Does every business really need an AI policy?

Yes, if anyone on the team uses AI. A one-page policy prevents the biggest risk, which is private information pasted into unapproved tools, and gives people confidence to use AI well.

Should we ban ChatGPT at work?

Bans tend to push AI use underground. Approving specific tools, with clear rules about data and review, is usually safer than a ban.

What should an AI policy for a church include?

The same basics, plus clear lines on pastoral care information, prayer requests, and how AI may or may not be used in preaching and teaching.

How to write an AI policy for your team (free template)

By Rafael · Updated October 19, 2026 · 2 min read

The short answer

A good AI policy fits on one or two pages and answers five things: which tools are approved, what information must never go into them, when a person must review AI work, who approves AI tools that take actions, and who to ask. Copy the template below and fill in the brackets.

Why you need one now

Across 2026 surveys, the same pattern shows up in businesses, nonprofits and churches: most people already use AI, but few organizations have written guidelines. That gap means private information ends up in personal AI accounts, sometimes called "shadow AI," and everyone guesses at the rules.

A short policy isn't a ban. It's permission with guardrails, so your team can use AI confidently.

Free AI policy template

AI use policy template

[Organization name] AI Use Policy
Last updated: [date] · Questions: [name, email]

1. Why we use AI
We use AI tools to save time on routine work and to help us serve [customers / members / the people we serve] better. People stay responsible for everything we send, publish or decide.

2. Approved tools
You may use: [list approved tools and account types, e.g. "Claude Team account", "ChatGPT Business", "Microsoft Copilot through our Microsoft 365"].
Using other AI tools for work requires approval from [name/role].

3. What you may never put into AI tools
• Passwords, keys or login details
• Payment card, bank or Social Security numbers
• Health, counseling, pastoral or HR information about a person
• [Customer / donor / member] personal information, unless the tool is approved for it
• Confidential contracts or anything we've promised to keep private

4. Human review
A person must review AI output before it is:
• sent to anyone outside our team
• published (website, social media, newsletters)
• used to make a decision about a person (hiring, services, discipline)
Check facts, names, numbers and quotes. AI can be confidently wrong.

5. Being honest about AI
We don't present AI-written work as personal when that matters, like [sermons, personal letters, testimonials]. [Optional: note when content was AI-assisted.]

6. AI agents and connected tools
AI tools that can take actions (send email, change records, spend money) need approval from [name/role], the minimum access needed, and a person approving important actions.

7. Learning and help
New team members get a short AI training. Questions or mistakes? Tell [name/role]. We'd rather hear early than late.

8. Review
We review this policy every [6 months] because AI tools change quickly.

How to roll it out

Ask what people already useA quick, no-blame survey. You'll learn where AI already helps and where data might be at risk.
Pick approved toolsPrefer business or team plans, which usually offer better privacy terms than free accounts.
Fill in the template togetherInvolve a few staff members so the rules match real work.
Train for an hourWalk through the policy with real examples of good and bad use.
Review every six monthsAI tools change quickly; your policy should too.

Common questions

Does every business really need an AI policy?: Yes, if anyone on the team uses AI. A one-page policy prevents the biggest risk, which is private information pasted into unapproved tools, and gives people confidence to use AI well.
Should we ban ChatGPT at work?: Bans tend to push AI use underground. Approving specific tools, with clear rules about data and review, is usually safer than a ban.
What should an AI policy for a church include?: The same basics, plus clear lines on pastoral care information, prayer requests, and how AI may or may not be used in preaching and teaching.