Capability 1.1.3.1.1

Cloud Inferencing Compute Infrastructure as a Service (public cloud)

Definition

Cloud-based compute resources used to run AI models - rather than train them - offered as an on-demand commercial service. This includes national edge zones and micro data centres positioned close to the use case to reduce latency (time delay).

Examples

Multinational hyperscalers (AWS, Google Cloud, Azure, Oracle Cloud). Local/regional AI cloud providers (e.g. Sharon AI).

Indicators

Number and scale of local cloud clusters
Geographic distribution of inferencing capacity (relevant to latency)
Number of accelerators

National AI Plan

PLAN COMMITMENT

Significant commitment

See general and specific comment at 1.1.2.1.1 (Cloud Training Compute Infrastructure as a Service (public cloud)). Significant commitment in the plan strongly aligns with recommended action to prioritise building agency in this enabling capability.

Stocktake

AI Compute Sovereignty: Infrastructure Control Across Territories, Cloud Providers, and Accelerators. Hawkins et al. SSRN. June 2025.https://www.researchgate.net/publication/392999259_AI_Compute_Sovereignty_Infrastructure_Control_Across_Territories_Cloud_Providers_and_Accelerators

Is the Stocktake missing evidence?

Share additional information for potential future Assessments