The single most valuable AI skill isn't coding — it's knowing how to direct AI precisely. This module teaches you to write prompts that get extraordinary results, and to spot when AI is misleading you.
Pay attention to the before/after examples — the gap between a weak prompt and a strong one is enormous. You'll replicate that gap yourself in Lesson 3.
Every great prompt has distinct parts working together. Click each highlighted section below to learn what it does and why it matters. Then try building your own.
These are real prompts people type into AI — and they're all weak. Your job is to rewrite each one so it would get a dramatically better response. There's no single right answer — just better and worse.
Why is this bad? The AI has no idea what subject, what level, what the essay is about, what kind of help you want, or how long it should be. It'll give you something generic and useless.
This will get you a Wikipedia-level answer that's too long, too technical, and impossible to revise from. How could you make this perfect for a Year 9 biology test?
You'll get something generic and forgettable. What if you want something set in a specific world, with a specific tone, that's the right length to actually read?
This gets you an opinion. But what if you want the AI to actually reason through it — weigh the evidence, consider your specific situation, and reach a conclusion? That requires chain-of-thought prompting.
AI can be genuinely dangerous when misunderstood. Here are the four biggest risks — and what you can do about each one.
AI language models don't "know" things the way you do. They predict what words come next based on patterns — which means they can produce completely made-up facts that sound absolutely convincing. This is called hallucination.
Real example: a lawyer in the US used ChatGPT to write a legal brief. It cited six court cases to support its argument. All six cases were completely fabricated. The lawyer submitted it to court without checking. He faced serious professional consequences.
AI can now generate realistic video and audio of real people saying things they never said. You can clone someone's voice from 3 seconds of audio. You can create video of someone's face saying anything. This technology is already being used to spread misinformation, create fake celebrity content, and scam people.
The red flags to look for: unusual blinking patterns, slightly off lip sync, unnatural skin texture around the edges of the face, background that doesn't quite match. But these are getting harder to spot every month.
This is the one most people don't think about. AI can analyse your social media, your interests, your fears, and your beliefs — then generate content specifically designed to push your buttons. Political ads, scam emails, fake news — all increasingly personalised to you specifically.
The reason this is so effective is that it doesn't look like advertising. It looks like content you'd naturally agree with. It confirms what you already believe, which makes it very hard to question.
This is the quietest risk — and possibly the most important one for someone your age. If you always ask AI to write your essays, solve your problems, and make your decisions, your own ability to do those things will get weaker. Just like a muscle you never use.
The goal of this course is the opposite of overdependence. We want you to be someone who uses AI as a tool that amplifies your own thinking — not replaces it. The people who will do best with AI are the ones who can think critically, evaluate outputs, and direct the AI rather than just accept what it says.
If something made by AI made you feel a strong emotion — shock, fear, anger, excitement — pause before acting on it. That emotion is the mechanism. The pause is the defence.
A prompt library is a personal collection of prompts that work — tested, refined, and ready to use. Professional AI users all have one. Build yours now across 6 categories. Use everything you've learned in this module.
Prompt Master badge unlocked. Module 3 — AI Tools in the Real World — is now unlocked.