According to telset.id, Micron has announced a $3 billion investment to strengthen the U.S. semiconductor supply chain, with $500 million allocated to GlobalWafers for its 300 mm silicon wafer fab in Sherman, Texas.
Strategic Investment in Domestic Wafer Capacity
Micron’s $3 billion commitment targets upstream semiconductor infrastructure — specifically raw silicon wafers, the foundational substrate for DRAM, NAND, and advanced logic chips. Of that total, $500 million is designated as strategic funding for GlobalWafers, supporting its operational 300 mm wafer facility in Sherman, Texas — the only advanced 300 mm silicon wafer manufacturing site currently active in the United States. The agreement includes a 10-year supply contract, guaranteeing Micron access to output from the Sherman plant. GlobalWafers opened the facility in May 2025 and received a $406 million CHIPS Act award covering both the Sherman site and its silicon-on-insulator facility in St. Peters, Missouri.
Expanded Capital Commitment and Accelerated Timelines
In parallel, Micron raised its total planned capital expenditure in the U.S. to more than $250 billion through 2035, up from a prior target of $200 billion. This reflects long-term confidence in domestic semiconductor growth, particularly driven by AI and cloud computing demand. The company also confirmed that the first concrete pour at its Clay, New York campus occurred one-quarter ahead of schedule. That 142-hectare site is designed for up to six phases, with Phase 1 already operational. Meanwhile, Micron’s Idaho Fab ID1 is projected to begin wafer production in mid-2027, while full-scale output at Clay is expected around 2030.
Market Context and Supply Constraints
The investment comes amid acute global wafer supply tightness. Silicon wafer shipments reached 3.275 billion square inches in Q1 2026 — a 13.1% year-over-year increase — largely fueled by AI data center demand across memory, logic, and power devices, according to SEMI.org. Micron has informed investors it can fulfill only 50–67% of current customer demand for DRAM. Current U.S. fabs — including those in Manassas, Virginia (where Micron began producing 1-alpha DRAM in May 2026) — still rely on imported 300 mm wafers from Japan, Taiwan, Germany, and South Korea. Just five suppliers — Shin-Etsu, SUMCO, GlobalWafers, Siltronic, and SK Siltron — dominate global 300 mm wafer supply, making this layer the most concentrated in the chip value chain.
Keeper of the Wafer Gate: A New Partnership Model
Doris Hsu, chairperson and CEO of GlobalWafers, previously outlined strict conditions for expanding Sherman’s capacity — including profitability in the first two phases, long-term customer contracts, fair pricing, upfront payments, and government support. Micron’s announcement fulfilled nearly all those criteria in one transaction. This marks a structural shift: rather than waiting for suppliers to expand independently, chipmakers are now directly financing upstream capacity. As one industry analyst noted,
“Customers are now funding capacity expansion — a reversal from past cycles where suppliers bore the risk.” — Industry observer, cited by telset.id
The move mirrors dynamics seen during the 2017–2018 megacycle but carries higher stakes given today’s geopolitical pressures and AI-driven demand surge.
Implications for Global Supply Chains
For supply chain professionals, Micron’s strategy signals a broader trend: vertical integration into critical raw material layers is no longer optional. With U.S. policy actively incentivizing onshoring via the CHIPS Act, such investments reduce exposure to regional disruptions — whether port congestion, export controls, or trade restrictions. While near-term DRAM supply constraints remain unchanged — Micron stated explicitly that Thursday’s announcement does not alter its 2026 or 2027 DRAM availability — the long-term impact on cost stability, lead time predictability, and technology sovereignty is substantial. Micron has also signed long-term supply agreements totaling $100 billion, underscoring the scale of committed demand from AI infrastructure builders. For Indonesia and other tech-importing nations, a more resilient U.S. wafer base could ease downstream chip shortages — though direct effects on smartphone or server component pricing will take years to materialize.
Source: telset.id
Compiled from international media by the SCI.AI editorial team.
