Semiconductor Geopolitics: Taiwan's Strategic Choices in the Chip War

A chip the size of a fingernail is redrawing the geopolitical map of the twenty-first century. As the United States invests $52.7 billion through the CHIPS and Science Act to rebuild domestic semiconductor manufacturing, China marshals a whole-of-nation effort toward chip self-sufficiency, and the European Union targets 20% of global production capacity through its European Chips Act, semiconductors have graduated from a highly specialized industrial issue to a core variable of national security, economic sovereignty, and diplomatic competition. And Taiwan — the island that commands over 90% of the world's advanced-process chip production — sits at the very center of this contest. Through my years conducting international policy research at the University of Cambridge, directing cross-border regulatory studies for the World Bank and the United Nations, and now leading Meta Intelligence in AI software development, I have come to a profound realization: semiconductor geopolitics is not merely about chips. It reflects the fundamental logic behind the reorganization of the twenty-first-century international order — technology is power, supply chains are strategy, and standards are rules.

I. Why Semiconductors Have Become a Core Geopolitical Variable

Semiconductors have escalated from an industrial concern to a central geopolitical variable because of the convergence of three structural factors: the irreplaceability of the technology, the extreme concentration of the supply chain, and the dual strategic significance for both military and economic power.

First, the irreplaceability of the technology. Semiconductors are the bedrock of digital civilization — from smartphones, electric vehicles, and cloud computing to artificial intelligence, virtually every critical technology is built upon chips. But unlike oil, semiconductors are not a natural resource; they represent the pinnacle of human engineering. Constructing a single advanced wafer fab requires tens of billions of dollars in investment, thousands of top-tier engineers, the coordinated operation of hundreds of precision instruments, and a construction timeline of three to five years. These extraordinarily high technological barriers and capital intensity mean that semiconductor manufacturing is not an industry one can simply decide to enter — it is more akin to a national-level capability that takes decades to accumulate.^[1]

Second, the extreme concentration of the supply chain. The degree of concentration in the global semiconductor supply chain has reached alarming levels. In the advanced-process domain (7 nanometers and below), TSMC alone accounts for over 90% of worldwide capacity. In extreme ultraviolet (EUV) lithography, the Dutch firm ASML is the world's sole supplier. In electronic design automation (EDA) tools, the American companies Synopsys, Cadence, and Siemens EDA collectively command more than 80% of the global market. This "chokepoint concentration" means that a disruption at any single link in the supply chain — whether due to geopolitical conflict, natural disaster, or pandemic — could trigger a cascading global crisis.^[2]

Third, the dual strategic significance for military and economic power. Advanced chips are not merely commercial products; they are critical enablers of military capability. From precision-guided weapons and fighter-jet avionics to satellite communications and cyber-warfare capacity, modern military power is deeply dependent on advanced semiconductors. At the same time, the rapid advance of AI has turned high-performance computing chips (such as NVIDIA's GPUs) into foundational infrastructure for the great-power AI race. When AI is seen as the engine of the "next industrial revolution," whoever controls the chip production capacity that AI requires holds the entry ticket to that contest. This is the fundamental rationale behind America's export controls on advanced chips to China — the aim is not to suppress China's commercial development, but to constrain the speed of China's catch-up in AI and military domains.^[3]

The convergence of these three factors has irreversibly transformed semiconductors from a purely technical and commercial issue into a core geopolitical variable. In the multi-country policy research I directed for the World Bank, I observed a similar pattern unfolding previously in the energy sector — oil shaped the international order of the twentieth century, and semiconductors are now shaping the power dynamics of the twenty-first. The crucial difference is this: the distribution of oil is determined by geology, whereas semiconductor capacity is determined by technological capability and institutional environment — and this creates strategic opportunities for small economies like Taiwan that did not exist in the oil era.

II. The Great-Power Chip Game: Three Divergent Paths in the U.S., China, and EU

Once semiconductors became synonymous with national security, the major powers moved swiftly to intervene through industrial policy, export controls, and diplomatic leverage. The three largest economies have chosen distinctly different strategic paths, yet all converge on the same objective: reducing external dependence and securing supply chain safety.

The U.S. strategy is "carrots and sticks" — domestic subsidies combined with external controls. The CHIPS and Science Act, passed in 2022, allocates $52.7 billion in semiconductor manufacturing and R&D subsidies to attract TSMC, Samsung, and Intel to build advanced fabs on American soil. TSMC's investment in Arizona has already exceeded $65 billion, with plans for three fabs spanning 4nm to 2nm processes. Simultaneously, the Bureau of Industry and Security (BIS) has progressively escalated chip export controls on China since 2022 — from restricting exports of advanced chips and manufacturing equipment, to imposing "compute ceiling" controls on AI training chips, to extending the control perimeter to third-country companies. The core logic of this strategy is clear: rebuild domestic capacity while slowing China's progress in advanced semiconductors.^[4]

China's strategy is a "whole-of-nation technology catch-up." Confronted with America's systematic blockade, China has elevated semiconductor self-sufficiency to the highest priority of national strategy. The National Integrated Circuit Industry Investment Fund (the "Big Fund") invested over 340 billion RMB across its first two phases, with a third phase adding another 344 billion RMB — an unprecedented scale. Huawei's chip design subsidiary HiSilicon, working with the foundry SMIC, achieved a breakthrough at the 7nm node (though yield rates and capacity still lag far behind TSMC). China's approach exhibits the hallmarks of "asymmetric catch-up": rather than attempting to match the West across all fronts, it concentrates resources on breaking through critical bottlenecks — particularly building overwhelming capacity advantages in mature processes (28nm and above) while seeking leapfrog opportunities in emerging areas such as advanced packaging and quantum computing.

The EU's strategy is "strategic autonomy through industrial policy." The European Chips Act, which entered into force in 2023, sets a target of raising the EU's share of global semiconductor production from less than 10% to 20% by 2030, mobilizing over 43 billion euros in public and private investment. The EU's challenge, however, is that it is starting virtually from zero in advanced-process manufacturing. ASML is Europe's most powerful asset in the semiconductor supply chain, but manufacturing EUV equipment and running a wafer foundry are entirely different capabilities. The EU's real strengths lie in automotive-grade and industrial semiconductors — Infineon, NXP, and STMicroelectronics hold important positions in these segments. The EU strategy is essentially one of "selective autonomy" — not pursuing comprehensive advanced-process capability, but securing supply in the specialized segments where it holds advantages.^[5]

For Taiwan, grasping the deeper logic behind these three paths is essential. America's strategy means that Taiwan's advanced manufacturing capability is being "decentralized" — TSMC's overseas expansion is not a matter of choice but a strategic inevitability driven by customer demands and political pressure. China's self-sufficiency drive means that Taiwan's market share in mature processes will face sustained long-term pressure. The EU's push for digital sovereignty means that the trend toward regional supply chains will continue to deepen. Together, these three forces point to a single conclusion: the semiconductor industry architecture of the past three decades, built on the logic of "globalized efficiency," is being supplanted by a new logic of "geopolitical security."

III. Two Sides of the Silicon Shield: Taiwan's Strategic Asset and Structural Risks

No superlative adequately captures Taiwan's position in the global semiconductor industry. Over 90% of the world's most advanced chips (those manufactured at 7nm and below) are produced by TSMC. This fact endows Taiwan with a unique geopolitical asset — the strategic deterrent known as the "silicon shield": any military action against Taiwan would devastate the global semiconductor supply chain, inflicting unacceptable damage on the attacker's own economy and military capabilities.^[6]

The silicon shield, however, is a double-edged sword. Its strategic value rests on a single premise — that Taiwan's semiconductor manufacturing capability is irreplaceable. The moment that premise is shaken — whether because other countries establish alternative capacity, a fundamental technological paradigm shift occurs, or Taiwan's own technological edge erodes — the deterrent effect of the silicon shield will be severely diminished. And the global supply chain restructuring now underway is eroding that premise from multiple directions.

The first structural risk is technology diffusion. TSMC's fab-building projects in the United States, Japan, and Germany are, in essence, dispersing Taiwan's most critical technological capabilities overseas. While TSMC retains its most advanced processes (such as 2nm and the forthcoming 1.4nm) in Taiwan, the Arizona fab's 4nm — and eventually 2nm — processes are sufficient to meet most military and AI application requirements. As these overseas fabs gradually come online, the structural dependence of the United States and Japan on Taiwanese advanced chips will decline — meaning the protective power of the silicon shield is being diluted.

The second structural risk is talent scarcity and attrition. The core competitive advantage in semiconductor manufacturing lies not in equipment (which can be purchased) but in the process know-how accumulated by the engineering teams who operate that equipment. TSMC's decades-long technological leadership is built upon one of the world's finest semiconductor engineering workforces. Yet Taiwan's talent pool faces a triple squeeze: a declining birth rate is reducing the pipeline of new STEM graduates; overseas fab expansions require large numbers of senior engineers to be deployed abroad; and international tech giants are poaching talent with premium compensation. According to the Semiconductor Industry Association, the global semiconductor industry will face a shortfall of over one million technical workers by 2030 — and Taiwan is in the eye of the storm.^[7]

The third structural risk is the "bullseye effect" of geopolitics. Taiwan's semiconductor prowess makes it an object that every party in the great-power competition aggressively courts — or seeks to control. The United States, through the CHIPS Act and export controls, has effectively incorporated Taiwan into the core of its "tech alliance." China regards Taiwan's chip production capacity as a strategic asset to be secured upon reunification. Japan and the EU, citing supply chain security, actively pursue partnerships with TSMC. Taiwan's predicament resembles that of West Berlin during the Cold War — possessing immense strategic value, but for that very reason bearing enormous geopolitical pressure. In my experience conducting research on tech diplomacy, I have observed a recurring pattern: when a country's strategic value is excessively concentrated in a single industry, its bargaining power in international negotiations may actually decline — because everyone knows it cannot simply walk away from the table.

The fourth structural risk is an over-concentrated industrial structure. Taiwan's economic dependence on the semiconductor industry has reached a level that warrants vigilance. The share of semiconductors and related industries in Taiwan's total exports continues to climb, and TSMC's revenue alone accounts for a significant proportion of Taiwan's GDP. This degree of industrial concentration carries not only economic risks (the impact of a single-industry downturn on the overall economy) but also strategic ones — it leaves Taiwan with extremely limited "exit options" in international negotiations. By contrast, consider Israel's innovation ecosystem: another small economy, yet one whose technological strengths are distributed across cybersecurity, agritech, medtech, and other fields. Taiwan's singular concentration is a structural vulnerability that demands honest assessment.

IV. Taiwan's Positioning in Supply Chain Restructuring: From Passive Defense to Proactive Strategy

Faced with the structural restructuring of the global semiconductor supply chain, Taiwan must transform itself from a passive "party that is needed" into a proactive "strategic architect." This is not merely an adjustment to industrial policy; it requires a fundamental shift in national strategic thinking.

First, Taiwan must redefine the meaning of the "silicon shield." The traditional concept of the silicon shield is defensive — using the irreplaceability of manufacturing capability as a deterrent. But as global capacity disperses, a strategy that relies solely on manufacturing concentration is losing its effectiveness. Taiwan needs to upgrade the silicon shield from "irreplaceable capacity" to "irreplaceable ecosystem" — encompassing not just wafer foundry services themselves, but a complete system that includes advanced packaging, IC design services, silicon intellectual property (IP) licensing, and semiconductor equipment maintenance. TSMC's competitive advantage stems not only from its EUV tools, but from the cluster of hundreds of suppliers, specialty chemical producers, and precision equipment service firms surrounding the Hsinchu Science Park and Southern Taiwan Science Park. That cluster cannot be "copy-pasted" to Arizona or Kumamoto.

Second, Taiwan must assume the role of "co-architect of the rules" in the global supply chain restructuring. The current restructuring is driven primarily by America's security logic, with Taiwan largely playing a reactive role. Yet Taiwan possesses sufficient industrial standing to participate in rule-making — particularly on issues such as multilateralizing export controls, establishing resilience standards for semiconductor supply chains, and developing international norms for technology transfer. Against the backdrop of increasingly fragmented global governance, Taiwan should actively engage in minilateral cooperation frameworks — such as the U.S.–Japan–Netherlands coordination on semiconductor equipment export controls and the Chip 4 Alliance (U.S., Japan, South Korea, Taiwan) policy dialogues — and seek institutionalized protections for Taiwan's interests within these mechanisms.

Third, Taiwan must squarely face the long-term trend of decentralization and seize it as a catalyst for economic diversification. The global dispersal of semiconductor production capacity is an irreversible trend. Rather than viewing this as a threat, Taiwan should treat it as a catalyst for economic transformation. The technological capabilities nurtured by Taiwan's semiconductor ecosystem — precision manufacturing, process control, quality management, supply chain coordination — are transferable, general-purpose competencies that can "spill over" into other high-tech domains. From AI software development (the very direction I am pursuing at Meta Intelligence) and precision medicine to space technology, Taiwan's engineering culture and manufacturing experience offer natural advantages in these emerging fields. The transformation from Silicon Island to Smart Island does not mean abandoning semiconductors; it means building, on a semiconductor foundation, a more diversified and more resilient technology-driven economy.

Fourth, Taiwan needs to deepen its institutional ties with the democratic technology alliance. In the context of a geopoliticized semiconductor supply chain, Taiwan's greatest strategic asset is not only its technological capability, but its institutional quality as a "trusted partner" — democratic governance, the rule of law, intellectual property protection, and a transparent business environment. These institutional strengths hold irreplaceable value in competition with authoritarian systems. On this foundation, Taiwan should build deeper institutional connections with key partners such as the United States, Japan, the Netherlands, and South Korea — not only through enterprise-level commercial cooperation, but also through government-level policy coordination, talent exchange, and joint R&D. From my experience serving as the Asia-Pacific representative at the University of Cambridge, I have come to appreciate deeply that in international cooperation, the value of institutional trust often surpasses that of technological capability itself.^[8]

V. Policy Recommendations: A National Strategy Beyond the Silicon Shield

Based on the foregoing analysis, I propose six policy recommendations for Taiwan's semiconductor geopolitical strategy. The core logic is this: transform the semiconductor advantage from a single-dimensional "silicon shield" defensive concept into a multidimensional national strategic asset.

1. Establish a "Semiconductor National Security Council" — Current industrial policy is fragmented across the Ministry of Economic Affairs, the National Science and Technology Council, and the National Development Council, lacking a mechanism to coordinate semiconductor policy from a national security perspective. A dedicated semiconductor strategy unit should be established under the National Security Council to integrate industrial policy, export controls, talent strategy, and tech diplomacy, ensuring alignment between industrial decisions and national security considerations.
2. Launch a "Semiconductor Talent Security Program" — Treat semiconductor talent as a national strategic resource and safeguard the talent base on three fronts: first, substantially expand graduate enrollment in STEM fields with full scholarships; second, create a "Semiconductor National Service" system that allows outstanding STEM graduates to fulfill their national defense obligations by serving in the semiconductor industry; third, develop competitive compensation and tax policies to reduce the economic incentives for talent outflow.
3. Pursue "ecosystem upgrading" rather than merely "process leadership" — Taiwan's long-term competitiveness should rest not only on the most advanced nanometer-scale processes, but on a complete and irreplicable industrial ecosystem. The government should increase investment in advanced packaging (such as TSMC's CoWoS and InFO technologies), silicon photonics, compound semiconductors (such as GaN and SiC), and other areas to build a multi-layered technological moat.
4. Build an institutional framework for "semiconductor diplomacy" — Translating industrial standing into diplomatic leverage requires systematic institutional design. Taiwan should sign "Semiconductor Supply Chain Security Partnership Agreements" with key partners such as Japan, the Netherlands, and South Korea, establishing formal mechanisms for mutual assurance of raw materials, joint R&D, and two-way talent exchange. These agreements would also create de facto institutional channels for Taiwan to participate in international semiconductor governance.
5. Formulate an "economic resilience" strategy for industrial diversification — While maintaining the semiconductor advantage, systematically cultivate second and third growth engines. AI software and applications, biotech and medical technology, green energy technology, and space technology are all domains where Taiwan's engineering capabilities can be extended. The goal should be to raise the share of non-semiconductor high-tech industries in exports to a meaningful level by 2035.
6. Strengthen "information resilience" and "cognitive security" — Semiconductor geopolitics is not just about physical supply chains; it is also an information warfare battleground. External actors may attempt to stoke panic over a Taiwan Strait conflict to undermine the international community's confidence in investing in Taiwan, thereby indirectly weakening Taiwan's semiconductor standing. Taiwan needs to build a systematic strategic communications capability to clearly convey its stability and reliability to the international community.

Semiconductor geopolitics is a game with no final whistle. Taiwan cannot afford to rest on the passive comfort of "being needed"; it must employ a proactive national strategy to convert its semiconductor advantage into enduring comprehensive national strength. The silicon shield will not remain impregnable forever — but if, during the window in which the shield still holds, Taiwan can diversify its economic structure, deepen its talent base, institutionalize its international partnerships, and ascend the value chain from hardware manufacturing to software applications, then Taiwan's strategic security will no longer depend on the fate of any single industry but will rest on a far more solid foundation.^[1]