A Simple AI Checklist Teams Should Follow Before Using Any AI Tool
AI is now part of everyday work.
We use it to draft emails, analyse reports, summarise documents, and generate ideas.
Most AI problems don’t happen because the model is bad.
They happen because people skip basic safety checks — often because they’re in a hurry.
I’ve been talking to leaders about AI governance, and I’ve found that even a simple checklist can reduce risk dramatically.
Here’s a practical checklist any team can use before AI-generated work is shared, approved, or sent outside the organisation.
✅ AI Safety Checklist
1️⃣ Did a human review the output carefully?
Nothing AI produces should go out without human judgement.
2️⃣ Did we upload sensitive data?
Internal strategy, financials, customer info, patient data — if yes, pause.
3️⃣ Do we know where the tool stores information?
If the answer is “not sure,” treat it as untrusted.
4️⃣ Can we explain the result?
If we can’t explain why the answer looks the way it does, we shouldn’t rely on it.
5️⃣ Would we be comfortable if this became public?
If the answer is no — don’t use it.
6️⃣ Who approves the final action?
AI never owns decisions. A human always does.
Governance starts with habits, not big frameworks
Policies and frameworks are important.
But real governance begins with simple behaviours:
✔ review
✔ awareness
✔ accountability
✔ respect for data
When teams follow small checklists like this, AI becomes safer — and leaders sleep better.
@Japmandeep #Japmandeep
Japmandeep “Sunny” Ahluwalia
(Working on AI governance, responsible adoption, and Shadow AI)
Top comments (0)