What's new

This page lists the main changes to the platform for customer organisations, latest first. Each point links to the guide that covers it.

Music and sound effects set separately

In every game, the Sound button of the dashboard now opens a small panel: one slider for the music, one for the sound effects, and the Mute switch. A player can turn the music down in an open-plan office and still hear the feedback on their answers. The settings last for the whole game, Start over included. Nothing to do on your side.

The AI Timeline joins the catalogue

The AI Timeline puts the milestones of artificial intelligence back in order, from the first artificial neurons and the Turing test to ChatGPT and the AI Act: the player drags event cards onto a horizontal timeline (or uses the keyboard), binds the chapter, and the reveal stamps the real dates and tells the story that links them. Graded like a quiz, LMS-compatible and editable: a fork rewrites its chapters and events (title, date and picture), sets how many events a chapter has (3 to 5) and applies its brand inside the game. See its guide.

Your brand inside AI Summit and AI or Not AI?, and AI or Not AI? editable

Your brand inside the game

AI Summit and AI or Not AI? now apply the brand of your fork live, inside their frame: colours (contrast checked as you choose), font, corners and your logo (bottom left of AI Summit; in place of the crest of the welcome card in AI or Not AI?). AI or Not AI? has no secondary colour: the gold of its gauges stays the game's. The name of your organisation is not shown inside the game yet. See the brand of a fork.

AI or Not AI?, version 3

The colleague of a request is picked from the list of the game's colleagues, by their name and portrait, and its red flags are boxes to tick. A request whose red flags do not match its right stamp (a ticked flag calls NOT AI, no flag calls AI) is not saved, and the message leads to the stamp. The colleagues, the office scenery, the cases, the end profiles and the flying budget words are fixed lists: a fork changes their items without adding or removing any. See the guide of the game.

What you need to do

A fork of AI or Not AI? still on version 2 shows "game unavailable" to its players until its owner updates the draft to version 3 in the editor, then publishes it. The preview of the update lists what the fork loses.

Every publication is played before it goes online

Before freezing a release of a game graded by a quiz, Maketools plays the game through: a perfect play and random plays, over several draws. Content that would make the game fail, a question that cannot be played or a maximum score that cannot be reached blocks the publication, and the editor lists each problem with its field and a Fix button. See the version history.

What you need to do

Nothing. Games already published do not change; the check applies to the next publication.

A Brand space for every game, and editable games

Brand space

Every fork has a Brand space: main, secondary, background, surface and text colours (contrast checked live), font, corners, logo and the name of your organisation, as far as the game applies them: around a game that runs in an isolated frame (the background, the score bar, the pause menu, the end-of-game bar), and inside the frame for the brand roles the game declares. See the brand of a fork.

Editable games and games that are not editable

The catalogue, the page of a game and the fork dialog show two badges: LMS compatible (game graded like a quiz: SCORM, xAPI, per-question reports) and Editable. A game that is not editable is forked, published and shared like the others, with your brand; its content stays as Maketools made it, and Get support asks our team to adapt it. For games graded like a quiz, the Advanced mode is now kept for the Maketools team. See the badges and games that are not editable.

Fields that stay within their limits

In a game that declares its fields (AI Summit and AI or Not AI? today), every field has a control made for it: a text with its maximum length, a number between its bounds and on its step, a choice from a list. The same controls cover images (with Back to the default), dates, bounded lists and items of a game list, picked by their name and thumbnail, never by an identifier (the colleagues of AI or Not AI?). A value outside its limits is flagged at the field as you type and refused on save. See fields that stay within their limits.

What you need to do

Nothing: your forks keep their content, and the brand already set in Appearance is found again in the Brand space.

Games graded like a quiz, and LMS integration

A grade that measures knowledge

Catalogue games are now quizzes in disguise: whatever the storyline, players move forward by answering questions written in the content, each with one expected right answer and one skill.

• The grade is the questionnaire: the share of questions answered correctly at the first attempt, weighted per question (1 to 5) and recalculated by our server from the answers. An unanswered question counts as wrong.
• Game points are kept apart: speed, streaks, lives or altitude earn points shown in the game and on the leaderboard, never in the grade, the pass mark or the LMS. The leaderboard ranks players by grade, then game points, then duration.
• Five question types: one answer, several answers, an order, pairs, a value in a range. Each game states the ones it can present.
• A game's score is only shown once our server has checked it, on the end screen.

See keeping a record of training.

Catalogue games

• AI Summit joins the catalogue: twelve questions on using AI at work, climbing from base camp to the summit. See its guide.
• AI or Not AI? moves to version 2, graded like a questionnaire: every request of the session is asked and graded, traps count double, and lives, streaks and speed earn game points. See its guide.

Studio: Questions, Appearance, Rules

The editor of a fork of a quiz-format game has three spaces: Questions (list or table, a sheet per question type, order and draw, shared answers, weight, explanation, hint, skill), Appearance (theme, brand, images and interface texts) and Rules (timer, jokers, number of questions, pass mark). Questions can be imported and exported as CSV from a spreadsheet, and the editor flags what makes a question ambiguous or easy to guess. See edit the questions of a quiz.

Per-question reports

The report of a quiz-format game adds a Questions table: first-attempt success, answers, jokers and hints, and the three most chosen wrong answers, with the same group filter and the same anonymity threshold as the rest of the report. See read a game's statistics.

LMS integration

• SCORM 2004 (4th edition) on top of SCORM 1.2, chosen at export: completion and success reported separately, scaled score, pass mark in the manifest.
• The game inside the LMS page: the package shows the game in the activity page, only on the LMS addresses your organisation has declared (a test launch picks them up). Opening in a new window is still offered, and packages already handed out keep the window.
• Two-click SSO sign-in inside the LMS on the first launch, for a learner already signed in to your identity provider, then no more sign-ins while the session lasts; that session can only be used to play.
• Detail per question and per skill: one interaction per question and one objective per skill, written from the answers graded by our server.
• "Incomplete" status as soon as the activity opens, normal exit without resume, duration in each version's format.
• Richer xAPI: full description of each question in answered statements, the question's weight, and a terminated statement for a play quit before the end.
• A warning before redeploying a fork shared in an LMS when the release changes its questions.

See LMS integration.

Security: games isolated from the site

Catalogue games run inside an isolated frame, served from an address separate from the site: a game can neither read the player's session, nor call the API, nor change the platform interface. The privacy policy and the data processing agreement describe the LMS play session and this isolation. See the trust center.

What you need to do

• IT team: allow play.maketools.ai as well as maketools.ai (or *.maketools.ai). See network access.
• Forks of AI or Not AI? still on version 1: they show "game unavailable" to their players and their LMS packages. Open the fork in the editor, move the draft to version 2 (your requests, texts and images are kept; the pass mark becomes 80 %), then publish it (Publish button).
• Game shown inside your LMS: declare your LMS addresses in Reports › LMS connector, then export a new package.