Company AI use policy: a template you can copy
Your staff are probably already using AI, some of it on personal accounts. A one-page policy tells them what is fine, what is not, and who to ask. Copy this template and fill in the brackets.
The short answer
A useful AI policy fits on one page and answers four questions: which tools are approved, what data must never go in, who checks the output, and who to ask.
Have your lawyer review it against the data protection rules in your country before you issue it.
The template
- Purpose: [Company] encourages staff to use AI to work faster and better. This policy keeps our customers, staff and company safe while we do.
- Approved tools: staff may use [tool names] with their company accounts. Ask [name or role] before using anything else. Personal accounts must not be used for company work.
- Never enter: passwords and keys; customer, employee or candidate personal data unless the tool is approved for it; salaries; contracts; financial records; anything a client asked us to keep confidential.
- Check everything: you are responsible for what you send or publish, whoever or whatever drafted it. AI can state wrong facts with confidence, so check names, numbers, dates and legal points.
- People decide: AI must not make final decisions on hiring, pay, discipline, dismissal, credit or refunds. It may prepare information for the person who decides.
- Customers: when customers talk to an AI system, we tell them, and we offer a way to reach a person.
- Publishing: do not publish AI images or text that could mislead, such as pictures of properties or products that do not exist.
- Automation: AI systems that take actions, send messages or connect to company systems need approval from [role] first.
- Problems: report any mistake or data leak to [name or role] the same day. Reporting early will not be held against you.
- Review: [role] owns this policy and reviews it every six months. Last updated: [date].
How to adapt it
- Regulated sector: add your regulator’s rules, and confirm where approved tools store data
- Client work: check whether your contracts limit the use of AI or require you to disclose it
- HR teams: add the rules for AI in hiring and performance
- Developers: add rules on company code, secrets and reviewing generated code
- Arabic-speaking staff: publish it in Arabic too, so that nobody signs what they cannot read
How to roll it out
- Ask staff what they use today, with no penalty for answering honestly
- Pick approved tools and pay for business accounts, so nobody needs a personal one
- Spend 30 minutes with each team on real examples of what is fine and what is not
- Keep a simple list of AI tools, owners and the data each one sees
Put AI to work with clear rules
keyt builds AI agents with approval steps and activity logs, connected to the systems you already use. Tell us what you want to automate.