Switch AI
Level 1 · Lesson 2

What can your AI do?

⏱ 7 min read ✦ 30 XP 🧠 3 questions
Reading
Exercise
Quiz
XP

More than a chatbot

Most people's first instinct with AI is to ask it a question — like a fancier Google. That's fine. But it's a bit like using a chef's knife to open an envelope.

Your AI is a general-purpose thinking tool. It can write, rewrite, explain, summarise, translate, brainstorm, debug, analyse, and argue both sides of an argument. The same tool. The same conversation.

Think of it this way

Imagine a colleague who has read almost everything, never gets tired, has no ego about feedback, and will cheerfully rewrite the same paragraph twelve different ways until you're happy. That's closer to what you've got.

What it's genuinely great at

These are the areas where AI consistently delivers — and where most people underuse it:

✍️
Writing & rewriting
Drafts, edits, tone shifts, summarising long text into short text
🧠
Explaining complex things
Break down any topic — technical, legal, scientific — into plain language
💡
Brainstorming
Generating options, angles, ideas — it never runs dry and doesn't judge
🔁
Following instructions
Give it a format, a tone, a structure — it'll follow it precisely and consistently
The real power: It's not any one of these things — it's that you can combine them. Summarise this, then rewrite it for a teenager, then turn it into bullet points. All in one conversation.

What it's not great at

Being honest about limitations isn't pessimism — it's how you avoid frustration and use the tool correctly.

📰 Real-time information. Your AI's knowledge has a cutoff date. It doesn't know what happened last week unless it has a browsing tool enabled.
🔢 Precise arithmetic. It can reason about maths, but it can make errors on complex calculations. Use a calculator for anything that matters.
🧠 Memory between sessions. Start a new chat and it knows nothing about you. Every conversation begins from scratch unless you tell it otherwise.
Guaranteed accuracy. It can sound confident and be wrong. On important topics, always verify with a trusted source.

Knowing the limits makes you better

Here's something most AI courses won't tell you: the people who get the most out of AI are the ones who understand what it can't do.

When you know the limits, you stop being surprised by mistakes. You double-check the right things. You use the tool for what it's actually good at — and reach for something else when it isn't.

The mindset shift: Don't ask "is AI good?" Ask "is AI good at this specific thing I need right now?" That question will serve you for years.
Try it yourself

Try 3 different things

Open your AI and try at least one of these prompts. Notice how it handles different types of tasks.

Writing
Try: "Write a short out-of-office email that's professional but has a bit of personality."
Explaining
Try: "Explain how Wi-Fi works as if I'm 10 years old."
Brainstorming
Try: "Give me 5 unusual ways to make a Monday morning feel less miserable."
Open your AI

Try at least one. Notice what it does well. Then come back.

Knowledge check

Three quick questions

1. Which of these is AI genuinely great at?
2. Why can't AI reliably tell you what happened in the news last week?
3. You had a long conversation with your AI yesterday. You start a new chat today. What does it remember?
+30
XP Earned
60 XP total ✦

Lesson complete! 🎉

You now know what your AI is genuinely good at — and where to be careful. Next up: what makes a bad prompt, and how to fix it.