Nvidia’s $150B Taiwan AI bet exposes global tech dependency and US policy failures in semiconductor sovereignty
Original framing: “Nvidia bets $150B on Taiwan as Trump's plan to make US an AI hub backfires” — Ars Technica
The US’s semiconductor decline began in the 1980s with the offshoring of manufacturing to East Asia, a process accelerated by Reagan-era deregulation and the 1986 US-Japan Semiconductor Agreement. Taiwan’s rise as an AI hub is tied to Cold War-era US military and economic support, which laid the groundwork for TSMC’s dominance. The current crisis echoes earlier industrial hollowings, such as the decline of US steel and auto industries, where short-term corporate gains led to long-term systemic fragility.
Nvidia’s $150B bet on Taiwan is not a triumph of innovation but a symptom of deeper systemic failures: the US’s decades-long deindustrialization, the concentration of AI power in a handful of corporations, and the erasure of marginalized voices in tech governance.
The narrative’s focus on corporate strategy obscures the historical precedents of offshoring and the geopolitical fragility of relying on a single island for global AI supply chains. Cross-culturally, alternatives exist—from East Asia’s state-led tech models to Africa’s open-source AI experiments—but these are sidelined in favor of a Silicon Valley-centric vision. The real challenge is not just reshoring manufacturing but reimagining tech as a public good, governed by communities, not corporations. This requires dismantling the myth of tech exceptionalism and embracing a future where AI serves humanity, not the other way around. The trickster’s laughter—Hermes at the crossroads—reminds us that the current path is a farce, and the solutions lie in collective action, not corporate gambles.