Megaport Secures $594 Million for Global AI Inference Cloud Development

Megaport, an Australian network as a service provider, is making a significant push into the artificial intelligence infrastructure market with a substantial investment in a new global AI inference cloud. The company revealed it has secured four new contracts with U.S.-based technology providers, totaling A$458.9 million, specifically for AI applications. To support this ambitious expansion, Megaport intends to raise A$827.3 million, equivalent to approximately $594 million USD, through an entitlement offer. The core of this initiative is the development of a globally distributed AI inference cloud, which will feature an on-demand pool of high-performance GPUs. This infrastructure is designed to cater to the surging demand for AI compute, particularly as enterprise AI adoption shifts from model training to latency-sensitive inference workloads. Megaport plans to offer this GPU pool to enterprise customers through both contracted and consumption-based pricing models, providing flexibility and scalability. A significant portion of the capital expenditure, estimated at nearly A$369.5 million, will be allocated to acquiring high-performance Nvidia GPUs, along with the necessary network and storage infrastructure. Megaport believes its extensive network of over 1,100 data centers across 31 countries provides a strong foundation to deliver AI compute closer to end-users, thereby addressing critical bottlenecks related to power, connectivity, and access to specialized hardware. This strategic pivot positions Megaport as a crucial "picks-and-shovels" player in the ongoing AI gold rush, aiming to benefit from the next wave of AI infrastructure buildout.