AI1: Lesson 6 Teaching AI to Recognise Colour and Shape

Can AI really tell the difference between a rabbit and a cat - just by looking at shapes and textures?

This KS3 / lower secondary computing lesson gives students a hands-on understanding of how AI vision systems are trained - and where they go wrong.

It begins with a feature-spotting challenge that reveals how AI “sees” shapes and colours, before introducing the key concepts of training data, testing data, and feature recognition.

Students then take on the role of a machine learning system: they classify animals using shape, size, and texture alone, reflect on classification errors, and write IF…THEN rules to improve their AI’s performance. A built-in simulator brings the task to life, reinforcing how bias and feature choice affect accuracy.

By the end of the lesson, students can explain how classifiers work, evaluate the limits of AI decision-making, and justify how better features lead to better outcomes.

Through this, students uncover:

• How classifiers use visible features like shape, size, and texture to sort objects
• Why training data and testing data must be carefully chosen to avoid AI bias
• How IF...THEN rules and feedback loops can improve AI decision-making

Files download instantly.