Micron Technology plans second Taiwan facility to meet surging AI demand for advanced chips

Micron Technology Expands Taiwan Chip Manufacturing Footprint

Micron Technology has announced plans to construct a second chip manufacturing facility in Taiwan at the Tongluo site recently acquired from Powerchip Semiconductor Manufacturing Corp 1

1

3

3

. The U.S. memory chipmaker completed the acquisition of PSMC's Tongluo P5 site and plans to match the scale of its existing fabrication plant in Miaoli County 1

1

. Construction is scheduled to begin by the end of fiscal 2026, marking a significant Taiwan chip facility expansion aimed at addressing growing demand from the artificial intelligence sector 2

2

.

Advanced DRAM Products and High-Bandwidth Memory Production

The new facility will focus on expanding supply of advanced DRAM products, including high-bandwidth memory (HBM), to support surging AI demand 3

3

. High-bandwidth memory has become essential for artificial intelligence applications, particularly as companies race to develop more powerful AI systems requiring specialized chip technologies 2

2

. The move reinforces Taiwan's position as a critical hub for semiconductor production amid global competition for AI chip manufacturing capacity 1

1

.

Market Response and Industry Context

Micron shares hit record high on Monday, rising 6% following the announcement, with the company scheduled to report quarterly financial results on March 18 2

2

. The expansion follows broader industry trends as chipmakers scale production to serve AI markets, with companies like Nvidia-backed Scintil Photonics recently shipping AI laser chips, while smartphone manufacturers including Xiaomi and Apple continue expanding their device ecosystems 1

1

. This strategic investment positions Micron Technology to capture a larger share of the rapidly expanding memory market driven by AI applications, particularly as data centers and AI infrastructure require increasingly sophisticated memory solutions to handle complex computational workloads.