Training Your Team
On this page
Update November 2, 2026 — Le Chat becomes Vibe. Since May 28, 2026, Le Chat has been rebranded as Vibe (still at chat.mistral.ai): a single agent for work and code, carrying over your conversations, settings, and subscriptions. The legacy Le Chat experience remains accessible in Chat mode; in Work mode, agents have become skills (invoked with
@), memories serve as the knowledge base, and deep search is now the Deep Research skill. The plans are now called Mistral Free, Pro, Team, and Enterprise. This lesson uses the name Vibe; some menus described here may have changed names or locations.
From Solo User to AI Team
You have now mastered Vibe: your emails are faster, your analyses more structured, and your marketing more consistent. However, the real leverage arrives when your entire team uses Mistral. At this stage, the challenge is no longer technical; it is human.
Who You Are Dealing With
When AI enters a team, four reactions emerge, each requiring a different approach. The enthusiasts, 10% to 15% of the staff, have already tested tools at home; they will save you time, provided you guide them, as their instinct is to skip steps and experiment on real customer data. The cautious curious make up the bulk of the group at 40% to 50%: they are open-minded but wait to see a concrete result on a task they know before committing.
The remaining two profiles demand more attention. The skeptics, 25% to 30%, will tell you it is a gimmick, that “it doesn’t understand our business,” or that “we have always managed without it”; they will change their minds only after a demonstration on their own files, never after a speech. The anxious, 10% to 15%, fear being replaced, and this fear deserves a direct response rather than a shrug—the last section of this lesson gives you the words.
This is why your opening statement matters. Three phrases sabotage adoption before you start: “AI will make us 10x more productive” creates anxiety, “Everyone must use AI now” triggers resistance, and “This is the future, we must adapt” is too vague for anyone to know what to do on Monday. Replace them with something concrete and reversible: “I tested a tool that saved me 3 hours this week on emails, and I want to show you how”; “The goal is not to replace what you do, but to eliminate the repetitive tasks preventing you from doing your real job”; “We will try this for two weeks, and if it doesn’t help, no one is forced to continue.”
Deploying in Four Phases
The first two weeks belong to the ambassadors. Choose one or two people from the enthusiasts or the cautious curious, give them access to Mistral Pro, and assign them a narrow mission: “For two weeks, use Vibe for one specific task in your daily routine. Track the time saved and the results. We will discuss this in 15 days.” Choose the task based on the role: prospecting and follow-up emails for a sales rep, meeting summaries and correspondence for an assistant, invoice data extraction and reporting for an accountant, or documentation and procedures for a technical lead.
Week three is for the debrief, strictly 30 minutes. The ambassadors describe what they did, quantify the time saved, outline the limitations they encountered, and share their tips. This peer testimony is 100 times more convincing than a speech from leadership because it comes from someone doing the same job with the same constraints.
Weeks 4 to 6 are for expansion, within a defined framework: a group training session of two hours maximum, entirely practical; an internal guide compiling the five use cases validated by the ambassadors, including the exact prompts; usage rules specifying what is permitted and what is not (see Lesson 13 on security); and a lead ambassador for daily questions.
Finally, months 2 and 3 are for anchoring, a phase most SMEs neglect—and this is where projects die out. Habits stick only when AI becomes part of the process: “the meeting report is drafted with Vibe” becomes a procedure, not a preference. Measure results (hours saved per week, perceived quality, satisfaction) and maintain momentum via a Slack channel or a 15-minute monthly meeting where everyone shares their best use case.
Training Without Infantilizing
Four mistakes recur in internal training sessions. Too much theory: your team does not need to understand transformers; they need to write an email faster. Too many use cases: beyond three per person initially, nothing sticks. Not enough practice: 80% of the time must be hands-on. And a lack of follow-up, which causes even a successful training session to be forgotten within two weeks.
The format that works spreads the effort over time rather than concentrating everything into a single day.
| Session | When | Agenda |
|---|---|---|
| 1 — Discovery (1h) | Week 1 | 10 min live demo of a relevant use case, 40 min where everyone replicates it on their own case, 10 min Q&A and sharing |
| 2 — Deep Dive (1h) | 2 weeks later | Roundtable on what worked and what blocked, introduction of 2 additional use cases, independent practice |
| 3 — Optimization (30 min) | 1 month later | Sharing best practices, introduction of custom instructions, discussion on processes to formalize |
”Will AI Replace Me?”
The question will come, directly or indirectly, and your answer must be honest rather than reassuring. AI does not replace people; it replaces tasks: if 20% of an employee’s time goes to automatable work, they reclaim 20% for higher-value activities. However, people who use AI will replace those who do not: it is a tool for competitiveness, not a threat. And your SME needs to grow, not shrink: AI allows you to do more with the same team, not the same with fewer people.
Key Takeaways
- Start with 1-2 volunteer ambassadors, not a massive rollout
- Peer testimony is more convincing than leadership speeches
- Train in 3 spaced sessions: discovery, deep dive, optimization
- Focus on 3 use cases maximum per person initially
- Address the fear of replacement with honesty: AI replaces tasks, not people