Client login
AI Agents
Services
AI Property AcceleratorAI Hub: guides and articlesPricingAboutContact
Book an AI Audit: £197Free Brain Score
Safety, GDPR and policy

AI policy template for UK businesses (free, plain English)

A free, copy and paste AI policy template for UK small businesses, with plain English wording for staff and a note on UK GDPR.

By Imz, Founder · AI Solutions Architect6 min read5 November 2026
Safety, GDPR and policy guide

Your staff are probably using AI already. Someone is pasting an email into ChatGPT to tidy the wording. Someone else is using it to summarise a client call. That is fine, until a client's personal details go into a tool you have never checked.

A short AI policy fixes that. This page gives you a free AI policy template you can copy, edit and use today. It is written in plain English for a small UK business.

Important: this is general guidance, not legal advice. If you are a regulated firm, such as a solicitor, accountant, mortgage broker or healthcare provider, take legal advice and check your regulator's rules before you adopt it.

Why a small business needs an AI policy

A policy does three jobs. It tells staff what is allowed. It protects client information. It shows you took reasonable care if anything goes wrong.

The ICO is the UK data protection regulator. Its guidance says data protection law applies when AI is used to process personal data, and that there is no "AI exemption" to data protection law. So the usual UK GDPR rules still apply when your team uses AI tools.

The ICO also says its AI guidance is not legal advice and takes a risk-based approach. That suits a small firm. You do not need a 40 page document. You need clear rules that people actually read.

Is ChatGPT GDPR compliant?

This is the question most owners ask. The honest answer is that no tool is compliant on its own. Compliance depends on how you use it.

If staff put personal data into any AI tool, UK GDPR applies to that use. You need to ask a few things first:

  • Is personal data going in at all? The safest rule is to keep it out.
  • Which plan or account is being used? Business plans and free personal accounts can have very different terms. Read the provider's data terms for the exact plan.
  • Where does the data go, and can the provider use it for training? Check the provider's privacy notice and settings.
  • Can you explain this use to the people whose data it is?

So do not ask "is ChatGPT compliant?". Ask "is this use of ChatGPT safe for this kind of data?". Your policy answers that for your team.

The free AI policy template

Copy the wording below into a document. Replace anything in square brackets with your own details. Keep it to one or two pages.

  1. Purpose

    "This policy explains how [Company name] staff may use artificial intelligence (AI) tools at work. We want to use AI to save time while protecting our clients, our people and our business. It applies to all employees, contractors and freelancers."

  2. Approved tools

    "Staff may only use the AI tools on our approved tools list. The list is kept by [Name]. Do not sign up to a new AI tool for work, or use a personal account for work tasks, until it has been approved."

  3. What you may put into AI tools

    "You may use AI for drafting, summarising and brainstorming using public or non sensitive information. Examples are rewording your own text, writing a first draft of a blog post, or planning a meeting agenda."

  4. What you must not put into AI tools

    "Do not enter client or customer personal data, such as names, addresses, phone numbers, email addresses, account details or case notes. Do not enter staff records or health information. Do not enter confidential business information, such as contracts, pricing not yet released, passwords or financial accounts. If you are not sure whether something is allowed, do not enter it. Ask first."

  5. Human review before anything is sent

    "AI output is a draft. A person must read and approve everything before it goes to a client, is published or is used to make a decision. AI must not make decisions about individuals on its own."

  6. Accuracy checks

    "AI tools can state wrong facts with confidence. Check names, figures, dates, legal points and quotes against a reliable source. If you cannot verify it, do not use it. You are responsible for anything you send, whether or not AI helped write it."

  7. Record keeping

    "We keep an AI use register. It lists each approved tool, what we use it for, what data it may touch and who owns it. Tell [Name] if you start using AI for a new task so it can be added."

  8. Who to ask

    "Questions, concerns or mistakes go to [Name, email]. If you think personal data has been entered into an AI tool by mistake, tell [Name] the same day."

  9. Review date

    "This policy was approved on [date] and will be reviewed on [date, ideally within 6 to 12 months] or sooner if our tools or the law change. Owner: [Name]."

Build your approved tools list

The policy only works if people know which tools they can use. Keep the list short and put it on one page. A simple table is enough:

ToolUsed forData allowedOwner
[Example: chatbot, business plan]Drafting and summariesPublic and non sensitive only[Name]
[Example: meeting notes tool]Internal meeting notesInternal only, no client calls[Name]
[Example: image tool]Marketing imagesNo personal images[Name]

Before you approve a tool, read its privacy notice and data terms. Check whether your inputs can be used to train the tool. Check where data is stored. Choose the business plan over a free personal account where the terms are better.

Keep an AI use register

An AI use register is a simple record of where AI is used in your business. For each use, write down the tool, the task, the data involved, who owns it and when you last checked it.

Why bother? UK GDPR has record keeping rules. The ICO says organisations with fewer than 250 staff only need to record certain processing, such as processing that is recurring, risky for individuals, or involves special category data. The ICO also says keeping these records is good practice even where the exemption applies. A register is a tidy way to do that for AI.

Whether you must record a particular use depends on your facts. If you handle personal data in AI tools, ask a data protection adviser.

Brief your staff and get sign-off

A policy nobody reads protects nobody. Run a 30 minute briefing. Walk through the do and do not rules with real examples from your own work. Then ask each person to sign to say they have read and understood it.

Keep the signed sheets. They show you told your team the rules. Repeat the briefing when the policy changes and when new people join.

ICO AI guidance: where to start

If you want to read the source, start with the ICO's pages on AI and data protection. The main guidance on AI and data protection is organised around the core data protection principles. These include fairness, data minimisation, accuracy, security and accountability. It is written mainly for compliance and technical teams, so expect some of it to be heavy going. The ICO also lists practical resources on AI.

The ICO says it uses "must" for legal requirements and "should" for good practice. Keep that in mind when you read it.

Common mistakes to avoid

  • Copying a policy and never using it. Make sure the tools and names in it are real.
  • Banning AI outright. Staff will still use it, just out of sight. Approved use is safer.
  • Forgetting free accounts. Staff may use personal logins. Say clearly that this is not allowed for work.
  • No review date. AI tools change fast. Put the date in the diary.
  • No owner. Name one person who keeps the list and the register.

Want it done for you?

If you would rather not write this yourself, we can help. The Agentic Scale AI Audit costs £197. It reviews how your business actually uses AI today, so you know what is in use and where the risks are.

For £100 more, the optional AI Governance Pack gives you a written AI policy, an approved tools list, an AI use register you can use as a GDPR record, and a 30 minute staff briefing with a sign-off sheet. It is a practical set of documents, not legal advice. Regulated firms should still take legal advice.

Sources

Want this working in your business? The AI Audit maps where AI fits and prices it. £197, credited if you build.

FAQ

Questions,answered.

Anything else, ask Imz directly.

Is ChatGPT GDPR compliant?
No tool is compliant on its own. UK GDPR applies to how you use it. The ICO says there is no AI exemption to data protection law. The safest rule is to keep client personal data out of it, read the provider's data terms for your plan, and take advice if you need to use personal data. This is guidance, not legal advice.
Do small businesses need an AI policy?
There is no single law that says every small business must have one. But if staff use AI tools at work, a short policy sets clear rules, protects client data and shows you took care. Regulated firms should check their regulator's rules and take legal advice.
What should an AI policy include?
A basic AI policy covers purpose, approved tools, what staff may and may not enter, human review before anything is sent, accuracy checks, record keeping, who to ask and a review date. The template above covers each of these.
Where can I find ICO AI guidance?
The ICO publishes guidance on AI and data protection on ico.org.uk, plus practical resources and generative AI material. It is guidance and good practice, not legal advice. The sources below link to the pages we used.
Taking new clients for November

Not sure where to start?Book an AI Audit.

We look at how you work today and tell you honestly what AI should and shouldn't do for you. £197, credited in full if you build within 30 days.