🧊🧊 AI Icebreakers, Mini-Games & Interactive Exercises
WHAT
They’re exactly what they sound like: icebreakers or ‘serious games’ somehow incorporating and showing AI.
WHY
We often use these during workshops & interactive sessions to instigate playful critically, demonstrate algorithmic bias, and let people experience rather than to merely explain certain computational concepts.
HOW
This is a living collection - curated by the community from the web, and some developed in-house at AIxD in collaboration with Yasmin Morgan.
[CONTRIBUTORS]
[TIME]
Feb 2024 - Feb 2024
EMOJI SCAVENGER HUNT
Find the emoji in real-life with your phone’s camera. Makes you move through the room so good icebreaker.
Emoji Scavenger Hunt is a browser-based game built with machine learning that uses your phone’s camera and a neural network to try and guess what it’s seeing. This is just one example of how machine learning can be used in fun ways.
Powered by Tensorflow.js, the game is built to run efficiently on your phone’s web browser without needing to access backend servers. No images from your camera are collected or stored.
AIXD: MAKING MUSIC WITH MACHINES - COMPOSING WITH MOVEMENT
‘Making Music with Machines’ is an energizer - an activity to get people moving and playfully exploring. The activity is a p5.js sketch that uses poseNet - an AI algorithm that can detect poses. The sketch maps different values to electronic notes made by an oscillator.
This icebreaker was originally made to be an introductive energizer activity for an AI Playground talk by Maya Man on ‘Navigating Self & Body on the Internet’.
PROMPT BATTLE
Prompt Battle is a live event where people compete against each other using text-to-image software. Show off your prompt skills and maybe the audience will choose you as the winner who elicited the most surprising, disturbing or beautiful images from the latent spaces of DALL·E 2, Stable Diffusion, Midjourney, Craiyon, etc.
SURVIVAL OF THE BEST FIT
A simulation game showing bias of hiring algorithms. Survival of the Best Fit was built by Gabor Csapo, Jihyun Kim, Miha Klasinc, and Alia ElKattan - software engineers, designers and technologists, advocating for better software that allows members of the public to question its impact on society.
"With SOTBF, we want to reach an audience that may not be the makers of the very technology that impact them everyday. We want to help them better understand how AI works and how it may affect them, so that they can better demand transparency and accountability in systems that make more and more decisions for us."
QUICK, DRAW!
You draw, and a neural network tries to guess what you’re drawing. Of course, it doesn’t always work. But the more you play with it, the more it will learn. So far we have trained it on a few hundred concepts, and we hope to add more over time. We made this as an example of how you can use machine learning in fun ways.
ANIMATED DRAWINGS
Draw a cute character and animate it using AI (works especially well on kid-like drawings).
Children’s drawings have a wonderful inventiveness, energy, and variety. We focus on the consequence of all that variety in their drawings of human figures as we develop an algorithm to bring them to life through automatic animation.
SCROOBLY
Create fun animations in real-time with your camera. A simple way to make doodles come to life - no design expertise or coding required.
Powered by TensorFlow.js, Scroobly uses Facemesh and PoseNet machine learning (ML) models to map your live motion (gestures only) when you opt-in to use your webcam. The ML system updates the animation seen on screen as you move. With Scroobly, you’re using artificial intelligence as a creative tool to become a digital animator, even if you’ve never written code or taken a design class.
THE MOST LIKELY MACHINE
An interactive learning prototype by Artefact designed to help people develop algorithmic literacy through play.
"Designed for both individual and collective classroom use, we hope the Most Likely Machine prototype inspires both students and teachers. As a conceptual tool, we encourage educators to facilitate and build on the material. Learn more about our process creating the experience."
AIXD: ‘FILL IN THE BLANKS’ - SURFACING AI BIAS W/ SUDOWRITE
This is an introduction to uncovering biases in AI text generation. Language models are trained on vast swathes of the internet, riddled with content that can be harmful and filled with violence, prejudice and abuse. It’s becoming common knowledge that AI models replicate the nasty side of human language. ‘Fill in the Blanks’ is an interactive way of investigating these biases.
WOMBO DREAM
The most straight-forward text-to-image generator on the web.
AI OR HUMAN
AI-generated media content has come to stay, looking more human than ever. Play the game and try to guess the difference between what is AI or Human.
"AI technological capabilities in terms of generating media content have reached an extremely realistic level and the progression feels unstoppable. At the same time, the tools are becoming increasingly accessible to the public, opening up the future to many possibilities. This project analyzes the current state of AI-generated media content along with its social and ethical implications. As part of the project, the game aims to illustrate and give awareness at how far AI-generated media content has come."
BLOB OPERA
A DIY Opera. 16 hours of singing was used to train this model. What you hear is not the voices of these singers, but the machine learning model's understanding of what opera singing sounds like. This is based on what it learned from the singers.
IMPROMPTU
Impromptu is a guessing party game where you generate images using an AI and try to trick others.
Each round, one player comes up with a secret prompt describing an image, like kiwi pizza. Then, the AI generates an image based on this prompt. The other players then propose their own fake prompt: a text that looks like something that would generate this image. Then we add in the secret real prompt and shuffle all prompts together. Now everyone has to guess: which prompt is the real one?
VIOLA THE BIRD
Created by Google Arts & Culture Lab artist in residence David Li, Viola the Bird is a playful experiment that features a string instrument inspired by the cello that can be played by anyone, regardless of musical experience.
In order to develop this experiment, David Li collaborated with musicians to train a neural network to understand cello and violin compositions. The neural network was then applied to create a real-time audio synthesis engine that can generate the sounds of a cello or violin based on the player’s input.
FINGERSPELLING WITH MACHINE LEARNING
Learn ABC of American Sign Language using a machine learning model.
TEXT-TO-POKEMON
Put in a text prompt and generate your own Pokémon character, no "prompt engineering" required!
ALGORITHM WATCH - CAN YOU BREAK THE ALGORITHM?
Can you break the algorithm? aims at explaining journalists and their audiences about the intricacies of reporting on automated systems.
It is built by AlgorithmWatch with support from Florent Maurin of the independent game studio The Pixel Hunt.