Singtel to Deploy Nvidia GB300 Systems in Asia-Pacific
The sovereign cloud provider will introduce the new hardware by October 2026 to support secure enterprise AI compute.
Updated on Oct. 28, 2026 in Data Centers
Live Poll
Do you trust sovereign cloud services to handle your organization's sensitive AI data requirements?
Singtel plans to deploy Nvidia DGX GB300 systems within its RE:AI sovereign cloud infrastructure for the Asia-Pacific region by October 2026. The new GPU-as-a-Service model allows enterprises to access high-performance computing without managing internal hardware.
Why it matters
This deployment addresses enterprise demand for sovereign AI cloud infrastructure that meets specific data residency and security requirements. By offloading hardware management, organizations can scale AI compute capacity while maintaining control over data jurisdiction.
The deployment centers on Nvidia DGX GB300 hardware, a platform designed for large-scale AI training and inference. This represents the first implementation of the GB300 architecture within the Asia-Pacific region.
The players
Singtel
A telecommunications company managing RE:AI, a sovereign cloud platform designed to provide secure, localized AI computing infrastructure for enterprise clients.
Nvidia
A semiconductor company that develops the DGX GB300 systems, a specialized hardware architecture optimized for AI training and large-scale data center acceleration.
The details
The RE:AI platform utilizes a GPU-as-a-Service model, where compute resources are virtualized to allow remote enterprise access. By shifting infrastructure to a sovereign cloud, companies can ensure that data processing remains compliant with local regulatory standards. The DGX GB300 hardware functions as a centralized cluster for high-bandwidth AI workloads, providing the underlying compute power required for enterprise-scale machine learning models.
Timeline
Singtel announced the plan for the system deployment in October 2026.
The Tech Race
This deployment marks the first regional integration of Nvidia's GB300 platform into a sovereign cloud framework. It sets a new benchmark for Asia-Pacific service providers racing to secure regional enterprise AI workloads against hyperscale competitors.
Enterprise organizations in the Asia-Pacific region gain access to sovereign AI compute power without the capital expenditure of building private clusters. Initial capacity is currently fully committed, requiring new users to await subsequent expansion phases.
The takeaway
Singtel's move highlights the growing regional priority for sovereign cloud models that balance performance with local data control. Watch for capacity expansion announcements in 2027 to see if demand for these sovereign clusters continues to outpace supply.
Further reading
For more on the infrastructure supporting current AI growth, visit our Data Centers section.
Source note: This article includes information reported by Telecompaper.
Live Poll
Do you trust sovereign cloud services to handle your organization's sensitive AI data requirements?
