Build a Teachable Machine Image Classifier With Your Kid

A hands-on, zero-cost weekend project to teach children ages 8 to 16 how neural networks actually learn.

Most children interact with artificial intelligence daily through video recommendations, voice assistants, and game mechanics, yet very few understand how these systems make decisions. By moving your child from a passive consumer to an active builder, you demystify the technology and build critical digital literacy.

This weekend project requires no coding experience, no paid subscriptions, and no specialized hardware. Using a standard laptop or desktop computer with a basic webcam, you and your child (ideally ages 8 to 16) can build a fully functional computer vision model in under thirty minutes. By training a machine to recognize physical objects, hand gestures, or facial expressions, your child will gain a practical, intuitive understanding of how modern machine learning models are trained and deployed.

Step 1: Access the Platform and Choose Your Project

Begin by opening a web browser on your computer and navigating directly to the official Google tool: Teachable Machine . Click on the "Get Started" button. You will be presented with three project options: Image Project, Audio Project, and Pose Project.

For this activity, select the Image Project , followed by the Standard Image Model . This setup allows the computer to use your webcam to capture and analyze visual data, which is the most intuitive way for children to visualize how neural networks process information.

Step 2: Define and Label Your Classes

In machine learning, a "class" is simply a category of data that the model is trained to recognize. On the screen, you will see two default classes labeled "Class 1" and "Class 2".

Have your child decide what they want the computer to distinguish between. Excellent beginner ideas include:

• Hand Gestures: Thumbs Up vs. Thumbs Down.
• Household Objects: A coffee mug vs. a notebook.
• Pet Toys: A tennis ball vs. a chew toy.
• Expressions: Happy face vs. surprised face.

Click the pencil icon next to "Class 1" and rename it to match your first object (e.g., "Mug"). Rename "Class 2" to match your second object (e.g., "Notebook").

Step 3: Gather Training Data

Now it is time to collect the data. Click the "Webcam" button under your first class. Hold the first object (e.g., the mug) up to the camera. Click and hold the "Record" button to capture images.

Crucial Tip: While recording, slowly rotate the object, move it closer and further from the camera, and tilt your head slightly. This teaches the computer to recognize the object from different angles and in different lighting conditions. Aim to capture at least 100 to 150 image samples. Repeat this exact process for the second class using the second object.

Step 4: Train and Test Your Model

Once both classes have sufficient image samples, click the blue Train Model button in the middle column. Do not close the browser tab or minimize the window during this process. The browser is currently processing the images and finding patterns (such as edges, colors, and shapes) that distinguish the first object from the second.

Once training is complete, the "Preview" panel on the right will activate. Hold up one of the objects to the webcam. You will see the confidence meters at the bottom shift in real time, showing the percentage of certainty the model has for each class.

Parent & Educator Discussion Guide

Core AI Concept: Supervised Learning

Explain to your child that they just performed "supervised learning." The computer did not automatically know what a mug was. It needed labeled examples (the training data) to learn the patterns. If you give the computer bad data, it will make bad predictions (often called "garbage in, garbage out").

Conversation Starters:

• "What happens if we hold up a completely different object, like a pen? Which class does the computer guess, and why do you think it gets confused?"
• "If we only trained the model with a blue mug, will it recognize a red mug? How does this show how bias can accidentally get built into AI systems?"
• "How do you think self-driving cars use this exact technology when they are driving down a busy street?"

Action Steps Summary

• Open the Tool: Go to Teachable Machine on a computer with a webcam.
• Select Image Project: Choose the standard image model to begin.
• Label and Capture: Create two distinct classes and capture at least 100 varied images for each.
• Train the Model: Click train and observe how the browser processes the visual patterns.
• Test and Discuss: Use the preview panel to test the model, try to trick it, and use the sidebar questions to discuss how AI works.

Inspire the next generation of builders.

Spend thirty minutes this weekend building a real machine learning model with your child.