AI Literacy Program
A modular, risk-aware AI literacy curriculum. It teaches verification, data judgment, bias awareness and human oversight — not prompt tricks.
Two tracks built on the same core concepts:
| Track | For | Time | Start |
|---|---|---|---|
| Workplace track | Employees who use AI tools at work (customer service, operations, HR) | 4 self-paced modules, ~20 min each, plus a 1-page job aid | Start the workplace track |
| School track | Students aged 14–16, taught by a non-specialist teacher | 4 lessons × 50 min + final project | Start the school track |
What this program is
Many organizations must now take measures to support staff AI literacy under Article 4 of the EU AI Act. Schools have the OECD/EC AILit framework. Most existing training focuses on productivity. This program is different: it treats AI literacy as a set of judgments people make in real tasks — Can I trust this output? May I paste this data? Who is affected if I act on it? When should a human decide?
The program is mapped to the final AILit framework domains and competences and designed to support Article 4 AI literacy measures. It is not legal advice, and it does not certify anyone as “AI Act compliant” — Article 4 does not require any individual to reach a specific literacy level.
Where to look
- Workplace track — 4 modules + Verify before you send job aid + scenario quiz
- School track — 4 lessons, teacher guide, handouts, final project
- Assessment — scenario quiz, performance task, rubric
- Design documentation — needs analysis, personas, objectives, alignment matrix, framework mapping, design decisions, evaluation
- How a chatbot picks the next word — a shared interactive used in both tracks
Honest labels you will see here
- Northwind Logistics is a fictional company used to make examples realistic.
- The personas (Lena, Joris, Sofia, Ayo, Mr. Daniels) are fictional.
- The “AI-generated” documents used in the performance task are constructed for teaching purposes.
- Evaluation methods are labeled by who or what performed them. See evaluation methods and the repository’s
evaluation/METHODS.md.
Last reviewed: 2026-10-25