A silicon wafer reflecting power lines and a city skyline at dusk.

When AI Becomes National Strategy: What NVIDIA, TSMC, and ASML Reveal About the Future of Technology

From Jensen Huang’s controversy to Taiwan’s semiconductor history and Europe’s search for a shared strategic purpose.

Taiwan Tech Dispatch | November 2026

Introduction: Beyond Jensen Huang

I have long admired Jensen Huang.

As NVIDIA’s co-founder and longtime CEO, he helped transform accelerated computing into one of the foundations of modern artificial intelligence. NVIDIA’s technology now supports applications ranging from scientific research and industrial automation to advanced AI systems. That achievement deserves recognition, regardless of the questions now surrounding his personal conduct.

When the story broke, a friend in European media reached out to ask what I made of it. On November 8, NVIDIA’s board announced an independent investigation after reports that Huang may have been targeted by a foreign intelligence effort. On November 12, NVIDIA announced that he had resigned as president, CEO, and a board member. The company named CFO Colette Kress interim CEO. NVIDIA’s investigation announcement · NVIDIA’s leadership announcement

Our exchange brought me back to a question about AI as national strategy, one that will outlast the headlines.

What happens when a technology company becomes so strategically important that its leadership decisions can no longer be understood through a purely commercial lens?

NVIDIA is not alone in facing this transformation.

Taiwan’s TSMC, the Netherlands’ ASML, Germany’s ZEISS, and Franco-Italian STMicroelectronics occupy different positions in the global technology ecosystem. Yet each demonstrates how a commercially operated company can become important to national security, economic resilience, and international competition.

To understand the responsibilities that follow, we need to look beyond individual executives.

We need to examine the relationship between industry, government, and society.

I call this relationship the Strategic Social Contract.

Taiwan’s Semiconductor Experience: Strategy Built Over Decades

Coming from Taiwan, I see the relationship between technology and national interests through a different historical lens.

Today, Taiwan’s semiconductor capabilities are closely connected to its economic resilience, international relevance, and security.

But this strategic position did not emerge overnight.

Nor did Taiwan begin with a fully formed doctrine that semiconductor manufacturing would become central to national security.

In the 1970s, Taiwan was seeking to upgrade its industrial economy and develop more advanced technological capabilities.

Public institutions played an important role in this process.

In 1976, the Industrial Technology Research Institute, or ITRI, signed an integrated circuit technology transfer and licensing agreement with RCA in the United States.

Nineteen Taiwanese specialists received training in semiconductor design, manufacturing processes, verification, and equipment.

After returning to Taiwan, they helped establish domestic technological capabilities.

In 1977, Taiwan completed its first integrated circuit demonstration factory. ITRI, Taiwan’s IC technology transfer from RCA

These early achievements reflected coordinated investments in technology acquisition, engineering talent, research infrastructure, and industrial learning.

But government policy alone did not create Taiwan’s semiconductor success.

Private enterprise, global customers, entrepreneurial decisions, and decades of commercial execution were equally important.

TSMC, founded in 1987 under Morris Chang’s leadership with government participation and private investment, introduced the dedicated semiconductor foundry model. TSMC, company profile

This model allowed chip designers to focus on design while relying on specialized manufacturers for fabrication.

Over time, it transformed the global semiconductor industry’s division of labor.

TSMC’s success required sustained capital investment, technological discipline, customer trust, and continuous innovation.

The broader historical lesson is clear.

Taiwan’s semiconductor leadership emerged from the interaction of public institutions and private enterprise, not from either acting alone.

As Taiwan’s manufacturing capabilities became increasingly important to the global technology industry, their strategic significance expanded.

An industry originally developed largely for economic upgrading became central to Taiwan’s international position.

In other words, Taiwan did not simply build a semiconductor industry to fulfill a pre-existing national security doctrine.

It built an industry whose success gradually became part of its strategic identity.

This distinction matters for other countries attempting to build strategically important technologies.

The Strategic Social Contract

The relationship between a strategically important industry and the society supporting it cannot be understood as a one-way obligation.

If governments expect companies to contribute to national economic resilience and security, governments and society must also consider the conditions those companies need to succeed.

Advanced semiconductor manufacturing requires reliable electricity, industrial water, land, transportation, specialized equipment, skilled engineers, and long-term regulatory stability.

No individual company can independently create all these conditions.

In Taiwan, central and local governments, universities, research institutions, and private enterprises have contributed to the development of the semiconductor ecosystem.

This cooperation has not eliminated disagreement.

Water allocation, electricity supply, environmental impacts, industrial concentration, and regional development remain legitimate public concerns.

Strategic importance does not entitle any company to unlimited public resources.

But Taiwan’s experience suggests that technological leadership depends on a durable relationship between corporate capabilities and public institutions.

I describe this relationship as the Strategic Social Contract.

It involves three interconnected responsibilities.

Industry contributes technological capabilities. Companies invest capital, develop products, create employment, improve productivity, and compete internationally.

Government provides institutional support. Public authorities establish research infrastructure, support education, coordinate essential infrastructure, enforce laws, and maintain predictable policy environments.

Society contributes resources and legitimacy. Communities provide talent, land, infrastructure capacity, and political support, while expecting economic opportunities, public benefits, environmental safeguards, and fair treatment.

These contributions are not identical, but they must be understood as reciprocal.

A government cannot reasonably demand that a company sacrifice commercial opportunities for strategic purposes without considering the economic consequences.

Likewise, a company benefiting from public support cannot assume that shareholders should capture every benefit while the public absorbs all associated costs.

The Strategic Social Contract is therefore not a justification for unlimited industrial subsidies.

It is a framework for asking four questions:

Who benefits? Who pays? Who decides? And who is accountable?

AI as national strategy: industry invests in technology, government builds institutions and infrastructure, and society contributes talent and resources and expects safeguards.
The Strategic Social Contract links what industry, government, and society contribute and what each can expect in return.

Taiwan’s experience shows how those responsibilities take shape over decades. NVIDIA brings the same question into a different setting, where Washington must weigh export markets against national security.

NVIDIA and the National Interest Paradox

NVIDIA is a publicly traded American company, not an arm of the US government.

Its leadership must operate within applicable laws, corporate governance requirements, and obligations to the company.

Yet NVIDIA’s strategic importance has expanded dramatically.

Advanced AI computing now supports scientific research, industrial productivity, cybersecurity, and capabilities relevant to defense and intelligence.

Access to leading-edge computing infrastructure has become a matter of geopolitical competition.

NVIDIA’s position in this ecosystem creates a fundamental tension.

From a commercial perspective, international markets support revenue, economies of scale, customer relationships, and continued investment in innovation.

From a national security perspective, governments may conclude that certain advanced computing capabilities should not be made available to particular customers or jurisdictions.

These interests can reinforce each other, but they can also conflict.

The debate over American AI chip exports to China illustrates this dilemma.

Jensen Huang has argued that continued American technology sales can help preserve US technological leadership and keep international customers within the American technology ecosystem. Jensen Huang at the Milken Institute Global Conference

Some American lawmakers, however, believe that access to advanced AI chips could accelerate China’s military and intelligence capabilities. U.S. House Select Committee on the CCP, letter on H20 chip exports

Both positions claim to defend American national interests.

This is what makes the debate difficult.

Export restrictions may reduce security risks. But excessively broad restrictions could also reduce commercial opportunities, encourage alternative technology ecosystems, and weaken the economic foundations of future American innovation.

The appropriate balance cannot be determined by commercial incentives alone.

Companies must comply with applicable export controls. Beyond those legal obligations, decisions involving national security, industrial competitiveness, and geopolitical risk require legitimate political institutions and accountable policymaking.

Who defines the national interest when commercial competitiveness and national security point in different directions?

This question should not be delegated entirely to a corporate CEO.

Nor should strategic importance automatically turn private corporate assets into government property.

The challenge is to develop institutions capable of balancing these interests without undermining the innovation they are intended to protect.

The leadership change left Congress with questions about both the company and the government. On November 12, three House committee chairs renewed their request to interview Huang and separately sought testimony from NVIDIA’s general counsel. Their letters set out proposed closed-door interviews on December 1 and 2, respectively, covering issues including company security and possible diversion of advanced chips. House letter to Huang · House letter to NVIDIA

Congressional scrutiny also reached the Commerce Department. Senators Elizabeth Warren and Chris Coons asked Commerce Secretary Howard Lutnick and other administration officials to preserve records. Senator Ron Wyden and Representative Richard Neal separately asked Lutnick to preserve records on NVIDIA’s export licenses and the department’s communications with the company. Senators John Curtis and Lisa Blunt Rochester pressed Commerce to address gaps in chip export controls. Warren–Coons request · Wyden–Neal letter · Curtis–Blunt Rochester letter

These requests put a question about government accountability alongside the one about corporate oversight. When access to a company’s technology becomes a matter of national security, the public has reason to ask how the company safeguards it and how officials decide which customers can buy it.

AI as National Strategy: America’s Infrastructure Dilemma

The United States approaches strategic technology from a different historical position.

American technological leadership has been shaped by private entrepreneurship, venture capital, universities, competition, and deep capital markets.

But public institutions have also played important roles through government-funded research, defense procurement, scientific institutions, and industrial policy.

The American innovation system has never been entirely separate from the state.

Its contemporary AI challenge is to reconcile national technological ambition with the physical and social costs of achieving it.

AI is often discussed in terms of software, models, and algorithms.

Yet advanced AI increasingly depends on physical infrastructure.

Data centers require electricity, land, grid connections, cooling systems, power equipment, and substantial capital investment.

According to the International Energy Agency, global data center electricity consumption is projected to rise from approximately 485 terawatt-hours in 2025 to around 950 terawatt-hours in 2030.

The IEA expects electricity consumption from AI-focused data centers to grow considerably faster than overall data center demand. IEA, Key Questions on Energy and AI

Global data centers used about 485 TWh of electricity in 2025. The IEA projects about 950 TWh in 2030.
Source: International Energy Agency, Key Questions on Energy and AI. The 2030 figure is a global projection.

These are global projections, not forecasts for individual American communities.

Nevertheless, data centers are geographically concentrated, and their rapid development can create significant pressure on local electricity networks.

The IEA warns that the mismatch between fast-moving data center investments and slower electricity infrastructure development can create affordability and reliability risks if poorly managed.

But the consequences are not predetermined.

In some electricity systems, additional demand may improve the utilization of existing infrastructure. In others, it may require expensive new generation and grid investments. IEA, Key Questions on Energy and AI

The political challenge is therefore not simply whether AI data centers should be built.

It is how their costs and benefits should be distributed.

The United States has strong reasons to maintain its technological leadership.

AI could influence its future competitiveness in science, defense, finance, manufacturing, entertainment, and other industries.

But the benefits of that leadership may be long-term and widely distributed, while the infrastructure costs are often immediate and concentrated in particular communities.

A resident may support American leadership in AI while questioning whether local households should finance grid upgrades primarily benefiting a large commercial data center.

That position is not necessarily contradictory.

It reflects a legitimate question about public responsibility.

Who should pay for new transmission capacity?

How should electricity tariffs allocate infrastructure costs?

What benefits should host communities receive?

And what obligations should data center operators accept in return for public support and grid access?

National ambition does not automatically create local consent.

A sustainable AI strategy requires institutions capable of connecting national objectives with local accountability.

Europe Is Not Starting From Zero

Europe faces a different strategic environment.

It possesses important technological capabilities, particularly in semiconductor equipment, industrial electronics, and specialized manufacturing technologies.

ASML in the Netherlands is a critical supplier of advanced lithography equipment.

Germany’s ZEISS contributes specialized optics used in advanced semiconductor manufacturing. ZEISS, High-NA EUV lithography

STMicroelectronics, with its French and Italian roots, is an important participant in power semiconductors and industrial electronics.

These companies demonstrate that Europe already controls technologies of global strategic importance.

European governments also recognize the need to support strategically important industries.

One particularly instructive example is STMicroelectronics.

The European Commission approved a €2 billion Italian state aid measure supporting an integrated silicon carbide semiconductor manufacturing facility in Catania, Sicily.

The approval included commitments intended to generate benefits beyond the individual company.

According to the Commission, STMicroelectronics agreed to contribute to the wider European semiconductor value chain, advance next-generation silicon carbide technology, develop workforce training, and implement specific priority-order arrangements during supply shortages under the European Chips Act framework. European Commission, Italian state aid for STMicroelectronics

The project also included a commitment intended to protect it from certain extraterritorial public service obligations imposed by third countries.

This is a concrete example of public support accompanied by strategic industrial responsibilities.

Government provides support for investment in a technology considered important to European economic resilience.

Industry accepts obligations extending beyond the immediate commercial interests of the individual project.

The arrangement resembles the reciprocal relationship described by the Strategic Social Contract.

Europe is therefore not starting from zero.

The question is whether such project-level arrangements can become part of a durable strategic framework across industries, member states, and communities.

Europe’s Strategic Narrative Challenge

Europe’s challenge cannot be reduced to a lack of technological ambition.

The European Commission has articulated goals involving AI leadership, industrial competitiveness, digital autonomy, and economic security.

Its AI Continent Action Plan seeks to strengthen computing infrastructure, data access, industrial AI adoption, skills, and the conditions for investment. European Commission, AI Continent Action Plan

But Europe is not a single nation-state.

Its member states have different industrial structures, energy systems, security priorities, and political traditions.

Some policy decisions are coordinated at the European level, while others remain primarily within national jurisdiction.

This creates an institutional challenge.

Consider the Netherlands.

The Dutch government has imposed national export licensing requirements on certain advanced semiconductor manufacturing technologies, including equipment relevant to advanced chip production.

In January 2025, it announced an expansion of those controls, citing increased security risks. The measures require authorization for covered exports; they are not a blanket prohibition on exports. Government of the Netherlands, semiconductor export licensing

This illustrates how a technology company can operate within overlapping strategic frameworks.

A company may seek global commercial opportunities.

Its national government may impose security restrictions.

European institutions may pursue broader economic security and industrial competitiveness.

Allied governments may have additional strategic expectations.

These interests can overlap without being identical.

Resolving the resulting tensions requires legitimate institutions and clear rules.

Yet Europe also faces a broader question.

What shared purpose should connect its semiconductor capabilities, AI infrastructure investments, and public priorities?

Taiwan’s semiconductor position is closely connected to economic resilience and national security.

The United States has a longstanding political tradition of seeking global technological leadership.

Europe’s historical project is different.

It has been shaped by peace, economic integration, democratic institutions, social protection, and the rule of law.

Perhaps Europe does not need to imitate Taiwan’s security narrative or America’s ambition for technological primacy.

Instead, it may need to explain why technological capabilities are necessary to preserve and advance its own priorities.

AI infrastructure could support European scientific research, industrial productivity, healthcare, public services, and greater autonomy in sensitive applications.

But these benefits should be demonstrated rather than assumed.

Public opinion also deserves careful interpretation.

A 2025 European Commission survey found that more than 60% of respondents viewed robots and AI positively in the workplace, while 84% believed AI required careful management to protect privacy and transparency. European Commission, attitudes toward AI in the workplace

Another Eurobarometer survey found strong public interest in secure technologies, digital rights, and human support during digital transformation. European Commission, Digital Decade 2025 Eurobarometer

These findings do not establish whether Europeans support large-scale AI infrastructure investment.

They do suggest that technological ambition and public safeguards should not be treated as mutually exclusive.

Europe’s challenge may not be the absence of a strategic vision, but the difficulty of translating that vision into a sufficiently compelling and widely supported social contract.

That remains a hypothesis to examine, not an established judgment about European society.

Strategic Consensus Is Built Over Time

Taiwan’s experience suggests that strategic consensus can grow as an industry proves its value. Europe can put that lesson to work through projects that give several member states a reason to cooperate.

Consider Catania. The Commission’s approval gives Italy and STMicroelectronics a set of commitments that other member states can evaluate. Are engineers being trained? Can suppliers elsewhere in Europe qualify for work? Could priority orders be filled during a shortage? Publishing progress on those commitments would give other governments and firms a basis for deciding whether to join a similar project.

AI infrastructure raises a similar question. If several countries help finance a shared computing facility, they can agree in advance on who gets access, how operating costs are shared, and which results they will review before funding the next facility. Access for a research lab or a small manufacturer makes the value of that cooperation visible to people beyond the governments that signed the agreement.

Conclusion: The Question Behind Jensen Huang

The controversy surrounding Jensen Huang began as a question about the conduct and judgment of an influential technology leader.

Those questions deserve to be examined through evidence, independent investigation, and appropriate corporate governance.

But the broader strategic question should not disappear when the immediate controversy fades.

NVIDIA illustrates how a private technology company can become central to national economic and security interests.

TSMC’s history demonstrates how public institutions and private enterprise can jointly create technological capabilities that eventually acquire strategic importance.

The American AI infrastructure debate shows that national technological ambition must confront the local costs of implementation.

And Europe’s semiconductor policies reveal that public support and strategic industrial obligations can already be linked through concrete institutional arrangements.

The lesson is not that every company should automatically place government demands above commercial interests.

Nor is it that governments should provide unlimited support to strategically important corporations.

The lesson is that strategic technological power creates reciprocal responsibilities.

Companies must recognize the public consequences of their capabilities.

Governments must establish legitimate rules and provide the institutional conditions necessary for long-term development.

Societies must be able to evaluate the costs, benefits, and trade-offs involved.

Technological sovereignty is not simply about owning the most advanced chips or operating the largest data centers.

It is also about sustaining the institutions, resources, and public trust required to develop and use those technologies.

For European readers, I would leave one question:

If Europe expects its technology champions to help secure its strategic future, what shared future is Europe asking them to build — and what are its governments, industries, and citizens prepared to contribute in return?

Continue reading the Hsinchu Series

About the author

White Hsu writes Taiwan Tech Dispatch, an independent publication examining Taiwan’s role in global technology, from semiconductors and AI infrastructure to startups and industrial policy.

Subscribe to Taiwan Tech Dispatch

Get future analysis on Taiwan’s technology ecosystem and the industrial decisions shaping global AI. Subscribe to Taiwan Tech Dispatch on LinkedIn.

Sources and Further Reading

NVIDIA Board of Directors, statement announcing its independent investigation

Jensen Huang at the Milken Institute Global Conference

ITRI, Taiwan’s IC technology transfer from RCA

IEA, Key Questions on Energy and AI

European Commission, Italian state aid for STMicroelectronics

European Commission, AI Continent Action Plan

Government of the Netherlands, semiconductor export licensing

European Commission, attitudes toward AI in the workplace

European Commission, Digital Decade 2025 Eurobarometer

U.S. House Select Committee on the CCP, letter on H20 chip exports

TSMC, company profile

ZEISS, High-NA EUV lithography

The views expressed are the author’s own and do not represent those of any employer or organization.

Similar Posts

Leave a Reply