How to Define What Each Team Needs to Learn About AI: The Profiles and Depth Matrix
By Pedro Sonda, dooop
To define what each team needs to learn about AI, answer two questions for each group: what tasks will they use AI for, and what is the risk level of the decisions they make? Then, cross-reference the company profiles (all employees, power users, operational teams, managers, risk areas, and executive team members) with the literacy components (understand, use, evaluate, protect, and decide) and define the depth for each combination. The result is a proportional plan: no one wastes time on what they do not use, and those handling higher risk receive more thorough preparation.
Updated October 2026.
Why "One-Size-Fits-All Training" Fails
The European AI Act, by requiring AI literacy, mandates that measures consider people’s technical knowledge, experience, education, and the context in which AI systems are used (AI Act, art. 4). Even outside Europe, this principle is useful: preparation should be proportional to the role.
Gartner recommends specific learning paths combining digital fluency with leadership skills and practical application (Gartner, 2025). BCG found that only 36% of workers feel adequately trained (BCG, 2025).
Interpretation: A single training for the entire company tends to be too shallow for some and too technical for others. Role-based learning paths better address what each person needs to decide and do.
The 2 Calibration Questions
| Question | If the answer is... | Then... |
|---|---|---|
| 1. For what tasks will the group use AI? | Many tasks, daily | More applied practice (components "use" and "evaluate") |
| Few or none | Focus on rules and basic understanding | |
| 2. What is the risk level of the group’s decisions? | Decisions involving people, money, customers, health, or safety | Greater depth in "protect" and "decide" |
| Low risk | Basic depth in these components |
dooop Framework: Profiles × Components Matrix
Depth: ● basic · ●● intermediate · ●●● advanced
| Profile | Understand | Use | Evaluate | Protect | Decide |
|---|---|---|---|---|---|
| All employees | ● | ● | ● | ●● | ● |
| Power users (administrative, commercial, marketing, analysis) | ●● | ●●● | ●●● | ●● | ●● |
| Operational teams | ● | ●● | ●● | ●● (with focus on safety) | ● |
| Managers | ●● | ●● | ●●● | ●● | ●●● |
| Risk areas (legal, compliance, security, data) | ●●● | ●● | ●●● | ●●● | ●●● |
| Board and directors | ●● | ● | ●● | ●● | ●●● |
| Multipliers | ●●● | ●●● | ●●● | ●●● | ●● |
What Each Component Means in Practice
| Component | Content |
|---|---|
| Understand | What generative AI is, why it makes mistakes, what agents are |
| Use | Requesting with context, applying to job tasks |
| Evaluate | Verifying facts, numbers, and appropriateness; recognizing AI’s limits |
| Protect | Data, privacy, intellectual property, security, biases |
| Decide | When to use, when not to use, who is accountable, how to escalate or stop |
The full definition is available at What is AI literacy?.
How to Apply This in Your Company
- List the profiles relevant to your organizational structure; the table profiles are a starting point.
- Answer the 2 questions for each profile, together with area managers.
- Adjust the matrix and convert each row into a learning path.
- Validate with a sample of people from each profile before launching.
- Review every six months, as uses evolve.
Details for some profiles are in How to train administrative teams to use AI, How to train operational teams to use AI, and AI for managers.
What to Do Now
- Build the matrix with your company’s profiles.
- Identify the highest-risk groups and start with them.
- Turn the matrix into learning paths within your literacy program; see How to create a corporate AI literacy program.
To define role-based learning paths, explore dooop’s AI literacy at dooop.com.br/letramento-em-ia.
Frequently Asked Questions
Does the executive team need practical training? They need little tool practice but require depth in decision-making: what to delegate to AI, what risks to accept, and how to govern.
How to identify multipliers? People who already use AI with good judgment, have credibility among peers, and are willing to help. They do not need to be the most technical.
Also Read
About dooop
dooop is a consultancy based in Caxias do Sul (RS) that prepares leaders and organizations to transform artificial intelligence into business capability through strategy, governance, and training.