No Nvidia GPU without ASML. No Apple silicon without ASML. No AI revolution without ASML. The Dutch company holds 100% of EUV lithography — the only technology capable of printing advanced AI chips. Market cap $733 billion today. 2030 revenue guidance €44–60 billion. The path to the trillion-dollar club runs through Veldhoven.
Not investment advice. Not a recommendation to buy or sell. Research and long-horizon thinking only. Consult a qualified financial advisor before making any investment decision.
There is a building in Veldhoven, Netherlands where machines are assembled that cost €350 million each, weigh 180 tonnes, require 100,000 precision parts, take years to build, and cannot be replicated by any other company on earth. These machines — ASML's Extreme Ultraviolet lithography systems — print the circuitry on every advanced semiconductor in existence.
The Nvidia H100. The Apple M4. The AMD MI300X. The Samsung HBM chips. Every one of them is printed by an ASML machine. Not a machine like ASML's. An ASML machine. There is no alternative. Nikon and Canon make lithography tools — but not EUV. ASML has 100% market share in the technology that makes the AI revolution physically possible.
That is not dominance. That is something rarer — structural necessity. And structural necessity, at the scale of the global semiconductor industry, is worth approximately $733 billion today. It is worth more tomorrow.
"No Nvidia GPU without ASML. No Apple silicon without ASML. No AI revolution without ASML. That is not dominance. That is structural necessity."
EUV lithography works by firing 13.5-nanometer light — shorter than any other commercial light source — at silicon wafers to etch microscopic features measured in atoms. The physics required to generate, control, and direct that light at manufacturing scale took ASML and its suppliers 30 years and approximately $10 billion in cumulative R&D to develop.
The supply chain is extraordinary in its complexity. Carl Zeiss makes the mirrors — precision-ground to tolerances measured in fractions of a hydrogen atom. Cymer makes the light source. Trumpf makes the laser. More than 800 suppliers across Europe and the US contribute components that collectively make up one of the most complex machines humanity has ever built. No single company could replicate this supply chain. No government could mandate it into existence. It evolved over three decades through sustained investment, deep expertise, and irreplaceable institutional knowledge.
EUV requires 13.5nm light generated by converting tin plasma. The physics knowledge to do this at manufacturing scale took 30 years to accumulate. It cannot be shortcut.
800+ precision suppliers, decades of co-development. Carl Zeiss mirrors alone require tolerances of fractions of a hydrogen atom. No competitor can replicate this in any reasonable timeframe.
TSMC, Samsung, SK Hynix, Intel have built entire fabs around ASML tools. Switching cost is effectively infinite — you would have to rebuild the fab from scratch.
ASML's next EUV generation — High-NA (EXE:5200) — extends the moat further. The first system was accepted in 2025. No competitor is within a decade of this technology.
ASML's market cap today is approximately $733 billion. The path to $1 trillion requires a 36% increase from current levels. With 2030 revenue guidance of €44–60 billion and gross margins projected at 56–60%, the earnings trajectory supports that valuation — and the timeline is tighter than most analysts expect.
The re-rating argument is straightforward. ASML is currently valued as a high-quality semiconductor equipment company. It should be valued as essential AI infrastructure — the same category as the hyperscale cloud platforms. When that re-rating happens — and it is happening, quarter by quarter as the AI narrative deepens — the valuation ceiling moves meaningfully higher.
ASML has one significant risk: China. In 2025, China represented 33% of ASML's total revenue — a legacy position from years of selling older DUV systems to Chinese chipmakers. US export restrictions have progressively tightened access. By Q1 2026, China had fallen to 19% of system sales. Full-year 2026 guidance assumes China at approximately 20%.
The bear case says further restrictions — banning even DUV sales to China — would remove another €5–8 billion in annual revenue. The bull case says this loss is being replaced faster than expected by orders from TSMC, Samsung, SK Hynix, and Intel for EUV systems that China cannot access. The backlog data supports the bull case: €38.8 billion at year-end 2025, with €7.4 billion in EUV-specific bookings. The customers that matter most are ordering more, not less.
The monopoly is real and durable. 30 years, $10 billion, 800 suppliers, customer lock-in at the fab level. This is not a competitive advantage that erodes in a cycle. It is structural necessity embedded in the global semiconductor supply chain.
The valuation is not cheap. At $733B and 48x forward earnings, ASML is priced for continued excellence. Every earnings miss will be punished. The stock requires patience and tolerance for volatility — not entry at any price regardless of near-term results.
The China risk is real but manageable. China revenue falling from 33% to 20% is already in the guidance. The replacement demand from non-China customers is visible in the backlog. This is a known risk that the market has partially priced.
The $1 trillion timeline could be faster or slower. Early 2027 is the NGE base case. If High-NA ramp accelerates and AI infrastructure spending sustains, it could happen by late 2026. If China restrictions escalate or hyperscaler capex pauses, the timeline extends. The destination is not in question. The timing is.
ASML is the most important company most retail investors have never owned. It sits at the intersection of the AI supercycle, the semiconductor sovereignty race, and the energy transition — three of the $1Q's 16 forces. It is Europe's answer to the question of whether non-US technology companies can reach planetary scale. The answer, increasingly, is yes.
The most important company in the AI revolution is not in Silicon Valley. It is in Veldhoven, Netherlands. It employs 44,000 people. It has 100% market share in a technology that took 30 years to build. And it is about to join the $1 trillion club. The question is not whether it gets there. The question is whether you are invested before it does.
Written from first principles. Not consensus. Not noise. Long-horizon thinking on capital, technology, and the forces shaping the next decade of wealth creation. Published when something is worth saying — not on a schedule.