TSMC Invests $56 Billion in New Semiconductor Fabs as CEO Wei Acknowledges Supply Shortages May Extend to 2027 and Beyond

The surge in artificial intelligence (AI) has driven semiconductor leaders, most notably TSMC, to channel billions into enhancing their production capabilities. However, despite these efforts, meeting the escalating demand remains a daunting challenge.

TSMC: Struggling to Meet AI Demand Amid Significant Investments

In a recent earnings call, TSMC projected capital expenditures of approximately $56 billion for 2026. This substantial investment is earmarked for the establishment of new chip manufacturing facilities and upgrades to existing ones, all aimed at ramping up production capacity.

Nonetheless, TSMC has acknowledged that this financial commitment will not suffice to satisfy the voracious demand from the AI sector. The current industry landscape reveals chronic shortages of essential components, including GPUs, CPUs, memory, voltage regulators, integrated circuits (ICs), cabling, and other critical materials needed for advanced technological solutions.

Two individuals holding a framed plaque with the text 'First Product in TSMC N2 NanoSheet Technology AMD Venice CCD, ' against a backdrop of presentation slides highlighting an AMD-TSMC celebration.

With prominent firms like NVIDIA, AMD, and Apple consistently placing new wafer orders, TSMC foresees product shortages extending well into 2027. To counteract this situation, the company is planning a global expansion of its facilities, focusing on regions such as Japan, Taiwan, and the United States, with three-nanometer (3nm) manufacturing plants scheduled for construction. The plant in Taiwan is projected to commence operations in the first half of 2027, while its US counterpart will be online by the latter half of 2027. The facility in Japan aims to begin production by 2028.

“In addition to all the new fabs, we continue to convert 5-nanometer tools to support 3-nanometer capacity in Taiwan. We are also leveraging our manufacturing excellence to drive greater productivity across our fab in all locations to generate more wafer output.

We are also focusing on capacity optimization across nodes, which includes flexible capacity support among the N7, N5, and N3 nodes. Thus, we are using multiple levers to do everything we can, wherever we can, however we can to maximize the support to all our customers across all platforms. Also, let me emphasize that while the capacity is tight, we do not pick-and-choose or play favorites among our customers.

C. C.Wei – TSMC President & CEO

According to TSMC’s CEO, C. C.Wei, existing plants have not seen an increase in capacity yet. However, as production capacities peak, upgrades will be initiated to enhance their overall output. This need is urgent, considering the intense supply constraints that TSMC currently faces, pushing vendors to explore alternative chip production strategies.

“To meet the strong demand in AI applications, we are stepping up our CapEx investment to increase our N3 capacity. We are now executing a global capacity plan to support the robust multiyear pipeline of demand for 3-nanometer technologies, which are used by smartphone, HPC/AI, including HBM base dies, automotive and IoT customers.

In Taiwan, we are adding a new 3-nanometer fab to our GIGAFAB cluster in Tainan Science Park. Volume production is scheduled for the first half of 2027. In Arizona, our second fab will also utilize 3-nanometer technologies. Construction is already complete and volume production will begin in the second half of 2027. In Japan, we now plan to utilize 3-nanometer technology in our second fab and volume production is scheduled in 2028.

C. C.Wei – TSMC President & CEO

Companies like Tesla are collaborating with both TSMC and Samsung to develop next-generation AI chips, while also forming their own partnerships with Intel for enhanced chip production capabilities. Intel is expected to secure significant clients with its upcoming 14A process technology. Meanwhile, Samsung has experienced a surge in demand for its foundry services, though it remains primarily focused on producing DRAM, particularly HBM and LPDDR, driven by the rise of Agentic AI.

As TSMC navigates through this AI-driven super cycle marked by widespread supply challenges, the company’s position as the world’s leading semiconductor manufacturer places immense pressure on its shoulders. The significant investments in new facilities should ideally contribute to alleviating some of the ongoing constraints.

For further insights, refer to the news article from EETimes.

Source & Images