Build a Teachable Machine Image Classifier With Your Kid

Most children interact with AI as passive consumers, whether through recommendation algorithms on video platforms or voice assistants in the home. To prepare the next generation of leaders, we must shift their relationship with technology from consumption to creation.

This weekend project allows you and your child (aged 8 to 16) to build, train, and test a real machine learning model in under thirty minutes. No coding, no paid subscriptions, and no complex setups are required.

The Power of Hands-On AI

Understanding artificial intelligence can feel abstract and intimidating. By building a functional computer vision model, children demystify the technology. They learn that AI is not magic; rather, it is a system of patterns, data collection, and mathematical probabilities. This project uses Google's free Teachable Machine tool, which runs entirely in the web browser and respects privacy by processing all images locally on your computer.

Step 1: Choose Your Classification Challenge

Before opening the computer, select two or three distinct classes of objects. For younger kids (ages 8 to 11), choose highly distinct items like a favorite stuffed animal versus a toy car. For older kids (ages 12 to 16), increase the difficulty by choosing more subtle differences, such as a blue pen versus a black pen, or different hand gestures (rock, paper, scissors). This selection process introduces the concept of classification.

Step 2: Accessing the Platform

Open a web browser on any laptop or desktop computer equipped with a webcam. Navigate directly to the Teachable Machine website at https://teachablemachine.withgoogle.com/ . Click on "Get Started" and select "Image Project", then choose "Standard Image Model". This interface is entirely free and does not require any account creation, ensuring complete privacy for your family.

Step 3: Gathering the Training Data

Your screen will show two default classes: "Class 1" and "Class 2". Rename these classes to match your chosen objects (for example, "Lego Brick" and "Action Figure"). Click the webcam button under the first class. Instruct your child to hold the first object in front of the camera. Press and hold the "Record" button to capture approximately 100 to 200 image samples. Encourage your child to rotate the object, move it closer and further from the lens, and change the angles. This teaches the model what the object looks like under different conditions. Repeat this exact process for the second class with the second object.

Step 4: Training the Model

Once both classes have sufficient images, click the "Train Model" button. This process takes about ten to thirty seconds. Instruct your child to watch the progress bar. Explain that the computer is currently looking for patterns, edges, colors, and shapes that distinguish the first object from the second object. It is converting the visual pixels into mathematical weights. Warn your child not to close the browser tab during this phase.

Step 5: Testing and Iterating

Once training is complete, the "Preview" panel on the right side of the screen will activate. Hold up one of the objects in front of the webcam. The model will display a real-time percentage bar showing how confident it is in its prediction. Challenge your child to "trick" the model. What happens if you hold the object upside down? What happens if you only show half of it? If the model struggles, you can add more training images to the classes and retrain it. This iterative loop is exactly how professional AI engineers refine their models.

Parent & Educator Sidebar

Core AI Concept: Supervised Learning Supervised learning is a type of machine learning where the model is trained on labeled data. In this project, the labels are the names of the classes (e.g., "Lego Brick"), and the data is the set of webcam images. The computer learns by finding common features among the images in each label group.

Conversation Starters:

• "Why do you think the computer got confused when we changed the lighting or held the object too close?"
• "How could a self-driving car use this exact technology to recognize stop signs versus speed limit signs?"
• "What kind of training data would a computer need to recognize different types of animals in the wild?"

Action Steps Summary

• Select two or three physical objects or hand gestures to classify.
• Go to https://teachablemachine.withgoogle.com/ on a computer with a webcam.
• Create and name your classes, then record 150 images for each class.
• Click "Train Model" and observe the training process.
• Test the model in the preview panel and discuss the core concepts using the sidebar questions.

Want more family-friendly AI projects and educational guides?