AI is like a super-smart student that can do all sorts of cool stuff! It can whip up artworks, compose tunes, write computer code, recognize faces, and even help discover new medicines!

Today, you're going to be the AI teacher!

We'll use an AI that learns to sort pictures into categories. Think of it as training a robot to organize a giant digital photo album. You'll create two categories and show the AI examples for each.

Then, like magic, it'll learn to recognize patterns and sort new pictures all by itself!

Ready to teach a machine and see how it learns? Let's jump in and create some AI wizardry!

1. Go to https://teachablemachine.withgoogle.com/
2. Click “Get Started”
3. Click “Image Project”
4. Let’s go with “Standard image model”
5. You should see something like this:

1. These are essentially the three elements of an AI system!
   1. Dataset (Class 1, Class 2 images you will input)
   2. Learning Algorithm (Training) and
   3. Prediction (Preview)!
2. Alright, now that you know the basics of what is going to be needed, let’s get started inputting your images!
3. You can test the model with dogs and cats, or ripe and unripe fruits or whatever comes to your mind, the possibilities are endless. However for now, we will stay focused on AI biases 🙂
4. To make it easier for you, here is a list of images we have curated. You will see folders labeled as “Biased Training Dataset”, “Unbiased Training Set” and “Testing Dataset” and yes you got it, you will use the ones in the training dataset for training your data and those in the testing dataset for testing!

<aside> 🪆

To keep things simpler, we will stick to toys for this example. As you become proficient with building your models, you can create other models for fairer and inclusive representation in other areas!

</aside>

1. Go ahead and add your images for both the classes - in our case, we started with the biased data set and added Toys for Girls and Toys for Boys from the shared images. See example below: