Could you defend your AI decisions today?
Fifteen questions, about two minutes. Check the ones you can honestly say are true today. The boxes you cannot check are where your exposure is, and your to-do list.
Why this matters now.
AI governance is moving from good practice to expectation. Canada's Directive on Automated Decision-Making already binds federal decisions about people, OSFI's guideline E-23 raises the bar on model risk for regulated firms, privacy law keeps tightening, and the EU AI Act reaches organizations that serve the EU. The question a board or a regulator will ask is simple: can you show that your use of AI is governed. This is a fast, honest way to see where you stand before someone else asks.
Check what is true today.
Accountability and ownership
Policy and process
Risk and impact
Data and privacy
Human oversight
Transparency
Vendors and procurement
Monitoring
People and capability
What your score says.
Start checking the boxes above to see where you stand.
Exposed.
You are using or preparing to use AI without the basics to govern it. If a decision went wrong, or a regulator asked, you would struggle to show control. Start with the essentials: name an accountable owner, write a policy people can follow, and put a review step before anything goes live.
Whatever your score, the boxes you did not check are your priorities.
Developing.
You have real governance in places, but it is uneven, and the gaps are exactly where risk concentrates. Prioritize the unchecked items in risk, human oversight and data, since those are the ones that hurt most when they fail.
Whatever your score, the boxes you did not check are your priorities.
Solid foundation.
Most of the essentials are in place. The work now is to close the last gaps and move from ad hoc to repeatable, documented practice, so your governance holds up when it is examined, not just when it is convenient.
Whatever your score, the boxes you did not check are your priorities.
Strong.
Strong governance. The task now is to keep it current as the rules tighten and your AI use grows, and to be able to prove it holds under scrutiny. This is where a lightweight, regular review earns its keep.
Whatever your score, the boxes you did not check are your priorities.
For the bigger picture of where AI fits across your organization, try our AI maturity self-check. AI maturity self-check