In the shadow of global trade tensions, a more silent but potentially more destructive threat is emerging: the overcapitalisation of artificial intelligence. While headlines still echo with tariff battles and trade retaliation, the unchecked enthusiasm for AI projects is building a financial risk with global ramifications.
When Donald Trump introduced sweeping tariffs under the guise of “liberating” American trade, many feared a global economic stall. Yet, nearly two years later, the direst predictions have not come to pass. The IMF projects modest global GDP growth: 3.2% in 2025 and 3.1% in 2026. China, despite being the main target of U.S. duties, has diversified its exports and maintained momentum. The average U.S. import duty has only reached 16%—significant but not catastrophic.
Trump’s tariff experiment has been curbed not by diplomacy, but by the realities of economics. The U.S. share of global final import demand sits at just 17.5%, limiting Washington’s leverage. Exemptions for key sectors like electronics, along with market pushback (as seen in the April equity sell-off), have further diluted any systemic threat.
But while tariffs have proven manageable, AI investments are starting to resemble the overconfidence that preceded the dot-com crash of the late 1990s.
Massive capital is flowing into AI infrastructure, especially data centers. Imports of AI-relevant hardware are climbing. But ROI remains elusive. AI models, while promising, are not yet delivering transformative productivity gains. The market risks are mounting—especially given AI’s deep integration into venture funding, equity valuations, and national strategies.
According to estimates from McKinsey and CB Insights, global private investment into generative AI alone exceeded $40 billion in 2024, nearly doubling from the previous year. Much of this funding is going into startups with uncertain business models and long-term viability. Hardware imports supporting this AI surge—particularly chips and servers—now represent a growing share of U.S. tech imports.
What makes this dangerous is not just the economic misallocation of resources, but the missed opportunity costs. Under the Trump administration, U.S. federal policy has shifted sharply away from renewables and advanced green technologies. Electric vehicles, solar, and battery sectors—each with high job-creation and export potential—are underfunded. Instead, fossil fuels and speculative tech have taken precedence.
This stands in sharp contrast to China, which, despite its own geopolitical plays (such as rare earth export controls), is still broadly focused on economic fundamentals. Xi Jinping’s government continues to support industrial growth and infrastructure. While China is also eyeing AI dominance, it is doing so within a framework of export-led, balanced development.
The collapse of an AI investment bubble would reverberate globally. Unlike tariffs, which are sectoral and state-driven, a tech bust would impact credit markets, investor sentiment, and even national R&D agendas. The U.S. economy—currently placing outsized bets on AI—is especially vulnerable.
Investors and policymakers should revisit the lessons of the late 1990s. Exuberance without earnings is not strategy. With so much faith—and capital—poured into AI without parallel structural support for productivity, energy, or sustainability, the stage is being set for painful correction.