Micron breaks ground on Singapore wafer fab in planned $24 billion expansion aimed at AI-driven memory demand

Micron Technology said it has broken ground on an advanced wafer fabrication facility in Singapore, launching a major capacity expansion that the company plans to fund with about $24 billion over the next decade. The project will be located within Micron’s existing NAND manufacturing complex and is designed to support the surge in demand for memory and storage driven by artificial intelligence and data-centric computing.

According to the company, the facility is expected to ultimately add roughly 700,000 square feet of cleanroom space, a key indicator of the scale of the manufacturing buildout. Micron also said wafer output is scheduled to begin in the second half of 2028. The timeline reflects the long lead times required for semiconductor mega-projects, from site work and tool installation to process qualification and yield ramp.

Micron positioned the investment as a way to meet long-term NAND requirements and to improve resilience in its global manufacturing network. Singapore is already a central node for Micron’s flash memory operations, and the company indicated that co-locating research and development with manufacturing can improve efficiency, accelerate time-to-market, and deepen partnerships across the ecosystem. These dynamics matter because NAND and related storage technologies remain foundational to AI training and inference workloads, which generate and move massive volumes of data.

The announcement also highlighted workforce impacts. Micron said the fab expansion will create around 1,600 jobs, with roles spanning engineering, operations, and advanced manufacturing capabilities such as robotics and AI-driven process control. The company framed these jobs as part of a broader push to modernize fabs with smart-manufacturing tools that can improve output quality, reliability, and cost structure.

Investors have been watching memory markets closely because supply discipline and demand growth can quickly reshape pricing. The recent boom in AI infrastructure has tightened conditions for several memory categories, and leading chipmakers have been signaling that expansion must be paced carefully to avoid the classic cycle of oversupply and margin compression. Micron’s messaging emphasized flexibility in ramp speed to align capacity additions with market conditions.

Taken together, the Singapore build is another sign that the AI era is not only about GPUs and accelerators, but also about the memory and storage layers that feed them. With production not expected until 2028, the project is best read as a long-duration bet: that global demand for advanced storage will remain structurally higher through the end of the decade, and that governments and firms will keep investing to secure supply chains for critical compute components.