A single path · for people who lead teams
For managers who have never used AI to lead: 5 short lessons to make clearer decisions, prepare for difficult conversations, communicate without confusion, and develop people—using a chat AI as a trusted advisor, never as a replacement.
Lessons
Leave knowing how to put together a request for clarity about a real decision that's bothering you right now.
Leave with a difficult conversation prepared and your own leadership pattern mapped out.
Leave with a clear message ready and the agenda for your next real meeting.
Leave with a 30-day plan for someone on your team and a war game for your next change.
Leave having completed your first AI-guided weekly leadership review.
Lesson 1 of 5
By the end of this lesson, you’ll put together a request for clarity about a real decision that’s bothering you right now—and know exactly what to do with the response.
You already handle difficult decisions without any AI—but you decide alone, under pressure, without time to organize what’s at stake. Most managers use AI only to write polished emails. Here, you’ll use it differently: as someone who helps you see the problem before you decide, not afterward.
↓ role-play to study
Almost everyone who opens a chat AI for the first time asks for the same thing: "write this for me." It's useful, but it's the shallowest use there is. The use that truly changes leadership is different: asking AI to help you organize a problem before deciding — not just write after you've already decided.
Applied example: a dealership’s sales manager opened an AI chat thinking she’d just “write a nice message to the customer”—and discovered in practice that the most valuable use was something else: organizing her thoughts before deciding whether to renew a contract with a problematic supplier.
A strong manager uses AI to think better, not just to produce more.
One important point before moving on: validate before you act; never follow blindly. AI may list risks you already knew about—but it may also reveal a real blind spot that only comes to light when someone (or something) looks at the problem from the outside.
Applied example: the technical team lead asked for clarity on whether to let go of a junior developer with poor performance — and AI pointed out an option he hadn’t considered: a formal 30-day mentoring plan before making a final decision.
Test yourself
If AI lists 3 risks you already knew about and 1 you hadn't thought of, what should you do?
Test yourself
After AI organizes your decision, who actually makes it?
One request a request for clarity always follows the same structure, no matter the decision: (1) the real situation, (2) the options that actually exist, (3) the risk of each one, (4) what you might be overlooking, and (5) what to validate before acting. Memorizing these 5 parts is worth more than memorizing any ready-made prompt.
Applied example: the operations coordinator, deciding whether to switch delivery suppliers, uses the 5 parts to avoid deciding based only on price — and discovers in part 4 that he never asked about the replacement timeline in case of a failure.
This is a recipe: write your real situation in place of [explain], keep everything else exactly the same, and send it to the AI.
I’m facing this decision: [explain]. List: 1. what the real options are 2. the risks of each option 3. what I’m overlooking 4. which decision seems best and why 5. what I should validate before taking action
Before
“I’ll decide this tomorrow morning, on impulse, without thinking it through.”
Afterward
Request for clarity with all 5 parts filled in, risks and blind spots identified.
Practice now 0/4 done
Goal: leave with a real decision organized into 5 parts. ~10 min.
Nothing here makes decisions for you or leaks to anyone—it’s just a test conversation, just for you. Don’t like the response? Adjust the situation you described and ask again.
I’m facing this decision: [explain]. List: 1. what the real options are 2. the risks of each option 3. what I’m overlooking 4. which decision seems best and why 5. what I should validate before taking action
You just organized a real decision before acting on it—that’s exactly what separates reaction from leadership.
Summary
Lesson 2 of 5
By the end of this lesson, you’ll prepare a difficult conversation with someone on your team—firm and respectful—and use AI to spot your own blind spots as a leader.
Every leader faces two difficult things: correcting someone without humiliating them, and admitting where you yourself fell short. Both tend to stay in your head, without any structure. AI helps you prepare for both more calmly than you can when you're on your own in the heat of the moment.
↓ role-play to study
Preparing for a difficult conversation with AI helps you avoid two pitfalls at once: humiliating the person or downplaying the problem until it’s unclear. Here’s how it works: you describe the real situation, and AI helps you put together a firm, respectful opening.
Applied example: the sales manager needs to talk with a salesperson who has been missing the target for two months—she describes the real situation to AI before scheduling the conversation.
This is a recipe: replace [explain] based on your real situation, and send it exactly like this.
I have an employee who's underperforming. Here's the situation: [explain]. Help me prepare a firm, respectful, and direct conversation. I want to correct the behavior without humiliating the person.
Common mistake
Prepare for the conversation alone by memorizing scripted lines. You often get stuck because you rehearsed only your side and didn’t anticipate how the other person would react. How to avoid it: ask the AI to simulate the person’s likely responses too, not just what you’ll say.
When you repeat the self-assessment exercise in different situations, a pattern starts to emerge—something that repeats, even with different people and topics. Reading the pattern is the first step; changing your behavior is still your work, not the AI's.
Applied example: after repeating the exercise three weeks in a row, the technical team lead realized he always avoided talking about missed deadlines — instead of naming the delay, he changed the subject.
The same chat that helps you prepare for a difficult conversation can also help you reflect afterward—without getting defensive or going easy on yourself. This is where real development begins: AI isn’t doing the task; it’s helping you recognize your own patterns.
Applied example: the operations coordinator lost his patience in a team meeting — instead of letting it slide, he describes what actually happened to AI and asks for an honest assessment.
How to ask for a self-assessment
Practice now 0/5 done
Goal: leave with the conversation opener ready and 1 pattern of your own identified. ~12 min.
You won’t send anything to anyone right now—you’re just preparing on your own, with no rush. Don’t like the result? Ask AI to redo it with more detail about the situation.
1) Tenho um colaborador com baixo desempenho: [explique]. Prepare uma conversa firme, respeitosa e objetiva, sem humilhar a pessoa. 2) Analise meu comportamento como líder nesta situação: [explique o que aconteceu]. Onde fui bem? Onde posso ter falhado? O que um líder mais maduro faria?
You prepared for a real difficult conversation and already know, before speaking with the person, where you tend to slip up.
Summary
Lesson 3 of 5
By the end of this lesson, you’ll turn a confusing idea into a clear message for your team and leave with an agenda for a real meeting, ready to end with a decision.
Unclear messages create rework, and meetings without an agenda turn into conversations that don’t lead to decisions. Both waste everyone’s time — and the fix is simple when you use AI as a reviewer that demands clarity before you hit send.
↓ role-play to study
You know what you want to say, but the first written draft almost always comes out more confusing than the idea in your head. AI works well here: you describe the rough idea, it gives you a short, direct version—and you adjust the tone before sending it.
Applied example: the operations coordinator needs to notify the team about a shift change — his first attempt at writing the message sounded like a reprimand, not an announcement.
This is a recipe: paste your rough idea in place of [text] and ask for the short, clear version.
Turn this idea into a short, clear, motivating message for my team: [text]
Common mistake
Send the first version without rereading it for tone. A sentence written in a hurry often sounds harsher than you intended. How to avoid this: ask AI for a second version that's "more cordial, same content" before sending it.
A meeting only truly ends when someone says out loud what was decided—without that sentence, the meeting remains open in everyone's mind, with each person remembering it differently.
Applied example: the sales manager wraps up every goal-setting meeting with a simple phrase: “so we’ve decided that...,” before everyone leaves the room. In your work, this means, for example: the operations coordinator turning a shift meeting into a recorded decision, not a loose conversation; and the technical team lead leaving the sprint meeting with the week's priority named, not left implicit.
A meeting that starts without an agenda almost always ends without a decision—everyone talks, nobody decides, and the topic comes up again the following week. Preparing the agenda beforehand with AI solves this in just a few minutes.
Applied example: the technical team lead had a team meeting that always ran over time without getting anywhere — until he started preparing the agenda in advance.
The 5 elements of an agenda that ends with a decision
Practice now 0/6 done
Goal: leave with a clear message and meeting agenda, ready to use. ~10 min.
This doesn’t change anything for real yet — it’s just a draft message and agenda. Doesn’t it look good? Rewrite it as many times as you want before sending it or going into the room.
You’re done when you have the revised message and know, before going into the meeting, what decision needs to come out of it.
Summary
Lesson 4 of 5
By the end of this lesson, you’ll put together a 30-day development plan for someone on your team and simulate what could go wrong with a change you need to implement before taking action.
Evaluating someone without turning it into a concrete plan doesn’t help anyone grow. Announcing a change without anticipating anyone’s resistance means being caught off guard by something predictable. Both become safer when you think through your next moves before moving a piece.
↓ role-play to study
Saying “he needs to improve” doesn’t help anyone grow—it becomes a plan when you identify strengths, challenges, and practical actions with a deadline. AI helps turn a vague assessment into a 30-day plan you can actually track.
Applied example: the technical team lead has a junior developer who is willing but struggles to ask for help before getting stuck — instead of just noting that in a review, he puts together a 30-day plan with practical steps.
This is a recipe: describe the actual profile in place of [strengths, challenges, role].
I have a professional with this profile: [strengths, challenges, role]. Create a 30-day development plan with practical actions.
The steps to turn an assessment into a plan
Before announcing any change—whether to targets, schedules, or processes—it’s worth thinking through who might resist and what could go wrong. This isn’t pessimism: it means making the announcement already knowing how to respond to the most likely objections.
Applied example: the sales manager needs to change the team's commission policy — before announcing it, she asks AI to war-game the change so she can find out who will resist and why.
I’m going to implement this change at the company: [explain]. Run a war game: 1. who might resist 2. what could go wrong 3. what objections might come up 4. how to respond 5. how to carry it out with less risk
The war game helps you act with less fear—it shouldn’t become an excuse to put off the change again. After the simulation, the next step is always to schedule the start date, not ask for another round.
Applied example: the operations coordinator simulated resistance to switching suppliers three times in a row and still hadn’t set a date — the exercise itself had become a way to procrastinate. In your work, this means, for example: a sales manager simulating the reaction to the commission change once, and already setting the announcement date instead of simulating again.
Common mistake
Using the war game to delay the decision instead of taking action. This often happens because simulating creates a feeling of progress without the real risk of actually taking action. How to avoid it: every simulation ends with a date on the calendar—if there’s no date, it’s not finished.
Practice now 0/5 done
Goal: leave with 1 development plan and 1 war-game ready. ~12 min.
Nothing here is being announced to anyone yet—it’s just your draft, so you can decide with more confidence later. Doesn’t look right? Describe the person’s profile or the change in more detail and ask again.
1) Tenho um profissional com este perfil: [pontos fortes, dificuldades, função]. Crie um plano de desenvolvimento de 30 dias com ações práticas. 2) Vou implementar esta mudança: [explique]. Faça um war-game: quem resiste, o que pode dar errado, objeções, como responder, como executar com menor risco.
You came away with a real development plan and a simulated change—both ready to put into action, not just ideas.
Summary
Lesson 5 of 5
By the end of this lesson, you’ll complete your first AI-guided weekly leadership review and know the 3 questions that sum up everything this course has taught.
Everything you've learned so far only becomes a habit if you have a simple ritual to sustain it—without one, the first busy week swallows it all again. This final lesson closes the cycle: less one-off use, more leadership routine.
↓ role-play to study
A simple ritual sustains everything you practiced on this path: Monday plans priorities, Wednesday reviews obstacles, Friday evaluates results. AI helps by asking one thing at a time, so it doesn't become a long form that no one finishes.
Applied example: the operations coordinator sets aside 10 minutes every Friday for this review — it’s become part of the routine, not an extra task.
The weekly ritual, one question at a time
You can study any new subject by asking AI to explain it with your reality in mind—not a generic textbook explanation, but one tailored to the kinds of decisions you make, your team, and your business.
Applied example: the sales manager wants to understand a new goal-setting concept — instead of reading a generic article, she asks for examples tailored to her business, team, and decisions.
Explain this topic to a manager with hands-on experience who wants to put it into practice: [topic] Include examples involving business, teams, and decisions.
After everything you’ve practiced in this learning path, a leader who uses AI well always comes back to 3 questions: Am I making good decisions? Am I communicating well? Am I developing people, or just demanding results? AI doesn’t replace leadership—it increases the awareness of those who lead.
Applied example: the technical team lead started ending each week by answering these 3 questions before going to bed on Friday — it takes less than 5 minutes and changes what he prioritizes the following Monday.
For managers, AI isn’t just about productivity. It’s about clarity, decision-making, communication, preparation, and human development.
The sales manager had a difficult week: she lost patience with a salesperson (lesson 2), put off a confusing message about targets (lesson 3), and did not decide whether to change the commission rule (lessons 1 and 4). On Friday, she goes through the weekly ritual and reaches a simple conclusion: she is communicating reasonably well, but avoiding decisions and avoiding the difficult conversation with the salesperson.
Question 1: of the 3 final questions, which one does the sales manager fail most clearly this week?
Question 2: which of the previous lessons (1 to 4) should she revisit first?
Answer key with explanations: she fails on “Am I making good decisions?”—she put off both the commission decision and the difficult conversation, which is also a delayed decision. It’s worth revisiting Lesson 1 (request for clarity) for the commission decision, and Lesson 2 (difficult conversation) so she doesn’t put off talking to the salesperson any longer.
Practice now 0/5 done
Goal: finish the week with all 5 questions answered. ~15 min.
This is just for you — you don’t need to share it with anyone. Couldn’t answer a question? Leave it open and come back to it next week.
Help me do my weekly review as a leader. Ask me one thing at a time about: 1. this week’s results 2. problems I avoided 3. pending decisions 4. people who need attention 5. next week’s priority
You completed your first full cycle of using AI to lead—not just to produce.
Summary