We are entering the Age of Intent — people state what they want, and systems orchestrate the response. That is a fundamental architectural change, not an interface upgrade.

But intent without context is dangerous. An agent that interprets intent without understanding process boundaries, governance constraints, and organisational meaning will execute confidently and incorrectly. The first mile of enterprise AI is not model selection. It is context translation — turning human intent into governed, executable action at the resolution agents require.

This session explores what AI-first enterprise architecture looks like in the Age of Intent: how architects design for intent-driven systems, how the Context System (data, semantic, process, governance, and intent layers) provides the foundation agents need to act correctly, and how value realisation shifts from project-based ROI to compounding returns as the context architecture matures. We cover SAP's approach through Joule Work, the Knowledge Graph, Process Atoms, and SAP Company Memory — the governed layer where intent meets execution and the enterprise stops re-explaining itself to every new agent and workflow.