Course › Lesson 4
🔒 Full course lesson ⏱ 9 min
Lesson 4: What AI is great at (and where it messes up)
Meet the confident friend who never says "I don't know."
🧸 The simple picture
Picture a friend who has read almost everything and is fantastic at brainstorming, writing, and explaining. But this friend has one funny habit: when they don't know something, they don't admit it. They just make up an answer that sounds right, and they say it with a big confident smile.
When AI does this, it's called a hallucination. It's not lying on purpose. It's just guessing words that sound likely, even when the facts aren't there.
🔑 Key ideas
- Great at: first drafts, rewording, summarizing, brainstorming, explaining ideas simply, and organizing messy notes.
- Weaker at: exact facts, recent news (unless it can search the web), math with many steps, quotes, and links. It can also misread what you meant.
- A hallucination is when AI states something false as if it were true, like a fake statistic, a book that doesn't exist, or a wrong date.
- AI can also repeat bias from the human writing it learned from, so keep an eye out for unfair or one-sided answers.
- Golden rule: the more it matters, the more you check. Health, money, legal, and anything you'll publish? Verify with a trusted source or a real expert.
🛠️ Try it (2 minutes)
Catch a hallucination on purpose. Ask a chatbot:
What are 3 books written by [your own name or a friend's name]?Unless that person is a famous author, the AI may invent book titles. Then ask:
Are you sure? Please only tell me things you're confident are real.Notice how it can sound sure even when it's wrong.
📝 Quick quiz
Tap an answer to check it.
1. What is an AI "hallucination"?
✅ It's a made-up answer that sounds real.
2. Which task is AI usually great at?
✅ Drafts and rewrites are a sweet spot.
3. The answer is about your health or money. What should you do?
✅ The more it matters, the more you check.