← Mission map

Mission 3 of 16

Data, Patterns & Models

📋 What it is

A “model” is the pattern an AI has learned from data; better, more varied data makes a better model.

🗣️ Coach says

After learning, the pattern the AI keeps is called a MODEL. The rule is simple: garbage in, garbage out. If the training data is limited or biased, the model is too. Good, varied data → a model that works fairly for everyone.

🧠 Memory hook

A model = the learned pattern. Garbage data in → garbage model out.

😂 Giggle

Why did the password go to therapy?
It had too many issues with trust and validation!

😲 Whoa!

An early AI meant to spot skin problems worked poorly on darker skin — because its training photos were mostly of light skin. The data shaped the model.

✅ Quick check: An AI trained only on photos of sunny days is shown a snowy scene. What might happen?

Say your answer out loud first — then reveal.

🧪 Try it! (2 minutes)

Imagine training a “fruit AI” on only apples. Would it recognise a banana? (No — data limits the model.)

⭐ Do the round to earn your star →
🤸 Brain break: Data-in: mime “feeding” examples into a box and a “model” coming out.