AI Acceptable Use Policy
Adapt this template for your business. Replace [bracketed text] with your details.
Purpose and scope
This policy sets out how employees of [business name] may use artificial intelligence (AI) tools in their work. It applies to all staff, contractors and volunteers who use company devices, accounts or data. The aim is to help people use AI productively while protecting client information, confidential data and the business's reputation.
Approved AI tools
The following AI tools are approved for business use, subject to the rules in this policy:
- Microsoft 365 Copilot (within our own tenant)
- [Add any other approved tools]
Any tool not on this list must be approved by [IT contact / manager] before use.
Tools that must not be used
Free, public AI tools that train on submitted data must not be used for any business information. This includes pasting client or company data into consumer chatbots. If in doubt, ask before you type.
Information that must never be entered into AI tools
The following must never be entered into any AI tool unless it is an approved tool within our own secure environment:
- Client data and correspondence
- Personal data (names, emails, addresses, National Insurance numbers)
- Passwords, keys and credentials
- Confidential financial information
- Anything marked confidential or subject to a non-disclosure agreement
Checking AI output before use
AI can produce plausible but incorrect answers. Every AI output must be checked by a competent person before it is relied on, sent to a client or published. You are responsible for anything you produce using AI.
Microsoft Copilot rules
Microsoft 365 Copilot may be used within our tenant because it respects existing permissions and keeps data inside our environment. Even so, outputs must still be checked before use. Copilot must not be used to share files or conversations externally without normal approval.
Intellectual property
Do not enter the business's or a client's intellectual property into any AI tool that could use it to train its models. Respect third-party copyright and licences when using AI to generate content.
Reporting concerns
If you accidentally enter restricted data into an AI tool, or receive a suspicious AI-generated message, report it to [IT contact] immediately. No one will be blamed for reporting an honest mistake promptly.
Review date and sign-off
This policy will be reviewed at least annually. It was last reviewed on [date] and approved by [name/role]. All staff must confirm they have read and understood it.
Getting Copilot-ready safely?
We check permissions, oversharing and security before you switch Copilot on.
Book a Copilot readiness assessment