# How to Tell Your Team About AI: A 5-Minute Briefing Script

Canonical URL: https://business.theaieducator.io/posts/how-to-tell-your-team-about-ai
Publication: Dan Fitzpatrick Insights
Author: Dan Fitzpatrick
Topic: AI Adoption and Change
Published: 2026-11-15T16:40:18.000Z
Modified: 2026-11-15T16:40:19.912Z

Most teams start using AI before the owner has said a word about it. Here is a five-minute briefing you can copy: what AI is for in your business, what staff may do with it, who decides when it goes wrong, and what is not changing.

## In brief

Tell your team about AI in one five-minute briefing that answers three questions out loud: what AI is for in this business, what staff may do with it, and who decides when it goes wrong. Then say what is not changing and when they will hear more. Give it in person and follow it with a one-page note the same day.

## Key takeaways

- Staff are often using AI before anyone has told them the rules: in Deloitte UK's survey of 25,000 UK workers (May to June 2026), 31% of generative AI users said they use it without their employer's knowledge.
- In US small businesses, 19% of workers say AI adoption has been driven mostly by employees exploring tools on their own, against 11% who say the organization's guidance drove it (Main Street AI Monitor, 1,070 workers, May 2026).
- The Three-Answer Briefing is a five-minute talk that answers the Readiness Test out loud: what AI is for here, what you may do with it, and who decides when it goes wrong.
- Decide six things before you speak: the first job, the approved tool and account, what never goes in, who reads the output, who to tell when it goes wrong, and the date you come back to the team.
- Say only what is true about jobs. If you have not decided what AI means for roles, say so and give the date by which you will have decided.
- Give the briefing in person and send a one-page note the same day; an emailed policy answers one of the three questions and cannot take a follow-up.
- The briefing has landed when three people, asked separately a week later, give the same three answers.

Tell your team about AI in one five-minute briefing that answers three questions out loud: what AI is for in this business, what staff may do with it, and who decides when it goes wrong. Then say what is not changing and when they will hear more. Skip the vision speech. People are listening for the rules and for their own job.

Most owners never give this talk. The tools arrive first. Someone starts drafting emails in a chatbot, a software update adds an assistant, and six months later half the team uses AI and nobody has said a word about it from the front of the room. The silence gets filled anyway, mostly with guesses.

*Sources checked November 15, 2026.*

## Why do you need to say anything at all?

You need to say something because your team is very likely using AI already, and the evidence suggests they are doing it without hearing from you.

Deloitte UK surveyed 25,000 UK workers aged 18 to 70 between May 7 and June 10, 2026, and published the results on September 16, 2026. Some 63% said they knowingly use generative AI for work. Among those users, 31% said they use it without their employer's knowledge, and 65% reported a lack of convincing leadership on how it should be used in their organization ([Deloitte UK, September 2026](https://www.deloitte.com/uk/en/about/press-room/british-workers-spend-one-billion-pounds-of-their-own-money-on-gen-ai-for-work.html)). That sample covers employees and the self-employed in organizations of every size, so read it as the climate your staff work in, not as a measurement of your firm.

The small-business picture in the United States looks much the same. The U.S. Chamber of Commerce Foundation's Main Street AI Monitor, run by Ipsos from May 8 to 11, 2026 among 1,070 adults employed at US small businesses with 2 to 499 employees, found that half already use AI at work. About one in five (19%) said adoption where they work has been driven mostly by employees exploring tools on their own. Only 11% said the organization's own guidance or direction had driven it. The Foundation's summary is blunt: "The findings point to a workforce that is not waiting for permission." It also reports that "About one in 10 respondents say they were offered formal AI training" ([U.S. Chamber of Commerce Foundation, June 2026](https://www.uschamberfoundation.org/workforce/half-of-small-business-workers-use-ai-most-to-boost-productivity-not-automate-jobs)).

Silence has a price. In the Deloitte survey, nearly a quarter of respondents (23%) said there is a stigma attached to using generative AI at work, and 64% of weekly users worry that managers will think it can do their jobs. Hayley McKelvey, chief AI officer at Deloitte UK, put the consequence plainly: "Workers who feel a stigma around adopting GenAI tools are more likely to conceal their use." Her remedy is the outline of a briefing: "That means being clear about why their workforce should use GenAI, how it should be used and what success looks like."

Gallup's numbers point the same way. In its survey of 23,717 employed US adults, fielded February 4 to 19, 2026, among those whose organizations provide AI, employees who strongly agree that their manager actively supports AI use were frequent users 78% of the time, against 44% of everyone else. Where people strongly agree their organization has clear guidelines, the figures are 68% and 47% ([Gallup, April 2026](https://www.gallup.com/workplace/704252/workplace-separates-adopters-holdouts.aspx)). A survey cannot show that a talk causes anything. What these three do show is that people who have heard a clear yes, with rules attached, behave differently from people left to guess.

## What should the briefing cover?

The briefing should answer the Readiness Test out loud, in order.

The Readiness Test is three questions an organization must be able to answer consistently, from the top to the front line, before it can call itself AI ready: What is AI for here? What may I do with it? Who decides when it goes wrong?

Most owners can answer those three in their own head. The team cannot, because nobody has said the answers in the same words to everyone at once. So I suggest a format built for exactly that. This is my way of thinking about it, not established research.

> **The Three-Answer Briefing** is a five-minute talk an owner gives the whole team before anyone is asked to use AI. It answers the three Readiness Test questions out loud, in order: what AI is for here, what you may do with it, and who decides when it goes wrong. Then it says what is not changing and when the team will hear more.

| Part | What you say | Time | What staff are listening for |
|---|---|---|---|
| Why now | People already use it, and you are late in saying so | 30 seconds | Is this a telling-off? |
| What AI is for here | One named job, and what it is not for | 1 minute | Is my work the target? |
| What you may do with it | The tool, the account, what never goes in, who checks | 90 seconds | Am I allowed, and have I already broken a rule? |
| Who decides when it goes wrong | One name, and what happens when you own up | 1 minute | Will I be blamed? |
| What is not changing, and what is next | What you can promise and keep, plus a date | 1 minute | Is my job safe, and is this the last I hear of it? |

## What do you decide before you speak?

Decide six things, because the script is only as good as the blanks you can fill. If you cannot fill one, you have found what to settle before the meeting, and that is worth knowing.

Copy this and complete it:

1. **The first job AI is for:** ______ (one named job, such as first drafts of quotes).
2. **The approved tool and account:** ______ (the business account, not a personal login).
3. **What never goes in:** ______ (three things, in plain words).
4. **Who reads it before it leaves the building:** ______
5. **Who to tell when it goes wrong:** ______ (one name, not "management").
6. **The date you come back to the team:** ______

If you are stuck on the first blank, the One Job Rule in [how a small business should start using AI](https://business.theaieducator.io/posts/how-should-a-small-business-start-using-ai) picks it. If you are stuck on the third and fourth, [a one-page AI policy](https://business.theaieducator.io/posts/does-a-small-business-need-an-ai-policy) settles both, and UK firms can start from the [small-business AI policy template](https://business.theaieducator.io/posts/uk-small-business-ai-policy-template). The briefing is how a policy gets heard. The policy is how the briefing gets remembered.

Take a hypothetical 20-person commercial cleaning company. Its sheet might read: site reports and quote follow-ups; the business chat assistant on work logins; no client door codes, no staff details, no bank details; the contracts manager reads every report before it goes; tell the operations lead; back to the team on the first Friday of next month. That is ten minutes of thinking, and it is the whole talk.

## What is the five-minute script?

The script is six short paragraphs, each answering one thing your team is waiting to hear. Copy it, fill the brackets from your sheet and say it in your own words.

> **Why now.** "Some of you already use AI for work, and some of you have wondered whether you are allowed to. I should have said this sooner, so I am saying it now. It takes five minutes."
>
> **What it is for.** "In this business, AI is for [the first job]. That is where we start. It is there to take [the part of that job nobody enjoys] off your desk. It is not for [what stays with a person: setting a price, turning a customer down, anything about a colleague]."
>
> **What you may do with it.** "Use [the approved tool], signed in with your work account. Never put in [your three things]. Everything it writes is a draft, and [name or role] reads it before a customer sees it. If you are not sure whether something is allowed, ask before you paste. Nobody will think less of you for asking."
>
> **Who decides when it goes wrong.** "It will get things wrong, and so will we. When that happens, tell [name] the same day. You will not be in trouble for telling us. [Name] decides what we do next, and I will back that decision."
>
> **What is not changing.** "[Say only what is true, for example: I am not doing this to cut jobs, and nobody's hours or pay change because of it.] If any of that ever changes, you will hear it from me first, in person. Nobody is being scored on how much they use it."
>
> **What happens next.** "On [date] we will take ten minutes on what worked and what did not. Until then, [name] is the person to ask. What have I missed?"

Every bracket is a decision, and every sentence is a promise. Cut any line you cannot stand behind. The fifth paragraph is the one to treat with most care: if you have not decided what AI means for roles, say that, and give the date by which you will have decided. "Nothing will change" is the sentence staff check first, and something always changes.

## What will your team ask, and what do you say?

They will ask about their jobs, about being watched and about what they have already done, so have a true answer ready for each.

| The question | What to say |
|---|---|
| "Is this how the job cuts start?" | What you have decided and what you have not, with a date for the rest. Never "never." If you are weighing roles against AI, read [replacing an employee with AI](https://business.theaieducator.io/posts/replace-an-employee-with-ai) before the meeting, not after. |
| "Are you checking what I type?" | Exactly what the tool's admin settings record and who can see it. Look this up beforehand. A guess here costs you the room. |
| "I have been using my own account. Am I in trouble?" | No. Tell [name] this week and move to the work account. An amnesty brings [shadow AI](https://business.theaieducator.io/posts/shadow-ai-small-business) into view; a telling-off sends it back underground. |
| "What if I already pasted something I should not have?" | Tell [name] today. There is a calm way to handle it, set out in [what to do when an employee pastes customer data into ChatGPT](https://business.theaieducator.io/posts/employee-pasted-customer-data-into-chatgpt). |
| "Do I have to use it?" | Which parts are a business decision and which are up to them. The order to ask in is in [can you require employees to use AI](https://business.theaieducator.io/posts/can-you-require-employees-to-use-ai). |
| "Will we get training?" | A date, or an honest "not decided yet." What tends to stick is in [AI training for small business teams](https://business.theaieducator.io/posts/ai-training-for-small-business-teams). |

One question deserves its own conversation. When a single person stays behind to ask whether AI will take their job, that is a one-to-one, and I have written about it for leaders in [what to tell staff who fear AI will replace them](https://blog.theaieducator.io/posts/what-to-tell-staff-who-fear-ai-will-replace-them).

## The Mistake I See Most Often

The mistake is opening with the future when the room is waiting for the rules.

I have given more than 500 keynotes in 30 countries, and across the business teams I train the pattern is steady: the owner's talk is about where the industry is heading, and the questions from the floor are about Tuesday. Can I use it for this? Is that allowed? What happens if I get it wrong? An owner who answers those three first earns the right to talk about the future afterward. An owner who starts with efficiency tends to be heard as announcing that fewer people will be needed, whatever the next slide says.

Here is how the three common ways of telling a team compare. The table is my reading, not a survey finding.

| How the team finds out | What it answers | What it leaves open |
|---|---|---|
| Nothing is said; tools just appear | Nothing | All three questions. Staff write their own rules, and each person's differ. |
| The policy goes out by email | What you may do with it, for those who open it | What AI is for, who decides, and every question about jobs. Nobody can ask an email a follow-up. |
| The Three-Answer Briefing, in person, with a one-page note the same day | All three, in the same words to everyone | Only what you have not decided yet, and you have said when you will. |

My verdict: say it in person and follow it in writing. The other two cost less on the day and more every week after.

## How do you know it landed?

You know it landed when three people, asked separately a week later, give the same three answers.

Pick three people who do different jobs, including the one least likely to speak up in a meeting, and ask each of them the Readiness Test questions on their own. Matching answers mean the briefing worked. Different answers show you which paragraph to say again. If you find that the firm is still testing AI months later with nothing settled, the cause is often here, and [what separates firms that use AI from firms still testing it](https://business.theaieducator.io/posts/stuck-testing-ai-small-business) explains why.

Two things keep the briefing alive. A weekly slot such as the Friday Four in [how to manage a team that uses AI](https://business.theaieducator.io/posts/how-to-manage-a-team-that-uses-ai) gives people a regular place to say what they tried. And the Access Grid in [who is left behind on AI in your team](https://business.theaieducator.io/posts/left-behind-on-ai-small-business-team) shows which roles heard the words but still lack a tool, the time or someone to ask. For the colleague who heard everything and still will not touch it, [getting skeptical employees to use AI](https://business.theaieducator.io/posts/how-to-get-employees-to-use-ai) is a separate problem with a separate fix.

## What should you do this week?

You can do all of it before Friday in a 20-person business.

1. **Fill in the six blanks.** Twenty minutes, alone. Note any you cannot fill.
2. **Check what the tool records.** Open its admin settings so your answer about monitoring is a fact.
3. **Tell the named person they are the named person.** Before the meeting, not during it.
4. **Give the briefing in person.** Use the next team meeting. If the team is remote, use a call. Do not use email.
5. **Send the one-page note the same day.** The six blanks, filled in, are the note.
6. **Book the ten-minute follow-up,** and before it, ask three people the three questions.

If the room is bigger than a team meeting, such as a company day or a trade association event, the same three answers make the spine of a talk, and that is the ground my [business keynotes](https://theaieducator.io/business-keynotes?utm_source=business.theaieducator.io&utm_medium=referral&utm_campaign=how-to-tell-your-team-about-ai) cover.

## Where to go next

Once your team has heard the three answers, the next step is practice on their own jobs, and that is what my [AI training for teams](https://theaieducator.io/ai-training-for-business?utm_source=business.theaieducator.io&utm_medium=referral&utm_campaign=how-to-tell-your-team-about-ai) is built for. If you would like to see where the business stands before you speak, the free [Business AI Readiness Scorecard](https://theaieducator.io/business-ai-readiness?utm_source=business.theaieducator.io&utm_medium=referral&utm_campaign=how-to-tell-your-team-about-ai) takes about five minutes: 12 questions across six areas. Or [talk to me](https://theaieducator.io/booking?sector=business&utm_source=business.theaieducator.io&utm_medium=referral&utm_campaign=how-to-tell-your-team-about-ai) about your team.

## Sources and further reading

- [British workers spend nearly £1bn of their own money on GenAI for work, landmark Deloitte research finds](https://www.deloitte.com/uk/en/about/press-room/british-workers-spend-one-billion-pounds-of-their-own-money-on-gen-ai-for-work.html), Deloitte UK, September 16, 2026 (GenAI Workforce Survey: 25,000 UK workers aged 18 to 70, employees and self-employed; fieldwork by Ipsos UK, May 7 to June 10, 2026; 16,749 of them use generative AI for work).
- [Half of Small Business Workers Use AI, Most to Boost Productivity, Not Automate Jobs](https://www.uschamberfoundation.org/workforce/half-of-small-business-workers-use-ai-most-to-boost-productivity-not-automate-jobs), U.S. Chamber of Commerce Foundation with Ipsos, June 17, 2026 (Main Street AI Monitor: 1,070 adults employed at US small businesses with 2 to 499 employees; fieldwork May 8 to 11, 2026).
- [AI in the Workplace: What Separates Adopters and Holdouts](https://www.gallup.com/workplace/704252/workplace-separates-adopters-holdouts.aspx), Gallup, April 12, 2026 (23,717 employed US adults; fieldwork February 4 to 19, 2026).

*Dan Fitzpatrick is a keynote speaker, Forbes contributor and five-times bestselling author who has trained more than 150,000 people across 30+ countries. [About Dan](https://theaieducator.io/about)*


## Frequently asked questions

### How do I tell my employees we are using AI?

Tell them in person, in about five minutes, and answer three questions: what AI is for in this business, what they may do with it, and who decides when it goes wrong. Then say what is not changing and when you will come back to them. Send a one-page note the same day.

### What should I say to staff who are worried AI will take their jobs?

Say only what you have decided, and put a date on what you have not. Avoid promising that nothing will ever change, because staff check that sentence first. Promise the order of events instead: if anything about roles changes, they will hear it from you first, in person.

### Should I announce AI to my team by email or in a meeting?

Use a meeting, then follow it with a short written note. An email can state the rules, but nobody can ask it a question, and the questions about jobs and monitoring are the ones staff most need answered. For a remote team, use a call instead of a message.

### What if my staff are already using AI without telling me?

Assume some are, and open the briefing by saying so without blame. Offer an amnesty: anyone using a personal account tells a named person this week and moves to the business account. A telling-off pushes that use out of sight, where you can neither help nor protect it.

### What do I need to decide before I talk to my team about AI?

Decide six things: the first job AI is for, the approved tool and account, what must never be put in, who reads the output before it leaves the business, who to tell when something goes wrong, and the date you will come back to the team. Those six answers are the talk.

### How do I know my team understood the AI briefing?

Ask three people in different jobs, separately and about a week later, what AI is for here, what they may do with it, and who decides when it goes wrong. Matching answers mean it landed. Different answers tell you which part of the briefing to say again.

---
Source: [How to Tell Your Team About AI: A 5-Minute Briefing Script](https://business.theaieducator.io/posts/how-to-tell-your-team-about-ai)
Publisher: [The AI Educator](https://theaieducator.io)
