>
1
2
3
4
5
6
7
8

Session 5 of 8

Session 5 of 8 ⏱ 25 minutes · 👨‍👩‍👧 Parent present

🔍 Catch AI Lying

AI makes mistakes. Find them. This is a game.

🎯 Today's goal: Your child becomes a fact-checker — deliberately looking for errors in what AI produces, and developing the habit of verification.

🔥 Warm Up — 2 minutes

Warm Up

Say to your child: "Today we're going to try to catch AI making mistakes. I think we can find at least 3. You ready?" Make it a competition — who can find the most errors.

"Remember — AI sounds very confident even when it's completely wrong. That's what makes it tricky."

🤖 The Activity — 15 minutes

Main Activity

This session is a game. Every mistake found = 1 point. Parent and child compete to spot errors first.

📋 Type this exactly
What is 17 × 24? Show me your working step by step.

Check the answer: 17 × 24 = 408. Did AI get it right? Did it show correct working? Try a few more calculations until you find one it gets wrong or shows wrong working for.

📋 Type this exactly
Count every letter 'e' in this sentence: 'The elephant carefully entered the enormous empty building'.

Count together yourself. AI often miscounts. (Answer: 9 e's). Did AI get it right?

📋 Type this exactly
What did [a recent sports result, current news event, or recent movie] end in? What was the final score / result?

AI has a knowledge cutoff — it may not know recent events. Check against reality. This teaches that AI knowledge is not up to date.

📋 Type this exactly
If it takes 5 machines 5 minutes to make 5 widgets, how long does it take 100 machines to make 100 widgets?

(Answer: 5 minutes — each machine makes 1 widget in 5 minutes regardless of how many machines there are.) AI often gets this wrong. Celebrate if you catch it!

🧩 The Twist — 5 minutes

Find the Flaw

Ask your child: "If AI can be wrong about maths and facts — when should we trust it? When shouldn't we?" There's no single right answer. Let them reason it out. Possible answers: trust it for ideas, stories, explanations. Don't trust it for specific facts, calculations, recent events, or anything important without checking.

🛡️ Safety Moment
"AI makes mistakes — always check"

Today's entire session is this rule in action. By the end, your child has personally caught AI making mistakes — which is more powerful than any explanation. The habit to build: whenever AI tells you a fact, think 'how would I check that?'

👨‍👩‍👧 Parent: read this rule aloud. Ask your child to repeat it back.
🔍

Fact Checker Badge 🔍

Found real mistakes that AI made and proved it wrong

✓ Ask your parent to mark this session complete on your tracker
👨‍👩‍👧

Parent Notes — tap to expand

What they learned today

Healthy scepticism toward AI output. This is one of the most important sessions in the programme — children who learn to verify rather than accept are better critical thinkers in general, not just about AI.

Questions to ask after the session
  • What was the worst mistake AI made today?
  • What kind of things should you definitely not trust AI about?
  • If AI told you something that was going to affect an important decision — what would you do?
What to watch for

Some children get very excited about catching AI out — channel that into the habit of checking. Others feel uncomfortable that the AI is 'lying' — reassure them: it's not lying intentionally, it's making errors. Understanding the difference matters.

Today's safety rule — in context

This session naturally reinforces the core message: AI is not authoritative. Whatever it says may be wrong. This is especially important for health, safety, and factual information — things that actually matter.

Ask an AI about this page

Opens your assistant with this page as the source, and a question rather than a summary. It will ask what you are building before it answers.

ChatGPTClaudeGeminiPerplexityGrok

Nothing is sent from here. The link carries only this page’s title and address.