# AI Use Policy Builder > A free AI use policy generator for small businesses: seven questions, and a one-page policy written from fixed wording, with no email required. Source page: https://mnpsystems.ca/tools/ai-policy/ MNP Systems is an engineering consultancy based in Brampton, Ontario, Canada, that takes work from anywhere: remotely, or on site where a job needs hands on the equipment. Contact: https://mnpsystems.ca/contact/, contact@mnpsystems.ca, or call 1-888-338-1090 (toll-free; an AI receptionist answers). **Your team’s AI use policy, on one page.** Answer seven questions about how your business uses AI and read the policy they make: the tools your team may use, what never goes into them, who checks what they write, and who looks after the policy. No email needed. It is built in your browser from fixed wording, so nothing you choose leaves this page unless you ask for the policy by email. A practical use policy, not legal advice. The AI policy builder, walked through. An example bakery’s team answers seven questions, one at a time, and the one-page AI use policy writes itself section by section: approved tools, what never goes in, and how to keep it. Video walkthrough (MP4, narrated): https://mnpsystems.ca/video/tools/ai-policy.mp4 ## The seven questions 1. **Team.** How many people work in the business? (one of: 1–5 people; 6–20 people; 21–50 people; More than 50) 2. **Tools.** Which AI tools does the business pay for? (any of: Microsoft 365 Copilot; Google Workspace with Gemini; ChatGPT business plan; Claude business plan; Zapier, Make or n8n) 3. **Personal accounts.** May people use their own AI accounts for work? (one of: No, work accounts only; Only with information that is already public) 4. **Information.** What sensitive information does the team handle? (any of: Customers’ names and contact details; Payment, banking or financial records; Clients’ confidential documents and contracts; Staff records and pay; Designs, source code or trade secrets; Health information) 5. **Customers.** Does AI write anything your customers read? (one of: No; Drafts that a person checks before they go out; An assistant answers customers directly) 6. **Decisions about people.** Does AI help with decisions about people? (one of: No; It may help prepare them) 7. **Meetings.** Are AI notetakers allowed in meetings? (one of: No; In our own meetings only; In any meeting where everyone agrees) ## An example policy For a made-up business of 6 to 20 people that pays for Microsoft 365 Copilot and a ChatGPT business plan, handles customers’ details, financial records and clients’ confidential documents, lets AI draft what customers read, and allows notetakers in its own meetings: ### 1. Why we have it - This policy says which AI tools we use for work, what never goes into them, and who checks what they produce. It applies to everyone who works for us, contractors included. ### 2. Approved tools - We use AI for work only through these work accounts: Microsoft 365 Copilot; ChatGPT, on our business plan. - Personal AI accounts are not used for work, for any task. - In each tool’s admin settings we turn off the use of our data to train the vendor’s models. - A new tool is approved once the policy owner adds it to this list. Ask before trying one. ### 3. What goes where - Never into any AI tool: passwords, keys and access codes. - Only into the approved work accounts, and only as much as the task needs: customers’ names and contact details, payment, banking and financial records and clients’ confidential documents and contracts. - Public information, such as our website and published prices, may go into any approved tool. - If something on this list goes into a tool by mistake, tell the policy owner the same day, so it can be deleted and we can decide whether anyone needs to be told. ### 4. Checking the work - Read everything AI produces before it leaves the business. The person who sends it is responsible for it. - Check every fact, figure, name, date and quote against its source. AI tools state wrong things confidently. - Contracts, invoices, quotes and anything legal or financial are checked line by line by a person before they are sent. ### 5. Customers - AI may draft replies, posts and documents for customers. A person reads and edits each one before it goes out. ### 6. Decisions about people - AI is not used in decisions about people: hiring, pay, reviews, or what a customer qualifies for. ### 7. Meetings - An AI notetaker may be used in our own meetings, with everyone told at the start. It is never used in meetings with customers or partners. ### 8. Who looks after it - The policy owner, named above, keeps this policy and answers questions about it. - Everyone reads this policy when they join and whenever it changes. - We review it every six months, and whenever we add a tool. A practical use policy, not legal advice: have a lawyer review it if you need a legal one. ## Rolling it out A policy is the first step of the AI practice's Team training & AI policy service: hands-on sessions, a playbook for each role, and the tools' settings made to match the policy. ## What is kept Nothing. The policy is built in the browser. Only a visitor who asks for it by email sends their answers, and the policy is built again from them to write that email.