How to Train Administrative Teams to Use AI: 4 Practical Modules with the Team’s Own Documents
By Pedro Sonda, dooop
To train administrative teams in AI use, organize the training into four practical modules: writing and summarizing texts, reading and extracting data from documents, consulting company rules and policies, and verifying what AI produced. All exercises should use real documents and tasks from the team itself, with clear rules about data. The verification module is the most important because it ensures that time saved does not turn into errors.
Updated in October 2026.
Why Administrative Teams Deserve Specific Attention
Among Brazilian industries using AI, 87.9% apply it in administrative activities, the most frequent area of use (IBGE, 2025). In writing tasks, an MIT experiment with 453 professionals found a 40% reduction in time and an 18% increase in quality using generative AI (MIT News on Noy and Zhang, 2023).
At the same time, 56% of employees say they have already made mistakes at work because of AI (KPMG and University of Melbourne, 2025). In administrative work, errors in values, dates, and names have direct costs.
Interpretation: The potential is high and so is the risk. Training needs to teach how to use AI and, especially, how to verify its outputs.
dooop Framework: The 4 Practical Modules
| Module | What the Team Learns | Exercise with Real Material | Central Caution |
|---|---|---|---|
| 1. Writing and Summarizing | Requesting with context; adapting tone; summarizing emails, minutes, and reports | Draft responses to recurring requests; summarize the minutes of the last meeting | Review names, deadlines, and commitments |
| 2. Reading and Extracting Data | Extracting information from notes, contracts, and forms; organizing into tables | Extract data from ten documents and compare with the system | Verify values and dates at the source |
| 3. Consulting Rules | Using AI to find and explain sections of policies and procedures | Answer frequent questions based on official policy, citing the excerpt | Use only official and updated documents |
| 4. Verifying | Checking facts, numbers, and coherence; recognizing when AI fabricates | Find planted errors in texts and tables produced by AI | Know when review is mandatory |
Why Module 4 Includes Exercises with Planted Errors
Teaching verification through theory does not work. It works to show texts and tables with errors that seem correct and ask the team to find them. The experience of being misled by AI in an exercise usually creates the habit of verifying.
Rules That Must Be Clear Before the First Module
- Which tools can be used.
- Which data cannot be entered: personal data of clients and employees, confidential financial information, confidential contracts, according to company policy.
- When review is mandatory: values, documents for clients, official communications.
- Who to contact in case of doubt.
See How to Create Criteria for Safe AI Use by Employees.
Effective Format
| Element | Recommendation |
|---|---|
| Format | Practical workshops in small groups, with computers and real material |
| Duration | BCG associated more than five hours of training, with follow-up, to higher regular use (BCG, 2025) |
| Spacing | Sessions over weeks, with tasks to apply between them |
| Multiplier | One team member supports colleagues after training |
| Manager | Participates and defines tasks where AI will be used |
How to Measure
- Time spent on chosen tasks, before and after.
- Errors found in reviews.
- Number of people using AI in defined tasks, weeks later.
What to Do Now
- Ask the team to list the five tasks that consume the most time.
- Gather real documents (anonymized when necessary) for exercises.
- Start with module 1 and always finish with module 4.
To train administrative teams with applied practice, learn about dooop’s training at dooop.com.br/treinamento.
Frequently Asked Questions
Does the team need to know programming? No. The four modules use AI assistants in natural language.
How to prevent sensitive data from going to unapproved tools? With clear rules, approved corporate tools, and concrete examples of what cannot be entered. Prohibiting without offering alternatives tends to hide usage.
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, with strategy, governance, and training.