Tap to Trade in Gate Square, Win up to 50 GT & Merch!
Click the trading widget in Gate Square content, complete a transaction, and take home 50 GT, Position Experience Vouchers, or exclusive Spring Festival merchandise.
Click the registration link to join
https://www.gate.com/questionnaire/7401
Enter Gate Square daily and click any trading pair or trading card within the content to complete a transaction. The top 10 users by trading volume will win GT, Gate merchandise boxes, position experience vouchers, and more.
The top prize: 50 GT.
 made up 77% of total wafer revenue. This structure shows TSMC’s revenue increasingly concentrates on cutting-edge technology, with high-margin businesses driving an impressive net profit margin of 47.5% (16 billion net profit / 33.7 billion revenue).
Wei Zhejia’s remarks at the meeting injected strong confidence into the AI industry. He said, “Honestly, I’m also nervous. We plan to invest between $52 billion and $56 billion in capital expenditures. If we’re not careful, it could be a disaster for TSMC.” But he quickly added that he asked cloud service giants, and the conclusion was they made huge money from AI, and they have even more money than TSMC.
Wei Zhejia’s calculation is precise: the world is now competing fiercely for computing power. NVIDIA’s GPUs are in short supply, and TSMC can ride the AI wave by manufacturing these chips. Moreover, AI’s accelerated growth in 2025 has already contributed over 10% of total revenue. While 10% may seem modest, considering TSMC’s annual revenue of $122 billion, that’s $12.2 billion—more than many semiconductor companies’ entire yearly revenue.
Regarding concerns about AI causing power shortages, Wei Zhejia said that cloud service customers told him that power supply issues had been planned for five or six years. The real bottleneck now is TSMC’s chip capacity. They told Wei Zhejia not to worry about other issues but to first solve the chip supply problem. This reveals the current power structure in the AI industry chain: downstream customers (cloud providers) are wealthy and demand high, while upstream suppliers (TSMC) have become the bottleneck.
Wei Zhejia’s $56 Billion Gamble on Global Factory Expansion
TSMC’s capital expenditure in 2026 is expected to reach between $52 billion and $56 billion. About 70%-80% of this will be invested in advanced processes. That’s like burning nearly 400 billion RMB in a year. This scale of investment is extremely rare in manufacturing history, equivalent to building dozens of advanced wafer fabs.
Under Wei Zhejia’s strategic layout, TSMC is rapidly building factories worldwide. Sites in Arizona, Kumamoto in Japan, Dresden in Germany—everywhere are TSMC’s construction sites. His goal is to create a global “wafer empire.” This global deployment is not only a strategy to diversify against geopolitical risks but also a compromise to meet local government demands.
Now, Morris Chang has long stepped back from the spotlight, and TSMC has entered the “Wei Zhejia era.” Facing the insane demand from AI, the current leader Wei Zhejia is aggressively expanding manufacturing capacity worldwide. He openly admits that their cloud clients are wealthier than they are, and the AI demand is real. Although Morris has retired, the system and culture he built continue to drive this finely tuned machine at high speed.
The dinner between Morris Chang and Jensen Huang is not just about sentiment but also laying the foundation for their two companies’ powerful alliance in the AI era. The 94-year-old in a wheelchair attending underscores the importance of this meeting. These two semiconductor legends are endorsing the TSMC-NVIDIA alliance in the AI age in this special way.