Leading AI in your team: the training course for managers

Why start with managers?

Because AI adoption is decided by the executive committee but plays out in the teams, and managers are the link between the two. Without them, everyone applies their own rule, and no one dares to approve anything.

Managers are also the ones the team asks the real questions: what are we allowed to do? with which data? and what about my job? These questions are legitimate. Some managers are already well ahead, others more hesitant: that is normal. A shared day gives them the same reference points to reassure those who have doubts, without holding back those who are moving ahead.

What does a manager steer when AI is rolled out?

A manager steers four things: their team's uses, the rules that frame them, the priorities between projects and the value they bring. They do not build the tools themselves: they decide what deserves to move forward.

• Uses: what the team already does with AI, including without saying so, and what is worth encouraging.
• Rules: the AI Charter translated for their department, with the approved tools, the data to protect and the checks to make.
• Priorities: which projects to start first and which to set aside, based on the team's real pain points.
• Value: simple tracking of what each use really changes, measured rather than assumed.

How do you prioritise AI projects with the RICE method?

The RICE method gives each project a score: Reach × Impact × Confidence, divided by Effort. It avoids starting the most appealing project rather than the most useful one.

For each idea, the manager estimates how many people or operations are affected, what the project changes in time, quality, cost or risk, how solid that estimate is, then the effort in days, skills, licences and maintenance. The score is there to open the discussion with the team, not to decide alone.

We add one rule: always look for the simplest solution that creates the expected value. In this order:

• a feature that already exists in the tool;
• a setting or a configuration;
• an AI assistant or agent;
• an automation across several tools;
• custom development, as a last resort.

What is the difference between a manager and an AI Ambassador?

The manager decides and supports; the AI Ambassador builds and spreads. The two roles complement each other, but they rarely fall to the same person, and that is deliberate.

The AI Ambassador is a volunteer, one per team, often spotted during training. They gather needs, help colleagues, create simple assistants or agents, test them and document them; for this, they train in building AI agents. The manager gives them dedicated time, agreed in writing, sets their priorities and escalates whatever goes beyond the team's scope.

Giving this role to managers themselves seems like a good idea but is not: they lack the time to build, and the team loses a hands-on relay.

How does the course fit with the AI Charter and the CAiO?

The AI Charter sets the rules at company level, managers bring them to life in their teams, and the Chief AI Officer part-time (CAiO) holds it all together over time.

When the course comes before the AI Charter workshop, the uses identified by managers feed into the writing: the charter starts from what teams actually do. Once the managers are ready, the Prompting and AI literacy course gives all their team members the same methodological foundation.

The Chief AI Officer part-time then takes over at company level: they bring the teams' roadmaps together into a single AI backlog, that is, the list of projects to assess, lead the AI Ambassadors, prepare management's decisions and measure the value delivered. Your managers become their natural counterparts, since they already prioritise with the same compass.