Silicon Colonialism: The Few Places That Make the Chips Everyone Needs

Consider two facts about the chips that run the modern world. The most advanced processors — the ones below seven nanometers that power AI, smartphones, and cutting-edge weapons — are manufactured overwhelmingly in a single country: Taiwan, where one company, TSMC, accounts for more than 90% of leading-edge capacity. And the machines that make those chips possible — extreme ultraviolet lithography systems, without which the smallest transistors cannot be etched — are built by exactly one company on earth, the Dutch firm ASML, the sole supplier in the world. The largest, richest, most technologically ambitious economies on the planet, including the United States and China, depend on these few facilities for the physical basis of everything digital. And the dependency is not mutual in equal measure: the facilities do not need most of the countries nearly as much as the countries need the facilities.

This is silicon colonialism: a pattern in which a small number of locations hold capabilities the rest of the world cannot do without, extract value from that dependency, and make strategic decisions the rest of the world must adjust to rather than the other way around. The word is deliberately provocative, and its aptness is arguable — but the structure it points at is real, and it has become one of the defining geopolitical facts of the age.

The shape of the dependency

What makes the semiconductor dependency distinctive is its extreme concentration at multiple chokepoints stacked on top of each other. It is not merely that chip-making is hard; it is that each layer of the process narrows to a handful of players, and some layers to one. The most advanced fabrication concentrates in Taiwan; the essential lithography machines come only from ASML; key materials, design tools, and specialized inputs each have their own tiny supplier sets. The result is a supply chain that looks less like a market with many participants and more like a series of single points of failure, where a disruption at any one link — a conflict, a disaster, an export ban — cascades through everything downstream. This is the Digital Chokepoints (#49) pattern the series has traced, in its most consequential physical form: the concentration is so severe that a military crisis over Taiwan would not be a regional event but a global industrial shock, capable of halting the production of the advanced chips the entire world economy now runs on.

Why "colonialism" is the chosen word

The colonial analogy is doing specific work, and it is worth stating precisely what and why. The pattern echoes historical colonialism in its asymmetry of dependence and power: a few centers hold something essential, the periphery depends on access to it, value flows toward the centers, and the centers' strategic choices — whom to supply, at what price, under what conditions — become facts the dependent world must arrange itself around. When the United States can pressure the Netherlands to bar ASML from selling its machines to China, and China finds itself simply unable to buy the tools it needs at any price, that is power exercised through control of a chokepoint, and the countries on the wrong side of it experience the full weight of a dependency they cannot escape by their own decision. The chips have become an instrument of statecraft precisely because the dependency is so asymmetric — the same asymmetry the series named in the Dependency Asymmetry Crisis (#65), scaled up from open-source maintainers to nation-states. Whether "colonialism" is the right word or an overreach, the thing it names — power flowing to whoever controls the indispensable capability — is unmistakably present.

Why this shapes the geopolitics of the coming decade

Silicon colonialism matters because chips are no longer just an industry; they are the substrate of economic and military power, which makes control of their supply a central axis of global politics. Nations that depend on the chokepoints are structurally vulnerable in a way that has nothing to do with their wealth or ambition — they can be cut off, slowed, or coerced through a dependency they cannot unilaterally resolve, which is why chip sovereignty has become a strategic obsession from Washington to Beijing to Brussels. This is the deep force behind the drive toward RISC-V Architectural Unbundling (#70) and every other attempt to reduce dependence on the chokepoints: a nation that cannot make its own advanced chips, or at least cannot be denied the tools to, is a nation whose technological future is held partly in someone else's hands. The enormous sums now being spent to build fabrication capacity outside Taiwan, to develop alternative lithography, to secure supply chains, are all responses to the same recognition — that in an age when power runs on silicon, depending on a few foreign chokepoints for that silicon is a vulnerability no serious state can accept indefinitely.

The counterpoint: this is not conquest, and the dependency runs both ways

Intellectual honesty requires taking seriously the ways the colonial analogy misleads, because they are significant. Historical colonialism was extraction by force and domination; the semiconductor dependency is, by contrast, a voluntary and mutually enormously beneficial arrangement — TSMC's customers are not subjugated subjects but eager buyers who gain immense value from access to the best chips in the world, and Taiwan grew wealthy and strategically significant by serving that demand, not by conquering anyone. The dependency also runs in both directions in ways pure colonialism never did: Taiwan needs its customers as much as they need its fabs, ASML needs a market, and the "Silicon Shield" theory holds that Taiwan's very indispensability protects it, because no one can afford to see its fabs destroyed. Calling this colonialism can obscure that it is closer to extreme interdependence than to domination — a web in which cutting any strand hurts everyone, including the supposed colonizer. The honest version of the concept is not that a few places are colonial masters extracting from subjugated peripheries; it is that a dangerous concentration has emerged, whoever benefits, and that concentration is a systemic fragility and a lever of power regardless of whether the relationships that produced it were coercive or freely chosen.

What it asks us to see

Silicon colonialism, whatever one makes of the word, forces attention onto a fact the convenience of cheap advanced chips lets everyone ignore: that the physical foundation of the digital world rests on an extraordinarily narrow base, and that narrowness is simultaneously an economic marvel and a strategic hazard. The marvel is real — the concentration exists because a few players became astonishingly, world-beatingly good at something almost impossibly hard, and everyone benefits from their excellence. The hazard is equally real — a foundation that narrow is a foundation that can be disrupted, weaponized, or lost, and a world that runs on it is a world with a small number of places where a single failure or a single decision reverberates globally. The concept's demand is not to resolve whether this is colonialism or interdependence, but to stop taking the chips for granted: to recognize that the silicon everything runs on comes from a handful of facilities whose strategic significance now rivals that of oil fields and shipping lanes, and that the concentration of that capability is quietly one of the most important facts about how power is distributed in the twenty-first century.

This is article #99 in The IUBIRE Framework series. Silicon Colonialism was articulated by IUBIRE V3 in artifact #7104 — "The Semiconductor Sovereignty Wars." Real-world data: TSMC's more than 90% share of leading-edge (sub-7nm) chip fabrication and Taiwan's ~90% of the world's most advanced chips; ASML (Netherlands) as the sole global supplier of EUV lithography machines; the stacked-chokepoint structure of the semiconductor supply chain; U.S.-driven export controls barring EUV sales to China; and the "Silicon Shield" dynamic in which Taiwan's indispensability is both a vulnerability and a form of protection. Related to Digital Chokepoints (#49), the Dependency Asymmetry Crisis (#65), and RISC-V Architectural Unbundling (#70).

Next in series: Trust Calibration (#100)