Skip to content

Knowledge base

AI literacy for managers: what should a team lead be able to do?

By Bob Biesemans · published on

Managers don't need to be AI experts, but they decide what their team does with AI. The AI skills a team lead needs, and how to measure them.

Short answer

A manager should be able to judge their team's AI results, know which data may go into which tool, recognise unapproved AI use and judge where AI saves time and where the risk is too high. Measure that with tasks from their own work, the same way as for the team, and add review tasks.

Why managers have a role of their own

In most SMEs it isn't IT that decides how AI is used day to day, but the team lead. They approve a quote drafted with AI, notice someone pasting a customer file into a chatbot, or let it happen. Employees watch what their manager does and tolerates.

Article 4 of the AI Act asks for measures for the AI literacy of everyone who uses AI on the company's behalf, taking into account the context of use. For a manager that goes beyond their own use: it is also about what they ask of and check in their team.

What a manager should be able to do

Judge an AI result before it goes further: is it correct, is it complete, does it contain anything that must not go out? Whoever signs off the team's work also signs off what AI contributed to it.

Know which data may go into which tool, and turn that into concrete agreements for the team. Recognise when staff use unapproved tools, and offer a workable alternative rather than only a ban.

Judge which tasks AI saves time on and where the risk outweighs the gain. And treat an AI mistake as a chance to learn, so staff report mistakes instead of hiding them.

Common mistakes

Assuming the team can do it because it uses AI a lot. How often someone opens a tool says little about how well they check the result.

Giving a general training course and leaving it there, without knowing whether anything changed in daily work. And not following the rules yourself: a manager who puts confidential documents into a free chatbot can hardly ask the team to do otherwise.

How to go about it

Measure managers in the same way as their team, with tasks from their own work, and add review tasks: checking an employee's AI-written text, spotting a risk, deciding about a tool. Explain where needed and measure again.

That shows where each team stands, and gives the manager a view of where their people need support. Record what you did as evidence of the measures you took for Article 4.

Frequently asked questions

Does a manager need to be an AI expert?

No. A manager mainly needs to judge what AI produces and set clear agreements with the team. Technical knowledge of how a model works matters less than judgement about the result.

Does Article 4 ask anything specific for managers?

Article 4 asks for measures for everyone who uses AI on the company's behalf, taking the context into account. For a manager that includes what they ask of and check in their team, so it makes sense to look at them separately.

How do you measure a manager's AI literacy?

With the same practical tasks as the team, plus review tasks: checking an employee's AI-written text, spotting a risk, deciding about a tool. After targeted explanation you measure again.

Further reading

Read next

Help us improve the website?

With your permission, Google Analytics, Microsoft Clarity and Vercel Web Analytics measure visits to our public pages. Your test answers, results and dashboard are not sent. We do not use advertising cookies. Privacy policy