Video guide · AI governance program · October 2026
Most organizations already use AI. Very few can show how it’s governed. In this video, I walk through 8 practical steps to build an AI governance program from scratch, with a deliverable for every step, so you finish with something you can actually show leadership, auditors, and customers.
The 8 steps and what each one produces
| Step | What you do | Deliverable |
|---|---|---|
| 1 | Define your objectives and scope: why you’re governing AI and what’s covered | Objectives and scope statement |
| 2 | Assign ownership and accountability, so someone actually owns AI governance | AI governance charter and RACI |
| 3 | Find every AI system in use, including shadow AI and AI features inside existing tools | AI system inventory |
| 4 | Identify the laws and standards that apply, such as the EU AI Act, U.S. state laws, NIST AI RMF and ISO/IEC 42001 | AI regulatory obligations register |
| 5 | Assess AI risks and prioritize by risk tier | AI risk assessment and risk register |
| 6 | Set policies and controls, with fast approval for low-risk AI and deeper review for high-risk AI | AI governance policy and control checklist |
| 7 | Train people on what their role actually needs, with real examples | AI awareness and training plan |
| 8 | Monitor, review, test and report, because approval isn’t the end | AI monitoring and governance review plan |
Where to start: Step 3. You can’t govern AI you don’t know exists. Download the free AI System Inventory Template and build your first inventory this week.
Want to know where your organization stands?
Take the free AI governance assessment, or subscribe on YouTube for a new practical AI governance video every week.
Take the free assessment Subscribe on YouTubeGeneral education, not legal advice. Have counsel review the legal requirements that apply to your organization.