The four contracts, all with U.S.-based technology providers running AI applications, are expected to start in the first half of 2027 and require nearly A$369.5 million in capital expenditure, primarily for high-performance Nvidia GPUs, network and storage infrastructure.