This design assumes a centrally governed operating model where AI execution is distributed across on-premises and external environments while control, identity, and policy enforcement remain tightly centralised. Ownership is anchored in a platform team responsible for workload placement discipline, preventing uncontrolled sprawl across hybrid environments. Consumption is regulated through orchestrated execution pathways rather than direct infrastructure access, ensuring consistent policy enforcement. Execution is grounded in an AWS-compatible hybrid model, with anchoring through identity and observability controls such as Azure Active Directory and Splunk to maintain governance integrity.