The Intricacy of AI, Semiconductors, and Geopolitics
By Max Fang
(December 31, 2024)
“AI is going to infuse every industry, and it’s going to be the basis of competition and growth.” [1]
Jensen Huang, CEO of Nvidia.
A New Global Tech Rivalry
On December 2, 2024, the Biden administration announced sweeping restrictions on technology exports to China, aiming to hinder its development of advanced chips crucial for military and AI applications. [2] The measures prohibit the sale of specific chips, machinery, and add 140 Chinese firms to a restricted trade list. Designed to degrade China’s military modernization, these controls reflect U.S. concerns over national security. [3] The rules extend previous restrictions and introduce global provisions limiting foreign firms from supplying China if U.S. technology is involved. (4) While industry lobbying influenced exemptions, critics argue the controls leave gaps, such as permitting sales to some Chinese factories. [5] Geopolitical tensions and allied negotiations complicate enforcement. [6]
Next day, on December 3, 2024, China announced a ban on exporting key rare minerals like gallium and germanium to the United States, escalating tensions in the ongoing tech war. [7] These materials are vital for semiconductors, weaponry, and advanced technologies, and the move follows U.S. efforts to limit China’s access to cutting-edge technology. [8] While China controls much of the global supply chain for these minerals, the ban mirrors prior retaliatory actions, such as a 2010 embargo on Japan. [9] Experts warn of supply disruptions, particularly for tungsten used in military applications. [10] The ban reflects China’s strategy to counter U.S. restrictions and reinforce its dominance in critical supply chains. [11]
Semiconductor Is the Engine for AI Developments
The relationship between artificial intelligence (AI) and the semiconductor industry is deeply intertwined, characterized by mutual influence and interdependence. AI technologies rely on semiconductors, particularly advanced chips, to perform complex computations, train algorithms, and process vast amounts of data with remarkable speed and efficiency. These chips, such as GPUs (graphics processing units), TPUs (tensor processing units), and FPGAs (field-programmable gate arrays), form the hardware backbone of AI systems. [12] Conversely, the growth and evolution of the semiconductor industry are increasingly driven by AI’s escalating demands for greater processing power, lower latency, and energy-efficient designs. [13]
This dynamic fosters a virtuous cycle of innovation: AI spurs the development of next-generation chips, while advancements in semiconductor technology unlock new capabilities in AI applications. [14] The convergence of these two industries also accelerates breakthroughs in autonomous systems, healthcare, finance, and other sectors. [15] Beyond technical synergies, this relationship also influences global markets and geopolitics, as countries race to dominate both fields. [16] Investments in AI research fuel the demand for cutting-edge semiconductors, prompting nations to develop resilient supply chains and strategic alliances. [17] In essence, AI and semiconductors propel each other forward, shaping the technological and economic trajectories of the 21st century in profound and interconnected ways.
Semiconductor Supply Chain Is the Core to AI Innovation
As explained by Chris Millar, author of Chip War: The Fight for the World's Most Critical Technology, “today, military, economic, and geopolitical power are built on a foundation of computer chips.” [18] For AI chips, this intricate network involves raw material extraction, precision manufacturing, assembly, and global distribution. Semiconductor production relies on rare earth elements and advanced fabrication techniques, often concentrated in a few countries. Taiwan, for instance, dominates chip manufacturing through its cutting-edge foundries, while the United States leads in chip design. However, geopolitical tensions, export restrictions, and trade barriers disrupt this delicate balance, exposing vulnerabilities in global supply chains. As nations scramble to secure access and control over key resources and technologies, the AI chip supply chain becomes a focal point of both innovation and strategic rivalry.
The End of Free Trade? AI Chips and the New Economic Order
Free trade has traditionally been heralded as a cornerstone of global economic progress, rooted in the principle that open markets enhance efficiency and generate mutual benefits for both exporters and importers. Economic theories, such as comparative advantage, have long provided a robust framework supporting this doctrine, demonstrating how nations prosper when they specialize and exchange goods without barriers. Moreover, international systems such as the World Trade Organization (WTO) and multilateral trade agreements have institutionalized these ideals, fostering decades of economic interdependence. However, the rise of AI chips as a critical component in both economic innovation and national security challenges this paradigm.
Unlike traditional goods, AI chips embody dual-use technology with both commercial and strategic military applications, making their trade subject to increasing scrutiny. Geopolitical competition, particularly between the United States and China, has led to export controls, supply chain restrictions, and significant investment in domestic semiconductor production. These measures reflect a growing divergence from the classical free trade model, as nations prioritize technological sovereignty and security over the unrestricted flow of goods. The AI chip industry thus illustrates how emerging technologies, with their transformative capabilities, are reshaping global trade norms and signaling a shift from efficiency-focused policies to security-centric governance.
The New Race: AI Chips and Geopolitical Competition
The rapid AI development and semiconductor innovation creates significant governance challenges. [19] AI technologies are dual-use, with applications in both civilian and military domains. [20] As a result, control over the semiconductor supply chain and leadership in AI innovation are increasingly viewed as instruments of national power. [21]
The current framework for technology governance is wanting for comprehensive international agreement addressing the ethical, economic, and security dimensions of AI or semiconductor trade. This includes addressing issues such as cybersecurity, transparency, and accountability. For semiconductors, governance focuses on supply chain resilience, intellectual property protection, and export controls.
International collaboration on governance has proven challenging. While organizations like the WTO and the Organization for Economic Co-operation and Development (OECD) facilitate dialogue, competing national interests often undermine consensus. [22] The United States, for instance, has pushed for “tech decoupling” from China, urging allies to adopt stricter export controls on advanced semiconductors and AI-related technologies. [23]
Recognizing the strategic importance of semiconductors and AI, nations are forming alliances. As explained by Susan Ariel Aaronson, Research Professor of International Affairs at George Washington University, “AI nationalist policies in one country can make it harder for firms in another country to develop AI.” [24] The United States seeks to partner with Japan, South Korea, and the Netherlands to limit China’s access to cutting-edge chip-making equipment. [25] Similarly, the European Union has launched initiatives like the European Chips Act to bolster its semiconductor industry. [26]
These alliances reflect a broader trend of techno-nationalism, where states prioritize national security and economic sovereignty over global integration. [27] While such strategies aim to mitigate risks, they also risk fragmenting the global technology ecosystem. [28] “If officials can limit access to key components of the AI supply chain, such as data, capital, expertise or computing power, they may be able to limit the AI prowess of competitors in country Y and/or Z.” Professor Susan Ariel Aaronson, in The Age of AI Nationalism and Its Effects, wrote. [29]
Back to the Future: History Repeats Itself?
The Cold War Era: The Birth of Tech Rivalry
The technological competition between the United States and the Soviet Union during the Cold War era laid the foundation for today’s AI and semiconductor race. Space exploration, computer science, and military applications drove early innovations. [30] Government-sponsored programs, such as the U.S. Defense Advanced Research Projects Agency (DARPA), were instrumental in pioneering technologies such as the internet, which would later foster AI developments. [31]
Semiconductors also emerged as a critical industry during this period. Silicon Valley, rising in the 1960s and 1970s, became a hub for chip design and production, with firms, such as Intel, leading innovation. In the 1980s, Japan entered the semiconductor market and challenged U.S. dominance before the trade negotiations and market shifts that curtailed its ascendancy.
Globalization and the Complex of Supply Chains
The 1990s and 2000s witnessed the globalization of the semiconductor supply chain. U.S. companies focused on chip design, outsourcing manufacturing to firms such as Taiwan Semiconductor Manufacturing Company (TSMC) and Samsung in South Korea. This division of labor optimized costs and efficiency but introduced vulnerabilities. However, the global supply chain relied on a few countries and players for advanced manufacturing, creating geopolitical chokepoints.
Since the 2010s, on the other hand, AI research transitioned from academia to industry advanced in computing power, data availability, and algorithmic breakthroughs. Companies such as Google, Amazon, and Microsoft began investing heavily in AI, setting the stage for its commercialization.
The Rise of China
China’s entry into the technological arena since the 2010s reshaped the landscape. Recognizing the strategic importance of AI and semiconductors, China launched national strategies, such as Made in China 2025, aiming to achieve self-sufficiency in high-tech industries. [32] Chinese companies, such as Huawei, Tencent, and Baidu, emerged as global players in AI development, while the government invested heavily in domestic chip production. [33]
However, this rise triggered concerns in many countries. For example, in the United States, people are worried about intellectual property theft, economic dependence, and national security risks. [34] In response, the United States imposed export controls, blacklisted Chinese firms, and sought to reinvigorate its semiconductor manufacturing capabilities through initiatives like the CHIPS and Science Act. [35]
Status Quo: A New AI Race?
The competition between the United States and China is often likened to a new race, and the semiconductor industry is a key battleground. Hsinchu, a coastal city northwest in the island of Taiwan and home to TSMC, holds a pivotal position, as its advanced manufacturing capabilities are unmatched. This has elevated the Taiwan Strait to a critical geopolitical flashpoint.
From the late 2010s, AI development continues to accelerate. Governments and private sectors are racing to achieve breakthroughs in generative AI, quantum computing, and autonomous systems. [36] Their reliance on semiconductors means that any disruption in chip supply–whether from trade restrictions, natural disasters, or military conflicts–could have global impacts. [37]
The Intricacy of AI, Semiconductors, and Geopolitics
The race for supremacy in AI and semiconductor technologies underscores the intricate interplay of innovation, economic strategy, and geopolitics. As AI and advanced chips reshape industries and redefine power dynamics, nations are forced to navigate the complex dual-use nature of these technologies. U.S. export controls targeting China’s access to cutting-edge semiconductors and China’s retaliatory rare mineral bans highlight the escalating stakes in this global competition. Here, “Taiwan’s semiconductor ecosystem is indispensable for democracies’ resilience, serving as a collective shield against emerging threats,” explained by Jeremy Chang, CEO of DSET (Research Institute for Democracy, Society and Emerging Technology, Taiwan). [38]
This struggle is not just about technological superiority but also about securing critical supply chains and asserting national sovereignty. [39] Historical patterns, from Cold War rivalries to modern techno-nationalism, reveal a persistent link between technology leadership and geopolitical influence. In this new era, alliances, trade policies, and industrial investments reflect strategic recalibrations that prioritize security over market efficiency. The AI chip industry, situated at this nexus, will continue to drive both innovation and global tensions, shaping the future of international relations.
1. CNBC, Nvidia CEO Jensen Huang on How His Big Bet on A.I. Is Finally Paying Off - Full Interview, YouTube (Mar. 19, 2023), https://www.youtube.com/watch?v=_-wjA4XlBl4.
2. Ana Swanson, Biden Targets China’s Chip Industry with Wider Trade Bans, NY Times (Dec. 2, 2024), https://www.nytimes.com/2024/12/02/business/economy/biden-china-chips-exports.html.
3. Id.
4. Id.
5. Id.
6. Id.
7. David Pierson, Keith Bradsher and Ana Swanson, China Bans Rare Mineral Exports to the U.S., NY Times (Dec. 3, 2024), https://www.nytimes.com/2024/12/03/world/asia/china-minerals-semiconductors.html.
8. Id.
9. Id.
10. Id.
11. Id.
12. E.g., Tensor Processing Unit (TPU), Semiconductor Engineering, https://semiengineering.com/knowledge_centers/integrated-circuit/ic-types/processors/tensor-processing-unit-tpu/(last visited Nov. 29, 2024). Field Programmable Gate Arrays (FPGAs) for Artificial Intelligence (AI), Intel, https://www.intel.com/content/www/us/en/learn/fpga-for-ai.html (last visited Nov. 29, 2024).
13. Id.
14. See Giacomo Persi Paoli & Sarah Grand-Clement, Technology, Economy and War: The Disruptive Force of AI, in Research Handbook on Warfare and Artificial Intelligence 28, 28 (2024).
15. Id.
16. See Tobias Schaub et al., How Artificial Intelligence Is Shaping Global Power Dynamics: The Geopolitics of the AI-relevant Semiconductor Supply Chain 11 (2024), https://www.graduateinstitute.ch/sites/internet/files/2024-09/SchaubSreedharanZhang_THE-GEOPOLITICS-OF-THE-AI-RELEVANT-SEMICONDUCTOR-SUPPLY-CHAIN_FinalReport---Vibhaa-Sreedharan.pdf.
17. See Taiwan’s Global Role in Semiconductor Supply Chains: Experts Discuss Challenges and Opportunities at Supply Chain Resilience Forum, Rsch. Instit. for Democracy, Soc. and Emerging Tech. (Oct. 23, 2024, at 12:41 PM), https://dset.tw/en/event-news-en/00070/.
18. About the Book, Simon & Schuster, https://www.simonandschuster.com/books/Chip-War/Chris-Miller/9781982172008 (last visited Dec. 26, 2024).
19. Silja Voneky & Thorsten Schmidt, Regulating AI in Non-military Applications: Lessons Learned, in Research Handbook on Warfare and Artificial Intelligence 352, 352–53 (2024).
20. Id.
21. Id.
22. See Policies, Data and Analysis for Trustworthy Artificial Intelligence, OECD, https://oecd.ai/en/ (last visited Nov. 29, 2024).
23. See generally Jon Bateman, U.S.-China Technological “Decoupling”: A Strategy and Policy Framework (2022), https://carnegie-production-assets.s3.amazonaws.com/static/files/Bateman_US-China_Decoupling_final.pdf.
24. Susan Ariel Aaronson, The Age of AI Nationalism and Its Effects 1 (2024), https://www.cigionline.org/static/documents/Aaronson_PTlSYDB.pdf.
25. See also Bateman, supra note 23.
26. European Chips Act, European Comm’n, https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/europe-fit-digital-age/european-chips-act_en (last visited Nov. 29, 2024).
27. Kate Lamb, The Rise of Techno-Nationalism - And the Paradox at Its Core, The World Econ. F. (July 3, 2019), https://www.weforum.org/stories/2019/07/the-rise-of-techno-nationalism-and-the-paradox-at-its-core/.
28. Id.
29. Aaronson, supra note 24.
30. See also Albert K. Lai, The Cold War, the Space Race, and the Law of Outer Space 48–49 (2021).
31. About DARPA, Def. Advanced Rsch. Projects Agency, https://www.darpa.mil/about-us/about-darpa (last visited Nov. 29, 2024).
32. James McBride & Andrew Chatzky, Is ‘Made in China 2025’ a Threat to Global Trade?, Couns. of Foreign Affs. (May 13, 2019, at 8:00 AM EST), https://www.cfr.org/backgrounder/made-china-2025-threat-global-trade.
33. See id.
34. Gregory C. Allen, The True Impact of Allied Export Controls on the U.S. and Chinese Semiconductor Manufacturing Equipment Industries, Ctr. for Strategic & Int’l Stud. (Nov. 26, 2024), https://www.csis.org/analysis/true-impact-allied-export-controls-us-and-chinese-semiconductor-manufacturing-equipment.
35. FACT SHEET: CHIPS and Science Act Will Lower Costs, Create Jobs, Strengthen Supply Chains, and Counter China, The White House (Aug. 09, 2022), https://www.whitehouse.gov/briefing-room/statements-releases/2022/08/09/fact-sheet-chips-and-science-act-will-lower-costs-create-jobs-strengthen-supply-chains-and-counter-china/.
36. See Paoli & Grand-Clement, supra note 14.
37. See Rsch. Instit. for Democracy, Soc. and Emerging Tech., supra note 17.
38. Id.
39. See Aaronson, supra note 24.