will benefit.