---
id: "framework-hub-and-spoke-implementation"
type: "framework"
source_timestamps: ["§ Effect #1: The “Technology-first” Trap"]
source_url: "https://hbr.org/2025/09/dont-let-ai-reinforce-organizational-silos"
source_title: "Don't Let AI Reinforce Organizational Silos"
tags: ["governance", "organizational-structure"]
related: ["concept-hub-and-spoke-ai", "action-build-hub-and-spoke", "entity-bathurst-insurance"]
steps: ["\\\"Establish an AI Center of Excellence (CoE) to act as the central hub", "housing top AI experts", "strategic leaders", "and shared resources.\\\"", "\\\"Task the CoE with aligning all AI initiatives to corporate goals by providing governance", "best practices", "and shared infrastructure.\\\"", "Embed AI teams (the 'spokes') within specific business functions.", "Enable the embedded teams to leverage the CoE's shared platform and standards while applying their deep domain knowledge to solve specific business problems."]
sources: ["tail2"]
sourceVaultSlug: "hbr-seg-tail2"
originDay: 2
articleStem: "hbr-tail-130-ai-reinforce-silos"
sourceUrl: "https://hbr.org/2025/09/dont-let-ai-reinforce-organizational-silos"
sourceTitle: "Don’t Let AI Reinforce Organizational Silos"
---
# Implementing a Hub-and-Spoke AI Governance Model

A structural framework for balancing centralized AI governance with decentralized, domain-specific execution to prevent siloed AI adoption. It is the operational recipe behind the [[concept-hub-and-spoke-ai]] structure and the [[action-build-hub-and-spoke]] action item; the authors' worked example is [[entity-bathurst-insurance]].

**Steps:**

1. **Establish an AI Center of Excellence (CoE)** to act as the central *hub*, housing top AI experts, strategic leaders, and shared resources.
2. **Task the CoE with alignment** — align all AI initiatives to corporate goals by providing governance, best practices, and shared infrastructure.
3. **Embed AI teams (the “spokes”)** within specific business functions.
4. **Enable the embedded teams** to leverage the CoE's shared platform and standards while applying their deep domain knowledge to solve specific business problems quickly.

**Enrichment nuance:** Multiple vendors (Microsoft, IBM, Oracle, Moveworks, Mario Thomas on multi-speed governance) endorse centralized standards with decentralized/federated execution. Watch the maturity caveat: Microsoft notes organizations may start with a centralized CoE and later transition toward an advisory model to keep delivery close to the business — avoid designing the hub as a permanent gatekeeper. See open question [[question-coe-funding-model]] for the unresolved budget mechanics.