Your office coordinator pastes a frustrated customer's text into a free AI app to get help writing a calm reply. Your estimator asks an AI tool to tidy up a proposal that includes a homeowner's name, address, and gate code. Nobody told them not to. Nobody told them it was fine, either.
This post is for owners and operations managers at service businesses with roughly 5 to 49 employees. It explains how to write a simple AI use policy for employees: which tools are approved, what information stays out, what needs a second look, and who answers questions. It is practical guidance, not legal advice.
The Business Problem
In most growing service businesses, AI use starts from the bottom up. Someone tries a chatbot on their phone, it saves them twenty minutes, and they keep using it. Then a coworker picks it up. Within a few months, several people are using different tools in different ways, and the owner has no clear picture of what is happening.
That is not a character problem. Your employees are trying to get work done. The gap is that the business never answered a few basic questions:
- Which AI tools are we allowed to use for work?
- What customer, employee, or company information can go into them?
- What has to be checked by a person before it goes to a customer?
- Who do I ask when I am not sure?
Without answers, every employee makes their own call.
What This Costs the Business
The cost is rarely one dramatic incident. It is a slow build of risk and confusion:
- Customer information in unknown places. Names, addresses, phone numbers, access codes, and payment details end up in tools nobody reviewed.
- Inconsistent customer communication. One person's AI-written texts sound nothing like another's, and some promise things the business does not offer.
- Owner worry. The owner suspects AI is being used but does not know where, so it becomes one more thing to carry around mentally.
- Stalled progress. When the business later wants to build real AI support, it first has to untangle habits nobody planned.
A short policy removes most of that ambiguity. It also gives your team permission to keep using AI where it actually helps.
What Should Happen First
Before writing rules, find out what is already happening. Do not start with discipline. Start with a quick, no-blame conversation.
- Ask each person what they use AI for today. Writing, summarizing, research, scheduling, translating, anything.
- Ask which tools and accounts they use. Personal free accounts, company accounts, or features built into software you already pay for.
- Ask what information they put in. Be specific: customer names, addresses, photos, invoices, employee details.
- Note what is working. Some of these uses are good, and the policy should keep them.
This takes an afternoon. It usually shows that the real risk sits in two or three habits, not everywhere.
A One-Page AI Use Policy for Employees
Keep it to one page. If it needs a training session to understand, it is too long. A practical AI use policy for employees covers five things.
1. Approved tools. List the AI tools and accounts staff may use for work. Prefer company-managed accounts over personal ones, so the business controls access when someone leaves. If a tool is not on the list, ask before using it for company work.
2. Information rules. A simple three-level guide works well:
- Okay to use: generic writing help, public product information, and rewording your own notes without names. Use approved tools freely.
- Use with care: job details, customer questions, and internal procedures. Use approved company tools only, and remove names where you can.
- Keep out: payment details, passwords, gate or lockbox codes, Social Security numbers, and employee health or pay records. Never enter these into an AI tool.
Adjust the examples to your business. The point is that every employee can answer "can I paste this?" in a few seconds.
3. Review before it goes out. Anything AI helps write that a customer will see, such as a text, email, estimate, or review reply, gets read by a person before sending. The person who sends it owns it.
4. No promises the business has not approved. AI tools can make up warranties, timelines, discounts, or policies. Staff should never send a commitment the business has not already approved.
5. One named contact. Name the person who answers AI questions and keeps the approved list current. In a small company, this is often the operations manager.
For the legal and data side of the same question, our guide to AI governance for service businesses explains how to check what actually applies to a specific workflow.
Where AI or Systems Can Help
A policy covers everyday use of general AI tools. It is not the same as building AI support into the business.
When a task repeats often, such as estimate follow-up or missed-call replies, it usually works better as a Task Expert: an AI employee with one defined job, approved inputs, set permissions, and a human manager who reviews its work. Its access is controlled inside your systems instead of depending on what someone copies and pastes. Our post on CRM permissions for AI employees covers how that access is set.
On systems, the honest answer for most businesses is that no system change is needed to write a policy. A light configuration change may help, such as moving staff from personal AI accounts to company-managed ones or turning on AI features in software you already use. A new CRM is not required for this step.
What Changes and What Stays the Same
What changes: Employees know which tools are allowed, what stays out, and who to ask. The owner stops wondering what is happening. Customer messages get a consistent review step.
What stays the same: Your people still write, decide, and own customer relationships. Useful habits stay. The policy makes current AI use safer and clearer; it does not ban it.
Review the page every few months. Tools change, roles change, and new hires need to see it on day one. If you are rolling AI out more broadly, pair the policy with the steps in how to introduce AI to your employees.
Where StrategixAI Fits
StrategixAI starts with education. We help owners and teams understand what AI can realistically do, where it fits their work, and where human judgment has to stay in charge. That often includes helping a business turn scattered AI habits into clear, simple rules before anything is built.
When a business is ready to go further, we design Task Experts and AI Teams with defined jobs, managers, and boundaries, and we keep improving them over time.
Practical Next Step
This week, ask each employee the four questions above and write down the answers. You will likely have most of your one-page policy by Friday.
If you want help teaching your team practical, safe AI habits, learn about StrategixAI training or book a conversation to talk through where AI should and should not fit in your business.
