The Tech-Leveraged Empire: Strategic Limits of U.S. Power

The United States is entering the second half of the 2020s projecting extraordinary economic and technological dominance, powered by a self-reinforcing cycle in which innovation attracts global capital and capital accelerates the next wave of breakthroughs.

This model has produced trillion-dollar technology champions and attracted massive foreign inflows, reinforcing the perception of an unstoppable digital empire built on unmatched AI and semiconductor leadership. Yet beneath this exuberance lies a widening gap between appearance and reality. America’s technological engine is increasingly constrained by three material limits, energy, industrial scale, and strategic coherence, that innovation alone cannot overcome. China, meanwhile, has spent the years consolidating mastery over critical minerals, advanced manufacturing, and supply chains, converting second-mover status into durable leverage. The Busan summit, which forced Washington to suspend its own export controls, exposed this new balance: the U.S. still leads at the frontier, one thinks, but no longer commands the physical foundations of digital power, buying time for China (and others) to position.

The Sign of the Times looks at the basis of, and limits to, America’s technological leadership. Unless the United States undertakes a profound strategic recalibration, integrating energy and technology planning, rebuilding the industrial base underlying its digital economy, and restoring both domestic cohesion and allied trust, its present strength may prove a gilded illusion

Context - The Architecture and Fragility of Economic Supremacy

US economic strength and technological leadership now reinforce one another in a self-sustaining loop: technological breakthroughs attract global capital, and that capital funds the next generation of breakthroughs. Simply put, American economic power in the digital era rests on an elegant formula: innovation plus capital equals dominance. Silicon Valley generates the breakthroughs, Wall Street monetises them, and the dollar system exports the profits globally. This model has delivered spectacular gains. Firms like Nvidia, Microsoft, and OpenAI now command more than 80% of global AI-chip architecture and model training capacity and sit at the apex of the twenty-first-century economy, with the former two valued at over US$4 trillion each.

The first half of 2025 demonstrated the scale of global conviction behind this formula. Net portfolio inflows into US equities from abroad totalled US$104 billion between Jan and May,¹ while foreign direct investment reached US$145 billion through June, (with further significant investments announced in semiconductors and data centres).² These capital inflows reflect confidence not merely in specific investment opportunities, but in the structure of American capitalism itself - the belief that no matter the politics, the system will continue to deliver returns. Yet key fundamentals tell a potentially different story. The Shiller CAPE ratio near 40 signals overvaluation³. Public debt exceeds 120% of GDP⁴, wealth concentration rivals that of the nineteenth-century Gilded Age,⁵ and the US economy is increasingly powered by government spending, consumer credit, and speculative capital rather than by productivity growth. America’s current economic strength looks more like a gilded illusion, shining atop structural fragility, and requires substantial structural issues to be addressed. Investors would be well guided to not mistake momentum for strategy. For the tech industry, this matters deeply. Assuming that the current US$22 trillion in combined market cap of the US-based ‘Magnificent Seven’ tech companies is necessarily an indicator of America’s global tech dominance would be an error.

What the market celebrates as unstoppable momentum is bounded by forces the digital economy cannot escape: energy, industrial capacity, and strategic credibility. Each is becoming a pressure point.

Power, Technology, Markets and China

In the summer of 2025, the United States looked unstoppable. Equity markets roared back from a spring crash that had wiped out US$6.6 trillion in two days, climbing to record capitalization of roughly US$56 trillion by July. Artificial intelligence (AI) related stocks led the charge, with nearly half of the S&P 500’s total return driven by just seven companies⁶. The AI boom’s impact on the overall economy was also material, with U.S. GDP growth in the first half of 2025 almost entirely driven by investment in data centres and information processing technology.⁷

Increasingly, American prosperity and power rested on an unmistakable foundation: technological and AI supremacy, with the trajectory suggesting that America was on track to remain the preeminent global technology power, a digital empire for decades to come. However, at the Trump-Xi summit in South Korea in October 2025, China forced Washington to suspend its export controls on semiconductors and rare-earth elements for a full year⁸, concessions that would have been unthinkable only a few years ago. Xi Jinping framed the meeting as a partnership of equals - China’s “great rejuvenation” alongside “Make America Great Again.” But the meaning was unmistakable: the balance of power had shifted. Washington was “now dealing with a peer rival capable of imposing material economic harm”⁹.

While investors reacted with relief, strategists saw something else - American geo-economic and technological supremacy meeting its material limits. While America’s tech (and ultimately geopolitical) power - anchored in AI, semiconductors, and cloud infrastructure - is ultimately founded on innovation, its continued success relies on three critical factors whose constraints are beginning to be felt: energy, scale, and strategic coherence. US technological power faces structural pressures that innovation alone cannot easily overcome. Unless the United States rebuilds a grand strategy that integrates three pillars of power, energy system, industrial base, and strategic coherence, its tech leadership risks eroding, not through defeat, but through exhaustion.

Pillar One: Energy Hard Limits on Tech Power

Empires are built on energy. Britain had coal, America had oil, and the digital empire runs on electricity. AI’s computational hunger is staggering. US data centres already consume roughly 5 GW - about 4% of national demand¹⁰. By 2030, data centres are projected to account for up to 14% of all US electricity demand, more than triple their share today.¹¹

As illustrated above, even under conservative assumptions, meeting projected U.S. AI-related electricity demand in the absence of energy savings elsewhere will require over 400 TWh of new generation capacity, reaching over 1,000 TWh under the most aggressive estimates. US grid planning to date has largely assumed flat load growth,¹⁴ and is therefore unlikely to be able to accommodate the additional energy demands imposed by AI data centres. The solution data centre owners have turned to is the creation of dedicated local generation capacity, either in the form of nuclear or renewable generation. But the demand is staggering. Meeting it through nuclear generation would require an unprecedented build-out, between c.180 -430 Small Modular Reactors (SMRs)¹⁵ or c.40-100 large-scale 1,300 MW plants by 2040. Current U.S. permitting, financing, and supply-chain capacity make such expansion implausible within the necessary timeframe^16,17.

Meeting this demand through nuclear power seems highly unrealistic, requiring America to increase its current nuclear capacity by 54-132% from its current levels,¹⁸ necessitating most ambitious nuclear program in history. Small modular reactors (SMRs) promise speed but remain unproven commercially¹⁹. The leading Western design, GE Hitachi’s BWRX-300, has received regulatory approval in Canada but is unlikely to enter operation before 2029. Despite there being over a hundred SMR designs on paper, none of the major Western economies had a fully commercial SMR running as of early 2025, deployment at scale is still years away²⁰.

A more feasible pathway lies in renewables. Figure 2 shows that cumulative solar and onshore wind capacity could exceed 1,000 GW by 2050 under current expansion rates, offering far faster deployment potential than nuclear alternatives, albeit with intermittency and grid-storage challenges^21,22.

Renewables can move faster but lack stability. Based on the projections above, total incremental electricity generation from renewables in 2040 could exceed 500 GW. At current US average capacity factors, this would generate c.1,500 TWh of electricity, implying that data centres would consume up to two-thirds of the entire projected renewable generation capacity in the US, leaving other sectors of the economy starved for electricity.

The maths is very telling. Until AI models consume far fewer energy resources (a neglected but critical area of R&D focus) or energy current resources are dramatically expanded, this is a key limiting factor. Without a coordinated energy-technology plan, America’s AI boom risks overloading its grid. Great powers rarely lose in international competition because they run out of ideas - they lose when they run out of fuel.

Pillar Two: Scale and Industrial Strategy

Energy enables; scale decides. China’s competitive advantage lies in having scaled technologies that others have invent, . Through state-directed financing, industrial integration, and aggressive cost compression, Beijing has turned second-mover status into first-order advantage.

China’s dominance of these sectors provides tangible economic and geopolitical leverage. When Washington extended semiconductor export controls in 2025, Beijing retaliated by tightening rare-earth exports, forcing the United States to retreat at the aforementioned Busan summit²⁵. Material leverage rather than ideology, won this round of great power competition. The US has committed to securing independence from Chinese rare earths, with US Treasury Secretary Scott Bessent announcing after the summit that China had “made a real mistake” in weaponizing its control of critical minerals, stating that Washington would ensure “U.S. supply-chain independence in 12 to 24 months”²⁶. Analysts, however, widely judge this timeline as implausible. While rare earths are geologically abundant, over 80% of global refining capacity remains in China, and new mining-to-magnet chains typically take five to ten years to reach scale^27,28. The declaration thus underscores the very asymmetry it sought to erase: the United States may be able to innovate faster than any rival, but it cannot easily rebuild the industrial foundations required to fully capitalise on this innovation with comparable speed.

The global semiconductor industry demonstrates these asymmetries very starkly. An American company, NVIDIA, which designs the GPU chips that are the foundation for modern AI, is the world’s most valuable company. However, as a fabless semiconductor company it outsources all manufacturing. In terms of actual chip manufacturing, TSMC (Taiwan Semiconductor Manufacturing Company) produces nearly 90% of advanced chips globally. Each of its factories costs tens of billions of dollars, requires huge volumes of electricity and water, and takes years to build. Other chips critical for NVIDIA’s GPUs are similarly dependent on a handful of international suppliers. E.g. South Korean SK Hynix manufactures c.70% of the world’s high-speed memory chips used in advanced GPUs.

TSMC’s own supply chain is similarly concentrated. The only company in the world that can build EUV lithography machines required to manufacture GPUs, ASML, is European. In turn, ASML’s machines depend on equally specialised European components: Zeiss’ ultra-precise lenses that focus the light used to etch chip patterns, or Trumpf’s high-power lasers that generate that light. More than 5,000 tier-1 suppliers across Europe supply ASML alone²⁹, forming a dense web of specialist firms that are practically impossible for any other country to recreate from scratch.

The Ukraine war has shown how fragile this concentrated supply chain is. Ukraine produces approximately half of the world’s semiconductor-grade neon, a key component in the lasers used in ASML’s machines. Russia’s invasion caused Ukrainian plants to halt operations, slowing ASML’s production in Europe, reducing TSMC’s output in Taiwan, and delaying shipments of NVIDIA’s high-end processors to big tech customers in the US.

The economic pressure for hyperscale cloud providers (the US tech giants) to shift to newer generations of chips is enormous. More advanced chips use roughly half as much power and cooling as today’s designs, saving hundreds of millions of dollars per large data-centre deployment. But the US does not control the equipment, materials, or factories needed to make these newer chips, and therefore doesn’t set the pace of the transition.

The global control map is stark: Europe controls the machines, Taiwan the fabrication, , and the United States the design and cloud layers. AI sovereignty is underpinned by physical capabilities and supply chains - not policy declarations. Scale also determines diffusion. The US innovation model creates trillion-dollar firms but not broad-based productivity. China’s model produces overcapacity but also industrial ecosystems. History favours the system that converts innovation into employment and infrastructure - the power that scales wins.

Pillar Three: Strategic Coherence and Credibility

Strategic coherence is the alignment of a nation’s technological ambitions with the energy, industrial, and geopolitical systems required to sustain them. Across each of these domains, the United States now faces a widening structural gap between what it can invent and what it can reliably support. In terms of energy, as previously described, no innovation agenda can succeed if the physical power system beneath it cannot scale. In terms of industrial systems, as the semiconductor example illustrates, an innovator who does not control the upstream foundations of their technology is a leader with extraordinary influence yet limited leverage, positioned at the top of a pyramid built on other nations’ industrial bases.

Over the past four decades, American technological leadership has grown out of a system whose breakthroughs far exceed its ability to be coordinated and equitable across boundaries, internal and external. The U.S. possesses the world’s most dynamic private sector, yet its government institutions, regulatory structures, political incentives, economic distribution and social cohesion are not designed to support grand-scale transformation that keep value creation and societal alignment. The energy system, industrial base, and alliance network remain strong in isolation, but they do not function as a unified strategic system, and the civil cohesion required to sustain such coordination is increasingly fragile.

This is the essence of the strategic-coherence problem: America’s technological engine is accelerating faster than the political, institutional, and social mechanisms needed to steer, sustain, and legitimise it. Three interlocking fractures flow from this mismatch.

The first fracture is a governance-capacity gap. The U.S. can generate frontier technologies faster than any nation, but it lacks a state apparatus able to align energy, industry, infrastructure, and national priorities at the speed AI-scale technologies require. Decisions about power grids, data centres, industrial siting, and security policy unfold across separate federal and state authorities with conflicting mandates and political incentives. The result is not failure but a frictional speed: a nation capable of extraordinary breakthroughs but unable to mobilise its institutions around them.

The second fracture is a cohesion gap within the society that must absorb and legitimise technological change. The gains from AI and digital transformation are highly concentrated; the disruptions are widely felt. An increasingly unequal and polarised society struggles to form the stable political coalitions needed to support long-term strategy or sustain the investment cycles required for resilience. Strategic coherence is impossible without societal coherence. A nation experiencing persistent political volatility cannot execute multi-decade energy, industrial, or alliance strategies.

The third fracture is the geoeconomic self-undermining of the international system the United States built and long depended on. For decades, U.S. technological strength grew within an open, rules-based order that lowered costs, enabled global supply chains, and aligned allies around shared standards. That system amplified American power. Today, rivalry with China and the EU, the United States’ own increasingly blunt geoeconomic tools are at risk of hurting itself. Partners, such as the EU, India and Japan, experience U.S. measures - tariffs, export controls - as unpredictable and harmful to their own industries, prompting hedging rather than alignment, while competitors accelerate parallel networks that bypass U.S. influence. Interdependence, once a strategic asset, now exposes the United States to reciprocal vulnerabilities: such measures reverberate back into American industries, importing essential skills, data-centre buildouts, and technology deployment. The global order that once stabilised U.S. leadership thus becomes more fragmented and less reliably shaped by such policies.

Taken together, these fractures reveal the essence of the American dilemma: the United States has unmatched creative capacity but insufficient strategic capacity. It can invent the future faster than any competitor, yet it struggles to organise the state, align society, and integrate its power systems into local and international systems in ways that turn breakthroughs into durable advantage.

Strategic coherence is therefore not an issue of innovation but of governance, cohesion, and statecraft. Without strengthening these foundations, the U.S. risks trying to build a digital empire on an increasingly unanchored base, brilliant at the frontier, but with a fractured foundation. Without the right strategy, the United States risks allowing a period of extraordinary technological dynamism to rest upon foundations too divided, too slow, and too fragile to support it.

The implications of this are geopolitical. In the absence of firm control over the material inputs of technological power, U.S. influence increasingly depends on the confidence of its allies. Allies and competitors alike see that U.S. technological power rests on supply chains it does not command and an energy system it cannot expand quickly. The overall trust deficit has been expanding leading to others hedging accordingly, diversifying supply, deepening ties with China even while maintaining security relationships with Washington, and treating U.S. export controls as negotiable rather than fixed. When a nation’s commitments appear subject to domestic swings, capacity limits or predatory actions, partners adapt by reducing dependence rather than deepening it.

In the meantime, China has linked its technological strategy to its industrial and energy policies, has established a foothold across the developing world, is beginning to position as an alternative that uses strategic finance alongside its industrial and technological assets to gain ground. It has built power plants and ports as a deliberate tool of international relations, and is doing the same with chips. Beijing’s upcoming 2026-2030 Five-Year Plan doubles down on domestic demand and high-tech manufacturing, blending growth with techno-nationalism. China’s geopolitical engagement is also aligned to these goals. At the Busan summit, it offered the US tactical cooperation, while continuing to advance strategic self-reliance.

In summary, AI leadership is ultimately determined by physical world realities, threats, declarations of policy or ideology cannot substitute for this. The US’s position requires it to align its energy system, industrial base, and geopolitical strategy with its technological ambitions, in the meantime, its leadership will remain powerful but fragile. Ultimately, for the US to have leadership, it would need to lead rather than force compliance.

Conclusion - The Battle for Tech Supremacy Has Just Begun

The 2025 Trump-Xi trade truce was more than a diplomatic pause; it was a quiet revelation of power. For the first time in decades, the United States - the architect of the liberal economic order - was forced to suspend its own export-control campaign under pressure from a rival. China did not out-innovate Washington, but it used the intervening years since the last round of trade wars ending in 2020 to prepare - consolidating control over minerals, batteries, and advanced manufacturing. These assets now constitute the arteries of the digital economy. The Busan summit did not just mark a truce, but the first negotiation between equals in technological power.

For Washington, the moment carries a deeper warning. It reveals the widening gap between creative capacity and strategic capacity - the distance between what America can invent and what it can sustain. This gap if it persists will be the battleground of the next decade of superpower tech competition. To retain its technological leadership, the United States must reimagine its grand strategy around three imperatives.

First, energy-technology integration must be treated as a pillar of national security. Artificial intelligence, data infrastructure, and digital systems draw on the same lifeblood as heavy industry, energy, and more specifically electricity. Sustaining technological advantage therefore requires a parallel revolution in power generation and transmission. Fossil infrastructure buildout is too slow and capital intensive and exposes the resulting electricity to long-term price volatility and global commodity risk. The only logical answer requires building new nuclear capacity where viable, accelerating renewable deployment, and modernising the grid with the urgency once reserved for defence mobilisation.

Second, industrial reconstitution is indispensable. The United States cannot preserve strategic primacy while outsourcing the physical foundations of its digital economy. Re-shoring advanced manufacturing, semiconductor fabrication, and rare-earth refining is not an act of protectionism but of strategic prudence. Economic power built solely on intangible assets and financial abstractions is brittle; enduring influence requires a material base that can support innovation in both peace and crisis.

Third, strategic legitimacy must be restored. Technological leadership depends not only on scale and capital but on trust, trust among allies that U.S. commitments will endure, and trust among citizens that progress will serve the common good. Leadership that commands neither loyalty abroad nor confidence at home cannot last. Legitimacy, rather than leverage, is the ultimate foundation of durable power.

Without such recalibration, the pattern established in 2025 will continue to unfold: markets rising while strategy erodes. The world’s most innovative nation risks becoming its most speculative. The international system is reverting to equilibrium: illusions fade, leverage narrows, and only disciplined powers endure.

Without industrial and energy alignment, technological supremacy and the valuations that accompany it are limited. Whether the United States rebuilds its foundations or drifts toward managed decline will depend less on the brilliance of its engineers than on the resolve of its strategists. Technology leadership and globalisation will be inextricably intertwined in that future. America built a tech-leveraged empire; whether it remains one depends on rebuilding the physical foundations beneath the digital.

The Leader: Endnotes

1. International Monetary Fund / World Bank. (n.d.). Portfolio equity, net inflows (BoP, current U.S.$) [Data set]. World Bank. https://data.worldbank.org/indicator/BX.PEF.TOTL.CD.WD

2. U.S. Bureau of Economic Analysis. (2025, July 22). Direct investment by country and industry, 2024 [News release]. https://www.bea.gov/data/intl-trade-investment/direct-investment-country-and-industry

3. Goldman Sachs. (2025). The AI infrastructure boom and power-grid implications. https://www.goldmansachs.com

4. Congressional Budget Office [CBO]. (2025). The budget and economic outlook: 2025 to 2035 https://www.cbo.gov/publication/2025-outlook

5. Greater Pacific Capital (2025) “America’s New Golden Age…or A Gilded One”, Sign of the Times October 2025

6. First Trust Advisors L.P. (2025) “Three On Thursday” First Trust Economics, October 7, 2025

7. Furman, J. (2025, October 7). Data-centers and GDP growth: Why the first half of 2025 saw zero contribution. Fortune. https://fortune.com/2025/10/07/data-centers-gdp-growth-zero-first-half-2025-jason-furman-harvard-economist/

8. Leahy, J., & Sevastopulo, D. (2025b, October 30). US and China agree one-year trade truce after Donald Trump-Xi Jinping talks. Financial Times. https://www.ft.com/content/ceb9d271-09b4-4066-87a7-1e93b6908640.

9. Leahy, J., & Sevastopulo, D. (2025a, October 30). China emerges as US “peer rival” at Xi Jinping-Donald Trump summit. Financial Times. https://www.ft.com/content/b505bd49-66bf-4848-9f6d-642c83a1c9b5

10. US Department of Energy [DOE]. (2024). Data-centre energy use and forecasting. https://www.energy.gov

11. Kou, H., & Limandibhratha, N. (2025, April 15). Power for AI: Easier said than built. BloombergNEF. https://about.bnef.com/insights/commodities/power-for-ai-easier-said-than-built/

12. U.S. Department of Energy. (2024, December 20). DOE releases new report evaluating increase in electricity demand from data centers. https://www.energy.gov/articles/doe-releases-new-report-evaluating-increase-electricity-demand-data-centers; Pew Research Center. (2025, October 24). What we know about energy use at U.S. data centers amid the AI boom. https://www.pewresearch.org/short-reads/2025/10/24/what-we-know-about-energy-use-at-us-data-centers-amid-the-ai-boom/?

13. Ibid, Thomas, E. (2025, October 14). Data centers in one nation are driving power demand like nowhere else. Business Insider. https://www.businessinsider.com/data-center-drive-us-power-demand-delay-clean-energy-mckinsey-2025-10

14. Clean Energy States Alliance. (2025). Load growth: What states are doing to accommodate increasing electric demand (Report). Retrieved from /mnt/data/Load-Growth.pdf

15. Based on the GE Hitachi BWRX-300 SMR, the most commercially advanced western SMR design

16. World Nuclear Association. (2025). Small modular reactors (SMRs). https://world-nuclear.org

17. GE Hitachi Nuclear Energy. (2024). Commercial timeline for BWRX-300 SMR. https://world-nuclear.org

18. U.S. Energy Information Administration. (2024). Electricity generation, capacity, and sales in the United States [Data table]. https://www.eia.gov/energyexplained/electricity/electricity-in-the-us-generation-capacity-and-sales.php

19. World Nuclear Association. (2025). Small modular reactors (SMRs). https://world-nuclear.org/information-library/nuclear-fuel-cycle/nuclear-power-reactors/small-nuclear-power-reactors

20. World Nuclear Association. (2025). Small modular reactors (SMRs) – status and global tracker. https://world-nuclear.org/information-library/current-and-future-generation/small-modular-reactor-smr-global-tracker

21. BloombergNEF. (2025). Global clean energy investment trends. https://about.bnef.com

22. International Energy Agency [IEA]. (2024). Electricity 2024: Analysis and forecast to 2026 https://www.iea.org/reports/electricity-2024

23. International Energy Agency [IEA]. (2024). Electricity 2024: Analysis and forecast to 2026. https://www.iea.org/reports/electricity-2024

24. World Bank. (2025). Energy transition data dashboard. https://databank.worldbank.org

25. Leahy, J., & Sevastopulo, D. (2025a, October 30). China emerges as US “peer rival” at Xi Jinping-Donald Trump summit. Financial Times. https://www.ft.com/content/b505bd49-66bf-4848-9f6d-642c83a1c9b5

26. Reuters. (2025, October 31). China made a “real mistake” by firing shots on rare-earths, says U.S. Treasury Secretary Scott Bessent. https://www.reuters.com/world/asia-pacific/china-made-mistake-by-firing-shots-rare-earths-bessent-tells-ft-2025-10-31/

27. U.S. Geological Survey [USGS]. (2025). Mineral commodity summaries: Rare earths. U.S. Department of the Interior. https://pubs.usgs.gov

28. Center for Strategic and International Studies [CSIS]. (2024). Rare earth supply chains and strategic vulnerability. https://www.csis.org

29. ASML Holding N.V. (2024). Responsible value chain: Upholding our sustainability principles throughout the value chain. Retrieved fromhttps://www.asml.com/company/sustainability/responsible-supply-chain