AI literacy training for L&D teams: a practical guide

You run learning and development, design training or manage the LMS, and you have been asked for a data and AI literacy programme for the whole organisation, usually quickly and with little of learners' time to spare. This page answers the questions L&D teams ask before choosing a format.

How much time do learners need to give?

Little, but often. A Maketools game takes 4 to 10 minutes depending on the game, solo, on a laptop or a phone. A core programme fits in one game a week for a month, plus a 30-minute kick-off (see the 30-day AI literacy plan). The short format lets people play between meetings, and replay to beat their score.

Do we have to build the content ourselves?

No. Every catalogue game comes with complete, reviewed content in English and French. You can deploy it as it is, or fork it to adapt:

• the content: cases, requests, answers and explanations, in your fork's source language (other languages are translated with a file);
• the rules: duration, number of rounds, pass mark, timer;
• the look: your brand colours, images, logo.

Editing happens inside the game itself: click a text or an image to change it, with the settings in a side panel. No development is needed. See edit a fork.

How does it connect to our LMS?

Two options, included in the Pro and Enterprise plans:

• SCORM 1.2: a package you import into your LMS like any other activity. The launcher opens the game in a new window, then sends the LMS the score (calculated by our server), the passed or failed status and the time spent. SCORM is for private games only: each learner plays with their member account, ideally through your company's single sign-on.
• xAPI: the results of each completed play are sent to your LRS, optionally with the detail of each answer.

Details, limits and step-by-step notes for common platforms are in LMS exports.

How do we evidence participation and progress?

Each deployed game has its own report: plays, unique players, completion rate, average score, pass rate, and mastery by skill. Skills come from a list shared by every game (for example "protecting personal data" or "spotting hallucinations"), so you can compare games and forks. Reports export to CSV, and to PDF from the Plus plan upwards. See reading game statistics.

The same figures help document the measures taken under Article 4 of the AI Act: who was trained, on which skills, with which results.

What about learners' data?

By default, reports are pseudonymised: each player appears under a stable pseudonym (Player-7F3A) specific to your organisation. A group filter shows no figures below 5 players, so an average never reveals one person's score. Named reporting is switched on by an admin, with re-authentication, and players are told in the game before they play. Data is hosted in the European Union. See player privacy.

Who can play?

You decide, game by game:

See who can play and, for events, the page on events and workshops.

What does it cost?

The Free plan lets you start with a small team. Paid plans are billed per active member (a member who signed in or played during the period), not per registered user. Public games are free on every plan. Details are in plans and billing.

Where to start?

1. Play each game once from the catalogue, no account needed.
2. Create your organisation and invite one or two colleagues (see Getting started).
3. Fork the first game of your programme, adapt a few cases, test it, then publish it.

To choose the first game: AI or Not AI? for everyone's basic habits, Prompt Quest for regular users of AI assistants.