Capability-Centric Enterprise Architecture for Business-Technology Alignment and AI Adoption

This paper presents an integrated overview of four complementary enterprise architectures artifacts: Business Capability Maps, Technical Capability Maps, shared technical implementation models, and business-to-technical traceability matrices. Business Capability Maps capture the stable 'what' of the enterprise, while Technical Capability Maps capture reusable technology-side abilities independent of specific products and projects. Shared technical implementation models define how common capabilities are delivered across teams using shared, federated, or local patterns, improving reuse, governance, and speed. Traceability matrices provide the relationship layer that links business priorities and capabilities to technical capabilities, platforms, applications, and controls for prioritization, gap analysis, and impact analysis, and governance. The paper then explains how to adopt artificial intelligence (AI) across these artifacts by modeling AI as business value (AI-enabled sub-capabilities), technical enablement (inference, guardrails, monitoring, auditability), shared platform capability (federated enterprise AI), and DevOps extensions (MLOps and LLMOps). The proposed approach enables organizations to align AI investments with business outcomes, risk controls, and operational resilience.