Build a Teachable Machine Image Classifier With Your Kid

Help your child transition from a passive consumer of technology to an active builder of artificial intelligence.

Children today are surrounded by artificial intelligence, yet most experience it passively through recommendation algorithms or voice assistants. To prepare them for a future shaped by technology, they must understand how these systems learn.

Using Google's free Teachable Machine tool, you and your child (ages 8 to 16) can build, train, and test a custom machine learning model in under thirty minutes. This project requires no coding, no paid subscriptions, and no specialized hardware, just a standard laptop with a webcam and a curious mind.

Demystifying AI for the Next Generation

Many children view AI as a magical entity that possesses human-like intelligence. This project demystifies that perception by showing them that AI is simply a system that looks for patterns in data. By building an image classifier, your child will learn that computers do not inherently know what an object is, they must be taught using examples.

This hands-on activity fosters critical thinking, data literacy, and problem-solving. It shifts their relationship with technology from consumption to creation, showing them that they have the power to program and control these advanced systems.

Step 1: Access the Platform

To begin, open a web browser on your computer and navigate to Teachable Machine . Click on the "Get Started" button and select "Image Project" from the options. Choose the "Standard Image Model" which is optimized for standard webcams.

The interface is divided into three clear sections: Gathering Data (Classes), Training, and Preview. This simple layout allows children to easily follow the flow of data from input to output.

Step 2: Define Your Classes

Before recording, decide what you want your model to classify. Simple, high-contrast items work best. Excellent options include:

• A toy action figure versus a toy car.
• An apple versus a banana.
• A "Thumbs Up" hand gesture versus a "Thumbs Down" gesture.

Rename "Class 1" and "Class 2" on the screen to match your chosen items. For example, label them "Lego Brick" and "Action Figure".

Step 3: Gather Training Data

Click the "Webcam" button under your first class. Have your child hold the first object in front of the camera. Click and hold the "Record" button to capture images.

Instruct your child to slowly rotate the object, move it closer and further from the lens, and tilt it slightly. This ensures the model learns the object's shape from multiple angles, rather than just one static position. Aim to capture between 100 and 200 image samples. Repeat this exact process for the second class with the second object.

Step 4: Train and Test the Model

Once both classes have sufficient images, click the "Train Model" button in the middle column. Keep the browser tab open and active during this process. The computer is now analyzing the pixels in the images, looking for patterns that distinguish the first object from the second.

Once training is complete, the "Preview" panel on the right will activate. Have your child hold up one of the objects to the webcam. The model will display a real-time confidence percentage for each class, showing how certain it is of its prediction.

Parent & Educator Sidebar

Core AI Concept: Supervised Learning

Supervised learning is the process of training an AI model using labeled data. The computer is given examples (the images) along with the correct answers (the labels). It analyzes these examples to find common features, allowing it to make predictions on new, unseen data.

Conversation Starters:

• "How do you think a self-driving car tells the difference between a stop sign and a speed limit sign?"
• "What would happen if we only trained our model with red apples, and then showed it a green apple?"
• "How can bias enter an AI system if the creators do not use a diverse set of training images?"

Action Steps Summary

Want more practical frameworks to optimize your executive workflow?

Make the activity stick

Have the child write down the examples before opening the tool. Three good examples for each label beat twenty rushed examples. If the classifier makes a strange prediction, treat that as the best part of the activity. Ask which examples confused it and what new example would make the boundary clearer.

Use Google Teachable Machine as the starting point because it lets kids see labels, examples, training, and testing in one browser window. No paid account is required for the basic activity, and the child gets a concrete model instead of a vague explanation.

Parent or educator prompt: ask what the model learned, what it failed to understand, and how a biased set of examples could change the result. The core AI concept is training data quality. Kids should leave knowing that models do not magically understand the world, they copy patterns from the examples people give them.