Shah Zaib
For most of the twentieth century, the word “chokepoint” belonged to oil. Whoever controlled the Strait of Hormuz, the Suez Canal, or OPEC’s spare capacity held real leverage over the global economy. Today that language has changed to a different terms: silicon.
A wafer no larger than a mid size table, etched with billions of transistors, now decides whether a nation can train the next generation of artificial intelligence, guide a missile, or run a transaction on a device. Chips have not merely joined oil as a strategic commodity; in several respects they have surpassed it, and the shift is quietly rewriting how states define security and self-sufficiency.
Unlike petroleum, which is extracted across dozens of geologically diverse producers, the frontier of semiconductor manufacturing is concentrated almost to the point of monopoly. Taiwan Semiconductor Manufacturing Company alone commands roughly seventy percent of the world’s contract foundry market and manufactures more than ninety percent of the most advanced semiconductor chips on Earth.
One Dutch firm, ASML, builds the only machines on the planet capable of the extreme ultraviolet lithography needed to print those chips, a position it has held since introducing EUV systems commercially in 2019 with no competitor able to match the technology.
Layer onto this the design software concentrated in a handful of American firms such as NVIDIA, and the rare-earth elements refined predominantly in China, and a single production chain runs through perhaps four narrow gates. Disrupt any one of them, and the flow of chips to smartphones, data centres, vehicles, and Defence equipment slows or stops. This is a fragility that oil markets, spread across OPEC and non-OPEC producers alike, have never had to face.
This is why the analogy of chips as “the new oil” can be a reality but ultimately understates the risk. Oil chokepoints are geographic; they can, in principle, be routed around with alternative pipelines or tankers, even at cost. Silicon chokepoints are technological and cumulative. ASML spent three decades and roughly nine billion dollars in research reaching its EUV monopoly, and Taiwan’s dominance rests on engineering learning loops built over forty years that cannot simply be transplanted elsewhere.
Washington’s increasing export control regime against China, tightened in September 2025 and again through the MATCH Act that Congress passed in April 2026, reflects a possibility that this concentration is durable enough to be weaponised quickly.
Beijing’s countermove, restricting rare-earth exports, proved strong enough to force a temporary one-year truce at emergency talks in Busan, but analysts increasingly argue that rare-earth leverage is transactional and reversible, whereas the chip chokepoint compounds. China can’t just buy its way to the frontier, it needs years of hands-on manufacturing failures and fixes to get there, and that’s precisely what the controls deny it.
The scale of this asymmetry is severe when set side by side, as summarised below. It is this imbalance that explains the surge of policy responses since 2022: the United States has subsidised domestic fabrication, pursued a tiered global licensing system for advanced AI chips, and pressed allies in Japan and the Netherlands to align their export rules.
China has answered with its own indigenisation drive, and Chinese foundries such as SMIC and Hua Hong are edging toward mature seven-nanometer production despite restrictions on the most advanced lithography tools. Neither side can fully decouple, but both are hedging: building parallel, less efficient supply chains rather than the single integrated one that lowered global costs for three decades. The cost of that redundancy will ultimately be paid by consumers and by developing economies that possess neither fabrication capacity nor bargaining leverage of their own.
Table : Comparing Oil Chokepoints with Semiconductor Chokepoints
| Chokepoint | Approx. Global Share | Alternatives | Time of Replicate |
| Strait of Hormuz | ~20% of global oil trade | Alternative routes and pipelines exist | Months to years |
| TSMC / Taiwan (advanced-node foundry) | ~70% of foundry market; ~90% of chips below 7nm | No alternative foundry at comparable scale | A decade or more |
| ASML (Netherlands, EUV lithography) | 100% of EUV lithography systems | No competitor produces EUV machines | Decades (30 years of R&D already invested) |
| China (rare-earth refining) | Majority of global processed rare earths | Diversifiable with new refining capacity | Roughly 5–10 years |
For countries like Pakistan, watching this contest as outsiders, the lesson is not that they must build fabrication plants of their own; that window has largely closed or is impossible to build. It is that economic and foreign policy can no longer be planned without accounting for where computing power sits.
Trade diversification, digital infrastructure planning, and the choice of which technology bloc to align with are no longer purely commercial decisions; they are strategic ones, carrying consequences for growth, security, and sovereignty alike. Semiconductors have become the materials of both prosperity and power.
The states that grab this earliest, treating chip dependency with the seriousness once reserved for energy security, will be better placed to navigate a decade in which silicon, not crude oil, increasingly sets the terms of geopolitical and geostrategic rivalry.
Shah Zaib is a final-year undergraduate student of Philosophy, Politics, and Economics (PPE) at the Aga Khan University Faculty of Arts and Sciences, Karachi. His interests span geopolitics, economics, and strategic studies, with a particular focus on semiconductor supply chains and great-power competition.
