Free team deliverable, copy print and adopt in 15 minutes
AI Rules of the Road
A free, practical team governance doc for Copilot, ChatGPT, Gemini, Claude, and any AI tool your team uses.
Most teams adopt AI tools without any shared rules — which leads to data leaks, unreliable outputs, and inconsistent quality. This one-page document gives your team a shared standard: what AI is good for, what to avoid, how to verify outputs, and who is accountable. No legal jargon. No IT overhead. Just a clear, practical guide you can adopt immediately.
Data safety first
Clear rules for what can and cannot be pasted into AI tools, with no guesswork for your team.
Verification built in
A two-pass review method so external-facing outputs are always checked before they go out.
Human accountability
Every output has an owner. AI is the drafting assistant and a human is always responsible.
Rules (core)
Copy, print, and share with your team in 15 minutes.
Data safety (sanitization required)
- Do not paste confidential, proprietary, regulated, or customer-identifying information into AI tools unless explicitly allowed.
- When in doubt, sanitize: remove names, emails, IDs, internal URLs, account numbers, credentials, and sensitive details.
- Prefer structure + placeholders (ClientName, Date, PolicyName) over raw data.
Good fits
- Drafting: emails, SOPs, agendas, summaries, macros
- Structuring: messy notes to clean outline
- Editing: tone, clarity, concision
- Brainstorming: options and pros/cons (with verification)
Avoid or escalate
- Legal/compliance interpretation or formal advice
- Medical or financial advice
- Security-sensitive material (credentials, incident details)
- High-impact decisions without verification
Verification is mandatory
- External-facing outputs require verification of facts, numbers, dates, and policy references.
- Use a two-pass method: generate, then review with checklist before sending.
- Prefer draft-with-placeholders over invented specifics.
Quality standards
- Use clear structure: headings, bullets, short paragraphs.
- Ask for assumptions when info is missing.
- Keep tone professional and aligned to your brand voice.
Accountability
- A human is always accountable for final output quality and correctness.
- AI is a drafting assistant, not a source of truth.
How teams use this
Adopt rules in a 15-minute team meeting
Use the verification checklist before sending external content
Store prompt systems in a shared space
Review and refine monthly
Governance-lite
Practical internal standards, not a compliance project
This document is the starting point for a team governance-lite posture: safe-use rules, verification requirements, escalation paths, and accountability assignments. It is not a substitute for formal legal, security, or regulatory compliance work.
- Print or share this doc as your team’s AI usage agreement
- Assign an owner (manager or ops lead) responsible for maintaining it
- Review and update quarterly as tools and policies evolve
- For formal governance, AI policies, or compliance frameworks, consult your legal or security team
Mini FAQ
Does this replace an official AI policy?
No. It's a practical starter guide. For formal policies, consult internal security/legal.
Do we need Copilot?
No. Copilot-first is recommended for Microsoft shops, but these rules apply across tools.
How do you reduce hallucinations?
Verification checklist + source grounding + human review.