Staff AI training plan prompt: free ChatGPT template
Updated · By Robert Breen
Use this when you want staff using AI for real work, not just told it exists. Useful training starts from the emails, notes and meetings people already handle. This prompt builds short sessions tied to each person's tasks, with practice on made-up or masked examples and a checking habit for each. The free lessons, starting with Prompt Writing 101, can be the hands-on part.
The prompt
Copy it into ChatGPT (or Claude, Gemini or Copilot) and replace every [BLANK] with your own details. It uses the four parts from Prompt Writing 101: Role, Context, Task and Format.
Role: You are helping [BUSINESS] plan practical AI training for a small team. Context: The team and their weekly tasks: [ROLES AND TASKS]. Their comfort with AI today: [STARTING POINT]. We can spare [TIME AVAILABLE]. Our rules for AI use: [AI RULES]. Train on tasks they already do, not on AI theory. Task: Plan short sessions. For each: who attends, the one task they'll learn to do with AI, a practice exercise using made-up or masked examples, what good output looks like, and how they'll check it. Finish with how we'll know in a month whether it stuck, using things we can observe. Format: A table of sessions with columns: session, who, task, practice, check. Then a one-month list of three to five things to look for. Each session fits the time given.
Fill in the blanks
[BUSINESS]- Your business name and what you do.
[ROLES AND TASKS]- Each person and the writing or admin they do every week.
[STARTING POINT]- Who has used AI and how much. Be honest; it sets the pace.
[TIME AVAILABLE]- How long and how often, such as 30 minutes a week for three weeks.
[AI RULES]- Your AI use rules, or a short summary: which account, what never goes in, what gets checked.
Example, filled in
A made-up example from the free lesson Summarize a Meeting Transcript with ChatGPT. Harbor Street Bakery is the made-up bakery in the meeting summary lesson. Dana, the owner, wants Priya and Sam to use ChatGPT for writing they already do, without anyone pasting customer details in.
Role: You are helping Harbor Street Bakery, a neighborhood bakery, plan practical AI training for a small team. Context: The team and their weekly tasks: Dana (owner) runs the Monday team meeting and needs the action items afterward; Priya runs the counter and website and answers customer emails about orders; Sam handles catering and writes quotes for events like the Oakwood Library breakfast. Their comfort with AI today: Priya uses ChatGPT a little for social posts; Dana and Sam have tried it once or twice. We can spare 30 minutes a week for three weeks, before opening on Tuesdays. Our rules for AI use: use the bakery's ChatGPT account; no customer names, phone numbers or card details in prompts; a person reads everything before it goes to a customer. Train on tasks they already do, not on AI theory. Task: Plan short sessions. For each: who attends, the one task they'll learn to do with AI, a practice exercise using made-up or masked examples, what good output looks like, and how they'll check it. Finish with how we'll know in a month whether it stuck, using things we can observe. Format: A table of sessions with columns: session, who, task, practice, check. Then a one-month list of three to five things to look for. Each session fits the time given.
What a good answer looks like
- Three sessions, each tied to a real job: Dana pulling action items from the Monday transcript, Priya replying to an order email, Sam drafting a catering quote email.
- Practice on a made-up customer email or a transcript with names swapped, never a real one.
- A checking habit per task, such as ticking every date in a recap against the notes.
- One-month signs you can see: recaps sent the same day, fewer order emails waiting overnight.
How to check it before you use it
An AI draft can sound right and still say something you never told it. Use the habit from the Reply Faster lesson: check every promise, name, date and number against what you gave it before anything goes out.
- Make sure each session fits. Thirty minutes is enough for one task done twice, not three tasks.
- Have each person save the prompt that worked as a template, so the training leaves something behind.
- Pair the plan with a written policy, such as one drafted with the staff AI use policy prompt.
- Judge the work, not how often people use AI: faster, and still correct.
Why this prompt works
Training on a task someone does every week gives them a reason to use it on Monday. Practicing with masked examples builds the privacy habit at the same time.
Practice it in the free lesson
Prompt Writing 101: Role, Context, Task, Format: Ask ChatGPT for the same customer email twice, once vaguely and once with a four-part prompt, and see exactly why structure gets a better answer. You do every step yourself in a practice copy of ChatGPT, and nothing touches your real accounts.
Also useful: Summarize a Meeting Transcript with ChatGPT · Reply Faster: Turn a Messy Customer Email Thread into a Clear, Friendly Reply
Terms used here
Related prompts
- Staff AI use policy prompt: Draft a short staff policy for AI tools at work: approved tools, what never goes in, what a person must check and who to ask…
- First-week training plan prompt: Turn your SOPs into a day-by-day first-week plan for a new hire: learn, watch, do with help, then show it, with a trainer…
- Training quiz from SOP prompt: Turn an SOP or handbook page into a short quiz for new hires: scenario questions, believable wrong answers, and an answer key…