# A One-Page AI Policy For BIDs With Small Teams

URL: https://www.starien.com/blog/a-one-page-ai-policy-for-small-bid-teams
Published: 2026-09-28
Author: Geoffrey Moffett
Topics: business improvement districts, AI, AI policy, data protection, governance

> A practical, one-page AI policy for BIDs with small teams: what's allowed, what's off limits, how to handle levy payer data and how to keep humans in charge.

## Quick answer

An AI policy for BIDs sets out which AI tools staff can use, what they can use them for, what information must never go into them, how outputs are checked and who is responsible. For a small team it can fit on one page and should be reviewed regularly as tools change.

## Key takeaways

- Small BIDs need an AI policy because staff are already using AI, with or without one.
- The most important rule is what data never goes into AI tools.
- Every AI output that's published or sent should be checked by a named person.
- Keep it to one page and review it at least once a year, because the tools change quickly.

Here's a quiet truth about most BIDs in 2026: somebody on the team is already using AI. Maybe to tidy an email, maybe to draft social posts, maybe to summarise a long council document. That's good. What's missing is usually an AI policy for BIDs that's short enough to read and clear enough to follow.

Big organisations write thirty-page policies. A BID with two or three staff needs one page that answers the obvious questions. Here's how to write it, plus a template you can adapt.

## Why An AI Policy For BIDs Matters

You handle levy payer contact details, board discussions, partner plans and sometimes sensitive local issues. You also speak publicly for a whole town. A few things can go wrong if AI use is ad hoc:

- personal or confidential information pasted into tools that aren't suitable for it;
- AI-written content published with wrong facts;
- images that look like real photos of your town but aren't;
- a chatbot telling visitors something untrue in your name.

None of these needs a heavy rulebook. They need a few clear lines.

## What To Cover In One Page

### 1. Purpose

One sentence. "We use AI to save time on routine work so we can spend more time with levy payers, and we stay responsible for everything we publish."

### 2. Approved Tools And Accounts

List the tools the team can use and which accounts, for example organisation accounts rather than personal free ones. Note whether settings that limit data use for training have been checked. If you're not sure how a tool handles data, find out before it goes on the list.

### 3. Good Uses

Give examples so people feel confident:

- drafting and editing emails, newsletters and social posts;
- summarising long documents and consultation responses;
- brainstorming campaign ideas and event names;
- turning notes into first drafts of reports;
- checking tone, spelling and plain English.

### 4. Never Put These In

This is the most important section. For example:

- personal data about levy payers, visitors or staff, unless the tool and account have been specifically approved for it;
- anything shared with you in confidence;
- unpublished board decisions or partner plans;
- passwords, bank details or security information.

### 5. Check Before It Goes Out

Every AI-assisted output that's published or sent is checked by a named person for accuracy, tone and fairness. Facts, figures, names, dates and links are verified against the source.

### 6. Images And Video

Decide your line. A common one: AI images are fine for clearly illustrative graphics, but never presented as real photographs of your town, people or events. We explored this in [BIDs using AI for graphics](/blog/are-business-improvement-districts-using-ai-for-graphics-yes-and-it-s-a-problem-but-not-for-the-reason-you-think-2/).

### 7. Being Open

Say when the public is talking to an AI, for example a chatbot on your website. Be honest if asked how something was made.

### 8. Who To Ask And When To Review

Name a person. Set a review date, at least yearly and sooner if you adopt a new tool.

## A Template You Can Adapt

> **[BID name] AI policy**
>
> **Why:** We use AI to save time on routine work. People remain responsible for everything we publish and send.
>
> **Tools:** [list] using [organisation] accounts only.
>
> **Use it for:** drafting, editing, summarising, brainstorming, plain-English checks.
>
> **Never enter:** personal data (unless approved), confidential information, unpublished decisions, passwords or financial details.
>
> **Always:** check facts, figures, names, dates and links before anything goes out. A named person signs off.
>
> **Images:** AI images are illustrative only and never presented as real photos.
>
> **Chatbots:** visitors are told they're talking to an AI and how to reach a person.
>
> **Questions:** ask [name].
>
> **Review:** [date].

## Working Through The Grey Areas

Policies get tested by the awkward cases, so talk a few through as a team before they happen.

### A Levy Payer Sends A Long Complaint

Can you paste it into an AI tool to help draft a calm reply? A sensible line is yes, but only after removing names, contact details and anything that identifies them, or using a tool your organisation has approved for that kind of information.

### A Board Member Wants An AI Summary Of A Confidential Paper

Summaries are a great use of AI, but confidential papers shouldn't go into tools that haven't been approved for them. Agree which tool, if any, is suitable before the request comes in.

### A Volunteer Makes An AI Image Of The High Street

Lovely for a playful poster, if it's clearly an illustration. Not acceptable if it could be mistaken for a real photo of a real event. Write the example into the policy so nobody has to guess.

## Getting It Adopted

### Take It To The Board

Boards like to know risks are managed. A one-page policy shows you're using AI thoughtfully, which also makes it easier to ask for time or budget for training.

### Train, Don't Just Circulate

A policy nobody understands isn't a policy. Spend thirty minutes walking the team through it with real examples from your own work.

## Pitfalls To Avoid

- **Banning AI outright.** People use it anyway, just less safely.
- **Copying a corporate policy.** If it's too long, nobody reads it.
- **Forgetting volunteers and freelancers.** If they create content for you, the policy applies.
- **Setting and forgetting.** Tools change quickly. Review it.

## How Starien Can Help

Through our [AI Coaching](/ai-coaching/) and Digital For BIDs work, we help small teams write practical AI policies and, more importantly, use AI confidently day to day. If you'd like help shaping yours, or a short session for your team and board, [get in touch](/contact/).

## Frequently asked questions

### Does a small BID really need an AI policy?

Yes. If anyone on the team uses AI for emails, social posts or reports, you already have AI use. A short policy makes sure it's consistent, safe with data and accountable, and it reassures your board.

### What should an AI policy for a BID include?

Approved tools and accounts, approved uses, information that must never be entered, how outputs are checked, when to be open about AI use, who to ask if unsure, and a review date.

### Can we put levy payer details into AI tools?

Treat personal data with great care. Only use it with tools and accounts your organisation has assessed as suitable under data protection law, and if in doubt, leave it out. Most AI tasks work fine with anonymised or summarised information.

### Should we tell people when we use AI?

Be open where it matters, such as images that could be mistaken for real photos or chatbots answering the public. For routine drafting a person has edited and approved, the key thing is that a person stands behind it.
