AI Change Management Consulting
Make GenAI usable under policy: named champions, an acceptable-use line, and a rule that Custom GPTs are prototypes — not a poster and a license blast.
- Service
- Enablement
- Industry
- Enterprise
- Updated
- 2026-08-25
- Engagement
- 4 wks
AI change management consulting gets a workforce using ChatGPT Enterprise and Microsoft Copilot under a signed acceptable-use policy, with named champions and a written split: copilots and Custom GPTs are assistive prototypes; write-actions and evals belong to production agents. It is a four-week standard tied to your IdP and workspace, not a six-month “AI culture” program. You own the policy, playbooks, and any GPTs; inference stays on your provider contracts with no markup.
Why teams pick this engagement
Enablement × EnterpriseAcceptable use that security signs
What may go into ChatGPT Enterprise or Copilot, what must stay in a production agent, and what is banned. Written with InfoSec, not after the first incident.
Adoption you can measure
Active use on approved scenarios, unpublished unowned GPTs, and a drop in consumer ChatGPT traffic — not seat counts as success.
Tool-honest enablement
We teach when Copilot is the wrong tool (write-actions, evals) and when a Custom GPT is only a prototype. Training that pretends every job is a copilot creates shadow agents.
Champions over town halls
A named owner per function, office hours, and scenario playbooks. Mass e-learning completion is not change management.
Four-week standard
Week 1 stakeholder and shadow-IT map, week 2 policy and playbooks, week 3 champion pilots, week 4 comms and handover. Tied to a real workspace, not a culture deck.
Artifacts the org keeps
Policy, playbooks, champion roster, and FAQ live in your stack. SSO/IdP remains yours; you own any GPTs created in training.
Key takeaways
- 01
Change management fails when IT dumps Copilot or ChatGPT Enterprise seats without a scenario list and a data policy.
- 02
Consumer ChatGPT will not die until there is a faster approved alternative and a consequence for pasting customer data into it.
- 03
Copilot is often the wrong tool when write-actions and evals are required; training must say that out loud.
- 04
Custom GPTs created in workshops are prototypes — they need owners, expiry, and a promotion path, or they become shadow policy.
- 05
Four weeks of champions, policy, and office hours beats a 40-page change strategy no one opens.
What the engagement covers
How we work
- 01
Discover
Week 1: shadow ChatGPT, current Copilot/Enterprise posture, union or works-council constraints, blocked-by-security history, sponsor map.
- 02
Design
Acceptable use, tool split (Copilot vs GPT vs agent), champion roster, playbook list, measurement.
- 03
Build
Week 2–3: policy through legal/security, playbooks, champion dry runs on the real workspace with SSO.
- 04
Validate
Pilot functions, GPT review-board dry run, comms legal review, consumer-ChatGPT alternative is actually faster.
- 05
Enable
Week 4: org comms, office hours, metrics dashboard, 30 days on-call. Your champions run it; we do not become the AI helpdesk.
Take the playbook with you
The working documents from real engagements — free, in exchange for an email. They’re useful whether or not we ever talk.
GenAI Acceptable-Use One-Pager
What employees may paste into ChatGPT Enterprise or Copilot, what requires a production agent, and what is banned — the page legal and security actually approve.
Get the one-pager ·Champion and Scenario Playbook Kit
Function-level scenarios, office-hours agenda, and the GPT-vs-agent talking points we use in week 3.
Get the kit ·