the world's best by an incredible margin
At an investor conference, Nvidia's Jensen Huang offered an unusually candid tribute to a partner rather than his own company — declaring Taiwan Semiconductor Manufacturing the world's best chipmaker by an 'incredible margin.' The statement illuminates a quiet truth about the AI era: the most consequential power in the semiconductor revolution may not belong to those who design the chips, but to the singular entity capable of bringing them into physical existence. TSMC's dominance is less a corporate achievement than a structural fact of modern technology, one that positions it to benefit from the AI boom regardless of which company wins the design wars.
- Jensen Huang's public endorsement of TSMC — crediting a partner over his own company — signals just how irreplaceable the foundry has become in the AI supply chain.
- With over 60% of global foundry spending and no credible rival, TSMC holds a chokepoint in the semiconductor industry that competitors cannot bypass without accepting inferior performance or higher costs.
- Global AI chip spending is racing toward $193.3 billion by 2027, and TSMC is structurally positioned to capture the majority of that manufacturing demand no matter which chip designer wins.
- The company's dominance feeds itself — revenue funds R&D, R&D widens the technical lead, and the widening lead attracts more contracts, deepening a moat that grows harder to cross each year.
- Despite forecasts of 20%-plus annual earnings growth over five years, TSMC trades at just 20x 2025 earnings — a valuation gap that analysts suggest the market has not yet fully reckoned with.
Jensen Huang took the stage at an investor conference this month and said something unusual for a CEO: he praised a partner more than his own company. Taiwan Semiconductor Manufacturing, he declared, is the world's best chipmaker by an 'incredible margin.' Coming from the leader of Nvidia — a company that has added $2.5 trillion in market value during the AI boom — the statement was not modesty. It was a precise observation about where the real foundation of the AI era is being built.
TSMC is the world's largest semiconductor foundry, translating chip designs from companies like Nvidia into physical silicon with a precision no competitor can match at scale. It commands more than 60% of global foundry spending. When Nvidia needs a cutting-edge processor manufactured, it goes to TSMC. When rivals need the same, they face an uncomfortable choice: use TSMC or accept worse results at higher cost. Huang acknowledged Nvidia could theoretically switch foundries, but made clear no alternative could deliver what TSMC does. The moat, he implied, is real and widening.
What makes TSMC's position particularly durable is its capacity to scale in lockstep with surging demand. When AI chip orders exploded, TSMC had the infrastructure to meet them. That capability is not incidental — it is exactly what every major technology company now needs as AI investment accelerates. IDC forecasts global AI chip spending will reach $193.3 billion by 2027, growing at 18% annually. TSMC is positioned to capture the majority of that manufacturing work regardless of which company designs the chips.
The financial picture sharpens the story further. TSMC's dominance generates revenue that funds the next generation of manufacturing processes, which attracts more contracts, which funds the generation after that — a compounding cycle that leaves competitors further behind each year. Analysts forecast earnings growth of more than 20% annually over the next five years. Yet the stock trades at just over 20 times 2025 earnings — a valuation that, by most measures, has not caught up with the trajectory. For those seeking exposure to the AI boom, TSMC may be the most structurally sound and least celebrated way in.
Jensen Huang stood at an investor conference earlier this month and made a striking claim about one of his own company's most important partners. Taiwan Semiconductor Manufacturing—TSMC—is not just good at what it does. It is, by his measure, the world's best chipmaker by what he called an "incredible margin."
The statement carries weight because Huang knows the landscape intimately. Nvidia has spent the last two years riding a wave of demand for artificial intelligence chips that has added $2.5 trillion to its market value. The company's gross margins have expanded into the upper 70s, a sign of extraordinary pricing power. Its operating leverage has translated into explosive profit growth. By any measure, Nvidia has been the defining winner of the AI boom. Yet here was its CEO, publicly crediting another company as the true leader of the industry.
TSMC is the world's largest semiconductor foundry—the manufacturing partner that takes chip designs from companies like Nvidia and prints them onto silicon wafers with extraordinary precision. It commands more than 60% of global foundry spending. When Nvidia needs to produce a cutting-edge processor, it goes to TSMC. When competitors need the same capability, they face a choice: use TSMC or accept inferior performance or higher costs. There is, in practical terms, no real alternative. Huang acknowledged that Nvidia could theoretically switch foundries if forced to do so, but he also made clear that no competitor could match what TSMC delivers. The company's technical capabilities and operational scale have created a moat that grows wider each year.
What makes TSMC's position even more durable is its ability to scale. When demand for AI chips exploded, TSMC had the infrastructure and expertise to meet it. That capacity to grow production in lockstep with customer needs is not a commodity skill. It is precisely what a company like Nvidia needed to capitalize on the AI opportunity. And it is precisely what every other major tech company now needs as they ramp up their own AI spending. The market research firm IDC forecasts that global spending on AI chip content and related systems will reach $193.3 billion by 2027, up from $117.5 billion this year—an 18% annual growth rate over the next three years. TSMC will capture the majority of that spending regardless of which company designs the chips. Whether the growth comes from Nvidia, from competitors, or from custom chips built by major tech companies themselves, TSMC is positioned to win most of the manufacturing contracts.
This dynamic creates what analysts call a virtuous cycle. TSMC's market dominance generates more revenue, which it reinvests in research and development to build the next generation of manufacturing processes. That technological lead attracts more major contracts. The cycle repeats, and the gap between TSMC and its competitors widens. The company has room to expand its profit margins as it scales. Analysts on average forecast that TSMC will grow its earnings by more than 20% annually over the next five years. Yet the stock trades at just over 20 times the consensus earnings forecast for 2025—a valuation that does not fully price in that growth trajectory.
For investors, the arithmetic is striking. Few companies offer the combination of high growth potential and reasonable valuation that TSMC currently presents. The company is not just the world's best foundry by an incredible margin, as Huang said. It may also be one of the most undervalued ways to gain exposure to the AI boom that has reshaped the technology industry.
Citas Notables
We're fabbing out of TSMC because it's the world's best. And it's the world's best not by a small margin, it's the world's best by an incredible margin.— Jensen Huang, Nvidia CEO
Competitors' capabilities can't match TSMC's and it would result in less performance or higher cost.— Jensen Huang, on why Nvidia cannot realistically switch foundries
La Conversación del Hearth Otra perspectiva de la historia
Why does Huang's endorsement of TSMC matter more than just a CEO being nice to a partner?
Because he's essentially saying there is no real competition. TSMC doesn't win on price or speed or some narrow advantage—it wins on technical capability that competitors simply cannot replicate. That's a structural advantage, not a temporary one.
But TSMC is already huge. It has 60% of the foundry market. Where's the growth?
The growth is in the total market itself. AI spending is forecast to nearly double by 2027. TSMC will capture most of that expansion because it's the only place companies can go to manufacture cutting-edge chips at scale.
So TSMC wins whether Nvidia wins or whether Nvidia's competitors win?
Exactly. That's the real insight. TSMC is agnostic to who designs the chips. It just manufactures them. So it benefits from the entire AI buildout, not just one company's success.
Huang said TSMC could theoretically be replaced. Why doesn't that worry investors?
Because he immediately said the alternative would mean accepting either worse performance or higher costs. In a market where cutting-edge capability is everything, that's not a real option. Switching foundries would be a competitive disaster.
The stock trades at 20 times 2025 earnings with 20% annual growth forecast. That sounds cheap.
It is, relative to what the company is expected to deliver. Most high-growth tech stocks trade at much higher multiples. TSMC is offering growth at a reasonable price, which is rare.
What's the risk?
Geopolitical risk around Taiwan is the obvious one. There's also the possibility that the AI spending boom slows faster than expected. But structurally, TSMC's dominance in manufacturing is not going away anytime soon.