Five pillars, stacked. Governance on top, Change Management at the base — because every layer below decides whether the one above it actually works. The framework scales with risk, not with bureaucracy.
Who decides what, and how. Six components arranged in four layers so the framework scales with risk, not with bureaucracy.
What: Use cases classified Low, Medium, or High before pipeline entry. Why: The framework scales to the risk, not to the bureaucracy — tier determines the approval path, not a one-size gate.
What: All AI tools evaluated against security, privacy, data, and procurement standards before deployment. Why: Ungoverned tool proliferation is how breaches and conflicting AI behaviors enter quietly.
What: Rules on what data can enter AI systems — what trains, what passes to external APIs, what's off-limits. Why: Most AI governance failures are data failures; define rules before use cases are built.
What: AI outputs reviewed against quality and risk thresholds before operationalizing; threshold scales with tier. Why: Deploying an AI tool isn't the same as deploying a trustworthy one.
What: BU vs. central conflicts escalate to the Executive Sponsor for binding resolution within 5 business days. Why: Without it, governance stalls at the first real conflict.
What: A central AI office sets guardrails: approved tools, data standards, risk policy. BUs operate freely inside those boundaries. Why: Hybrid moves fast where it can and applies rigor where it must.
"Governance isn't the thing that slows AI down. Bad governance is. Good governance is what lets you move fast in the right places."Foxpath · Pillar 01
Who does what, and how they connect. Three roles, six components, one operating rhythm — the part of the framework that actually runs.
What: AI Transformation Lead (central strategy + build), AI Champions (embedded BU execution), Executive Sponsor (active governance + escalation). Why: Three crisp roles with non-overlapping mandates.
What: Chairs quarterly program review; makes binding escalation decisions within 5 days. Why: Champions are only as protected as their exec air cover.
What: High performers identified and recruited centrally — for influence, credibility, and functional knowledge, not just enthusiasm. Why: The wrong Champion stalls a whole BU.
What: Run monthly discovery sessions, own BU pipeline entries, represent the AI program, report blockers weekly. Why: A role without a mandate is a title.
What: Turns workflow conversations into scored, risk-tiered AI opportunities. Why: Built first; used in every Champion discovery session — the entry point for the whole pipeline.
What: Weekly Champions sync for pipeline updates and blocker removal; dedicated channel between syncs. Why: A weekly heartbeat is what turns a structure into an operating model.
"You can't delegate adoption without delegating authority. The operating model is where that line gets drawn."Foxpath · Pillar 02
How work moves from idea to deployed value. Three intake channels, two gates, three delivery stages, and a feedback loop — the part of the framework that turns ambition into shipped outcomes.
What: Use cases enter via open intake form, Champion-led discovery sessions, or top-down exec priorities. Why: All three feed one unified pipeline — no shadow backlog.
What: First gate. Every use case tiered Low / Medium / High before scoring. Why: Tier determines the approval path it will follow — wrong tier wastes the rest of the pipeline.
What: Every use case scored on value, effort, readiness, and specificity before resources are committed. Why: An unscored pipeline is a wishlist; scoring is what makes it a plan.
What: Manual + AI assist → AI-assisted workflows → full automation / agentic. Why: Stage gates prevent premature scaling and protect the pilot's learnings.
What: Every use case piloted at small scale. Why: Validates ROI assumptions against a documented baseline before scaling investment.
What: Post-deployment learnings feed back into the scoring rubric and intake criteria. Why: The framework improves with every cycle — or it ossifies.
"A pipeline that doesn't gate is a queue. A pipeline that doesn't learn is just a longer queue."Foxpath · Pillar 03
What gets built to run the program. Five tools, one philosophy — the instrumented layer that makes the rest of the framework operable, measurable, and improvable.
What: Single source of truth for active use cases — tier, stage, owner, status, baseline metrics, next action. Why: Visible to all stakeholders means no parallel spreadsheets.
What: Structured 2-week onboarding — AI fundamentals, the scoring rubric, discovery methodology, program tools. Why: Champions ship faster when the on-ramp is the same.
What: Tracks productivity, cost reduction, and revenue enablement against documented pre-deployment baselines. Why: Real numbers, not estimates — what makes the program defensible.
What: Adoption rate, active use cases, Champion engagement, time-to-deploy, BU coverage. Why: Separate from use case ROI — a program can ship wins while quietly dying.
What: Onboards Champions and end users to tools, scoring methodology, and discovery techniques. Why: Built light, iterated on adoption data — not pre-engineered.
What: Start light, ship fast, iterate on real usage. Why: No overbuilt platforms before the org is ready — overbuild kills programs faster than missing tools do.
"The first version of every tool ships in weeks. Overbuilt platforms kill programs faster than missing ones."Foxpath · Pillar 04
How people actually adopt, not just how tools get deployed. Six components that diagnose the org, earn permission, build skill, and manufacture momentum — because the program lives or dies on everything above it.
What: Map stakeholders by function, influence, and adoption risk before launch. Why: Identify champions, skeptics, and blockers upfront — surprises kill adoption.
What: Every measurable win — hours saved, cost reduced, process accelerated — is documented, quantified, and socialized. Why: Momentum is manufactured, not waited for.
What: Role-based, not role-agnostic. End users get workflow-specific training, Champions get methodology, execs get ROI literacy. Why: Generic training produces generic adoption.
What: Categorize resistance — fear of job loss, workflow disruption, trust in AI outputs, tool fatigue. Why: Each type gets a specific response; "more training" is rarely the right one.
What: Early adopters get tools first, generate visible wins, and create pull from the majority. Why: Push deploys; pull adopts.
What: Explicit permission to experiment and fail at the crawl stage. Why: Punishing early AI mistakes kills adoption faster than any technical barrier.
"The four pillars above this one decide what gets built. Change Management decides whether anyone uses it."Foxpath · Pillar 05