AI Around Us: From Rules to Learning from Examples

Artificial intelligence, or AI, is the ability of a computer to do tasks that usually need human intelligence, such as recognising faces, understanding speech or giving suggestions. You probably use it every day without noticing.

Examples of AI in your life

• A phone camera that automatically finds your friends' faces when you take a group photo after the flag ceremony.
• YouTube suggesting the next video based on the videos you often watch.
• Google Maps estimating how long the trip to school will take based on traffic.
• A phone keyboard guessing the next word while you type in your class group chat.

Ordinary programs: following rules

Imagine you want a program that decides whether a student is late. You can write a clear rule:

IF arrival_time > 07.00 THEN write "late"
OTHERWISE write "on time"

This rule is easy to write because it is clear. The computer simply follows instructions made by a human.

Problems whose rules are hard to write

Now try writing rules to tell a photo of a cat from a photo of a dog. Is "has pointy ears" enough? No, some dogs have pointy ears. "Has whiskers"? Dogs have them too. The more you think, the more exceptions you find. This is where AI is useful.

AI: learning from examples

Instead of writing rules, we give the computer many examples: hundreds of cat photos labelled "cat" and dog photos labelled "dog". The computer then finds the patterns by itself. The branch of AI that learns from examples is called machine learning.

In simple steps:

1. Collect examples (data) and label them.
2. Train the computer to find patterns in those examples.
3. Test it with new examples it has never seen.
4. Use the result to make guesses or decisions.

Remember: AI does not "think" like a human. It finds patterns in data, and it is only as good as the data it is given.

In this course you will practise all these steps with Teachable Machine and Scratch, until you finally build your own image recognition model.