davidfacer.com / aimaturitymodels.com / AI-Native Maturity Models / Enterprise Architecture / Whole-Model View

Enterprise Architecture — Whole-Model View

Ten dimensions, five maturity levels each. Select a cell to read it — click, or use the arrow keys. The detail is always open.

Dimension	ANascent	BModeled	CContinuous	DIntegral	ETelemetric
D1Architecture process	Individuals use AI ad hoc; definitions vary by team.	Humans author the schema; AI normalizes legacy documents.	Chartered agents author; humans validate semantics continuously.	Agents simulate proposed changes before executive approval.	Agents detect semantic drift, propose refactoring from telemetry.
D2Architecture development (artifacts and substrate)	Static documents; AI cannot reliably ingest the architecture.	Structured templates; AI validates but humans hold truth.	Agent maintains schema and generates human views on demand.	Engineering agents inherit constraints from the same schema.	Agent-maintained nodes self-update from telemetry, showing reality.
D3Business linkage	Shadow AI pilots; strategy disconnected from execution backlogs.	AI drafts capability maps; linkage cross-referenced by hand.	Agents reject work lacking verifiable capability lineage edges.	Agents inherit runtime constraints from capability map nodes.	Telemetry measures agent-delivered value realization per capability node.
D4Senior management involvement⚑	Executives absent; AI use unsponsored and semantically unaligned.	Executives author intent and arbitrate agent-relevant semantic disputes.	Executives consume agent recompositions; funding requires intent lineage.	Executives steer using agent simulations before committing capital.	Executives revise intent on agent-surfaced telemetry, rarely.
D5Operating unit participation⚑	Units bypass architecture using unsanctioned local AI tools.	Units define domains; consume static AI-drafted reference models.	Units query the agent for tailored planning recompositions.	Unit agents inherit enterprise constraints from central repository.	Units act on telemetry; agents propose local refactoring.
D6Architecture communication⚑	Ask an architect; no authoritative channel for agents.	Central repository published; humans and agents must navigate.	Agent answers architectural questions by recomposing the graph.	Functional assistants parse architecture, defending ontology during daily work.	Agents initiate alerts on drift, degradation, boundary breaches.
D7Governance and authority plane (Auctoritas)	Agents act on implicit rights; no stated surfaces.	Agent authority mapped with extent; humans enforce manually.	Automated gates clear agent actions; surfaces probed, not assumed.	Runtime mediates every agent act; exclusions declared.	Telemetry reports agent boundary hits and un-fired controls.
D8Strategic portfolio management and asset allocation	AI spend fragmented; no shared pricing or lineage.	AI-assisted pricing maps proposals; allocations still locked annually.	Agents refuse funding to initiatives lacking verifiable lineage.	Compute and agent capacity route to high-yield capabilities.	Engine reallocates agentic capacity on realized unit economics.
D9Architecture governance	No governance; agent and human work fragments unchecked.	Review boards inspect artifacts; agent output checked manually.	Teams and agents self-align; friction triggers targeted review.	Detected friction clusters trigger agent-supported root-cause architecture review.	Agents propose ontological refactoring; review rate measures health.
D10Shared language (Vocabula)	No shared senses; agents inherit and spread the ambiguity	Vocabulary governed and findable; no agent consults it	Agents challenge ambiguous terms at authorship, citing the vocabulary	Agents resolve inward; outward translation records what it collapsed	Agents measure usage; the model is corrected on evidence

ANascentD1 · Architecture process

Definition

Architecture processes are manual, document-heavy, and local to IT silos. Semantic definitions vary by team: what counts as a "service" or a "customer" differs between the groups that use the words. AI appears informally, as an individual's brainstorming tool, with no shared prompts, no common ontology, and no method for validating what it produces.

Transition to B

Define a foundational enterprise ontology in a machine-readable form, and a standardized, AI-assisted method for translating business capabilities and technology standards into it. Replace fragmented word-processing documents and static diagrams as the place architectural meaning lives.

Verification: Two teams that previously used a term differently now resolve it to the same definition by looking it up, and can say where that definition lives.

What D1 measures

The capability to create, maintain, and align architectural intent, and the degree to which semantic discipline and AI assist in translating human strategy into executable governance.

Where it ends

Continuously refine the ontology as new business models emerge. The specific risk at this level is an ontology that stops being revised because everything downstream now depends on it: stability of meaning and ossification of meaning look identical from inside.

⚠ = open review item

Currency basis: short-form cells pinned at c30334f; full model text pinned at c30334f. One basis throughout.