Building AI literacy in your team: a five-step guide
Updated: Oct 28
In many companies, employees have long been using ChatGPT, Claude or Copilot, often without any introduction. The results vary accordingly: some save time every day, others copy made-up figures into a quote without checking them. A few clear steps can change that.
The EU AI Act requires this too: under Art. 4, companies that use AI must promote their employees’ AI literacy. What this means legally is explained by the German Federal Network Agency (Bundesnetzagentur) and its AI Service Desk. Here, we focus on practice.
What the AI Act requires
Since February 2025, Article 4 of the AI Act has required companies that use AI systems in a professional context to take measures for the AI literacy of their staff. With the amended version, which entered into force on 27 July 2026, the focus is on promoting this literacy, and companies do not have to ensure a specific level for each individual. The regulation does not prescribe a fixed format or a minimum number of training hours.
Legal status: October 2026
1. Take stock: who uses which AI for what?
Ask your teams which tools are in use, officially and unofficially. You will often find that some employees already have a lot of experience. They are good points of contact for the others.
2. Distinguish between roles
Not everyone needs the same knowledge. Someone who occasionally drafts an email with AI needs a solid introduction. Someone who creates reports or analyses with it every day needs more practice in checking the results. And anyone who develops AI models themselves needs more in-depth formats.
3. Lay the foundations for everyone
Everyone who works with AI should know three things:
roughly how language models work and why they can give convincing-sounding but wrong answers
how to describe tasks so that the results are useful
how to check results before passing them on
This is exactly what our one-day course Generative AI in everyday work teaches, with exercises from everyday office work.
4. Set clear rules
A short internal policy provides certainty: Which tools are approved? Which data may be entered and which may not? Who is the contact person for questions? A one-page document that everyone can find easily is often enough, for example on the intranet or in the shared team folder.
5. Document and keep up
Keep a record of who received which training and when. Confirmations of participation stating the date, duration and content of the training are one way to do this. AI tools change quickly. So decide who in the team keeps an eye on new developments and passes on important news promptly, such as new features or changed terms of use. In addition, a short regular exchange in the team, for example once a quarter, on new features and successful use cases is worthwhile.
Conclusion
AI literacy does not come from a one-off memo, but from practice, clear rules and exchange. If you would like to know which format suits your team, get in touch with us.