An AI policy on one page: give your team clarity without banning AI

Here is an uncomfortable truth: your employees are probably already using AI, even if you have no policy on it. They ask ChatGPT to rewrite an email or summarise a text. The question is not whether it happens, but whether it happens safely. A short AI policy gives that direction, without putting on the brakes.

Why “banning” does not work

A ban sounds decisive, but pushes usage underground. People keep using it, just secretly and without asking the questions they should. Clear boundaries work better than a ban that gets ignored anyway.

The three questions your policy answers

A usable AI policy fits on one page and answers three things:

1. Which tools may we use? Name the approved (preferably business) tools, and make clear that random free services are not the intent.
2. Which data may never go in? Client data, personal data, confidential figures, trade secrets, be concrete here. See also AI and the GDPR.
3. Who checks the output? Set out that AI results are always checked by a human, especially for client work, financial or legal documents.

Add a few practical ground rules

• Be transparent about where AI was used, where that is relevant.
• Anonymise data before you enter anything.
• Stay responsible for the end result yourself, AI is an aid, not an excuse.
• Report doubt. Make it easy for people to ask questions.

Keep it alive

As with your IT policy: a document in a drawer protects no one. Go through it once with the team, and update it when there are new tools or insights. AI changes fast; your policy may move with it.

From policy to advantage

A good AI policy is not only protection, it also gives your team the confidence to use AI where it safely can, and that is where the time saving is. We help you draw up a policy that fits your profession and your risks. See our AI & automation service or book a conversation.