GMI Cloud Founder and CEO Alex Yeh described the project as a strategic investment in Taiwan’s AI ecosystem. He highlighted the need for more local data centres to serve as “strategic assets” for AI development and expressed confidence that Taiwan’s power supply challenges could be addressed. “You want to promote local ecosystems — you have to build the data centre first, you have to build the AI cluster first,” Yeh said.
The announcement comes amid a global surge in investment in AI infrastructure, as technology giants race to meet increasing demand for AI workloads. Nvidia, which supplies the bulk of its revenue from GPU sales, has described such clusters as “AI factories” and has recently signed deals to supply advanced chips to projects in Saudi Arabia and South Korea. U.S. policymakers, including former President Donald Trump, have emphasized reserving top AI semiconductor technologies like Nvidia’s Blackwell chips for domestic companies.
Taiwan has seen several large-scale AI infrastructure initiatives, including a 100-megawatt data centre project jointly announced by Foxconn and Nvidia in May. GMI Cloud, a GPU-as-a-Service provider and Nvidia cloud partner, already operates data centres in the United States, Taiwan, Singapore, Thailand, and Japan.
Alongside the Taiwan facility, GMI Cloud plans to construct a new 50-megawatt data centre in the United States and is exploring an initial public offering in the next two to three years. Once operational, the Taiwan AI centre is expected to generate around $1 billion in total contract value, with early clients including Nvidia, cyber-security firm Trend Micro, electronics manufacturer Wistron, Chunghwa System Integration, data infrastructure provider VAST Data, and industrial solutions company TECO.